Page 1

HHPR Harvard Health Policy Review

Volume 15 Issue 2

CRISPR ‘Broad Societal Consensus” on human germline editing See Page 19


Regional Health

Setting Prescription Drug Prices: A Comparison of Strategies in the US, UK, Canada, Australia, and Germany

Neighborhood Disparities in Health: Why Do We See Them and What Can We Do?

See Page 4

See Page 27


Editor’s Note Pharmaceuticals Setting Prescription Drug Prices: A Comparison of Strategies in the US, UK, Canada, Australia, and Germany


Eunah Lee


Jing Luo, M.D., Aaron S. Kesselheim, M.D., J.D., M.P.H.

Differential Pharmaceutical Pricing for 10 F.M. Scherer, M.B.A., Ph.D. Low-Income Nations Pricing of Pharmaceuticals in the Supply 14 Thani Jambulingam, Ph.D. Chain CRISPR ‘Broad societal consensus’ on human 19 Francoise Baylis, Ph.D. germline editing The CRISPR Revolution: Technical and 24 Dana Carroll, Ph.D. Societal Aspects of Current Genome Editing Technologies Regional Health Neighborhood Disparities in Health: Why 27 Mariana Arcaya, MCP, ScD, and S.V. Subramanian, M.A., M.Phil., Ph.D. Do We See Them and What Can We Do IMPACT: the Role of Peer to Peer 31 Anjali Chandra education and the Life-course Perspective in Infant Mortality California Summit on Preterm Birth 33 Leslie Kowalewski; Michael Curtis, Ph.D.; Connie Mitchell, M.D., M.P.H.; Larry Rand, M.D.; Scott D. Berns, M.D., M.P.H., F.A.A.P.; Gary M. Shaw, DrPH; and David K. Stevenson, M.D.


Harvard Health Policy Review

editor’ s note

Editor-in-Chief Eunah Lee Managing Editor Melinda Song Senior Design Editor Hueyjong Shih Senior Business Manager Eric Li Senior Online Content Editor Patric Cao Senior Publishing Editor Kamran Jamil Web Design & Technology Jiahui Huang Associate Editors Anjali Chandra Neil Davey Henna Hundal Jesper Ke Michael Liu Samuel Oh Courtney Okwara Bovey Rao Scott Xiao Publishing Associates Bettina Edelstein Courtney Lewis Business Associates Marcus Gutierrez Anna Valuev Benjamin Zheng Web Design & Technology Associate Marcia Soviak The Harvard Health Policy Review is an undergraduate publication of Harvard College. The Harvard name and Veritas Shield are trademarks of the President and Fellows of Harvard University.

Harvard Health Policy Review

Harvard University Student Organization Center at Hilles Box #40 59 Shepard Street Cambridge, Massachusetts 02138

Historically and continuously important health policy issues have resurged, and new research technologies have emerged, as the country entered 2016. Popular media, such as the New York Times, have recently published widely about rising drug prices, generating ripples of public debate about the ethics of drug pricing. The discovery of CRISPR/Cas9 genome editing technology in early-mid 2015 and its subsequent rapid development amidst little ethical or policy framework has motivated scientific conferences about how to regulate the use of this groundbreaking technology. Especially given the controversy of Chinese scientists’ editing of human embryos published in April 2015, and clinical applications looming possibly in the near future, health policy discussion is of urgent necessity. Continuously, our diverse nation is plagued with rising inequalities in healthcare, one fascinating factor of the problem being regional inequalities across states and neighborhoods. While there is always a multitude of interesting health policy issues, we at HHPR chose to focus this Spring 2016 issue around these three topics: Pharmaceuticals, CRISPR, and Regional Health Inequalities. We begin our health policy discussion of pharmaceuticals with a piece that highlights the rising prices of prescription drugs for Americans, including drugs for infectious diseases. Notably, the author discusses the response to the increased price of an anti-parasitic drug – from $13.50 when it was first approved in 1963 to $750 today. The piece then explains how prices are set and pharmaceutical spending negotiated, with country-bycountry analysis. This article is followed by two more in this topic; one delves into how drugs should be priced in low-income nations while still providing profit in richer nations, and another delineates the pharmaceutical supply chain to inform cost-control policy recommendations. Our CRISPR section opens with an article that analyzes requirements for a broad societal consensus on human germline genome editing, converging on principles including responsibility, self-discipline, respect, and cooperation. The second piece expounds on the applications of CRISPR technology, the history of genome editing, and the major questions facing the field today. Finally, our Regional Health section begins with an article advocating for more affordable housing, fair housing laws, and equitable access to high-opportunity neighborhoods, to address regional health inequalities. The second article investigates the role of peer-to-peer education and the life course perspective in infant mortality, examining disparate health outcomes across racial lines. The author cites startling statistics, with African-American babies in California more likely to be pre-term, and a particular county in the state – Hamilton County – having an infant mortality rate of 8.84% behind countries like Cuba and Chile. This section closes with an informative article explaining the developments from a California summit on preterm birth, and the cooperation of public and private institutions to improve infant and maternal health. Our carefully curated Spring 2016 issue examines the developments in health care policy around these three important topics. This exciting project would not have been possible without the dedication and hard work of our contributors, our advisers, our managing editor and the rest of our diligent staff on our editorial review, design, technology, publishing and business boards. We hope you enjoy this latest issue of our journal. Sincerely, Eunah Lee Editor-in-Chief 2015-2016

about the cover


Cover illustrations modified from stock images via Flickr.

The Harvard Health Policy Review would like to thank the Harvard Undergraduate Council and The Institute of Politics at Harvard University for their generous support of this publication.

©2016 by President and Fellows of Harvard University. All rights reserved. No part of this publication may be reproduced in any form without express written consent from the publisher.

Spring 2016 Volume 15, Issue 2



Setting Prescription Drug Prices: A Comparison of Strategies in the US, UK, Canada, Australia, and Germany By: Jing Luo, M.D. and Aaron S. Kesselheim, M.D., J.D., M.P.H.

of AWP for physician-administered drugs paid for by Medicare Part B because data suggested that these prices were artificially inflated and failed to accurately reflect true acquisition costs.11,12 Still, these list prices represent the starting point for drug price negotiations between drug manufacturer and payor in the US. For example, many state Medicaid programs continue to use these list prices in formulas to calculate pharmacy reimbursements.13,14 Massachusetts’ Medicaid program pays pharmacies WAC plus 5% for ingredient costs, while Arizona’s program pays AWP minus 15%. It is important to note that a portion of these outgoing payments will return to state programs and CMS through mandatory rebates and any supplementary rebates that the state negotiates. These rebates are calculated as a percentage of another (unpublished) benchmark price called the Average Manufacturer Price (AMP). The AMP is the average price paid to the manufacturer by wholesalers for drugs available in retail pharmacies. Medicare uses yet another benchmark price, the Average Sales Price (ASP), to calculate payments to physicians for drugs covered by Part B. The ASP is defined as net sales to purchasers divided by total number of units sold.

In 2015, Americans experienced dramatic increases in the prices of prescription drugs, affecting many areas of medicine including infectious disease (hepatitis C), oncology (multiple myeloma), rheumatology (rheumatoid arthritis), endocrinology (diabetes), cardiology (hyperlipidemia), and neurology (multiple sclerosis).1-8 Since the Affordable Care Act was enacted in 2010, patients have had to shoulder increasingly higher portions of prescription drug costs through deductibles, coinsurance or co-payments.9 These trends appear set to continue. The Office of the Actuary in the Centers for Medicare & Medicaid Services (CMS) projects that percapita pharmaceutical spending will increase an average of 6.3% per year from 2015 to 2024.10 These price increases are not simply limited to new drugs. Pyrimethamine (Daraprim), an anti-parasitic drug first approved in 1963, increased in price from $13.50 to $750 per pill, while prices of the decades-old heart failure drug isoproterenol (Isuprel) and the anti-hypertensive drug nitroprusside (Nitropress) jumped 525% and 212%, respectively. In this review, we first provide context on these drug pricing trends and then place US drug prices in an international context, by detailing how prescription drug prices are handled in 4 high-income countries: Australia, Canada, Germany and the UK.

How are drug prices set in the US? In the US, list prices for newly approved drug products are set by the manufacturer, and subsequently negotiated by public and private payors (As we will discuss later, different US payors’ abilities to negotiate successfully are limited in numerous ways). Partly because of the fragmented nature of how Americans pay for healthcare services, the US market has a dizzying array of published and unpublished drug prices (Table 1). The list price, also known as the Average Wholesale Price (AWP), is a self-reported manufacturer price published in catalogs such as the Drug Topics Red Book. The Wholesale Acquisition Cost (WAC) is also a published price, but since 2003 has been defined by federal statute as the price charged to wholesalers, excluding discounts or rebates. Although the WAC is not independently verified or validated, it is often used in cost-effectiveness studies or in published articles to approximate the cost of a drug. Neither it nor the AWP is based on data from actual sales, invoices or transactions. Thus, payors have changed their practices to avoid directly relying on these list prices. Starting in 2003, the CMS phased out the use

Two major themes of this discussion are relevant. First, there is little to no accountability regarding how list prices are determined for drugs in the US. Second, there is a total lack of transparency regarding how much is actually spent on individual drugs. When critically important data are not available to the public, it increases the risk for inefficiency and frustrates attempts to fashion policy solutions.

What are some of the current approaches for controlling pharmaceutical spending in the US?

Table 1: Characteristics of selected benchmark pharmaceutical prices in the US *Although pricing data submitted to CMS used to calculate the ASP is confidential, CMS does publish reimbursable amounts for Part-B covered drugs on a quarterly basis.

The majority of Americans—about 175 million—receive prescription drug coverage through employer-sponsored health benefits, which are managed primarily through three large pharmaceutical benefits managers (PBMs): ExpressScripts, Caremark, and UnitedHealthcare. PBMs negotiate with drug manufacturers to secure discounts or rebates in exchange for lower patient copays, lower co-insurance amounts, or preferred formulary status. Unlike government payors, PBMs operating in the private market do not have to provide coverage for any specific class of drugs. Many PBMs do so anyway but increasingly discourage the use of high-cost drugs through cost-sharing practices such as high co-pays, deductibles or co-insurance amounts. Cost-sharing measures within the context of pharmacy benefit management reduce overall prescription drug costs because they incentivize the use of lower-priced, generic drugs, when they are available.15 Data on exact drug prices neSpring 2016 Volume 15, Issue 2


gotiated by PBMs are not available as they are protected by confidentiality agreements and laws surrounding trade secrets. Among public payors, Medicaid, the federal- and state-funded health insurance program for the poor, covers prescription drug costs for another 60 million Americans. Medicaid programs receive a set rebate on drug prices, but since the rebate is based off of an already high price, these rebates are often insufficient to control Medicaid prescription drug budgets. Medicaid programs are able to negotiate separate supplemental rebates, but their negotiating power is limited since they are generally required by federal law to provide access to FDA-approved drugs. Since federal law prohibits Medicaid from charging more than a minimal co-payment for prescription drugs, state Medicaid programs are unable to use cost-sharing or tiered co-pays to save money. Instead, in the face of rising drug expenditures, they tend to use prior authorization (in which a patients must first obtain written or telephone approval from a payor or PBM before they can receive a drug) or step-therapy requirements (where patients must have previously failed more cost-effective therapies). Recently, many state Medicaid programs have limited access to direct acting antiviral medications used to treat hepatitis C. For example, some states have limited access to sofosbuvir (Sovaldi) to those presumed most likely to benefit from cure, such as patients with advanced fibrosis or cirrhosis, even if there is no medical basis for such limitations.16 The next largest public plan, Medicare, covers about 40 million adults, most aged 65 and older, for outpatient (Part D) and inpatient (Part B) drug costs. Among the large public payors, Medicare often pays the highest drug prices.17 Medicare Part D benefits are administered through thousands of individual commercial plans, each with its own formulary. Beneficiaries enrolled in these commercial prescription drug plans (who do not qualify for low income subsidies) are responsible for premiums, deductibles, copays/co-insurance, and until 2011, the entirety of drug costs in the so-called “donut hole.� The Part D donut hole is the coverage gap between standard coverage and a spending limit in which beneficiaries are responsible for the full cost of prescription drugs. Since 2011, the Medicare Part D Coverage Gap Discount Program has required manufacturers to pay 50%-55% of plan-negotiated drug prices, inclusive of sales tax, for


Harvard Health Policy Review

Table 2: Comparing features of pharmaceutical pricing and coverage in Australia, Canada, Germany and the UK beneficiaries in the donut hole. Plan sponsors may negotiate with manufacturers individually or through PBMs; however, as with Medicaid, negotiating power is limited by external factors—in this case, CMS regulations governing coverage. For example, plans must cover at least 2 drugs from 148 classes and 1 formulation of every FDA approved drug in 6 protected classes. Additionally, although Medicare spends about $70 billion annually to subsidize commercial plans, the CMS has been prohibited from negotiating for lower overall prices on behalf of individual sponsors.18 Smaller public payors include the Veterans Health Administration, individual state prison systems, and the federal employee health benefits program. Because it has a centralized pharmacy benefits manager and a single national formulary, the Veterans Affairs Administration receives some of the lowest drug prices in the country.19,20 Unlike payors in the commercial market, the VA can access substantially discounted drug prices. For example, a 1-month supply of sofosbuvir from January to April 2016 was $16,620, or approximately 41% less than the list price (a full treatment course requires 3-6 months). For drugs anticipated to be a large portion of drug spending, the VA negotiates separate national contracts to obtain additional price concessions in exchange for purchasing commitments or preferred status.

How does the US compare against other high-income countries with respect to pharmaceutical pricing? Although the US federal government sets reimbursement amounts for physician services, hospitalizations and other healthcare services, it is not involved in setting pharmaceutical prices. This is not the case in other high-income countries. Australia, Canada, Germany, and the UK all use some combination of cost-effectiveness and comparative effectiveness when setting reimbursement amounts for pharmaceuticals. With the exception of the UK, these countries also reference the price of existing products when determining the prices of a new products.21

Australia Australia has provided universal prescription drug coverage to residents holding a Medicare card since the creation of the Pharmaceutical Benefits Scheme (PBS) in 1953. Since 1993, the PBS has considered value for money as an important component of the decision making process when evaluating applications for new prescription drug coverage. Manufacturers seeking subsidization for a new drug product by the PBS must receive a positive listing recommendation by the Pharmaceutical Benefits Advisory Committee (PBAC), which meets

three times a year. After the Australian drug regulatory agency has determined that a drug is safe and effective for use, the PBAC will make a listing recommendation based on effectiveness, safety, cost-effectiveness, projected usage, and overall costs to the healthcare system. A new drug is compared against one main comparator using data from either randomized trials or observational studies. Cost-utility studies, cost-effectiveness studies or cost-minimization (resembling reference pricing) are required depending on whether the new product is determined to be superior or non-inferior to the comparator. Overall budgetary impact is also considered, which involves projected drug utilization and net costs based on the epidemiology of the disease being treated and the indication. After evaluating the evidence, PBAC makes an overall recommendation to list, not list, or restrict usage of the new drug. Drug applications may be rejected for a variety of reasons. For example, the use of bevacizumab (Avastin) in advanced cervical cancer recently received a negative recommendation because of increased adverse events and an unfavorable incremental cost-effectiveness ratio of AUD$ 75,000 - AUD$ 100,000 per quality-adjusted life-year (QALY) gained. A QALY is a year of life adjusted by society’s preference (or utility) for a particular health state relative to full health; it has become a widely used measure of cost-effectiveness. A unique feature of the PBS is that manufacturers may resubmit their applications an unlimited number of times after modifying the proposed price or indication. In some cases, the PBAC has recommended a specific price per QALY that it considers reasonable to the Minister of Health for the purposes of negotiations. For example, in its March 2015 decision on the direct-acting antiviral combination Hepatitis-C treatment ledipasvir/sofosbuvir (Harvoni), the PBAC suggested that the drug would be cost effective at ~AUD$ 15,000 per QALY. For many high cost drugs, the PBS can secure additional concessions through financial risk sharing agreements. For example, the manufacturer may be responsible for the costs of treatment above a certain cap in cases of use beyond the recommended indication or for longer than expected treatment durations. One potential disadvantage of this highly centralized and aggressive system for evaluating and pricing new drugs is that access to new treatments may be delayed when com-

pared against market entry dates in the U.S. or Europe. Another contributor to any delay is the fact that drug companies tend to submit their products for regulatory approval first in other, larger markets. Still, because of its single payor prescription drug scheme and a legislative requirement to use value for money (especially when compared against existing treatments) as a criterion when considering new drug applications, Australia has some of the lowest prescription drug prices in the world.22,23

Canada Canada’s system for controlling drug prices is far less centralized than Australia’s. Apart from the one million people (First Nations and Inuit, military or veterans and inmates) who receive prescription drug coverage through the federal government, the majority of Canadians receive fragmented coverage through provincial public drug plans (42% of total spending on prescription drugs in 2014) or privately through employer sponsored drug plans (36% of total spending). Each of these plans makes formulary decisions and conducts price negotiations separately. Canada has two national systems aimed at reducing excessive prescription drug prices: the Common Drug Review (CDR) process and the Patented Medicines Prices Review Board (PMPRB). The PMPRB was created in 1987 to ensure that the prices of patented medicines (about 60% of drug sales) sold in Canada are not “excessive.” A committee first evaluates the degree of therapeutic innovation and then conducts a price review. Prices sold to wholesalers, hospitals and pharmacies are determined to be excessive if they exceed the median international price in 7 countries (France, Germany, Italy, Sweden, Switzerland, the UK and the US Federal Supply Schedule), or if they exceed the prices of comparator drugs in the same therapeutic class already being sold in Canada. The PMPRB can order investigations into drug prices considered excessive and order either price reductions or payments to the federal government for excessive revenue. However, in practice, these measures are uncommon. For example, only 61 investigations were conducted out of over 1300 patented drugs in 2014. An alternate process aimed at reducing the administrative burden of reviewing new drugs and standardizing access to prescrip-

tion drugs among the 18 publicly funded drug plans is called the Common Drug Review. The CDR started in 2003 and covers all new drugs except anticancer drugs or those sold in Quebec. Like Australia’s PBAC, the CDR’s expert committee makes reimbursement recommendations to public drug plans based on comparative effectiveness and cost-effectiveness studies. However, unlike in Australia, the CDR’s recommendations are not binding. Individual plans may make their own reimbursement decisions and negotiate their own pricing agreements. One study found that participating drug plans agreed with the CDR’s recommendation 61% to 96% of the time.24 In some ways, Canada’s system of relying on individual plan sponsors or provincial drug plans to determine drug prices is similar to the decentralized system seen in US, which relies heavily on state Medicaid programs and commercial Medicare Part D sponsors. One important difference is that increases in brand-name drug prices are limited by the Consumer Price Index in Canada. This ceiling on price inflation and the existence of the CDR and PMPRB may explain why retail prices for brand-name medications are about 24% lower in Canada than they are in the US.25 However, Canadians still pay more for patented medicines than most of their European counterparts.26 Some researchers suggest that if Canada adopted a universal system of prescription drug coverage using the purchasing power of a single national formulary, it could save approximately $7.3 billion per year.27

Germany To contain rising prescription drug spending, Germany overhauled its system of controlling prescription drug prices in 2011 and created the Arzneimittelmarkt-Neuordnungsgesetz (AMNOG) procedure. Under AMNOG, new prescription drug products may enter the German market at any price set by the manufacturer. This initial price is valid for 12 months. During this time, an early benefit assessment is carried out by the Federal Joint Committee (G-BA). This committee (or Germany’s health technology assessment agency, the Institute for Quality and Efficiency in Healthcare [ICWiG]) determines the level of benefit of a drug compared with existing therapies. In this early benefit assessment, only meaningful clinical outcomes are considered. For example, the use of progression free survival, which is a Spring 2016 Volume 15, Issue 2


surrogate outcome frequently used in cancer treatment trials, is not sufficient to demonstrate added benefit within this early benefit assessment. This practice of relying only on overall survival or quality of life outcomes for new cancer therapeutics is a key feature of the German system. Drugs found to have no additional benefit compared to existing therapies have their prices adjusted to a fixed-rate based upon the most economic comparator within a therapeutic class. For example, if a new diabetes drug is shown to not offer improvement compared against an existing therapy, its initial market entry price may be adjusted to a fixed rate according to the lowest-priced drug within a basket of antidiabetic treatments. Companies may charge more than this fixed rate; however, this rarely happens because patients are directly responsible for any additional charges above the fixed rate. If a new drug is found to offer additional benefit over existing therapy, the sponsor will enter into negotiations for additional reimbursement amounts with the national association of statutory health insurance funds (approximately 70 million covered lives). If no agreement is reached within 6 months, the proceedings go to an arbitration panel, which may examine the sale prices across 15 European countries to ensure comparable prices for Germany. The AMNOG procedure is reported to have saved over 180 million euros in 2012 and 2013.28 It is still too early to assess the effect of the recently implemented procedures on overall comparative drug prices in Germany. As more benefit assessments are conducted, the government anticipates that the streamlined procedures will save over 2 billion euros per year.28

United Kingdom The UK’s system of pharmaceutical pricing determination is highly centralized and very effective at controlling drug prices. Like Australia, one single agency – the National Institute for Health and Clinical Excellence (NICE) – is responsible for drug reimbursement decisions for the entire National Health Service (NHS). If NICE recommends a new drug, the constitution of the NHS states that it must be made available within 3 months of publication. As in Germany and Australia, an independent academic center in the UK evalu-


Harvard Health Policy Review

ates a manufacturer’s evidence submission and completes an evidence report which considers both the clinical and cost-effectiveness of a new drug product. An advisory committee then produces a final recommendation document. In general, NICE is unlikely to recommend any new treatment with a cost-effectiveness ratio above £20,000 – £30,000. However, some high-cost cancer drugs that are not recommended by NICE for cost-effectiveness reasons may still be accessed through a federally-funded Cancer Drugs Fund. Like Australia, the UK also engages in risk sharing agreements with manufacturers. For example, in exchange for a positive recommendation for the use of lenalidomide (Revlimid) in multiple myeloma, the manufacturer agreed to pay for the drug costs for patients who remain on treatment beyond 2 years.

What can the US learn from international pharmaceutical price control mechanisms? While no country has a perfect system which guarantees timely access to available therapies while minimizing budgetary impact to payors and financial risk to patients, the US system is designed in a way that favors high prescription drug prices, especially with regards to brand-name drugs. It also is the least value-based when compared to the pricing systems of other high-income countries. Each of the 4 markets described above have significantly more effective and centralized systems for controlling pharmaceutical prices than the US. All use expert committees to evaluate the degree of therapeutic or clinical innovation for each newly approved prescription drug compared against existing therapy. These determinations are then used to help inform coverage decisions and/or price negotiations by the dominant national payor (as in the cases of the UK and Australia) or by collective bodies representing important regional or local payors (as in the cases of Canada and Germany). Under centrally organized pharmaceutical pricing systems outside the US, new drug products that represent substantial improvements over existing therapies usually receive positive recommendations, often at premium reimbursement rates. Less innovative or follow-on therapies will either not be recommended, or only recommended in a subset

of patients with substantial price controls in place (e.g. internal reference pricing, international reference pricing and/or risk-sharing agreements). By contrast, US patients often end up paying high prices for marginally innovative new products, such as the proton-pump inhibitor esomeprazole (Nexium) or the cholesterol-lowering drug rosuvastatin (Crestor), both of which offer little improvements over other drugs currently available as generics. Many of the features of international pharmaceutical markets that contribute to their lower prices are not easily transportable to the US. Political constraints make it highly unlikely that we will move to a single prescription drug payor (such as Medicare or VA for all), which would allow Americans to tap into the savings achieved in the UK or Australia. It is also unlikely that we will soon have a single body which will be responsible for both evaluating the comparative clinical and cost-effectiveness of a new prescription drug and making reimbursement decisions on behalf of important payors. An attempt was made in 2010 to create a non-governmental body based in Washington D.C. to fund research, comparing different drug treatments as part of the Affordable Care Act. However, in part due to a unfounded fear that access to high-cost medical treatments would be rationed as a result of the studies it commissioned, the Patient Centered Outcomes Research Institute (PCORI) was statutorily prohibited from funding cost-effectiveness research. Furthermore, while PCORI does fund clinical comparative effectiveness research, it does not make any recommendations with respects to reimbursement decisions by insurance companies. These significant limitations mean that PCORI, in its current form, is unlikely to have any meaningful effect on prescription drug prices. In recent years, a number of non-governmental expert bodies skilled in comparative effectiveness studies have begun to operate in the US, including Institute for Clinical and Economic Review, the Independent Drug Information Service, Oregon’s Drug Effectiveness Review Project, and Consumer Reports Best Buy Drugs. These groups generally work similarly to Germany’s AMNOG procedures, assigning products to their level of additional benefit relative to existing standard-of-care therapies. Private payors in the US—and government payors, provided there is sufficient public outcry about

prescription drug prices to provide the rule changes necessary—could use these evaluations to help decide on appropriate pricing levels. Prescription medications can have a major positive impact on the health of individuals as well as populations, but unnecessarily high prices can limit patients’ ability to benefit from them.

References 1. Schulman KA, Balu S, Reed SD. Specialty pharmaceuticals for hyperlipidemia—impact on insurance premiums. New England Journal of Medicine. 2015;373(17):1591-1593. 2. Shrank WH, Barlow JF, Brennan TA. New therapies in the treatment of high cholesterol: An argument to return to goal-based lipid guidelines. Jama. 2015. 3. Tischner JR, Hartung DM, Bourdette DN, Whitham RH, Rittenhouse BE, Ahmed S. The cost of multiple sclerosis drugs in the US and the pharmaceutical industry: Too big to fail? Neurology. 2015;85(19):1727-1728. 4. Luo J, Avorn J, Kesselheim AS. Trends in Medicaid reimbursements for insulin from 1991 through 2014. JAMA internal medicine. 2015;175(10):1681-1686. 5. Kantarjian H, Rajkumar SV. Why are cancer drugs so expensive in the United States, and what are the solutions? Paper presented at: Mayo Clinic Proceedings2015. 6. Polinski JM, Mohr PE, Johnson L. Impact of Medicare Part D on access to and cost sharing for specialty biologic medications for beneficiaries with rheumatoid arthritis. Arthritis Care & Research. 2009;61(6):745-754. 7. Mailankody S, Prasad V. Five years of cancer drug approvals: Innovation, efficacy, and costs. JAMA oncology. 2015;1(4):539-540. 8. Keehan SP, Cuckler GA, Sisko AM, et al. National health expenditure projections, 2014–24: spending growth faster than recent trends. Health Affairs. 2015;34(8):1407-1417. 9. Martin AB, Hartman M, Benson J, Catlin A, Team NHEA. National Health Spending In 2014: Faster Growth Driven By Coverage Expansion And Prescription Drug Spending. Health Affairs. 2016;35(1):150-160. 10. Centers for Medicare & Medicaid Services OotA. National Health Expenditure Projections 2014-2024. 2015. 11. Mullen P. The Arrival of Average Sale Price. Biotechnology healthcare. 2007;4(3):48. 12. Gatesman ML, Smith TJ. The shortage of essential chemotherapy drugs in the United States. New England Journal of Medicine. 2011;365(18):1653-1655. 13. Services CfMM. Medicaid Covered Outpatient Prescription Drug Reimbursement Information by State: Quarter Ending December 2015. 2016. 14. Alpert A, Duggan M, Hellerstein JK. Perverse reverse price competition: Average wholesale prices and Medicaid pharmaceutical spending. Journal of Public Economics. 2013;108:44-62. 15. Shrank WH, Stedman M, Ettner SL, et al.

Patient, physician, pharmacy, and pharmacy benefit design factors related to generic medication use. Journal of general internal medicine. 2007;22(9):1298-1304. 16. Barua S, Greenwald R, Grebely J, Dore GJ, Swan T, Taylor LE. Restrictions for Medicaid reimbursement of sofosbuvir for the treatment of hepatitis C virus infection in the United States. Annals of internal medicine. 2015;163(3):215-223. 17. General OoI. Medicaid Rebates for BrandName Drugs Exceeded Part D Rebates by a Substantial Margin. 2015. 18. Grootendorst P. How Effective Is the Medicare Part D Drug Plan? Annals of internal medicine. 2015;162(12):869-870. 19. Good CB, Valentino M. Access to affordable medications: the Department of Veterans Affairs pharmacy plan as a national model. American journal of public health. 2007;97(12):21292131. 20. Gellad WF, Donohue JM, Zhao X, et al. Brand-name prescription drug use among Veterans Affairs and Medicare Part D patients with diabetes: a national cohort comparison. Annals of internal medicine. 2013;159(2):105-114. 21. Ruggeri K, Nolte E. Pharmaceutical pricing: The use of external reference pricing [Internet]. Cambridge: RAND Corporation; 2013 [citado 2 nov 2014]. 22. Vogler S, Vitry A. Cancer drugs in 16 European countries, Australia, and New Zealand: a cross-country price comparison study. The Lancet Oncology. 2016;17(1):39-47. 23. Kanavos P, Ferrario A, Vandoros S, Anderson GF. Higher US branded drug prices and spending compared to other countries may stem partly from quick uptake of new drugs. Health Affairs. 2013;32(4):753-761. 24. Gamble J-M, Weir DL, Johnson JA, Eurich DT. Analysis of drug coverage before and after the implementation of Canada’s Common Drug Review. Canadian Medical Association Journal. 2011:cmaj. 110670. 25. Quon BS, Firszt R, Eisenberg MJ. A Comparison of Brand-Name Drug Prices between Canadian-Based Internet Pharmacies and Major U.S. Drug Chain Pharmacies. Annals of Internal Medicine. 2005;143(6):397-403. 26. Board PMPR. Annual Report 2014. Ottawa, Ontario2015. 27. Morgan SG, Law M, Daw JR, Abraham L, Martin D. Estimated cost of universal public coverage of prescription drugs in Canada. Canadian Medical Association Journal. 2015;187(7):491-497. 28. Spitzenverband G. AMNOG - evaluation of a new pharmaceutical. 2015; https:// Accessed February 22, 2016.

Jing Luo, M.D. is a research fellow in the Program On Regulation, Therapeutics, And Law (PORTAL) at the Division of Pharmacoepidemiology and Pharmacoeconomics in Brigham and Women’s Hospital. His research interests are in prescription drug prices, access to medicine, generic competition and non-communicable diseases. He completed a residency in primary care internal medicine at Yale-New Haven Hospital. He practices clinically at the adult internal medicine clinic at the Brigham and Women’s Hospital and at Harvard University Health Services. Aaron S. Kesselheim, M.D., J.D., M.P.H., is an Associate Professor of Medicine at Harvard Medical School and a faculty member in the Division of Pharmacoepidemiology and Pharmacoeconomics in the Department of Medicine at Brigham and Women’s Hospital. He graduated from Harvard College in 1996, where he was an editor of the Harvard Lampoon and a co-author of the 149th Hasty Pudding Theatricals show. Dr. Kesselheim leads the Program On Regulation, Therapeutics, And Law (PORTAL), an interdisciplinary research core focusing on intersections among prescription drugs and medical devices, patient health outcomes, and regulatory practices and the law. Dr. Kesselheim’s work on drug prices is funded by the Laura and John Arnold Foundation. Dr. Kesselheim is also a Greenwall Faculty Scholar in Bioethics and is supported by the Harvard Program in Therapeutic Science. He is a practicing primary care physician at Brigham and Women’s Hospital and a Visiting Associate Professor of Law at Yale Law School.

Spring 2016 Volume 15, Issue 2



Differential Pharmaceutical Pricing for Low-Income Nations By: F. M. Scherer, M.B.A., Ph.D. Differential Pharmaceutical Pric- usually high pricing (mitigated by discounts demand for the illustrated medicine is much to some insurers and sales at “compassion- lower at nearly all prices in Nation B than in ing for Low-Income Nations Most newly-developed pharmaceutical products are patented, which strengthens the drug producer’s ability to set a monopolistic sales price. In rich nations like the United States without government price controls, this often means that the drug can be sold at a price far above production and distribution costs. But how should that same drug be priced in low-income nations, where a high price could mean that only the most affluent citizens can afford it? The solution that has gained prominence in recent decades is so-called “differential pricing” (also called “equity” or “tiered” pricing), entailing high prices in rich nations and low prices in poor nations. A dramatic example attracted public attention in 2015.1 In December 2013 U.S. biopharmaceutical specialist Gilead Sciences launched a revolutionary new drug, sofosbuvir (brand name Sovaldi), at a price per daily pill of $1,000, implying a total price of $84,000 for the 12-week course of treatment believed to reduce mortality rates for the largest class of Hepatitis C sufferers by more than 90 percent.2 Other combination variants followed for more resistant genotypes. There were complaints in the United States, inter alia through a U.S. Senate Finance Committee investigation concluded in December 2015, about what was perceived to be un-


Harvard Health Policy Review

ate” prices). But pre-existing therapies were much less effective and more prone to adverse side effects. Thus, Sovaldi remained from a broader perspective an attractive option. Gilead recognized nevertheless that at those prices, very few residents of low-income nations, where the disease was more prevalent, could afford the treatment. India, for example, was home to an estimated 11 million Hepatitis C sufferers, Egypt to 9 million. In 2014 Gilead addressed this problem, first through direct sales in some high-incidence, low-income nations at $10 per pill -- a 99 percent discount -- and then by licensing generic drug producers in India to distribute Sovaldi in 91 (later, 101) low-income nations at sharply reduced prices, on which a royalty of only 7 percent accrued to Gilead.

Differential Pricing: The Theory The recent Sovaldi experience provides an extreme case of what economics textbooks call price discrimination (differential pricing and other less pejorative names in public discourse). Figures 1a and 1b illustrate (with considerable simplification) the underlying economic theory, which dates back to the 19th Century.3 Figure 1(a) portrays demand and supply conditions in rich Nation A, assumed to be a monopolist’s home market; Figure 1(b) the same in low-income Nation B. Because of the adverse “income effect,”

Nation A. The marginal cost of production is assumed to be constant and identical at $18 per Rx in both nations.4 A profit-maximizing monopolist will choose to price-discriminate, charging the high price $59 per Rx in the rich nation and a much lower price ($26.50) in the low-income nation. In this way the monopolist serves both markets, rather than choking off all demand in the low-income nation with the $59 price. And through this differential pricing, the monopoly increases its total profits (before deduction of fixed and sunk costs) from $16.8 million per month (($59-18) x 410,000) selling only in the rich market, adding (($26.50-18) x 243,000) = $2.07 million) by selling in both markets. Among other things, this strategy yields more net revenue to repay sunk research and development investments (or in the case of Sovaldi, acquisition costs) than the non-discriminatory single-market alternative. As always, things can go wrong. The conventional obstacle to successful discriminatory or differential pricing is what is called “arbitrage” in the general case and “parallel trade” in the drug pricing example. If independent agents buy up large quantities of the illustrated drug in poor nation B at the $26.50 price and export (or re-export) it to rich nation A, the demand remaining for the original drug producer in the rich nation is squeezed to the left, the original produc-

er will sell less output, quite possibly at depressed prices, and profits in the rich nation will fall. Recognizing the threat of parallel trade, drug companies may choose not to offer their products in low-income nations at discriminatorily low prices. Differential pricing may also be discouraged by other conditions. For one, if the low-income nation contains two separable classes, e.g., rich consumers, inter alia with generous health insurance, and poor consumers without insurance, drug companies may choose to serve only the rich consumers at high (non-discriminatory) prices and leave the low-income consumers without viable options. Second, many high- and medium-income nations practice what is called “external reference pricing,” constraining the prices of drugs in their home markets to the average of prices charged in a handful of

markets: “You charged only $18 in Nation B, so give us an equivalent discount.”

TRIPS and Its Offspring These conflicting forces came to a focus during the 1990s for two main reasons. First, many less-developed nations chose not to grant any patents, or selectively to refrain (as did India and Brazil) from extending patent rights on pharmaceutical product inventions. The ratification of the TRIPS (Trade-Related Intellectual Property Rights) agreement, as part of the creation of the World Trade Organization in 1995, required the time-phased extension of drug product patent rights to all member nations -- first in rich nations, then in medium-income nations, and then (by 2006) in the least-developed nations.5 Second, an epidemic of HIV-AIDS was escalat-

prices. Thus, the low-income nations looked to the United States and Europe for relief, but they found it difficult to obtain the needed supplies at prices they could afford. A statistical analysis of the prices charged at wholesale by multinational pharmaceutical companies for 15 AIDS anti-retroviral drugs in low- and medium-income countries for the years 1995-1999 revealed very little tendency toward income-correlated differential pricing.6 The correlation between prices charged and national GDP per capita explained only 1.6 percent of the observed cross-nation price variation. Average prices did fall on average relative to those charged in the United States toward the end of the five-year period, suggesting, given that other wealthy nations were excluded from the sample, that the multinational companies were moving toward a less encompassing form of

“...if the low-income nation contains two separable classes, e.g., rich consumers, inter alia with generous health insurance, and poor consumers without insurance, drug companies may choose to serve only the rich consumers at high (non-discriminatory) prices and leave the low-income consumers without viable options.” reference nations. If the reference nations include those in which the drug producer would otherwise charge low prices, the producer will be inhibited from offering those low prices. And drug producers may be fearful of less formal pressure in their high-price

ing, and the novel drugs created to combat it were patented. Many of the low-income nations suffering most extensively from HIVAIDS lacked the technological capabilities to produce those drugs, even when there were no patent barriers to their sale at competitive

Figure 1: Price Discrimination between Markets of Differing Wealth

differential pricing. As time passed, however, the situation began to change -- eventually, radically. As noted earlier, pharmaceutical product patents could not be obtained at first in India and (with a change in the late 1990s) Brazil. A systematic effort was organized by representatives of U.S. generic producers, Indian pharmaceutical firms, Medecins sans Frontieres, and a unit of Ralph Nader’s consumer-oriented organization to have anti-retrovirals produced in India and then exported at highly competitive prices to needy low-income nations, notably, in Africa. Brazil followed suit in South America. Also, in South Africa, which did allow drug product patents, a private antitrust action induced the producers with patents on the leading triple therapy drugs to license their patents to generic suppliers at low royalties. Analogous compulsory licensing actions followed in Thailand, among other nations. In the early years of the 21st century, multinational organizations such as UNAIDS, the Global Fund to Fight AIDS, Tuberculosis, and Malaria; Medecins Sans Frontieres, and Oxfam as well as PEPFAR (the U.S. President’s Emergency Plan for AIDS Relief) used their bargaining Spring 2016 Volume 15, Issue 2


power and moral suasion to procure drugs at low prices from the leading pharmaceutical companies and organize their distribution in needy nations. Providing an analytic foundation for these changes, the World Health Organization and the World Trade Organization organized in April 2001 a conference in Høsbjør, Norway, bringing together the leaders of national, multinational, and philanthropic health agencies with pharmaceutical company executives, economists, lawyers, other scholars, and AIDS problem activists. The delegates explored the law and economics of differential drug pricing and concluded inter alia that:7 1) “Differential pricing could, and should, play an important role in ensuring access to existing essential drugs at affordable prices, especially in poor countries, while allowing the patent system to continue to play its role in providing incentives for research and development into new drugs.” 2) “Markets for differentially priced drugs need to be tightly segmented to prevent leakage of differentially priced drugs to high-income markets.” 3) “Developing-country prices should not be used, either directly or indirectly, as references in developed countries’ reference-price systems.” 4) “... although affordable prices are important, actually getting drugs, whether patented or generic, to the people in poor countries who need them will require a major financing effort ... [mostly] from the international community.” Already-existing tendencies to implement differential pricing were strengthened. To add incentive and implement recommendation 2, the conference participants recommended stronger controls by high-income countries’ customs authorities to prevent the importation of products marketed elsewhere at differential prices. And indeed, in 2002, European customs officials carried out a well-publicized confiscation of drugs re-imported from low-income countries into European markets.8 With the World Trade Association as co-sponsor of the Højsbør conference, the conference’s recommendations fed into the Doha Round of international trade nego-


Harvard Health Policy Review

tiations, which opened in November 2001. The Doha Round was noteworthy for its lack of progress, but drug pricing was an exception. The original TRIPS treaty allowed nations to declare compulsory patent licenses in cases of national emergency, when goodfaith license negotiations had broken down, and to remedy anti-competitive practices. In what came to be known as “the Doha declaration,” the WTO authorities stated that:9 Each member has the right to determine what constitutes a national emergency or other circumstances of extreme urgency, it being understood that public health crises, including those relating to HIV/AIDS, tuberculosis, malaria and other epidemics, can represent a national emergency or other circumstances of extreme urgency. A problem brought to light at the Højsbør meeting was that even though less-developed nations might issue compulsory licensing decrees, their domestic producers often

As a consequence of these initiatives from the WTO and through AIDS activists and multinational organizations, the supply of HIV/AIDS medications in low-income nations increased greatly, and prices fell sharply. The cost of providing three-drug AIDS therapy in the poorest nations dropped from approximately $15,000 for a year’s treatment in 2001 to an average of $127 in 2012.11 The price reductions in turn led to an increase in the number of low- and middle-income nations citizens being treated with anti-retrovirals from the low hundreds of thousands in 2001 to an estimated 10 million in 2012. After AIDS mortality levels peaked in 2005, the number of individuals dying from AIDS then fell considerably.12 The patent, supply, and price changes that occurred in the case of HIV/AIDS represented a triumph of collective action, driven mainly by governmental and public interest groups. The more recent history of Sovaldi and other Hepatitis C drugs13 seems to

“Differential pricing could, and should, play an important role in ensuring access to existing essential drugs at affordable prices, especially in poor countries, while allowing the patent system to continue to play its role in providing incentives for research and development into new drugs” lacked the technological capability actually to produce the needed drugs. At Doha in 2001, it was recognized that: We recognize that WTO members with insufficient or no manufacturing capacities in the pharmaceutical sector could face difficulties in making effective use of compulsory licensing under the TRIPS agreement. A formal solution was postponed until August 2003, when the WTO officially declared that border-hopping compulsory licenses -- i.e., covering firms in supplying nations as well as beneficiaries in recipient nations -- could be issued “in good faith to protect public health.”10

The Spread of Differential Pricing

imply an even sharper break in the policies of multinational pharmaceutical companies. How much the new and more enlightened policies were influenced by the HIV/AIDS history, the activism of multinational entities such as the World Trade Organization and the World Health Organization, and formal amendments in international intellectual property law, as distinguished from more idiosyncratic changes in the relevant suppliers’ public-spiritedness, might be illuminated by further research. Gilead’s annual report for the year 2014, published in early 2015, provides only limited insight into the company’s motivations. Its chief executive officer stated simply in his letter to shareholders that the company’s “ultimate goal is to offer [hepatitis] patients a cure” and articulated a “continued focus on enabling worldwide access to our life-saving medications.” Its 10-K

report for 2015 (signed February 24, 2016) adds mainly (p. 3) that its tiered pricing programs “enabled access to our medicines for people who need them around the world.” That Gilead did learn something from the early HIV/AIDS experience is suggested by its unique approach to preventing parallel trade --i.e., arbitrage re-export of low-priced drugs into rich nations. The Høsjbør conference report suggested inter alia that manufacturers should use different trademarks or at least different packaging for their lowpriced exports. Gilead went farther, requiring that the consumers of low-priced Sovaldi unseal and open the package they receive from a pharmacist or clinic and consume the first pill in the sight of the dispenser, thus rendering the package much less suitable for resale and re-export.14 Granted, this poses a risk that Hepatitis C sufferers in high-priced nations might travel to a low-price site to obtain prescriptions, buy the pills, and save money. But the risk may be small, given that a full course of treatment requires 84 days and the cost of medical tourism is hardly trivial.

Conclusion In sum, differential pricing offers a way to make the newest patented drugs available to low-income nations’ consumers even while much higher prices in rich nations yield profits that among other things, help reimburse past research and development cost and maintain incentives to invest in further R&D. And when it fails, compulsory patent licensing to generic suppliers provides an alternative consistent with world trade rules. Implementing differential pricing and compulsory licensing is not without difficulties. But the experience with HIV/AIDS and now with therapies effective against Hepatitis C -- even more widespread than AIDS -- shows that solutions can be achieved.

References 1. See e.g. Donald G. McNeil Jr., “Fighting Hepatitis by Slashing Drug Price, in Lab Size of Egypt,” New York Times, Dec. 16, 2015, p. A1; “Lowering the Bar on Hep C Meds,” Bloomberg Business Week, January 11, 2016, pp. 18-19; and “Gilead Faces Fight Over Price of Its Hepatitis C Drug and Patents on H.I.V. Drugs, New York Times, January 28, 2016, p. B5. 2. Sofosbuvir entered Gilead’s portfolio with the $11.2 billion acquisition of a Princeton, N.J., biotech startup, Pharmasset Inc., in January 2012. Following ongoing clinical trials and (still uncertain) FDA approval, Pharmasset planned at the time to introduce sofosbuvir at an annual per-patient cost of $36,000. 3. The diagrams are taken from F. M. Scherer and Jayashree Watal, “Post-TRIPS Options for Access to Patented Medicines in Developing Nations,” Journal of International Law, December 2002, p. 926. 4. This assumption simplifies the analysis. If marginal cost is rising with increased volume, serving Nation B at discriminatorily low prices raises prices in Nation A; if it is falling, prices in Nation A are reduced with discrimination.

licenses. See “AIDS Activists Issue Grades to Drug Companies,” New York Times, Sept. 10, 2009, p. B3. Abbott’s grade was “F.” 14. See again McNeil, op. cit.

F. M. Scherer is Aetna Professor Emeritus at the John F. Kennedy School of Government, Harvard University. His undergraduate degree was from the University of Michigan. He received MBA (1958 ) and Ph.D. (1963)degrees from Harvard. His research specialties are industrial economics and the economics of technological innovation.

Note too that the $28 marginal cost assumed in Figure 1 is much higher in relationship to the demand curve than it must be in the Sovaldi case. 5. The deadline for the least-developed countries was later extended to 2016, and still more recently to 2033. 6. Scherer and Watal, “Post-TRIPS Options,” pp. 930-933. 7. See the conference summary report, written by rapporteur Jayashree Watal, “Differential Pricing and the Financing of Essential Drugs,” in Brigitte Granville, The Economics of Essential Medicines (England: Royal Institute of International Affairs, 2002), Chapter 11. 8. See “Nearly $18M in Discounted AIDS Drugs Allocated for Africa Diverted by Wholesalers and Sold on European Markets,” Kaiser Health News, October 3, 2002, p. 1. 9. Decision of the WTO General Counsel, August 30, 2003 (WTOL/540, amended in WT/L/641). 10. Ibid. 11. See Stephane Luccini et al., “Decrease in Prices of Antiretroviral Drugs in Developing Countries,” in Jean-Paul Moati, ed, Economics of AIDS and Access to HIV/AIDS Care in Developing Nations (Paris: 2003); and “A New Approach to Solicitations for a Troubled AIDS Charity,” New York Times, July 10, 2012, p. D6. 12. “The 30 Years War,” The Economist, June 4, 2001, pp. 89-91; and “A Dispatch from the Front,” The Economist, September 28, 2013, pl 79. 13. Similar price reductions and generic licensing actions were taken for a three-entity Hepatitis C remedy by AbVie, successor to Abbott Laboratories. See McNeil, op. cit. During the first decade of the 21st century, Abbott Laboratories also marketed an important HIV/AIDS drug, Norvir. However, it was slow in embracing differential pricing or granting generic

Spring 2016 Volume 15, Issue 2



Pricing of Pharmaceuticals in the Supply Chain By: Thani Jambulingam, Ph.D. The complexity of pricing in the United States pharmaceutical supply chain is explored. First, a brief description of the supply chain for pharmaceuticals is described followed by a discussion on drug pricing, reimbursement and mark up of different channel members in the supply chain. In the changing regulatory and political environment for pricing, key policies issues are discussed on how they would affect the future of drug pricing in US.

Introduction Pharmaceutical pricing is of heightened interest in United States among all stakeholders including consumers, providers, payers, regulators, and policy makers alike. Pharmaceuticals are essential components of the health care system, and these products are distributed through a network of supply chain partners to reach the patient. In this paper an attempt is made to decipher the complexity of product and payment flows and explain the supply and demand side pricing of pharmaceuticals and mark-ups in United States. The paper concludes with a discussion on policy changes that could impact the supply chain. The United States is the largest pharmaceutical market, attributing to a third of the total global sales of pharmaceuticals.1 About 92% of all retail prescriptions dispensed in United States are covered by some form of insurance plan. About 53% of the prescriptions are covered by private third party insurance, and 39% covered by public program (26% by Medicare and 13% by Medicaid).2 Unlike consumer goods and services, insurance contributes to the complexity and hence the pricing of pharmaceuticals.


Harvard Health Policy Review

For example, consumers buy goods and services where they would evaluate, pay, buy, and use the product. In the case of prescription drugs, the consumer who uses the product does not evaluate or choose the product, the physician who chooses (prescribes) the product does not use or pay for it, and finally the payer or insurer, who pays for the drug (benefit) influence but does not choose or use the product. This creates a level of price insensitivity of consumers to prices and information asymmetry to all the stakeholders (patients, providers, payers, and the government) in the decision making process. Prescription drugs could be classified into single source and multi-source drugs. Single source drugs are branded drugs that are under patent protection and primarily compete in an oligopolistic market with differentiated products due to differences in efficacy, safety, and mechanism of action within a therapeutic area.3 Certain new drugs are exceptions to this condition, as they are the only treatment choices available and operate in a monopolistic market model. Multisource drugs are generics or bio-similars on the other hand, and they enter the market upon patent expiration of the brand and compete primarily on channel access and price.

Pharmaceuticals are formulated into different dosage forms (for example, tablets, capsules, injectables) and can be self-administered or administered by a professional in an outpatient (retail, clinic) or inpatient (hospital) settings. Self-administered drugs are reimbursed as a pharmacy benefit and in Medicare it is the Part D program. Those pharmaceuticals that are administered in a clinic, hospital, home health or specialty pharmacy settings where a professional is involved in the administration of the drug would primarily be part of the medical benefit. In Medicare this is the Part B reimbursement program. The classification of a drug for reimbursement as medical benefit varies by the type of payer. The drugs covered under these two different benefit programs will have different payment method and pricing benchmarks. Also the payments would vary by the type of payer, i.e., cash paying consumer, third party commercial payer, Medicare, Medicaid or Federal facilities (Example: Veteran Administration). Before discussing the pricing by type of payer, a brief description of the supply chain for the pharmaceuticals is warranted.

Pharmaceutical Supply Chain In 2014, an estimated $374 billion worth of pharmaceuticals were distributed in the US supply chain.4 Majority of the prescriptions were generic (87%) and only 13% were for brands. Ninety-one-percent of the total 4.3 billion prescriptions that were dispensed in the outpatient setting (retail) accounts for 72% of the total value.5 Fig.1 shows the US pharmaceutical supply chain with product and payment flows.6 About 90% of phar-

maceutical products primarily get distributed from the manufacturers to the wholesalers and 10% directly to large purchasers such as regional wholesalers, large chain pharmacies, mail order, food stores and non-retail providers (Hospitals, HMOs, Clinics, Home Health Care Providers, Nursing homes and Federal Facilities).7 Out of the 90% of the pharmaceutical products that are distributed through wholesalers, 80% go through traditional wholesalers and 10% through specialty distributors.8 The top three traditional wholesalers namely, McKesson Corporation, Cardinal Health Inc., and AmerisourceBergen generate about 85% to 90% of all revenues in pharmaceutical distribution in US.9 The wholesalers distribute the products to retail pharmacies (independents, chains, mail order pharmacies, supermarkets and mass merchandisers) and non-retail outlets (hospitals, HMOs, clinics and nursing Homes). The specialty distributors supply specialty products (example: oncology, multiple sclerosis etc.) and services to physician offices, clinics, home care providers, hospital pharmacies, and specialty pharmacies. For the independents, 90% of the revenue are from prescriptions. Large Chains drugstore such as Walgreens and CVS generate two thirds of their revenue from prescriptions. The prescription revenue only attribute to less than 10% for supermarkets and mass merchandisers. Consumers have access to their medications from retail and non-retail outlets.

Pricing: Single Source Drugs Pricing of pharmaceuticals can be classified as supply side and demand side pricing. Supply side pricing is what the pharmaceutical manufacturers price their products through the supply chain via wholesaler, pharmacy to patient. The demand side pricing is what the different types of payers would pay. For single source drug, average wholesale price (AWP) and wholesale acquisition cost (WAC) are the most commonly used supply side benchmarks in pricing drugs. Average Wholesale Price (AWP) is a list price that a manufacturer suggests wholesalers charge pharmacies and is used in calculating reimbursements for pharmacies by payers, specifically for brand drugs mainly because AWP is readily available, easily updated, and regularly maintained. Third-parties companies (Ex. Medi-Span) publish this price for public information. The AWP is not the actual price that wholesalers charge but is more like a sticker price of a car. On

Figure 1: Pharmaceutical Supply Chain an average AWP is estimated to be about 20% higher than the Wholesaler Acquisition Cost (WAC) prices. Since it is easily available, payers use AWP as the basis for setting payment to pharmacies. Published by the manufacturers for sale via a wholesaler, WAC is what a wholesaler would pay manufacturer for drugs before discounts. The price the pharmacy pays to acquire drugs for their inventory is usually based on the listed WAC price. WAC pricing does NOT exist for all drugs. Since this is generally a “Wholesaler� price, drug manufacturers who only sell their drugs directly to pharmacies sometimes do not publish a WAC. WAC is also used as a benchmark for payment calculation by the payers. The pharmacies buy products either through the wholesaler or directly. Pharmacies based on their size have different buying power and they negotiate and acquire at prices termed as actual acquisition cost (AAC). But the payers may not know the actual acquisition cost (AAC) by different purchasers unless they have a contract for disclosure of those prices. So the individual payers might survey the pharmacies and wholesalers and create an Estimated Acquisition Cost (EAC) with the objective of estimating the AAC for reimbursement. The payer would use the AWP or WAC as the benchmark and calculate a formula (ex. AWP-15%) that would reflect EAC.10 Since the reimbursement is anchored to the AWP, when the AWP changes so does the reimbursement unless the payer modifies the percentage that it is reduced by

for reimbursement. Alternatively the payer can use the WAC plus model where the reimbursement that would be anchored to WAC plus a percentage as reimbursement to the pharmacies (ex. WAC+4%). So the manufacturers, wholesalers and pharmacies use the supply side pricing for the pharmaceuticals distributed through them. The demand side pricing is what a payer pays for the pharmaceuticals. When a patient fills a prescription at the pharmacy, the reimbursement to the pharmacy varies by the insurance status of the patient. For example if the patient has no insurance he or she would pay the full price, which can also be called usual and customary (U&C) price and it can be up to the AWP prices plus a mark up (ex. AWP + 4%). But when the patient has insurance, for example, private insurance through an employer, the insurer would use a Pharmacy Benefit Management (PBM) to contract with the pharmacy with the objective of creating a network of pharmacies for service to the membership. The pharmacies get the physical products from the wholesalers but negotiate the payment contract with the PBMs who represent the insurers in processing the claims. The negotiated contract would stipulate the reimbursement terms for the drug (ex. AWP-15%) and a dispensing fee (ex. $2).11 The dispensing fee is a fix service fee that the payer would pay to the pharmacy for the professional services provided. The patient would pay a co-pay or co-insurance depending upon the formulary status of the drug. The formulary is a list of drugs that the Spring 2016 Volume 15, Issue 2


health plan would use to cover prescription drugs under the pharmacy benefits. If the patient belongs to Medicare, based on the 2003 Medicare Modernization Act (MMA), the enrollee would obtain the drug benefit from the assigned private prescription drug plan (PDPs) offered by private insurance companies. Center for Medicare and Medicaid Services (CMS) uses the competitive nature of the private insurers to negotiate better reimbursement terms for their members using the market forces. CMS also stipulates special coverage requirements that the private insurers should adhere to. Depending on the payer who pays for the patient, the pricing varies, hence it is called demand pricing.

tion drug, regardless of package size, by any purchaser. Best price(BP) is reported by the manufacturers to the CMS and states only. Medicaid prices can be increased annually up the Consumer Price Index (CPI) rate. Any excess of price increase over CPI needs to be offset with additional Medicaid discounts off AMP. For a multisource generic drug, the rebate is 13% of the AMP.13 Unlike Medicare drug benefit, which uses private health care plans to negotiate prices, Medicaid uses rebate policy to ensure that no private payer would pay less than what Medicaid pays for a drug.

“Specialty drugs are primarily single source drugs that are dispensed by physicians in their clinics such as injections and infusions for cancer, arthritis, etc. These drugs are primarily biologics and might have cold chain requirements” Pricing: Medicaid

Pricing: Federal Facilities

Medicaid is the program for the indigent populations and it is partly funded by federal government and states’ fund that manage the program. With the enactment of Omnibus Budget Reconciliation Act (OBRA 1990), congress created the Average Manufacturer Price (AMP) for the purpose of calculating rebates to be paid by manufacturer to states for drugs dispensed to their Medicaid beneficiaries. The AMP is the average price paid to manufacturers for drugs by the wholesaler or direct purchasers for drugs distributed to the retail pharmacy class of trade distributed through retail pharmacies excluding customary prompt pay discounts extended to wholesalers. Single source drugs supplied to Medicaid patients are subjected to a mandated discount of 23.1% percent (it was 15.1 percent until 2010) off the Average Manufacturer Price (AMP) or the difference between AMP and the best price offered to any private purchaser.12 The best price is the lowest manufacturer price paid for a prescrip-

Department of Veterans Affairs (VA), Department of Defense (DoD), Public Health Services (PHS), and Coast Guard are called the Big 4 pharmaceutical purchasers within the federal government. They have the right to purchase the pharmaceuticals from the Federal Supply Schedule (FSS) that is negotiated by the VA based on the prices that the manufacturers charge their “most favored” non-federal customers.14


Harvard Health Policy Review

Pricing: Specialty Drugs Specialty drugs are primarily single source drugs that are dispensed by physicians in their clinics such as injections and infusions for cancer, arthritis, etc. These drugs are primarily biologics and might have cold chain requirements. These drugs would be reimbursed under medical benefit unlike outpatient pharmacy benefit. Medicare reimburses these specialty drugs under Part B as opposed on Part D, which is outpatient

prescription benefit. Under the current CMS rules these products are reimbursed based on manufacturer’s Average Selling Price (ASP) plus a 6 percent mark up.15 ASP is based on manufacturer reported actual selling price data and includes the majority of rebates, volume discounts, and other price concessions offered to all classes of trade. Biosimilar drugs also use the ASP benchmark to price the products as well.16 There is incentive for the physicians to use products with higher ASP since the mark up is related to the value of the ASP. Payers see that there is a misalignment of incentives to the physicians to control costs. Payers have less control on prices for these drugs since the providers are unwilling to accept price control that would compromise clinical options. More drugs are becoming specialty products, and this is a rapidly growing segment of pharmaceuticals.

Pricing: Multisource Drugs - Generics When single source brands lose patents, multiple manufacturers enter the market. If the drug is a small molecule (chemical based) then it is called a generic but if it is a biologic then is called a biosimilar. A biosimilar product is a biological product that is approved based on a showing that it is highly similar to an already-approved biological product, known as a reference product. In US out of the total number of prescriptions dispensed, generics account for 86% of the total prescriptions by volume. The success of a generic manufacturer is dependent in its ability to access the distribution channel (i.e., wholesalers and pharmacies) providing market access. Generic companies provide significant discounts to get channel access and so payers have to devise a better method for reimbursement especially when there are multiple generics in the market. In comparison to single source branded products, the AWP and/ or WAC prices are not a good proxy, hence payers developed a payment mechanism called Maximum Allowable Cost (MAC) for multi-source generic drugs specific to their plan. Manufacturers of multi-source generic drugs might have different established AWP prices.17 The payers would like to curtail the pharmacies from selecting the most expensive generics and also enable easier reimbursement model for generics. So the payers established the MAC by comparing prices of the different generics and using a formula to establish the maximum allowable cost for all

those generics. This pricing would allow the payers to pay the same price of the drug regardless of the manufacturer. The formula used might vary by payer and often they are proprietary. For Medicaid, MAC is defined by each state. In addition, the federal government that funds for state Medicaid programs establishes a Federal Upper Limit (FUL) for payment for multisource generic drugs.18 FUL is the maximum amount determined by CMS that a state Medicaid program can pay for a multiple source pharmaceutical product. The objective of MAC and FUL are to control the cost of drugs. These prices are determined by the payers or by PBMs. Payers generally set the MAC at a relatively low price, say 27% of AWP. Because the pharmacy captures the margins between the MAC and the acquisition cost of the drug (AAC) the generic manufacturers compete by offering discounts on the acquisition cost to increase the margins. Figure 2: Hypothetical Pricing for Single Source Drugs

Hypothetical Example: Margins aFederal Facilities include Department of Veteran Affairs, Department of Defense (DoD), Public Health Services (PHS) and Cost Guard. The 4 federal agencies have the right to purchase their pharmaceuticals and Mark Up from the Federal Supply Schedule (FSS).

Research-based pharmaceutical firms over the past five years had, on an average, a gross margin of 70.2% and net profit of 20.9% based on the analysis of top 20 pharmaceuticals firms in the industry.19 Generic firms in the last five years on an average had lower gross margins of 41.1% and net

wholesaler is 4.4% and the average net profit is 0.82%.21 In the case of retail pharmacies, overall average gross margin is 20-25%. For drug chains the five-year average gross margin is 22% and the net profit is 3.2%.22 The discounts are moderate from list prices for brand-name prescriptions but deep dis-

are anchored to AMP prices, the lower the AMP gets the better the pricing to these systems.

“Since AMP is only available to CMS for calculating rebates for Medicaid, the data is not available to the commercial third party insurers and they are increasingly demanding transparency.� profits of 5.1%.20 For branded companies, the net profit margins have remained stable over a 5-year period whereas the margins for generics are declining steadily due to competition. Wholesaler margins for single source branded drugs are 3-5% for drug distribution and obtain their revenues from price increase by manufacturers, prompt pay discounts (2%), and spread between the published WAC prices and the WAC after discounts. The average gross margins for the top 3

counts for generics. Pharmacies profit margins are much higher for generics (50-60%) due to the spread (EAC-AAC), prompt pay discounts and other service charges and fees. The service charges are applicable to specialty pharmacies and not so much for traditional pharmacies. Fig 2 summarizes the pricing of the pharmaceutical in the supply chain and by payer type. The patient who does not have insurance pays the highest, followed by those who have third party insurance (including Medicare), followed by Medicaid and Federal facilities such as VA system. Since the prices to Medicaid and federal facilities

Policy Implications There are several policy implications of pharmaceutical pricing in the US. Since AMP is only available to CMS for calculating rebates for Medicaid, the data is not available to the commercial third party insurers and they are increasingly demanding transparency. Will AMP become publically available? What are the implications to the pharmaceutical industry? Second, the Medicare drug benefit proSpring 2016 Volume 15, Issue 2


gram (Part D) is implemented through private third party insurance companies for the enrollees. The market competition among the insurance companies drives the prices down for Medicare patients. But there is an outcry from policy makers that Medicare should directly negotiate prices with pharmaceutical companies.23 The key policy question is whether the CMS is equipped to do the negotiations and implementation of the program, or let the market forces drive better pricing for the enrollees? What are the pros and cons of such a policy change? What impact will it have on new drug development and pricing?24 Finally, new benchmarks are being created to replace AWP price referencing to calculate reimbursement.25 The benchmark National Average Drug Acquisition Cost (NADAC) is being used to get much closer to the actual acquisition costs (AAC) so that reimbursements to pharmaceuticals will reflect and be in alignment to the true price of the drugs.26 Yet the newer benchmarks are slow to be adopted as the AWP reference pricing is still used commonly. What impacts will the new benchmarks have on profitability of wholesalers and pharmacies and how will it impact primarily the independent pharmacies whose primary business is prescription pharmaceuticals?

Conclusion Pricing of pharmaceuticals in the supply chain is complex. This paper is an attempt to unravel this mystery. In summary, consumers with no drug coverage pay the highest price for their medications if they do not avail the patient assistance or drug discount and coupon programs available to them if they qualify for the programs. Pricing and mark up for branded drugs differ from generics and wholesalers, and pharmacies get higher margins for generics versus. branded drugs. Policy makers are constantly finding newer methods to control costs by fostering competition and by regulatory polices.


Harvard Health Policy Review

References 1. Medicine Use and Shifting Costs of Healthcare: A Review of the Use of Medicines in the United States in 2014, IMS Institute, April 2015 2. ibid. 1 3. Berndt, Ernst, Thomas G. McGuire and Joseph P. Newhouse, “ A Primer on the Economics of Prescription Pharmaceutical Pricing in Health Insurance Markets,” NBER Working Paper Series, Working Paper 16879, 2011 4. ibid. 1 5. ibid. 1 6. AMCP Guide to Pharmaceutical Payment Methods, 2013 Update, Version 3.0, Academy of Managed Care Pharmacy, p. 11. 7. Fri, Perry. “Understanding the Pharmaceutical Supply Chain,” HDMA, July 2015 8. ibid. 7 9. ibid. 7 10. CBO, Prescription Drug Pricing in the Private Sector, January 2007 11. ibid. 10. 12. ibid. 6 13. US Department of Veteran Affairs, http:// - Accessed January 28th 2016. 14. Ibid. 12 15. EMD Serono Specialty Digest, Managed Care Strategies for Specialty Pharmaceuticals, 11th Edition, 2015 16. CMS. gov: Medicare-Fee-for-Service-Part-B-Drugs/ McrPartBDrugAvgSalesPrice/Part-B-Biosimilar-Biological-Product-Payment.html, Accessed January 29th 2016. 17. ibid. 7 18. Medicaid-CHIP-Program-Information/ By-Topics/Benefits/Prescription-Drugs/ Federal-Upper-Limits.html - Accessed January 29th 2016. 19. Factset Research System Inc., United States: Pharmaceutical Major, Accessed January 29th 2016 20. Factset Research System Inc., United States: Generics, Accessed January 29th 2016 21. Factset Research System Inc., United States: Medical Distributors, Accessed January 29th 2016 22. Factset Research System Inc., United States: Drug Store Chains, Accessed January 29th 2016 23. Sanger-Katz, Margot, “The Real Reason Medicare is a Lousy Drug Negotiator: It Can’t Say No,” The New York Times, Feb. 6 2016. 24. CBO Report. Competition and the Cost of Medicare’s Prescription Drug Program, July 2014. 25. Bruen, Brian and Katherine Young, “Paying for Prescribed Drugs in Medicaid: Current Policy and Upcoming Changes,” Kaiser Family Foundation, May 2014 26. ibid. 22

Dr. Thani Jambulingam is Arrupe Research Fellow and Professor of Pharmaceutical and Healthcare Marketing at Erivan K. Haub School of Business, Saint Joseph’s University in Philadelphia, whose research interests include pharmaceutical and healthcare strategy and policy. He is widely regarded as a pharmaceutical and healthcare industry expert and his research has been published in journals in marketing, management, and economics.


‘Broad societal consensus’ on human germline editing By: Françoise Baylis, Ph.D. CRISPR (‘Clustered Regularly Interspaced Short Palindromic Repeats’) is a new gene editing technique that can be used to change the genome of living cells by deleting, repairing or replacing genes.1 This technology can be used to change somatic cells (i.e., body cells whose genomes are not transmitted to subsequent generations) or germ cells (i.e., sperm and eggs whose genomes are transmitted to subsequent generations). To date, no CRISPR-edited human cells have been transferred to humans. In the near future, the hope is to move to clinical trials using CRISPR-edited human somatic cells. In the distant future, there is the prospect of using CRISPR-edited human gametes or early human embryos for reproduction. The genetic modification of gametes or early embryos would result in germline editing, as the genetic changes would be passed on to offspring and subsequent generations. At the time of writing, there is common knowledge of two basic science projects involving gene editing of early human embryos in a research setting. In April 2015, the journal Protein & Cell published work by a research group in China at Sun Yat-sen University in Guangzhou (led by Canquan Zhou and Junjiu Huang) that involved making genetic alterations to nonviable human embryos. 2 The research aimed to repair a genetic defect that causes beta thalassemia (a potentially fatal blood disorder). No genetically modified human

embryos were transferred to initiate a pregnancy. More recently, in February 2016, the United Kingdom’s Human Fertilisation and Embryology Authority (HFEA) approved a research license renewal application submitted by Kathy Niakan from the Francis Crick Institute. The license application was for human embryo research that would include knocking out

The first of these two human embryo projects spurred considerable ethical debate and angst, as the research demonstrated both the potential to modify the human genome across generations, and the inherent risks in doing so. In the months preceding the publication of the research (and according to some ‘in anticipation of the publication of the research’),6 there were calls for a voluntary moratorium on modifying the human germline7,8 to allow for careful deliberation on the risks and benefits of the technology and “the attendant ethical, social, and legal implications of genome modification”.9 In the ensuing debate, many argued that the research, though not

“At the time of writing, there is common knowledge of two basic science projects involving gene editing of early human embryos in a research setting.” the OCT4 gene in healthy embryos to better understand embryonic development with the hope of eventually contributing to advances in pregnancy and fertility treatment.3 The HFEA approval was subject to ethics approval “from an appropriately constituted research ethics committee”.4 At the time of approval, the HFEA underscored the fact that “as with all embryos used in research, it is illegal to transfer them to a woman for treatment.”5

intended for use in pregnancy, nevertheless crossed an ethical rubicon and would lead to the creation of ‘designer babies’ and the introduction of a new eugenics. In response to this burgeoning debate, in December 2015, the U.S. National Academies of Science, the U.S. National Academy of Medicine, the Royal Society, and the Chinese Academy of Science hosted an International Summit on Human Gene Editing. At the close of the International Summit, the Organizing Committee of ten Spring 2016 Volume 15, Issue 2 19

scientists and two bioethicists (of which I was a member) issued a formal statement.10 This statement – On Human Gene Editing: International Summit Statement – included four discrete conclusions. In this article, I briefly outline each of the four conclusions. I then elaborate on the third conclusion which includes two clear thresholds for moving forward with human germline editing – namely, (i) evidence of safety and efficacy, and (ii) ‘broad societal consensus’. Taken together, these two thresholds for acceptability create a potentially useful policy-making framework. Next, I move to a discussion of the fourth conclusion, which calls for an ongoing international forum – broadly inclusive of a diversity of nations, perspectives and expertise – to discuss the potential merits and harms of engineering humans. I suggest that such an ongoing forum is a sine qua non for achieving ‘broad societal consensus’, and then offer a model for decision-making by consensus. This model embraces work done by women activists in the 1980s and calls on scientists to support the consensus building process through honest brokering of policy options. In this way, this article begins the project of fleshing out whether and, if so, under what circumstances, human germline engineering might proceed.

International Summit Statement: four conclusions11 First, members of the Organizing Committee concluded that, in their view, there was no reason to curtail basic and preclinical research on human cells. Labbased research could continue in accordance with “appropriate legal and ethical rules and oversight”. In this way, the Committee endorsed laboratory research involving human somatic cells as well as human sperm, eggs and early embryos. This conclusion would have been reassuring to those involved in human embryo research, as they could interpret it to mean ‘business as usual.’ Conversely, those who object to any and all human embryo research would not have agreed with this conclusion. Further, among those who might otherwise cautiously support some human embryo research, there would be those who object to this conclusion on the grounds that it leaves the door open to possible misappropriation of genetically modified embryos to initiate a pregnancy. As well, depending upon the focus of the embryo research, disability activists would


Harvard Health Policy Review

have had serious reservations about the ways in which specific research might reinforce flawed and harmful assumptions about what kinds of lives should be prevented. Second, the Organizing Committee concluded that gene editing involving human somatic cells could proceed in both a research and a therapeutic context, always with careful attention on the part of both the researchers and the regulators to the risks and potential benefits of such research. This conclusion would have been welcome news to researchers working to develop therapeutic interventions for identifiable patients, who feared that disproportionate attention to the controversy surrounding germline editing would negatively affect their ability to proceed with clinical trials. This conclusion would also have satisfied patients and patient advocacy groups eager for the science to move forward with a view to improving human health. Some disability rights activists, however, would have been concerned with this conclusion, which could reasonably be perceived as uncritical endorsement of a technology that would further contribute to both geneticization (understanding humans primarily in terms of their DNA)12 and ableism (discrimination that favours able-bodied individuals).13 As well, there could have been concerns about the ways in which this conclusion would further undermine important distinctions between normal variation and disability.14 Lastly, some would have been concerned about the possible use of gene-edited somatic cells for enhancement purposes. Third, the Organizing Committee addressed the use of gene editing technology in human gametes and early human embryos destined for reproductive use. Gene editing in these cells would result in genetic alterations to offspring and subsequent generations. The Committee concluded that “[i]t would be irresponsible to proceed with any clinical use of germline editing unless and until: (i) the relevant safety and efficacy issues have been resolved, based on appropriate understanding and balancing of risks, potential benefits, and alternatives, and (ii) there is broad societal consensus about the appropriateness of the proposed application.” Those who hoped for as wide as possible a marge de manoeuvre would have been pleased with this conclusion, as they could reasonably interpret it as a recommendation to ‘proceed with caution’. On this view, the first three conclusions follow in step-wise

fashion one from the other: first, do the lab work in human somatic and germ cells; second, proceed to clinical trials involving the transfer of gene-edited human somatic cells; and third, by the time there is sufficient evidence of safety and efficacy with the transfer of genetically modified somatic cells, expect public awareness and acceptance of the potential therapeutic benefits of germline editing to have shifted sufficiently as to provide a ‘broad societal consensus’ on human germline editing for therapeutic purposes. On the other hand, those who hoped for either a ban or a moratorium on human germline editing would have been disappointed with the Committee’s failure to take a stronger stance in support of what might reasonably be described as a ‘broad societal consensus’ against this use of gene editing. Evidence of this consensus could be found in a number of countries with legislation or guidelines prohibiting human germline editing,15 and in quasi-governmental and professional organizations’ statements condemning human germline editing.16 Either of these two mechanisms – a ban or a moratorium – would have served to temper the enthusiasm of researchers like George Church who is reported to have described his lab as “the center of a new technological genesis—one in which man rebuilds creation to suit himself ”.17 Fourth and finally, the Organizing Committee called on the sponsors of the Summit to create an ongoing forum for discussion to encourage thoughtful conversation among individuals with a wide range of knowledge, expertise, experience, and values. Participants in this conversation were to include “not only biomedical scientists, policymakers, regulators, research funders, and industry representatives, but also health-care providers, patients and their families, people with disabilities, ethicists, lawyers, social scientists, faith leaders, public interest advocates, and members of the general public”.18 Some would have been particularly pleased with this conclusion for at least two reasons. First, it clearly aimed to create a legitimate space for additional voices to contribute to the global discussion. Second, it arguably recognized that the work that needed to be done to flesh out the two threshold elements would benefit from discussion among persons with diverse perspectives. Others would have been deeply concerned about who ultimately would have authority to make what decisions.

A moratorium by any other name19 genome engineering of the human germline, The claimed benefit of human germline editing is its potential to cure serious inherited diseases not only in individuals, but in their children and in subsequent generations. A second potential benefit, from the perspective of some, is the prospect of enhancing human traits and capabilities. The widely acknowledged potential harms of human germline editing include: the risk of introducing genetic changes with longterm harmful consequences for individuals, families, and future generations; the risk of exacerbating social inequalities; the risk that the technology might be used coercively; the risk of a new eugenics; and the risk of human enhancement. While some are ever so keen to co-author the human evolutionary story and thus see human enhancement as a benefit, others question the audacity of those who embrace volitional evolution in seeking to improve the human condition. In the months leading up to the Summit, and at the Summit, there were prominent calls for a ban or a moratorium on human germline

at least as long as the safety and efficacy of the procedures are not adequately proven as treatments.”23 And, during the Summit, on December 2, 2015, the Council of Europe Committee on Bioethics issued a “Statement on Genome Editing Technologies”24 in which it recalled the prohibition in Article 13 of the Convention on Human Rights and Biomedicine (commonly known as “the Oviedo Convention”) on any intervention that would affect the germline.25 In very general terms, those who advocate a complete ban on human germline editing typically advance one or more of the following arguments. There are arguments about the inability of children born of genetically altered embryos to consent to such alterations and the resulting threat to their right to an open future. There are arguments about the difficulties of long-term follow-up given that the results of germline editing could not be fully analyzed for generations to come. There are arguments about unbridled hubris and the attendant risk of catastrophic

“The claimed benefit of human germline editing is its potential to cure serious inherited diseases not only in individuals, but in their children and in subsequent generations. A second potential benefit, from the perspective of some, is the prospect of enhancing human traits and capabilities” editing, not only from individuals20,21 but also from professional organizations.22 Notable among these was the statement issued by the International Bioethics Committee (IBC) of UNESCO on October 2, 2015, the same day as the public information gathering meeting hosted by the International Summit Organizing Committee. On that day, UNESCO released the “Report of the IBC on Updating its Reflection on the Human Genome and Human Rights”. Taking into account the Universal Declaration on the Human Genome and Human Rights (1997), the International Declaration on Human Genetic Data (2003), and the Universal Declaration on Bioethics and Human Rights (2005), the IBC called on states and governments to “Agree on a moratorium on

irreversible biological consequences. And, there are arguments about unacceptable social consequences and inevitable human rights abuses resulting from new forms of eugenics, unfair discrimination and prejudice, and stigmatization. The purpose of a ban is to entrench a permanent prohibition. Meanwhile, those who advocate a moratorium on human germline editing tend not to be troubled by arguments suggesting that consent on the part of children born of genetically altered embryos is required. Rather, they typically worry that the risk of failure may be too great to warrant proceeding, or that the anticipated benefits may be too few to warrant proceeding, or that preferable alternatives may be available

(or in the offing). In this context, some focus narrowly on harmful biological or medical consequences, others worry about negative ethical and social consequences – including a new kind of bottom-up eugenics shaped by dominant economic, social, and political forces. The hope with a moratorium is for a stay to allow for careful reflection, discussion and debate (during which time, available facts and social mores inevitably will change). The worry with a moratorium, which essentially ‘stops the clock’, is that it might nonetheless function like a ban – as when a temporary prohibition becomes ‘frozen in time’. The Organizing Committee did not endorse a ban, and it eschewed the language of a moratorium in favour of language that clearly communicated ‘not now’. Why ‘not now’? Because of serious concerns about safety and efficacy, and because of lack of agreement on legitimate (ethically sound) goals for the use of this technology.26 Importantly, these two reasons for the ‘not now’ pronouncement/verdict (i.e., for the ‘actual, but not so-called, moratorium’) form the basis of a policy-making framework that allows for moving forward (i.e., ‘not now’ but ‘maybe later’). The framework is simple insofar as it only includes two threshold elements: (i) demonstrated safety and efficacy (taking into consideration risks, potential benefits and alternatives); and (ii) broad societal consensus about acceptable uses of the technology. Paradoxically, the framework is also quite complicated because the substance (meaning and the scope) of each of these elements is unclear and very likely to be contested. As the science of human gene editing continues to develop, we may be able to negotiate a common understanding of appropriate standards for safety and efficacy for human germline editing, but this will not be without considerable (and perhaps vociferous) ethical and policy debate. For example, is ‘reversibility’ – whatever that might mean – a facet of safety? In any case, regardless of how easy or difficult it will be to agree on appropriate standards for safety and efficacy, in all likelihood it will be more difficult still to negotiate agreement on what is required for ‘broad societal consensus’. What could this mean? And, more importantly, what should this mean?

‘Broad societal consensus’

If ‘broad societal consensus’ is to be a meaningful criterion for moving forward Spring 2016 Volume 15, Issue 2


with science for the benefit of humankind, we need a clear and robust answer to the normative question, “What should this mean?” Notably, the Organizing Committee’s fourth conclusion calls for an ongoing international forum for discussion. Helpfully, this sets the stage for continued learning about the science of human germline editing. It also sets the stage for continued deliberation about the meaning of ‘broad societal consensus’, and about how such consensus might best be achieved. Here it is worth recalling Ruha Benjamin’s caution that we not create a forum for discussion that appears public, but really only serves to insulate science from criticism.27 In the summer of 1983, thousands of women camped out at Romulus, in Seneca County near the Seneca Army Depot to stop the deployment of Cruise and Pershing II nuclear missiles to Europe. These women – participants in the Seneca Women’s Peace Encampment – stood together in opposition to violence and oppression. As part of this collective effort, the women developed a statement on decision-making by consensus which they included in their Resource Handbook. This statement is reprinted below in its entirety: Consensus does not mean that everyone thinks that the decision made is necessarily the best one possible, or even that they are sure it will work. What it does mean is that in coming to that decision no one felt that her position on the matter was misunderstood or that it wasn’t given a proper hearing. Hopefully, everyone will think it is the best decision; this often happens because, when it works, collective intelligence does come up with better solutions. Responsibility: Participants are responsible for voicing their opinions, participating in the discussion, and actively implementing the agreement. Self-discipline: Blocking consensus should only be done for principled objections. Object clearly, to the point, and without putdowns or speeches. Participate in finding an alternative solution. Respect: Respect others and trust them to make responsible input. Cooperation: Look for areas of agreement and common ground and build on them. Avoid competitive, right/wrong, win/lose thinking. Struggle: Use clear means of disagreement – no putdowns. Use disagreements and


Harvard Health Policy Review

arguments to learn, grow and change. Work hard to build unity in the group, but not at the expense of the individual who are its members.28 What is perhaps most noteworthy about this particular understanding of consensus is that it does not set the impossible standard of unanimity, nor does it reduce consensus to majority rule (which clearly would be ethically suspect in this context). Rather, it spells out clear responsibilities for all interested and willing participants in a democratic decision-making process. All are to assume responsibility for active, principled, respectful, and cooperative engagement. The consensus building process does not rely on hierarchy, does not privilege elites, and does not denigrate different types of knowledge. As different perspectives are discussed and debated, participants are enjoined to look for common ground on which to build consensus. When there is no common ground to be found, participants are to critically examine their contributions to discussion and debate, and, as appropriate, to recognize when they are an outlier. If they have had a fair hearing and others have not been swayed by their arguments, then they ought to recognize their own fallibility and step back instead of blocking emerging consensus for personal as contrasted with principled reasons. In this way, the consensus building process valorizes compromise as evidence of commitment to procedural fairness (which is necessary in a democracy), but eschews compromise that results in an erosion of personal moral integrity, leaving the individual with the experience of having ‘been compromised’. This is the difference between compromise as a conciliatory process and outcome, and compromise as betrayal. As an important feature of the consensus building exercise on human gene editing is broad-based participation by persons from around the world with a range of perspectives and interests, an important question arises as to the proper role of scientists in the deliberations. Roger Pielke Jr. outlines four discrete idealized roles for scientists vis-à-vis policy-making.29 These are: (i) the Pure Scientist who is narrowly interested in knowledge production and who takes no responsibility for how policy-makers do or don’t use the knowledge produced; (ii) the Science Arbiter who stands ready, willing, and able to answer questions policymakers deem relevant (having no particular

interest in the values or priorities informing the questions asked); (iii) the Issue Advocate who is committed to a particular policy option and who accordingly tries to inform the preferences of policy-makers; and (iv) the Honest Broker of Policy Alternatives who provides policy-makers with an informed overview of a wide range of policy options and who trusts the policy-maker to make worthy science-informed policy choices. According to Pielke, the first two of these idealized roles – that of Pure Scientist and Science Arbiter – rest on an outdated linear model of science according to which science moves along a (mythical) continuum from basic research, to applied research, to development, to application, to societal benefit. On his view, if the linear model of science has any purchase, it is in allowing scientists to present themselves as Pure Scientists or Science Arbiters while effectively taking on the role of stealth Issue Advocates. These are scientists who, unlike their doppelgangers, have a clear interest in ‘helping’ policy-makers to ‘see’ which policy alternatives they ‘prefer’. 30 Contrastingly, in an ideal world, scientists would pride themselves on being Honest Brokers of Policy Alternatives – persons committed to expanding the policy options and empowering decision-makers.

Conclusion In closing, courtesy of the Organizing Committee of the International Summit on Human Gene Editing, we have a potentially useful policy-making framework for human germline gene editing that has two threshold elements – evidence of safety and efficacy, and ‘broad societal consensus’. The framework is simple in terms of structure, and complex in terms of substance. As we set about exploring this complexity, I propose that we: (i) endorse an understanding of consensus building that, at minimum, is grounded in a commitment to the principles of responsibility, selfdiscipline, respect, cooperation and struggle; (ii) we enjoin scientists to embrace the role of honest brokers of policy alternatives, and (iii) work together towards a common understanding of the science and the ethics of human germline editing.

References 1. Doudna, J.A., and Charpentier E. The new frontier of genome engineering with CRISPR-Cas9. Science 2014; 346: 1077. 2. Liang, P., Xu, Y, Ding C., et al. CRISPR/ Cas9-mediated gene editing in human tripronuclear zygotes. Protein & Cell 2015; 6: 363-372. 3. HFEA approves license application to use gene editing in research. Available at: http://www. 4. HFEA License Committee. Minutes. January 14, 2016. Available at: uk/guide/ShowPDF.aspx?ID=5966 5. HFEA approves license application to use gene editing in research. Available at: http://www. 6. Collins, K. Researchers ‘edit’ human embryo genes for the first time ever. Wired April 23, 2015. Available at: news/archive/2015-04/23/gene-editing-human-embryos-first-controversial-study 7. Lanphier, E., Urnov, F., Haecker, S.E., et al. Don’t edit the human germ line. Nature 2015; 519: 410-11. 8. Baltimore, D., Berg, P., Botchan, M., et al. A prudent path forward for genomic engineering and germline gene modification. Science 2015; 348:36-8. 9. Ibid. 10. Organizing Committee for the InternationalSummit on Human Gene Editing. On human gene editing: International summit statement. December 3, 2015. Available at: http://www8. aspx?RecordID=12032015a 11. Baylis, F. Human gene editing: A global discussion. Impact Ethics February 12, 2016. Available at: https://impactethics. ca/2016/02/12/global-response-to-human-gene-editing/ 12. Lippman, A. Prenatal genetic testing and geneticization: mother matters for all. 1993. Fetal Diagnosis & Therapy 8 (Suppl. 1), 175–188. 13. Wolbring, G. Gene editing: Govern ability expectations. Nature 2015; 527: 446. 14. Shakespeare, T. Gene editing: Heed disability views. Nature 2015; 527: 446. 15. Araki, M., and Ishii, T.: International regulatory landscape and integration of corrective genome editing into in vitro fertilization. Reproductive Biology and Endocrinology 2014; 12:108. Available at: content/12/1/108 16. For example, Friedmann, T., Jonlin, E.C., King, N.M.P., et al. ASGCT and JSGT joint position statement on human genomic editing. Molecular Therapy 23(8): 1282; International Bioethics Committee, UNESCO. Report of the IBC on Updating its Reflection on the Human Genome and Human Rights. October 2, 2015, p. 3. Available at: http://unesdoc. pdf; and Committee on Bioethics, Council of Europe. Statement on genome editing technologies. December 2, 2015. Available at: https:// documentId=090000168049034a 17. Regalado, A. Engineering the perfect baby. MIT Technology Review. March 5, 2015 Available at: https://www.technologyreview.

com/s/535661/engineering-the-perfect-baby/ 18. Baylis, F. and Rossant, J. This CRISPR moment: Editing human DNA the way we edit text – are we ready? The Walrus. April 2016: 15-17. Available at: 19. Borrowing from William Shakespeare’s “A rose by any other name would smell as sweet”, I contend that “A moratorium by any other name would have the same effect”. 20. Lanphier, E., Urnov, F., Haecker, S.E., et al. Don’t edit the human germ line. Nature 2015; 519: 410-11. 21. Kaebnick, G.E., A moratorium on gene editing? Bioethics Forum. March 27, 2015. Available at: 22. Friedmann, T., Jonlin, E.C., King, N.M.P., et al. ASGCT and JSGT joint position statement on human genomic editing. Molecular Therapy 23(8): 1282. 23. International Bioethics Committee, UNESCO. Report of the IBC on Updating its Reflection on the Human Genome and Human Rights. October 2, 2015, p. 3. Available at: 24. Committee on Bioethics, Council of Europe. Statement on genome editing technologies. December 2, 2015. Available at: https:// 25. Council of Europe. Treaty No. 164. Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine. Available at: 26. Organizing Committee for the International Summit on Human Gene Editing. On human gene editing: International summit statement. December 3, 2015. Available at: http://www8. aspx?RecordID=12032015a 27. Benjamin, R. Interrogating equity: a disability justice approach to genetic engineering. International Summit on Human Gene Editing. Washington, DC December 3, 2015. 28. Women’s Encampment for a Future of Peace and Justice, Seneca Army Depot, NY, Resource Handbook, 42. 29. Pielke, R. Jr. The honest broker: Making sense of science in policy and politics. Cambridge: Cambridge University Press, 2007. 30. Ibid., p. 76.

Françoise Baylis, Professor and Canada Research Chair in Bioethics and Philosophy at Dalhousie University in Canada, is an elected fellow of the Royal Society of Canada and the Canadian Academy of Health Sciences. She has particular expertise on the ethics of heritable genetic modification. She was an external reviewer for the Institute of Medicine report “Mitochondrial Replacement Techniques”. She was also a member of the 12-person Organizing Committee for the December 2015 International Summit on Human Gene Editing.

Spring 2016 Volume 15, Issue 2



The CRISPR Revolution: Technical and Societal Aspects of Current Genome Editing Technologies By: Dana Carroll, Ph.D. The advent of the CRISPR methods for making targeted changes in chromosomal DNA sequences has generated unprecedented levels of excitement in the laboratory and attention from the media. In many ways this “new” technology simply builds on methods that came before, but the ease and simplicity of the CRISPR approach make genome editing much more accessible to researchers around the world. The same features raise genuine concerns about how, where and when the technology will be applied. This article provides a description of how CRISPR works, illustrates some current and potential uses, and suggests why a cautious approach to those uses is warranted.


Genome Editing Technology

When I tell people about our current capabilities for genetic manipulation, I often begin by saying that nature has been editing genomes for a very long time. Genetic variation – the slow change of chromosomal DNA sequences – is the material on which evolution works, allowing for selection of genomes that provide organismal fitness in particular environments. Over many decades, scientists have developed methods to increase the rate of variation; then methods to introduce specific genetic changes in a few organisms; and most recently, methods to edit DNA sequences of genes in their natural locations in essentially any animal, plant or microbe. Surprisingly, the secret of efficient, precise gene editing is the ability to make a unique break in chromosomal DNA at the desired target1. This is what the CRISPR tools do, and it was true of previous platforms as well.

DNA cutting enzymes are called nucleases, and the first truly targetable genome editing reagents were the zinc-finger nucleases (ZFN), which were shown to have this property in cells and organisms in 2001 and 20022,3. The ZFNs were followed by TALENs in 2010 and 20114,5. The programmed specificity of both platforms depends on the assembly of DNA recognition modules that are linked to a non-specific DNA cleavage domain from a natural protein. The ZFNs were challenging to produce for new targets because there was not a simple recognition code to rely on. TALENs were a substantial improvement, offering a robust code of one protein module for each base pair in the DNA target. While these platforms opened entirely new prospects for modifying cellular genomes, each required the design and production of two new proteins for each new target.


Harvard Health Policy Review

A key feature of the CRISPR system is that it depends on a single protein, Cas9, that does not vary from gene to gene6,7. Recognition is provided by a small RNA molecule, called the guide or single-guide RNA. To address a new target, one simply relies on the Watson-Crick pairing rules to generate an RNA that is complementary to one strand of the desired DNA. Not only does this simplify the process of nuclease design, it is possible to attack multiple targets simultaneously by providing multiple guide RNAs. Furthermore, the production of Cas9 protein and guide RNAs requires only basic skills in molecular biology. Once a break is made in genomic DNA by any of these nucleases, subsequent events are determined by cellular processes designed to repair that break1. One mode of repair involves sticking the ends back together, sometimes including small errors – insertions or deletions of base pairs – at the junction. If this occurs in protein coding sequences or other critical regions, it can result in a gene knockout mutation. Another type of repair employs closely related sequences as a template. Researchers can provide a synthetic template that carries the desired sequence change, and the cell will incorporate this change at some frequency. Thus, genes can be disabled or altered in specific ways via nuclease-mediated genome editing. Remarkably, genome editing can be very efficient – in very favorable circumstances, the targeted sequence is altered in essentially every cell attacked. It can also be very specific, with induced sequence changes limited to the desired target, although there is still concern about low-level off-target effects. It is worth pointing out that all of these

powerful nuclease platforms arose from unpredictable places. The DNA recognition modules of ZFNs and TALENs were found in natural transcription factors: the first in eukaryotic organisms, the second in plant pathogenic bacteria. The nuclease domain used in both ZFNs and TALENs came from a natural, bacterial restriction endonuclease. The CRISPR tools are part of a natural system of viral immunity that was discovered in bacteria and archaea based on an initial description of odd sequence repeats in their genomes. It was only broad based research into the natural world that revealed the components of the technologies that provide our current capabilities.

Applications of Genome Editing We now can make targeted changes in essentially any gene in any organism. The list I maintain of species that have had their genomes successfully modified with the three platforms now runs to more than 80 organisms. In many cases, the objective is to investigate the function of a gene by disabling it and noting the consequences. More am-

age-induced browning, and removing allergenic proteins from peanuts8,9. In the realm of livestock, a current application is to genetically dehorn dairy cattle by introducing a DNA sequence from naturally hornless beef breeds10. This will obviate the invasive physical dehorning methods in very broad use now. As the world population continues to grow and climate change threatens historical agricultural practices, these and other uses in food organisms can enhance the security of the world food supply. While the edited agricultural products are technically genetically modified, the genomic changes are very precise, very localized, and involve no introduction of DNA from other species. The modifications are often natural ones identified in one breed or cultivar and transferred to others with molecular technology in place of standard breeding. This approach is not only faster, but it is much cleaner in the sense that no genomic mixing occurs. Furthermore, the molecular introgression of desirable traits can be performed in any existing breed or cultivar without disrupting existing favorable characteristics or running the danger of monoculture11.

“We now can make targeted changes in essentially any gene in any organism. The list I maintain of species that have had their genomes successfully modified with the three platforms now runs to more than 80 organisms.” bitiously, many researchers have introduced human disease mutations into the genomes of model organisms in order to study their effects in a more accessible context. It is certainly easier to see how a muscular dystrophy mutation affects muscle degeneration in a mouse than in a human. In addition, potential therapies can be tested in an experimental situation before being tried in the clinic. Many applications of genome engineering to agriculture are being undertaken. Examples include making individual gene mutations that produce a healthier oil content in soy seed, generating potatoes that resist stor-

There are additional benefits of applying genome editing to pigs, in particular. One is that human disease mutations can be introduced and studied in an organisms that is much closer to humans in its anatomy and physiology. Another relates to the long sought goal of using porcine organs for transplantation. Using genome editing, it is possible to modify the pig genome to eliminate some of the barriers to transplantation, and procedures are being tested to grow largely human organs in pigs.

Medical applications A few examples of genome editing are already in clinical trials, and more are sure to follow. The approved trials involve ZFNs, which have been around long enough to undergo adequate development and pre-clinical testing. The one instance that has been published relies on ex vivo modification of T cells to eliminate the CCR5 protein, a necessary co-receptor for the most common HIV-1 strains12. Interestingly, this therapy is based on the observation that there are individuals who completely lack CCR5 due to a natural mutation. These people can become HIV-infected, but they do not develop AIDS because their T cells are protected from the virus. Additional therapies are in development in model systems, many using the CRISPR platform. The most obvious applications rely, as in the CCR5 case, on ex vivo treatment of cells from affected individuals. Delivery of the editing components – the nuclease and sometimes a template DNA – can be accomplished relatively easily with cells in the laboratory. Such approaches are applicable to some situations, including ones that might be treated via hematopoietic stem cells from the bone marrow. As other stem cell methods are developed, genome editing will be applicable to these. Getting the editing materials into cells in a target organ in vivo is much more challenging, although approaches using engineered viruses as vectors are showing promise.

Societal considerations In the flush of success with genome editing, the facility of the CRISPR platform, and the rush to take advantage of its benefits, many researchers have taken a step backward to consider where the technology may be heading13,14. As noted above, current approaches focus on somatic therapies for current patients. As demonstrated in many organisms, however, genome editing is also very effective when applied to germ cells or very early embryos. This means that we are capable of making heritable changes, not just in human disease genes, but potentially in targets that contribute to other traits. Experience in many cells and organisms, and one published study using inviable human embryos15, has shown that we do not have complete control over the genomic alterations induced by CRISPR or the other Spring 2016 Volume 15, Issue 2


targetable nucleases. In simple research studies and even in carefully monitored somatic therapies, some level of off-target effects and unpredictable on-target modifications are tolerable. The standards for efficacy and safety need to be much higher for any human germline applications, however, and the technology needs to advance significantly before these are met. It would be unethical to attempt germline modifications with the current tools. Therapeutic applications of genome editing are inherently designed to be personalized, involving the treatment of a patient’s own cells, whether ex vivo or in vivo. Each instance will be expensive. How will it be decided who receives the treatment, and how will it be paid for? One approach to making these therapies available more broadly is the development of universal donor cells that could be used for many individuals based on a single source. It would be sad, however, if only those able to pay for advanced treatments ultimately enjoy their benefits. A question perhaps not quite as obvious is: What conditions will be treated with the genome editing approaches, particularly in the germline? The elimination of some devastating genetic diseases, like Huntington’s disease or inherited immunodeficiency, seems well justified. In other cases there is a danger of reinforcing social biases. Many people with hereditary deafness or short stature, for example, do not consider themselves disabled and in need of correction. And it would be discouraging if cosmetic changes, like choice of eye color, rose to the top of the list. Clearly broad and thorough examination of this issue is warranted. Even in the cases of serious genetic diseases, there are alternatives to making intentional genomic changes, including preimplantation screening, therapeutic abortion, and adoption. This has led some people to oppose germline, or reproductive, human genome editing altogether. Currently most governments ban or disapprove of attempts to produce live births of edited children, and most scientists support this stance. As progress is made with the technology and doctors, patients and advocacy groups make their case, a more permissive climate will likely develop. My view is that human germline genome editing will happen. Our job as scientists and citizens is to ensure that it is done safely and


Harvard Health Policy Review

in conditions that reflect our social values and ethical standards. This will require both technical advances and broad discussions of complex issues involving all levels of society.

References 1. Carroll D. Genome Engineering with Targetable Nucleases. Annu Rev Biochem. 2014;83:409-39. 2. Bibikova M, Carroll D, Segal DJ, Trautman JK, Smith J, Kim Y-G, et al. Stimulation of homologous recombination through targeted cleavage by chimeric nucleases. Mol Cell Biol. 2001;21:289-97. 3. Bibikova M, Golic M, Golic KG, Carroll D. Targeted chromosomal cleavage and mutagenesis in Drosophila using zinc-finger nucleases. Genetics. 2002;161:1169-75. 4. Christian M, Cermak T, Doyle EL, Schmidt C, Zhang F, Hummel A, et al. Targeting DNA double-strand breaks with TAL effector nucleases. Genetics. 2010;186:757-61. 5. Miller JC, Tan S, Qiao G, Barlow KA, Wang J, Xia DF, et al. A TALE nuclease architecture for efficient genome editing. Nat Biotechnol. 2011;29:143-8. 6. Doudna JA, Charpentier E. Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science. 2014;346(6213):1258096. 7. Hsu PD, Lander ES, Zhang F. Development and applications of CRISPR-Cas9 for genome engineering. Cell. 2014;157(6):1262-78. 8. Clasen BM, Stoddard TJ, Luo S, Demorest ZL, Li J, Cedrone F, et al. Improving cold storage and processing traits in potato through targeted gene knockout. Plant biotechnology journal. 2015. 9. Haun W, Coffman A, Clasen BM, Demorest ZL, Lowy A, Ray E, et al. Improved soybean oil quality by targeted mutagenesis of the fatty acid desaturase 2 gene family. Plant biotechnology journal. 2014;12(7):934-40. 10. Tan WS, Carlson DF, Walton MW, Fahrenkrug SC, Hackett PB. Precision editing of large animal genomes. Adv Genetics. 2012;80:37-97. 11. Carroll D, Van Eenennaam AL, Taylor JF, Seger J, Voytas DF. Regulate genome-edited products, not genome editing itself. Nature biotechnology. 2016:in press. 12. Tebas P, Stein D, Tang WW, Frank I, Wang SQ, Lee G, et al. Gene editing of CCR5 in autologous CD4 T cells of persons infected with HIV. N Engl J Med. 2014;370(10):901-10. 13. Baltimore D, Berg P, Botchan M, Carroll D, Charo RA, Church G, et al. Biotechnology. A prudent path forward for genomic engineering and germline gene modification. Science. 2015;348(6230):36-8. 14. Lanphier E, Urnov F, Haecker SE, Werner M, Smolenski J. Don’t edit the human germ line. Nature. 2015;519(7544):410-1. 15. Liang P, Xu Y, Zhang X, Ding C, Huang R, Zhang Z, et al. CRISPR/Cas9-mediated gene editing in human tripronuclear zygotes. Protein and Cell. 2015;6:363-72.

Dana Carroll is Distinguished Professor of Biochemistry at the University of Utah School of Medicine. He pioneered the use of ZFNs as genome editing tools, contributed to research with TALENs and CRISPRs, and received several awards for his contributions, including the Novitsky Prize from the Genetics Society of America.


Neighborhood Disparities in Health: Why Do We See Them and What Can We Do? By: Mariana Arcaya, MCP, ScD, and S.V. Subramanian, M.A., M.Phil., Ph.D. Nearly 200 years ago, French physician Louis-René Villermé showed that mortality rates differed across the arrondissements of Paris1. Since that time, geographic differences in health have been observed all over the world and across myriad health outcomes2. Within countries, meaningful health differences have also been documented at multiple geographic scales. Domestically, for example, mortality rates vary substantially across counties (3), and we see staggering life expectancy differences across short distances within cities. For example, the Robert Wood Johnson Foundation’s Commission to Build a Healthier America has shown that, in New York City, babies born in one neighborhood are expected to live ten years longer than babies born just six subway stops away4. In Richmond, Virginia, neighborhoods separated by just five miles have life expectancy gaps of up to twenty years, and in Chicago, a few stops on the L can mean up to a sixteen-year difference4. Life expectancy differences by place are not just influenced by current conditions, but also historical factors. For example, trends in county-level life expectancy estimates from 1961 to 19993 suggest that counties exposed to 1930s state policies related to legal racial segregation (‘Jim Crow’ laws)5 continue to have life expectancies that are notably lower than those counties not exposed to these state mandated discriminatory laws (Figure 1). The fact that these geographic disparities

exist is not disputed. How to explain them, and more importantly, what we can do about them, are much more complex issues. Explanations for why we observe geographic health disparities generally fall into three categories: geographic clustering of people who share non-geographic health risk factors, neighborhood effects, and health selection into neighborhoods 6. The first explanation says that health outcomes vary by place because many of the same social factors that influence health, such as income, for example, also influence where people live. Secondly, neighborhood effects explanations view geographic health disparities as resulting from the causal effects places have on people’s health. Finally, health selection arguments posit that health status differentially sorts people into residential environments. Although these three explanations are conceptually distinct, evidence suggests that all are at play and may interact with each other7.

Segregation Persistent residential segregation by race/ethnicity and socioeconomic status contribute to geographic health disparities through multiple pathways. Not only are segregated neighborhoods thought to causally affect health8,9, but also they create geographic clusters of people who share nongeographic health risks, including, but not limited to, exposure to targeted marketing, discrimination, and inadequate medical care10.

Examining the makeup of the country’s metropolitan regions, Massey and Tannen found that about a third of black residents lived in “hypersegregated” areas as recently as 201011, reflecting the United State’s legacy of racist housing and urban development policy12. Economic segregation, while less severe than racial segregation, is a growing problem in the United States. Looking at the country’s major metropolitan areas over the past thirty years, the percentage of lowincome households living in predominantly low-income neighborhoods has risen from 23% to 28%, and the share of highincome households living in high income neighborhoods has doubled from 9% to 18% 13. Health inequities driven by individual socioeconomic status become geographic disparities under economic segregation and can become further compounded by interactions between individual- and neighborhood-level socioeconomic status12,14. We note that racial/ethnic and socioeconomic composition of a place is not a proxy for an individual’s race or socioeconomic status. This idea is illustrated in Figure 2. Using the historical data from the 1930 US Census (a time when racial composition was also a strong marker for systemic discriminatory policies), we show state percentage of black population (“racial context”) was differentially associated with the probability of being illiterate depending on individual race. Thus, blacks living in predominantly black states (largely the Jim Crow states) were substantially more likely to be illiterate than blacks living in states with lesser concentration of blacks5.

Spring 2016 Volume 15, Issue 2


Neighborhood effects on health Second, there is strong evidence that the natural, built, and social environment can harm or protect health. Observational evidence has linked neighborhood factors to outcomes as diverse as physical activity15, birth outcomes16, and mortality17, among others. Experimental evidence also implicates neighborhoods in shaping health. The opportunity to move from high to low poverty neighborhoods has been shown to reduce the prevalence of extreme obesity and diabetes18 in a randomized social experiment, for example. Natural experiments that change individuals’ exposures to neighborhood factors in ways that are outside their control

also show the importance of place for health. When Atlanta hosted the Olympics in 1996, the city succeeded in drastically reducing traffic congestion during the games. In responses, health care visits for acute asthma events decreased 42%, only to rise again when normal traffic patterns resumed19. In another natural experiment, Hurricane Katrina survivors involuntarily displaced to more sprawling, less walking-friendly areas showed signs of weight gain, suggesting a built environment effect on body mass index 20 .

Health as a driver of neighborhood outcomes Finally, a small but growing body of evidence frames health as a predictor of neighborhood outcomes, influencing whether and where people move. We have examined this process in a cohort of Hurricane Katrina survivors, many of whom were displaced from their homes by the disaster, showing that poor health prior to the storm predicted residence in a higher poverty neighborhood years later6. We have also used data from the national Moving to Opportunity experiment18 to examine whether differences in health influenced whether families that were offered a chance to move from high- to low-poverty neighborhoods with a housing voucher actually did so. We found that families caring for a child with health challenges were significantly less likely to participate in the experiment, and those who did participate moved to less affluent areas, on average, than families not reporting that a child in the home had a health challenge21. While more research is needed to test the role of health in shaping neighborhood outcomes, and to establish the mechanisms behind these examples of “health selection,� it could be that coping with illness reduces mental bandwidth, lowers risk tolerance, depletes financial resources, or makes people more dependent on neighborhood-based services and social ties6, thereby constraining mobility. So what can be done? The complex reality that health and residential outcomes share prior common causes and also interact with each other bi-directionally may help explain why geographic health disparities are so persistent. Tackling geographic differences in health may mean taking on many types of mechanisms that link place and health. We offer potential targets of intervention that policymakers might consider.

Policy and Research Agenda

Figure 1: Life Expectancies for Females (Top) and Males (Bottom)


Harvard Health Policy Review

Programs, policies, and regulations that help poor and minority families access a broader set of neighborhoods are key to tackling geographic health disparities as well as many other types of social inequities. Creating more affordable housing in a wider variety of neighborhoods, planning for equitable community development, and enforcing fair housing law are just some ways that housing and neighborhood choices


neighborhoods, and neighborhoods causally impact health. Combating geographic health disparities means addressing pressing urban planning and public policy challenges related to residential segregation, social and health services, community development and design, environmental protection, and policymaking in non-health sectors.


Foreign-born White

Native White

Figure 2: Literacy in States could be expanded22. More equitable access to high opportunity neighborhoods9 would not only address some of the compositional drivers of geographic health disparities but could also intervene on neighborhood characteristics that are thought to have causal effects on health, including concentrated poverty and segregation. Even more fundamentally, policies that address income inequality and those that improve conditions in poor neighborhoods could also help narrow gaps in access to neighborhoods that support health. Measures to reduce the impact of illness on financial resources, employment, and care giving burden, such as paid sick leave policies, improved access to affordable health insurance, and social services and transportation assistance for sick and disabled individuals, may also intervene on geographic health disparities. Mitigating costs and stressors associated with poor health could reduce the extent to which health problems differentially sort people into poorer neighborhoods. Finally, creating healthier environments is a critical component not only of eliminating geographic health disparities but also of raising overall levels of health. When we have good evidence for what makes places

healthier, we should use “health in all policies” approaches23 to integrate these considerations into how non-health sectors, such as transportation, housing, education, and others, do business. Health Impact Assessment methodology, which analyzes how decisions made in non-health sectors could affect health and makes evidencebased recommendations to maximize health benefits while minimizing risks24, could be more widely applied in order to promote healthier environments. Where we do not have good evidence, we should expand our research efforts to better understand neighborhood effects on health. Many useful research agendas have been put forward to this end 25-27, stressing the need to consider relevant spatial and temporal scales, implement stronger research designs, including the testing and evaluation of place-based interventions, and develop richer conceptualizations and measures of residential environment. In summary, geographic health disparities appear to stem from multiple and overlapping causes. Health and place interact dynamically with each other and with other related social factors such that people facing higher health risks tend to live near each other, health may sort people differentially into

1. Julia C, Valleron A-J. Louis-René Villermé (1782–1863), a pioneer in social epidemiology: re-analysis of his data on comparative mortality in Paris in the early 19th century. J Epidemiol Community Health. 2011 Aug 1;65(8):666–70. 2. GBD 2013 Risk Factors Collaborators, Forouzanfar MH, Alexander L, Anderson HR, Bachman VF, Biryukov S, et al. Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet Lond Engl. 2015 Dec 5;386(10010):2287–323. 3. Ezzati M, Friedman AB, Kulkarni SC, Murray The reversal of fortunes: trends in county CJL. mortality and cross-county mortality disparities in the United States. PLoS Med. 2008 Apr 22;5(4):e66. 4. The Robert Wood Johnson Foundation. Maps to #CloseHealthGaps [Internet]. RWJF. 2015 [cited 2016 Feb 25]. Available from: http:// 5. Subramanian SV, Jones K, Kaddour A, Krieger N. Revisiting Robinson: the perils of individualistic and ecologic fallacy. Int J Epidemiol. 2009 Apr;38(2):342–60; author reply 370–3. 6. Arcaya MC, Subramanian SV, Rhodes JE, Waters MC. Role of health in predicting moves to poor neighborhoods among Hurricane Katrina survivors. Proc Natl Acad Sci. 2014 Oct 20;111(46):16246–53. 7. Auchincloss AH, Roux AVD. A New Tool for Epidemiology: The Usefulness of Dynamic-Agent Models in Understanding Place Effects on Health. Am J Epidemiol. 2008 Jul 1;168(1):1–8. 8. Acevedo-Garcia D, Lochner KA, Osypuk TL, Subramanian SV. Future Directions in Residential Segregation and Health Research: A Multilevel Approach. Am J Public Health. 2003 Feb;93(2):215–21. 9. Acevedo-Garcia D, Osypuk TL, McArdle N, Williams DR. Toward A Policy-Relevant Analysis Of Geographic And Racial/Ethnic Disparities In Child Health. Health Aff (Millwood). 2008 Mar 1;27(2):321–33. 10. Krieger N. Discrimination and Health Inequities. Int J Health Serv. 2014 Oct 1;44(4):643– 710. 11. Massey DS, Tannen J. A Research Note on Trends in Black Hypersegregation. Demography. 2015 Jun;52(3):1025–34. 12. Wilson WJ. More than Just Race: Being Black and Poor in the Inner City (Issues of Our

Spring 2016 Volume 15, Issue 2


Time). W. W. Norton & Company; 2010. 194 p. 13. Taylor P, Fry R. The rise of residential segregation by income. Spec Rep Pew Soc Demogr Trends August. 2012; 14. Berkman LF, Kawachi I. Social epidemiology. Oxford University Press, USA; 2000. 15. Ding D, Sallis JF, Kerr J, Lee S, Rosenberg DE. Neighborhood environment and physical activity among youth a review. Am J Prev Med. 2011 Oct;41(4):442–55. 16. Vos AA, Posthumus AG, Bonsel GJ, Steegers EAP, Denktas S. Deprived neighborhoods and adverse perinatal outcome: a systematic review and meta-analysis. Acta Obstet Gynecol Scand. 2014 Aug;93(8):727–40. 17. Meijer M, Röhl J, Bloomfield K, Grittner U. Do neighborhoods affect individual mortality? A systematic review and meta-analysis of multilevel studies. Soc Sci Med 1982. 2012 Apr;74(8):1204–12. 18. Ludwig J, Sanbonmatsu L, Gennetian L, Adam E, Duncan GJ, Katz LF, et al. Neighborhoods, Obesity, and Diabetes — A Randomized Social Experiment. N Engl J Med. 2011 Oct 20;365(16):1509–19. 19. Friedman MS, Powell KE, Hutwagner L, Graham LM, Teague WG. Impact of changes in transportation and commuting behaviors during the 1996 Summer Olympic Games in Atlanta on air quality and childhood asthma. JAMA. 2001 Feb 21;285(7):897–905. 20. Arcaya M, James P, Rhodes JE, Waters MC, Subramanian SV. Urban sprawl and body mass index among displaced Hurricane Katrina survivors. Prev Med. 2014;65:40–6. 21. Arcaya MC, Graif C, Waters MC, Subramanian SV. Health Selection into Neighborhoods Among Families in the Moving to Opportunity Program. Am J Epidemiol. 2015 Dec 10;kwv189. 22. Briggs X de S, editor. The geography of opportunity: race and housing choice in metropolitan America. Brookings Institution Press; 2005. 374 p. 23. Corburn J, Curl S, Arredondo G. A HealthIn-All-Policies Approach Addresses Many Of Richmond, California’s Place-Based Hazards, Stressors. Health Aff (Millwood). 2014 Nov 1;33(11):1905–13. 24. Harris-Roxas B, Viliani F, Bond A, Cave B, Divall M, Furu P, et al. Health impact assessment: the state of the art. Impact Assess Proj Apprais. 2012 Mar;30(1):43–52. 25. Oakes JM, Andrade KN. Methodologic Innovations and Advances in Social Epidemiology. Curr Epidemiol Rep. 2014 Jan 17;1(1):38–44. 26. Diez Roux A-V. Neighborhoods and health: where are we and were do we go from here? Rev Dépidémiologie Santé Publique. 2007 Feb;55(1):13–21. 27. Subramanian SV, Glymour MM, Kawachi I. Identifying Causal Ecologic Effects on Health: A Methodological Assessment. In: Macrosocial Determinants of Population Health [Internet]. Springer New York; 2007 [cited 2014 May 29]. p. 301–31. Available from: http://link.springer. com/chapter/10.1007/978-0-387-70812-6_15


Harvard Health Policy Review

S.V. Subramanian is a Professor of Population Health and Geography at the Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health (HSPH), and a senior core faculty at the Harvard Center for Population and Development Studies (http://www. He is also the Founding Director of the Interdisciplinary PhD program in Population Health Sciences at Harvard. He received his under- and post-graduate training at the University of Delhi, and completed his PhD in geography from the University of Portsmouth, UK in 2000. Subu has published around 475 articles, book chapters, and books in the field of social and contextual determinants of health, health inequalities in India, and applied multilevel statistical models. His current research interests include exploring the concept of variation for population health, the reciprocal association between neighborhoods and health; and understanding the causes and consequences of undernutrition among children in disadvantaged settings. As ane educator, Subu was the first to develop a course on the concept and application of multilevel statistical methods at Harvard, which he has been successfully teaching at HSPH since 2001, as well as around the world. He has advised over 100 masters, doctoral and postdoctoral students as mentor, academic advisor and dissertation committee member. Subu is the CoEditor-in-Chief for the international journal Social Science & Medicine (SSM), in addition to be being a Co-Senior Editor for the social epidemiology office of SSM. He is also the founding Co-Editor-in-Chief of a new journal SSM – Population Health. He is an editorial consultant to The Lancet, an international advisory board member for the Lancet Global Health.

Mariana Arcaya is an assistant professor of urban planning and public health at the Massachusetts Institute of Technology. She is a social epidemiologist and urban planner whose work explores dynamic relationships between geographic contexts, particularly neighborhoods, and health. Mariana conducts scholarly and policy-relevant research in two main areas: 1) bi-directional relationships between place and health, including how health considerations shape socioeconomic outcomes for individuals and communities, and, 2) applied and translational research on the social determinants of health, particularly health risk factors shaped by urban policy and planning decisions. Prior to coming to MIT in 2015, Mariana was a post-doctoral fellow at the Harvard Center for Population and Development Studies. She holds a Doctorate of Science from the Harvard School of Public Health, and a Master of City Planning from MIT’s Department of Urban Studies & Planning. Her professional work experience includes instituting and managing a Public Health Division within Metropolitan Boston’s regional planning agency, as well as designing and overseeing the implementation of healthy urban planning strategies under a federally funded Community Transformation Grant. Acknowledgements: The authors would like to acknowledge Justin Steil for his thoughtful comments on a draft of the manuscript.


IMPACT: the Role of Peer to Peer education and the Life-course Perspective in Infant Mortality twice as high as their Caucasian counterparts. To investigate the reason for this racial disparity, IMPACT members traveled to Madison, Wisconsin to meet with healthcare specialists at the University of WisconsinMadison to understand how they addressed a similar divide in their state.

The article by the California based research group also mentions that preterm birth is one of the leading causes of death among African American infants in California. So too, in Wisconsin, researchers look to premature birth. Thus they began exploring the factors that could affect the physical and mental health of the mother before and during pregnancy.

Part of the larger Partnership Program at the University of Wisconsin’s School of Medicine and Public health, the Life-course Initiative for Healthy Families is a program that aims to eliminate racial disparities in birth outcomes in Wisconsin. The Lifecourse perspective is similar into the statesponsored Black Infant Health program mentioned in the California-article in that it

As the California-based researcher group noted, there is a significant disparity between rates of pre-term birth, and downstream effects like infant mortality among women living in poverty. This is very often because of violent-stressful home and community environments, lack of prenatal care as well as environmental health hazards such as toxins and pollutants. For this reason the Life-

By: Anjali Chandra While inequity in healthcare and health outcomes between states, and rural and urban communities within states is appalling, there are a few organizations advancing models for catalytic change in these issue areas. One such organization is the Infant Mortality Prevention and Awareness Campaign of Tennessee (IMPACT). IMPACT was founded by the Governor of Tennessee in 2002 to address Tennessee’s exceedingly high infant mortality rate, and socio-economic disparities associated with it. In 2006, Girls Incorporated, an organization which provides in-school and enrichment programming for girls ages 6-18 took charge of the program. IMPACT is now funded by Blue Cross Blue Shield of Tennessee. IMPACT is a peer to peer media and outreach campaign dedicated to raising awareness about infant mortality and its causes in the Hamilton County Area. Infant mortality is defined as the number of babies who die before their first birthday. Hamilton County has the second highest infant mortality rate in Tennessee, which at 8.84% ranks behind developing countries such as Cuba and Chile. Nationally, Tennessee ranks 5th highest among all states for its rate of Infant mortality incidence. However, there is an even deeper issue that underlies these statistics. Just as African-American babies in California are more likely to be pre-term, black infants in Hamilton county die at a rate

and environment, something called the lifecourse perspective.

IMPACT is a peer to peer media and outreach campaign dedicated to raising awareness about infant mortality and its causes in the Hamilton County Area. Infant mortality is defined as the number of babies who die before their first birthday. acknowledges and seeks to address the nonimmediate factors which could affect the viability of the fetus and ultimately the health of the infant that results from a pregnancy by examining mother’s stress, nutrition,

course initiative sponsors community based programs aimed at reducing the negative environmental and socio-economic factors which a mother encounters before and during pregnancy. Spring 2016 Volume 15, Issue 2


The holistic approach employed by the Black Infant Health initiative mentioned in the article on “California Summit on Preterm birth” and the Life-course Initiative in Wisconsin illustrate a need for communityoriented, culturally-sensitive care, that is both preventative and reparative. While programs such as the Life-course Initiative need to be expanded to create a paradigm shift in U.S. public health, it is promising that across the country researchers are recognizing the need for a life-course perspective in order to address pre-term birth and Infant mortality. The Life-course model considers the racial background, socioeconomic status, disease exposure, nutritional intake, healthcare and food access, and general mental and physical health of a woman, in determining her risk for infant mortality. What the program discovered and has since incorporated into their multi-faceted approach is that racism and prejudice can increase the risk for infant mortality, posing a possible explanation for the elevated infant mortality rate amongst African Americans. Furthermore, as is seen in Hamilton county where IMPACT operates, most low income areas, including public housing projects, are primary inhabited by African Americans. These areas, often known as food deserts, generally only have a gas-station or a convenience store within their locality. When transportation is limited, and food options are few, many women have to resort to the nutrient deficient, processed food available within walking distance. Recognizing community specific needs has allowed the Life-course initiative to develop three initiatives designed for the communities of southern Wisconsin: a Prenatal care and education program, “Sister Friend”, an initiative that provides community-based support and transportation to medical appointments for the mother, and Young Dads- which seeks to strengthen father involvement. Each of these programs are community-based and as such they vary from city-to-city in the regions where the Life-course Initiative has been implemented. Through this experience, members of IMPACT came to realize the circumstances of at-risk populations is key to developing a solution that is tailored to fissures causing the high infant mortality rates. The same solution cannot be applied to every community because the dynamic of that community is different. With this in mind,


Harvard Health Policy Review

they began developing a comprehensive action-plan targeted to reducing infant mortality in Hamilton County that was community specific. As a part of IMPACT at the time, I was particularly passionate about addressing the disparate health outcomes of infants across racial lines. So, I spearheaded a media campaign, which entailed having billboards in ethnically-dense areas, and urban centers with high levels of African-American populations, as well as less racially-specific areas that presented bold statistics like “African American babies die at twice the rate of Caucasian babies” with graphics of black and white pregnant women facing each other. I hoped that this somewhat, subversive campaign would help the community realize what a profound discordance we have in the health of our African American and white communities. IMPACT, by inspiring girls to be peereducators, and moreover strategists for community health, furthers Girls Inc’s mission to inspire girls to be strong smart and bold. Many of IMPACT’s members come from at-risk communities, and have had several family members whose babies have died before their first birthday. As such, they are extremely passionate about this cause, and have insights into the underlying community dynamics which could be fueling the disparity. It is my hope that IMPACT members are integrating their experience in Madison, with their personal and community experiences- to build community health from the ground up.

Anjali Chandra is a freshman at Harvard College interested in psychology and global health. She is the Founder and Chairwoman of Global Excel TN, a non-profit dedicated to providing academic enrichment opportunities to underprivileged youth. During her last two years of high school, she was part of IMPACT, the Infant Mortality Prevention and Awareness Campaign of Tennessee. Through her time there, she found her passion for addressing racial and socio-economic barriers to health, and hopes to bring an integrated socio-cultural perspective to patient care.


California Summit on Preterm Birth By: Leslie Kowalewski; Michael Curtis, Ph.D.; Connie Mitchell, M.D., M.P.H.; Larry Rand, M.D.; Scott D. Berns, M.D., M.P.H., F.A.A.P.; Gary M. Shaw, DrPH; and David K. Stevenson, M.D. Preterm birth, associated with substantial individual and societal burdens, is a leading cause of death worldwide in children before five years of age. In 2013, California experienced a preterm birth rate of 8.4%, remarkable for a state with nearly 500,000 births each year and a socioeconomically and culturally diverse population. In March 2015, the California Prematurity Summit brought together forty-four key leaders from across the state, including providers, researchers, public health officials, hospital administrators and community representatives, to identify best practices contributing to California’s low preterm birth rate and to discuss other preventive measures that might be implemented to further reduce the preterm birth rate to 5.5% by 2030 or sooner. Some countries, including Japan, Norway, and Finland, have already achieved that target rate. This paper presents the key findings and recommendations based on the collaborative work of Summit participants.

Introduction In 2013, California experienced a preterm birth rate of 8.4% which is remarkable for a state with about half a million births each year and a socioeconomically and culturally diverse population. The burden of preterm birth differs greatly across the U.S. In 2013, the overall rate of preterm birth in the US was 9.6%, with the highest rates observed in Mississippi (13.1%) and Louisiana (12.5%), and the lowest rates observed in Oregon (7.9%) and Vermont (7.6%). The latter two states are much smaller and less diverse than California. Both the size and diversity of California make it a demonstration project for success in implementing large-scale, population-based changes in healthcare to address a major public health problem. Preterm birth is associated with

substantial individual and societal burdens. Preterm birth is a leading cause of death worldwide in children before 5 years of age.1 Among survivors, preterm birth is associated with substantially increased risks for developmental delay and disability, particularly for babies born before 32 completed weeks gestation.2-10 In addition to contributing to substantial psychological and financial burdens at the individual and familial levels, preterm birth extracts a heavy toll on communities and societies with respect to short- and long-term healthcare costs as well as the lost potential of those born too early. In 2006, the cost of preterm birth has been estimated at more than $26 billion annually in the U.S. alone.11 On March 18, 2015, the California Prematurity Summit brought together fortyfour leaders from across the state, including

providers, researchers, public health officials, hospital administrators and community representatives, to identify best practices contributing to California’s low preterm birth rate and to discuss other preventive measures that might be implemented to further reduce the preterm birth rate to 5.5%12 by 2030 or sooner. Some countries, including Japan, Norway, and Finland, have already achieved that target rate (March of Dimes Report Born too Soon). This paper presents the key findings and recommendations based on the collaborative work of Summit participants.

Measuring Preterm Birthrate in California One of the considerable challenges in monitoring and addressing rates of preterm birth is the metric by which gestational age is measured. Beginning with the 2014 data year, the Centers for Disease Control and Prevention’s (CDC) National Center for Health Statistics (NCHS) transitioned from using date of the last menstrual period (LMP) to obstetric estimate (OE) of gestation at delivery as the new standard for estimating gestational age. This change was made because evidence increasingly pointed to the greater validity of the OE-based compared with the LMP-based measure. Births were less likely to be classified as preterm using the OE than with the LMP. National and California data based on the obstetric estimate are available beginning with the 2007 data year.13 The data presented in this report on California and counties of focus for the California Summit on Preterm Birth (Fresno, Kern, Los Angeles, San Bernardino, San Diego, and Santa Clara) are based on the OE. Spring 2016 Volume 15, Issue 2 33

From 2007 to 2013, the preterm birth rate for California decreased 0.7% from 9.1% in 2007 to 8.4% in 2013 (Figure 1). In counties with more than 1,000 births, the highest rates were observed in Fresno (9.8%) and Kern (9.4%), and the lowest rates observed in Humboldt (6.8%) and Placer (6.5%). Since rates of preterm birth often differ greatly within counties as well, a greater focus on within-county patterns of preterm birth may be more informative. For example, when rates of preterm birth among singletons were examined from 2011 to 2013 within Fresno County by Medical Service Study Area (MSSA), preterm birth rates ranged from 6.0% to 9.9% [MSSAs are sub-city and sub-county clusters of census tracts used for analyses and resource allocation purposes by the California Office of Statewide Health Planning and Development (OSHPD) and are recognized by the U.S. Health Resources and Services Administration, Bureau of Health Professions’ Office of Shortage Designation as rational service areas for purposes of designating Health Professional Shortage Areas (HPSAs) and Medically Underserved Areas and Medically Underserved Populations (MUAs/MUPs)]. In 2013, nearly one in two (46.8%) births in California occurred in women receiving Medi-Cal assistance, California’s Medicaid

Program. This proportion was also observed for preterm birth. In fact, preterm birth among pregnancies receiving MediCal assistance for delivery or prenatal care was 8.6% compared to 8.3% for those not receiving Medi-Cal. The majority of preterm births occur in singleton pregnancies (78.6% in 2013) in California. However, owing to the high risk of preterm birth among pregnancies that are twins or greater, such pregnancies represent a disproportionate contribution to all preterm births and therefore may be especially relevant to efforts aimed at understanding and reducing preterm birth overall. Preterm birth rates have dropped by a larger percentage among multiple births for whom rates were 61.2% in 2007 and 55.5% in 2013 (a drop of 5.7%). During this same time, the drop among singletons was 0.7%. Race/ethnicity has been shown to be a potent risk factor for preterm birth, especially among African-Americans.14-16 Within California, patterns of preterm birth differ greatly by race/ethnicity and with respect to the burden of preterm birth. Statewide, the rate of preterm birth among AfricanAmerican women with a singleton pregnancy is 9.9%, whereas among Hispanic, Asian, or White women is 7.2%, 6.6%, and 5.7%, respectively (Figure 2). The contribution of African-American women to all preterm

births is 7.5%; whereas, the contribution from Hispanic, Asian, or White women is 50.9%, 13.2%, and 23.0%, respectively. Women of Pacific Islander race/ethnicity have a slightly increased risk of preterm birth with a rate of 8.3% and a total contribution to all preterm births across the state of 0.5%. Among women of Hispanic race/ethnicity, women born outside of Mexico had the highest rate of preterm birth in 2013 (7.7%), followed by those born in the U.S. (7.3%) and Mexico (6.8%). In addition, preterm birth is the leading cause of infant death, responsible for approximately 1 in 3 infant deaths and nearly half of all African-American infant deaths (Figure 4). Rates of preterm birth are highest among women living in areas of concentrated poverty. In California neighborhoods where more than 20% of the population had incomes below the federal poverty guidelines, women with singleton pregnancies had a rate of preterm birth of 7.6% compared to 6.0% among women who lived in neighborhoods where less than 5% of the population was in poverty (Figure 3). Often areas of concentrated poverty coexist with environmental factors known to contribute to increased risk of preterm birth, such as traffic-related air pollution.17 Possible Contributing Factors California’s Low Preterm Birthrate


Summit participants recognized that each of the following factors might be important and could have had an impact on California’s preterm birth rate: improved pregnancy dating; better identification of women at risk for preterm birth and early interventions (such as 17-hydroxyprogesterone); increased standards for fewer embryo transfers in in vitro fertilization technologies; reduced elective deliveries at 37 to 38 weeks, resulting in a spillover effect in the reduction of late preterm births at 34 to 36 weeks; expanded state and community programs, including home visiting and teen pregnancy reduction efforts; greater access to prenatal care through improving access to Medi-Cal; strict smoking regulations that increase and support cessation programs; and deeper consumer engagement and messaging. Figure 1. Rate of all preterm births by year in California, 2007 to 2013 Gestational age is based on the obstetric estimate. Includes gestational age range 17 to 47 weeks; preterm delivery <37 weeks. Data Source: California Birth Statistical Master Files, 2007to 2013. Prepared by the Epidemiology, Assessment and Program Development Branch, Maternal, Child and Adolescent Program, Center for Family Health


Harvard Health Policy Review

Statewide public health efforts may also have contributed, such as Title V funded local fetal infant mortality reviews and interventions, the strengthening of prenatal care through provider incentives

Figure 2. Preterm Birth Rate and Percent Preterm Births among Singleton Pregnancies by Race/Ethnicity, 2013 in California Gestational age is based on the obstetric estimate. Includes gestational age range 17 to 47 weeks; preterm delivery <37 weeks. Source: California Department of Public Health, 2007 to 2013 Birth Statistical Master Files;Prepared by the Epidemiology, Assessment and Program Development Branch, Maternal, Child and Adolescent Program, Center for Family Health as part of the California Perinatal Services Program and regionalized efforts to support hospital quality improvement through the Regionalized Perinatal Programs of

quality of maternal and infant healthcare: the California Perinatal Quality Care Collaborative (CPQCC) and the California Maternal Quality Care Collaborative (CMQCC). Both

“Rates of preterm birth are highest amoung women living in areas of concentrated poverty. In California neighborhoods where more than 20% of the population had incomes below the federal poverty guidelines, women with singleton pregnancies had a rate of preterm birth of 7.6% compared to 6.0% among women who lived in neighborhoods where less than 5% of the population was in poverty” California. California has also benefitted from two collaboratives focused on improving the

collaboratives have convened experts to develop quality improvement toolkits and supported hospital implementation with

demonstrated impact on preterm birth, infant and maternal health outcomes. 18-24 The CPQCC ( is an outgrowth of a 1997 initiative proposed by the California Association of Neonatologists and is supported by funding from the State and membership fees (135 NICUs). Over the last ten years, CPQCC has developed a network of stakeholders consisting of public and private obstetric and neonatal providers, healthcare purchasers, public health professionals, and private sector health industry specialists. The CMQCC ( emerged from CPQCC, with a mission to end preventable morbidity, mortality, and racial disparities in California maternity care. To achieve this, CMQCC 1) gathers, reviews, and organizes birth data and statistics, transforming them into actionable information; 2) creates and facilitates channels of communications and collaboration among all maternity stakeholders; 3) establishes quality and safety as the clear priority of every decision and every action; and 4) defines, disseminates, and implements clinical best practices and quality improvement principles and techniques. Although the discovery work to prevent preterm birth continues at the major institutions throughout the State and, in particular, at the March of Dimes Prematurity Research Center at Stanford University and the Preterm Birth Initiative at the University of California, San Francisco, the Summit unanimously agreed that public health interventions could proceed ahead of our full understanding of how the sociodemographic and environmental factors contribute to the changes in biology that predispose women to preterm birth or initiate the onset of preterm labor.

How to achieve further reductions in preterm births in California Even though California’s preterm birth rate is lower than 46 States in the US despite the large demographic and heterogeneous composition of its populace, there remain opportunities to drive the rate in California lower, such that all counties, communities, and ethnic groups experience equal birth outcomes. Participants identified 12 strategic implementation objectives at the Summit: 1. Maximize every encounter with pregnant women – Incorporate preterm birth prevention strategies and messaging, such as the importance of birth spacing Spring 2016 Volume 15, Issue 2


(e.g. optimizing at 18-23 months), into state and county social service programs, existing collaborative partnerships, and public health initiatives such as Women, Infants and Children’s Supplemental Nutrition (WIC) and Maternal, Child and Adolescent Health programs. Promote well-woman and welladolescent health care visits and monitor delivery of clinical preventive health services outlined in the Affordable Care Act, including reproductive life planning, when visits occur. 2. Apply the “Health in All Policies” approach – The American Public Health Association has developed a Health in All Policies toolkit that outlines how health, equity, and sustainability considerations can be incorporated into decision-making across all sectors and policy areas. Health in All Policies is a collaborative approach to improve the health of all communities and people.25 In the communities where mothers reside, address the social determinants of health, such as economic opportunities, social cohesion, safety, and structural racism. 3. Develop authentic community engagement partnerships – Listen and empower the community to drive their own local solutions. Transition to a community-

based participatory model, an applied collaborative approach that engages the community in identifying local stressors and preterm birth reduction action plans that lead to sustainable results. Community insiders and researchers should partner to combine knowledge and action to improve community health and reduce health disparities.26 Best Babies Zone (a national project funded by the W.K. Kellogg foundation and led by UC Berkeley, with the aim to change children’s trajectories by working with non-traditional partners to transform a historically disinvested neighborhood into one of health and economic vibrancy) is an example of one effort to foster community change that supports perinatal health. 4. Promote health equity – Because the African-American singleton preterm birth rate in California is 9.9% compared to the average rate of 6.8%, a task force should be established and charged to support and empower the African-American communities throughout the State to take action to reduce prematurity. Recruit and hire healthcare providers to mirror the patient population they serve. Tie physician incentives to patient satisfaction and provider cultural humility/sensitivity ratings. To focus on reducing disparities, utilize social support

Figure 3. Preterm Birth Rate and Percent Preterm Births among Singleton Pregnancies by Neighborhood Poverty Level, 2013 Gestational age is based on the obstetric estimate. Includes gestational age range 17 to 47 weeks; preterm delivery <37 weeks. Source: California Department of Public Health, 2007 to 2013 Birth Statistical Master Files;Prepared by the Epidemiology, Assessment and Program Development Branch, Maternal, Child and Adolescent Program, Center for Family Health


Harvard Health Policy Review

models, such as peer counseling, Centering Pregnancy® (a multifaceted model of group care that integrates the three major components of care: health assessment, education, and support to improve maternal and child health), and the Promotora model (a program created to address the lack of access to reproductive healthcare and sex education information in the Latino community). 5. Concentrate on what we do know and what we can do – Because we do not know why preterm labor occurs in approximately 50% of cases, it is difficult to prevent preterm birth. Develop and disseminate preterm birth prevention toolkits to educate all professionals (both health and social services) about implementing the known early risk assessment and preterm birth prevention strategies – such as reducing early elective deliveries, smoking cessation, single embryo transfer for infertility, cervical length measurement, early ultrasound, administration of progesterone for women with a short cervix, and 17-alpha hydroxyprogesterone and cerclage for women with previous spontaneous PTB. In addition, strive for the uniform implementation of evidence-based practices such as optimized use of low-dose aspirin to reduce preeclampsia, optimizing interpregnancy interval, and eliminating teenage pregnancy.27 These interventions should be part of social support models including Centering Pregnancy and Promotora provider training. 6. Increase preconception, interconception care attendance and interpregnancy interval spacing – Through policy change, establish billable codes for preconception, interconception, and postpartum visits to align incentives for providers. These visits are currently bundled into one global payment for all maternity care, eliminating any monetary incentives for the provider. Research has shown that 18-23 months is the optimal interpregnancy interval that can help reduce preterm birth. Clinicians can enhance family planning and birth spacing by promoting contraception such as long-acting reversible contraception (LARC) and by improving attendance of the postpartum and interconception care visits between pregnancies. 7. Increase public awareness about the critical issue of preterm birth – Meet patients where they are to educate them about prematurity, through support

– Explore varying preterm birth rates among populations, geographic regions, neighborhoods, race and ethnic groups, and gestational age. Encourage research to understand the role of stress, obesity, nutrition, environment, substance abuse, hypertension, diabetes, and poverty play in the pathophysiologic processes of preterm birth. Further research is needed to understand the mechanism of preterm birth, identify any potential biomarkers of risk, and test possible therapeutic interventions leading to prevention.

Figure 4. Percent and Number of Infant Deaths in California from Preterm-Related Causes by Race/Ethnicity, 2011 Gestational age is based on the obstetric estimate. Includes gestational age range 17 to 47 weeks; preterm delivery <37 weeks. Source: California Department of Public Health, 2007 to 2013 Birth Statistical Master Files;Prepared by the Epidemiology, Assessment and Program Development Branch, Maternal, Child and Adolescent Program, Center for Family Health

networks, trusted circles of influence, faith based organizations, community groups, and their healthcare provider. Creatively increase awareness about the serious and costly issue of prematurity among women, communities, and providers, utilizing social media (e.g.,Text4baby, a mobile information service designed to promote maternal and child health through text messaging). 8. Reduce obesity and related chronic disease –Overweight or obesity are observed in more than 40% of women, and obesity contributes to preterm birth risk as well as chronic illness. Improved management during the preconception and postpartum follow-up visit is a critical intervention to reduce preterm birth and maternal and child chronic illness. 9. Utilize incentives and disincentives to improve birth outcomes – Align reimbursement around evidence-based best practices, providing incentives to providers, who follow best practices and incentives to patients for improving health outcomes. Incentivize hospitals for meeting specific performance measures, such as early term elective deliveries, vaginal birth after

Cesarean section, and providing counseling for future pregnancies. 10. Ensure quality and fidelity of existing programs –Continue to measure and validate which methods of care improve health outcomes and reduce preterm birth, such as WIC, which provides federal grants to States for supplemental foods, healthcare referrals, and nutrition education for lowincome pregnant, breastfeeding, and nonbreastfeeding postpartum women, and to infants and children up to age five, who are found to be at nutritional risk; the California Home Visiting Program, a positive parenting program created as a result of the Patient Protection and Affordable Care Act of 2010 to help vulnerable families independently raise their children; and Black Infant Health, a state program which uses a group-based approach with complementary clientcentered case management to help women develop life skills, learn strategies for reducing stress, and build social support with the aim of reducing racial disparities in birth outcomes. 11. Conduct research to understand variables contributing to preterm birth

12. Develop a shared healthcare quality improvement data dashboard – Create a shared dashboard to track progress in reducing prematurity across the State. The envisioned system could track preterm birth rates, deliveries at appropriate levels of care, indications for preterm birth, use of antenatal corticosteroids, 17-alpha-hydroxyprogesterone utilization, gestational age data stratified by weeks of gestation, preterm birth rates by race, and the association with numerous social and community indicators of health. This dashboard could be used at the hospital, agency, neighborhood, county and state level, encouraging collaboration across agencies.

Conclusion California is an entrepreneurial state, known for its risk-taking in business, but also for its contributions to discovery science and multidisciplinary collaboration statewide. Public and private institutions are working together and across various healthcare systems, breaking through traditional boundaries to forge new partnerships to improve health outcomes for mothers and babies. The state’s declining preterm birth rate is a reflection of these efforts. Although it is likely that researchers throughout California will make important contributions to understanding the causes and prevention of preterm birth, health leaders throughout the state are already working together from a practical standpoint to implement a variety of interventions that impact most of the current factors that have been identified as contributing to the risk for preterm birth. In the end, we may identify the biological effects of each of these and understand better how social and psychological factors translate to causal biology, but in the meantime, much progress can be made in public health and in the delivery of high quality healthcare with respect to preterm birth. Spring 2016 Volume 15, Issue 2


References 1. Liu L, Johnson HL, Cousens S et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet 2012;379:21512161. 2. Vohr BR, Wright LL, Poole WK, McDonald SA. Neurodevelopmental outcomes of extremely low birth weight infants <32 weeks’ gestation between 1993 and 1998. Pediatrics 2005;116:635-643. 3. Tomashek KM, Shapiro-Mendoza CK, Davidoff MJ, Petrini JR. Differences in mortality between late-preterm and term singleton infants in the United States, 1995-2002. J Pediatr 2007;151:450-6, 456. 4. Swamy GK, Ostbye T, Skjaerven R. Association of preterm birth with long-term survival, reproduction, and next generation preterm birth. JAMA 2008;299:1429-1436. 5. Melamed N, Klinger G, Tenenbaum-Gavish K et al. Short-term neonatal outcome in low-risk, spontaneous, singleton, late preterm deliveries. Obstet Gynecol 2009;114:253-260. 6. Doyle LW, Roberts G, Anderson PJ. Outcomes at age 2 years of infants < 28 weeks’ gestational age born in Victoria in 2005. J Pediatr 2010;156:49-53. 7. Stoll BJ, Hansen NI, Bell EF et al. Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics 2010;126:443-456. 8. Crump C, Sundquist K, Sundquist J, Winkleby MA. Gestational age at birth and mortality in young adulthood. JAMA 2011;306:1233-1240. 9. Leone A, Ersfeld P, Adams M, Schiffer PM, Bucher HU, Arlettaz R. Neonatal morbidity in singleton late preterm infants compared with full-term infants. Acta Paediatr 2012;101:e6-10. 10. Parkinson JR, Hyde MJ, Gale C, Santhakumaran S, Modi N. Preterm birth and the metabolic syndrome in adult life: a systematic review and meta-analysis. Pediatrics 2013;131:e1240-e1263. 11. Institute of Medicine. Preterm Birth: Causes, consequences, and prevention. 2007. Washington DC, National Academy Press. 12. McCabe ER, Carrino GE, Russell RB, Howse JL. Fighting for the next generation: US Prematurity in 2030. Pediatrics 2014;134:1193-9. 13. Hamilton BE, Martin JA, Osterman MJK, Curtin SC. Births: Preliminary data for 2014. National vital statistics reports; vol 64 no 6. Hyattsville, MD: National Center for Health Statistics. 2015. 14. Martin JA, Osterman MJK, Kirmeyer SE, Gregory ECW. Measuring gestational age in vital statistics data: Transitioning to the obstetric estimate. National vital statistics reports; vol 64 no 5. Hyattsville, MD: National Center for Health Statistics. 2015.Rosenberg TJ, Garbers S, Lipkind H, Chiasson MA. Maternal obesity and diabetes as risk factors for adverse pregnancy outcomes: differences among 4 racial/ethnic groups. Am J Public Health 2005;95:1545-1551. 15. Christian LM, Glaser R, Porter K, Iams JD. Stress-induced inflammatory responses in women: effects of race and pregnancy. Psychosom Med 2013;75:658-669. 16. Braveman PA, Heck K, Egerter S et al. The Role of Socioeconomic Factors in Black-White


Harvard Health Policy Review

Disparities in Preterm Birth. Am J Public Health 2014;e1-e9. 17. Padula AM, Mortimer KM, Tager IB, et al. Traffic-related air pollution and risk of preterm birth in the San Joaquin Valley of California. Annals of epidemiology 2014;24:888-95e4. 18. Oshiro B, Branch W, Main E. Neonatal outcomes after implementation of guidelines limiting elective delivery before 39 weeks of gestation. Obstetrics and gynecology 2012;119:656; author reply 7. 19. Finer NN, Powers RJ, Ou CH, Durand D, Wirtschafter D, Gould JB. Prospective evaluation of postnatal steroid administration: a 1-year experience from the California Perinatal Quality Care Collaborative. Pediatrics 2006;117:704-13. 20. Hintz SR, Gould JB, Bennett MV, et al. Referral of very low birth weight infants to high-risk follow-up at neonatal intensive care unit discharge varies widely across California. The Journal of pediatrics 2015;166:289-95. 21. Main EK. Clues for understanding hospital variation among obstetric services. American journal of obstetrics and gynecology 2015;213:443-4. 22. Main EK, Morton CH, Melsop K, Hopkins D, Giuliani G, Gould JB. Creating a public agenda for maternity safety and quality in cesarean delivery. Obstetrics and gynecology 2012;120:1194-8. 23. Powers RJ, Wirtschafter D. Prevention of Group B Streptococcus early-onset disease: a toolkit by the California Perinatal Quality Care Collaborative. Journal of perinatology : official journal of the California Perinatal Association 2010;30:77-87. 24. Wirtschafter DD, Danielsen BH, Main EK, et al. Promoting antenatal steroid use for fetal maturation: results from the California Perinatal Quality Care Collaborative. The Journal of pediatrics 2006;148:606-12. 25. Rudolph, L., Caplan, J., Ben-Moshe, K., & Dillon, L. (2013). Health in All Policies: A Guide for State and Local Governments. Washington, DC and Oakland, CA: American Public Health Association and Public Health Institute. 26. Community-Based Participatory Research, National Institutes of Health Website. Retrieved from:

Leslie Kowalewski currently serves as the Associate State Director for the California Chapter. Her responsibilities include advocacy, c o m mu n i c a t i o n s, programs, IT, operations, and fundraising. She is a native Oregonian and graduated from Oregon State University with a Bachelor of Science degree in health promotion and education.

Dr. Michael Curtis serves a Chief of the Surveillance and Program Evaluation Section 
in the Epidemiology, Assessment & Program Development Branch of the CDPH. The EAPDB provides scientific evidence bases for public policy and design and administers the Maternal and Infant Health Assessment Survey and manages contracts for data systems. Dr. Connie Mitchell is an expert on health and domestic violence and works in health policy and system change. She currently works in policy development at the CDPH in the Maternal, Child and Adolescent Health Division and serves as volunteer faculty in the School of Medicine at UC Davis. Dr. Larry Rand has experience with highrisk pregnancies cares for women with fetal complications. He was a medical/ cultural anthropologist and studied biochemistry before entering the Medicine. He attends patients on Labor and Delivery, has a special interest in medical education and quality improvement, and heads our Perinatal Fetal Treatment team. Dr. Scott Berns joined NICHQ from the March of Dimes National Office, where he was the Senior Vice President of Chapter Programs and Deputy Medical Officer. There he provided direction in education and community services to all March of Dimes state-based chapters. He directed national initiatives implemented in communities. Dr. Garry Shaw has been conducting research on birth defects for over 20 years and is a recognized leader in birth defects research. He has produced numerous publications on birth defect causes related to diet, obesity, drugs, alcohol, stress, pollution, occupations, and genes. Dr. David Stevenson is the Senior Associate Dean for Maternal & Child Health, Director of the LPCH’s Center for Pregnancy & Newborn Services, and PI of the MOD Prematurity Research Center. His work in neonatal jaundice and prevention of preterm birth has led to improved outcomes for countless infants and families.


The Leading Forum on the Accountable Care and Related Delivery System and Payment Reform SPONSORED BY Accountable Care Learning Collaborative CO-SPONSORED BY CAPG and Integrated Healthcare Association (IHA) MEDIA PARTNERS: Harvard Health Policy Review and Health Affairs November 16 – 18, 2015 • Los Angeles, CA


The Leading Forum on Pay for Performance, Transparency and Value-Driven Healthcare CO-SPONSORED BY Center for Healthcare Quality and Payment Reform, Integrated Healthcare Association, Health Care Incentives Improvement Institute, National Committee for Quality Assurance, Network for Regional Healthcare Improvement and Premier, Inc. MEDIA PARTNERS: Harvard Health Policy Review and Health Affairs February 16 – 18, 2016 • San Francisco, CA


The Leading Forum on Innovations in Population Health & Care Coordination Featuring a Special Medical Home Track SPONSORED BY Jefferson School of Population Health COSPONSORED BY Population Health Alliance MEDIA PARTNERS: Harvard Health Policy Review, Health Affairs, Accountable Care News, Healthcare Innovation News, Medical Home News, Population Health News and Population Health Journal March 7 – 9, 2016 • Philadelphia, PA


The Leading Forum on Healthcare EDI, Privacy, Breach Notification, Confidentiality, Data Security and HIPAA Compliance MEDIA PARTNERS: Harvard Health Policy Review and Health Affairs March 21 – 23, 2016 • Washington, DC


SPONSORED BY Health Data Consortium MEDIA PARTNERS: Harvard Health Policy Review and Health Affairs May 8 – 11, 2016 • Washington, DC

All Are Hybrid Conferences & Internet Events

SPONSORED BY International Society of Healthcare Compliance Professionals (ETHICS) COSPONSORED BY Pharmaceutical Compliance Forum (PCF) May 10 – 12, 2016 • Warsaw, Poland


The Leading Forum on Developing and Implementing Patient- and Family-Centered Medical Homes SPONSORED BY Patient Centered Primary Care Collaborative (PCPCC) and Jefferson School of Population Health MEDIA PARTNERS: Harvard Health Policy Review, Health Affairs, Accountable Care News, Medical Home News, Population Health News and Population Health Journal June 6 – 7, 2016 • Washington, DC


The Leading Forum on the Role of Healthcare Payment Reforms with Special Focus on Bundled Payment Approaches Offered in Sequence with the Seventh National ACO Summit MEDIA PARTNERS: Harvard Health Policy Review, Health Affairs, Accountable Care News, Healthcare Innovation News, Medical Home News, Population Health News and Population Health Journal June 7 – 9, 2016, Grand Hyatt, Washington, DC


The Leading Forum on the Accountable Care Organizations (ACOs) and Related Delivery System and Payment Reform Offered in Sequence with the Sixth National Bundled Payment Summit MEDIA PARTNERS: Harvard Health Policy Review, Health Affairs, Accountable Care News, Healthcare Innovation News, Medical Home News, Population Health News and Population Health Journal June 9 – 10, 2016 • Washington, DC

Attend Onsite or via Webcast — In your own office or home live via the Internet with 24/7 access for six months


Profile for Harvard Health Policy Review

HHPR Spring 2016 Issue  

HHPR Spring 2016 Issue