Adverse drug reactions are an important

Transcription

1 Adverse Drug Events doi: /hlthaff HEALTH AFFAIRS 31, NO. 5 (2012): Project HOPE The People-to-People Health Foundation, Inc. By Dongyi Tony Du, John Goldsmith, Kathryn J. Aikin, William E. Encinosa, and Clark Nardinelli Despite 2007 Law Requiring FDA Hotline To Be Included In Print Drug Ads, Reporting Of Adverse Events By Consumers Still Low Dongyi Tony Du (dongyi.du@ fda.hhs.gov) is an epidemiologist in the Center for Biologics Evaluation and Research, Food and Drug Administration, in Rockville, Maryland. John Goldsmith is an economist in the Office of Planning, Food and Drug Administration. Kathryn J. Aikin is a senior social science analyst in the Office of Prescription Drug Promotion, Food and Drug Administration. William E. Encinosa is a senior economist in the Center for Delivery, Organization, and Markets, Agency for Healthcare Research and Quality, in Rockville, Maryland. Clark Nardinelli is the chief economist and director of the economics staff in the Office of Policy, Planning, and Budget, Food and Drug Administration. ABSTRACT In 2007 the federal government began requiring drug makers to include in their print direct-to-consumer advertisements information for consumers on how to contact the Food and Drug Administration directly, either by phone or through the agency s website, to report any adverse events that they experienced after taking a prescription drug. Adverse events can range from minor skin problems like itching to serious injuries or illness that result in hospitalization, permanent disability, or even death. Even so, current rates of adverse event reporting are low. We studied adverse event reports about 123 drugs that came from patients before and after the enactment of the print advertising requirement and estimated that requirement s impact with model simulations. We found that if monthly spending on print direct-toconsumer advertising increased from zero to $7.7 million per drug, the presence of the Food and Drug Administration contact information tripled the increase in patient-reported adverse events, compared to what would have happened in the absence of the law. However, the absolute monthly increase was fewer than 0.24 reports per drug, suggesting that the public health impact of the increase was small and that the adverse event reporting rate would still be low. The study results suggest that additional measures, such as more publicity about the Adverse Event Reporting System or more consumer education, should be considered to promote patient reporting of adverse events. Adverse drug reactions are an important public health problem. These reactions, also known as adverse events, range from minor skin problems such as rashes or itching to serious injuries and illnesses that can result in hospitalization, permanent disability, or even death. It has been estimated that serious adverse drug reactions caused percent of all hospital admissions and 106,000 deaths in The average cost of one adverse drug event for hospitalized Medicare patients was estimated to be about $2,000 in The Centers for Medicare and Medicaid Services pays out at least $887 million each year to cover treatment costs for patients with adverse drug reactions. 2 Because clinical trials have limited scale and scope, they do not reveal all of a drug s potential risks. In fact, some risks may not be discovered until the drug is on the market and has been used by a large number of patients. This fact makes postmarket surveillance of adverse events a critical component of drug safety. The Food and Drug Administration (FDA) established the Adverse Event Reporting System in 1997 to collect reports of adverse drug reactions and medication errors from pharmaceutical manufacturers, health care professionals, and 1022 Health Affairs May :5

2 consumers. 3 Currently, only manufacturers are required to report adverse events about their products to the FDA. The agency has encouraged health care providers, other interested parties, and consumers to report them voluntarily. Congress has attempted to address consumer reporting of adverse drug events with two statutes. First, section 17 of the Best Pharmaceuticals for Children Act of 2002 required the labels of new drug products that is, those drugs approved for marketing under section 505 of the Food, Drug, and Cosmetic Act of 1938 to include a statement informing people how to report adverse events directly to FDA. The statement must include a toll-free number maintained by the FDA for the purpose of receiving reports of adverse events regarding drugs. 4 This rule took effect on November 28, 2008, and had to be complied with by July 1, Second, Title IX of the Food and Drug Administration Amendments Act of 2007 required print direct-to-consumer advertisements for prescription drug products to include the following statement, printed in conspicuous text: You are encouraged to report negative side effects of prescription drugs to the FDA. Visit or call FDA This requirement went into effect on March 25, Thus, consumers are now exposed to the possibility of reporting adverse events through two channels: both the labeling and the advertising of prescription drugs and therapeutic biologic products, or biologics. We present here an evaluation of the potential effect of the toll-free number in direct-toconsumer print advertisements on adverse event reporting.we first evaluated the relationship between adverse events reporting from patients and direct-to-consumer print advertisement spending. Then we estimated the impact of the inclusion of FDA contact information on such adverse event reporting. Study Data And Methods Data Sources The data used in this study included the FDA Adverse Event Reporting System data, data on the cost of direct-to-consumer advertising from IMS Health, and drug utilization data from SDI Health. Both IMS Health and SDI Health supply the pharmaceutical and health care industries with market information and consulting services. The FDA data provided the adverse event counts for studied drugs. The IMS data provided monthly spending information on print direct-to-consumer advertising. The SDI data supplied the demographic information for the study, such as number of users of a specific drug in a given month, and their age and sex. We combined information from these three sources to allow comparisons among drugs by month and year. The study used data over a thirty-five-month period, from July 2006 through May The Adverse Event Reporting System database supports the FDA s postmarketing surveillance program for all approved drug and biologic products. 6 As noted above, manufacturers are required to send adverse event reports to the FDA, but adverse events reporting by others, such as consumers or health care providers, is voluntary. For this study, we analyzed only adverse drug events reported by consumers. Studied Drugs The 123 drugs selected for sampling met the following criteria: enough variation of patient-reported adverse events and print direct-to-consumer advertisement spending during the study period to avoid multicollinearity (a problem that occurs when two or more variables are highly correlated and thus affect the accuracy of regression analysis); and no major FDA regulatory action, such as a new boxed warning or recall, regarding the drug during the study period. Analysis We used multivariate regression models to study the monthly counts of adverse events reported by consumers. The main predictor variables were as follows: the monthly level of spending on print direct-to-consumer advertising; a dummy variable coded to indicate whether the adverse event was reported before or after the enactment of the requirement that a toll-free number be listed in the advertisements; and the interaction term, which is the product of the spending on print direct-to-consumer advertising and the dummy variable. The interaction term examined the additional impact of directto-consumer spending on adverse event reporting after the enactment of the requirement. We assumed that direct-to-consumer advertising affects people in a cumulative manner, so that as exposure to direct-to-consumer advertisements increased, the more likely a patient would be to remember the message in an advertisement and act on it. Therefore, we used cumulative spending on direct-to-consumer advertising in our analysis. We calculated cumulative spending for directto-consumer advertisements for the month in which the adverse event was reported as the direct-to-consumer advertisement spending during that month plus the sum of such spending from the previous five months. Empirical evidence supported this consideration of directto-consumer promotion. 7 9 Other important confounding factors controlled in regression models included the number of patients per drug per month and the May :5 Health Affairs 1023

3 Adverse Drug Events percentage of each drug prescribed to females or males in the following age groups: under 35, 35 50, 51 65, and over 65. To account for possible secular trends in the reporting rate during the study period, we included a time trend variable in the regression to indicate the year of measurement. Because reporting rates probably differed for the drugs included in the study, we included drug fixed effects to adjust for these differences. In this study we used zero-inflated negative binomial regressions to estimate the incidence rate ratios of monthly adverse drug events. 10,11 This method accommodated the high proportion of zero observations and overdispersed outcome variables, while controlling for the variables of interest, confounding factors, the time trend, and unconditional drug fixed effects. We used the log numbers of patients per drug per month as the measure of exposure. Time from approval could also affect the number of adverse events reported to the FDA and the amount of pharmaceutical promotion Based on the data distribution, we included a categorical variable to indicate how long a drug had been on the market at the observation time. We defined three time-on-market categories to control for the life-cycle effect. These three categories divided all observations into three equal groups. The three categories coincidentally corresponded to the following lengths of time that a drug had been on the market at the observation time: 0 5 years, 6 10 years, and more than 10 years. The data used in this study were deidentified. Our study was deemed exempt from review by the FDA Research in Human Subjects Committee. Limitations Our analyses were limited by a number of factors that influence adverse event reporting. First, federal law requires pharmaceutical companies to include a statement about a toll-free number in labeling. 4 The requirement for labeling did not go into effect until July However, some companies distributed labeling with the statement for their products before the enactment date. We do not know which these products were. Second, we examined the impact of print direct-to-consumer advertising within thirteen months after the enactment of the requirement in the FDA Amendments Act. As more consumers became aware of the adverse event reporting information from repeated exposure, more adverse events might be reported through this channel. In that case, the long-term effect could be larger than what we estimated here. Furthermore, we assumed that the adverse event reporting rate was constant for patients Adding the toll-free number to television advertisements could have a bigger impact than doing so in print advertising. using a specific drug, but that assumption may not hold. For example, one study showed that expanded medication use could lead to an increased mismatch between patient and drug, resulting in increased reporting of adverse medical events. 15 Finally, patients might be exposed to other information that eventually leads to a change in the amount of adverse event reporting. In the current study, we included only the spending on print direct-to-consumer advertising for drugs that patients used. Patients are likely to be exposed to print advertisements for drugs that they do not take, an exposure that the current study could not measure. Therefore, even the small impact that we estimated could be overstated. Study Results From July 2006 to May 2009 the FDA received 7,100 reports from patients of adverse drug events associated with the 123 drugs we studied. On average, as Exhibit 1 shows, patients reported more adverse events per month after the enactment of the FDA Amendments Act than before the enactment of the law (2.35 events per drug, compared to 1.17 events). However, this difference is not significant. Manufacturers spent less on direct-toconsumer advertising overall after the enactment of the law than before (Exhibit 1). The proportion of manufacturers spending on print direct-to-consumer advertising compared to nonprint direct-to-consumer advertising also declined. Our study shows that the number of patients, the spending on print direct-to-consumer ads, and the interaction term between that spending and the time since the law was enacted were positively associated with the number of adverse events reported by patients (see Appendix Exhibit 2 in the online Appendix) Health Affairs May :5

4 Exhibit 1 Major Variables Used In The Study Of Adverse Event Reporting To The Food And Drug Administration Before FDA Amendments Act After FDA Amendments Act Variable Average Standard deviation Average Standard deviation Adverse events reported per month Patients using drug a 473, , , ,119 DTC spending on print advertising (millions) $1.08 $2.25 $0.65 $1.54 DTC spending on nonprint advertising b (millions) $1.73 $3.53 $1.54 $3.47 Average years on market since launch 0 5 years ago years ago >10 years ago SOURCE Author s analysis.notes An expanded version of this exhibit is available in the online Appendix (see Note 16 in text). The Food and Drug Administration Amendments Act was enacted in DTC is direct-to-consumer. a The analysis included 123 drugs and 7,100 reportsofadversereactionstothosedrugsfrompatientstakingthem. b Nonprint advertising includes broadcast advertising and outdoor advertising. To account accurately for the additional impact of direct-to-consumer advertising on patients after the law s enactment, we estimated that impact with model simulations. 17 The simulation results indicated that before the law s enactment, if the cumulative spending on print direct-to-consumer advertising increased from $0 to $7.7 million per drug (the seventy-fifth percentile of all observations), there were 0.08 more reports each month of adverse drug events per drug. After the law s enactment, the same increase in spending on print direct-to-consumer advertising resulted in 0.24 more monthly reports of adverse events per drug. Of that increase, 64.8 percent was attributable to the requirement that manufacturers include toll-free reporting numbers in print direct-to-consumer advertisements (data not shown). The number of patients was positively associated with adverse drug event reporting in all models (Appendix Exhibit 2). 16 More patients led to more adverse event reports. The simulation results imply that adverse drug event reporting increased approximately 40 percent, from 1.28 monthly reports per drug to 1.74 monthly reports per drug, when the number of patients increased from the twenty-fifth percentile of all observations (59,000 patients) to the seventyfifth percentile of all observations (613,000 patients). Discussion Although the number of reports has been increasing since the creation of the Adverse Event Reporting System, the reporting rate is still low. Depending on the drug involved, the severity of the adverse reaction, and the time since drug launch, the estimated rates of reporting range from 0.3 percent to 33.0 percent of all adverse events. 18,19 About 6 percent of the adverse event reports received by the FDA came directly from health care professionals, and fewer than 2 percent came directly from consumers. 20 We estimated how much adding a reporting statement with a toll-free number and a website in print direct-to-consumer advertising for prescription drugs encouraged patients to report adverse events directly to the FDA.We found that the requirement did increase adverse drug event reporting from patients. However, the increase was small compared to the magnitude of the amount spent on print direct-to-consumer advertising. After the enactment of the law, our model estimated that if the cumulative monthly spending on print direct-to-consumer advertising increased from $0 to $7.7 million, there would be only 0.24 more monthly adverse event reports per drug. Television Versus Print Advertising It should be noted that the law also required the Department of Health and Human Services to determine if the reporting statement is appropriate for inclusion in direct-to-consumer television advertisements. If the inclusion of such a statement is determined to be appropriate in these ads, the FDA must issue regulations requiring that the ads include the statement. The FDA is currently studying whether including a tollfree adverse events phone number in such advertisements would detract from the important risk information that the ads are already required to contain. 21 If the positive relationship between spending on direct-to-consumer advertising and adverse event reporting holds, adding the toll-free number to television advertisements could have a May :5 Health Affairs 1025

5 Adverse Drug Events There is room to increasetherateat which patients report adverse events to the FDA. bigger impact than doing so in print advertising. Pharmaceutical companies spend more on direct-to-consumer television advertising than on print advertisements. For the 123 drugs included in this study, for example, the average monthly spending on television direct-to-consumer advertising was 77.7 percent higher than for print advertising. The average American television viewer was exposed to sixteen hours of direct-to-consumer broadcast advertising in Television advertisements are subject to different regulations than print advertisements. Print advertisements for prescription drugs must disclose all of the product s risks in a brief summary, including side effects, warnings, and precautions. Television advertisements must either disclose all of the product s risks or disclose the product s major risks typically, the most serious and most frequent ones and make adequate provision for letting the consumer know how to obtain the full risk information. This obligation is known as the adequate provision requirement. 23 Manufacturers almost exclusively choose to follow the adequate provision option. The reduced information load in television advertising compared to that in print advertising may increase consumers opportunities to understand and remember the messages carried in television advertisements. However, these advertisements are not self-paced: Consumers do not have unlimited time to process the information being presented. In contrast, consumers can read a print advertisement at their own pace, which gives them more opportunity to process the information presented. Moreover, the reader does not have to remember or jot down a toll-free number that is on a printed page, in contrast to a broadcast number. Therefore, we cannot conclude that the response to toll-free numbers in television advertisements would necessarily be greater than the response we found for print advertisements. One potential solution to avoid the interference between the reporting statement and the important risk information is to decouple the two and disseminate information on the availability of this reporting avenue separately from direct-to-consumer advertising. Currently, the FDA does not have the financial resources to disseminate its adverse event reporting information at the level of direct-to-consumer advertising. The Value Of Reports From Consumers For several reasons, the adverse event reports from consumers have value beyond that in reports provided by health care providers and pharmaceutical companies. First, consumers reports are usually proportional to the number of medication users. Whether a patient reports an adverse event or not is less likely to be directly influenced by other patients who experienced the same adverse events if the adverse event is rare and there is no media coverage of it. In contrast, providers reports of adverse events for a specific drug may decrease with time. 13 The decline results from the fact that health care professionals are generally not likely to report the same adverse event when they encounter it in additional patients. Pharmaceutical companies are required to report adverse events, but they can sometimes obtain waivers from the FDA for reporting certain nonserious adverse events. Second, studies suggest that patients reports of adverse drug events might help to identify safety issues earlier. 24,25 For example, one study showed that twelve out of twenty-three safety issues were identified earlier, eight issues were identified at the same time, and only three issues were identified later when an analysis of adverse event reports included reports from consumers as well as providers. 24 Third, patients reports of adverse events may provide different information than those from health care professionals. In an analysis of 6,319 adverse drug event reports, another study found that patients were more likely than other sources to report adverse drug reactions associated with nervous system disorders, psychiatric disorders, and reproductive system and breast disorders. 26 Although evidence points to the added benefit of patient reporting of adverse events, the issue remains controversial. Some studies have suggested that patients reports are of similar quality to those from health care professionals. 27,28 But other studies have revealed the downside of consumer reports, 29 including one study showing that patient reports require more time and resources to evaluate than reports from health care professionals. 28 Policy Implications The direct policy implications of these results are not clear. The results 1026 Health Affairs May :5

6 show that the presence in direct-to-consumer advertisements of language about the toll-free number for reporting adverse events increases the likelihood that patients will report adverse events directly to the FDA.With the toll-free number, the FDA has greater access to important and unique data in the form of patients reports on adverse events. Moreover, our results imply that more widespread availability of FDA contact information could increase reporting rates further. It is possible that by encouraging consumers to report adverse events directly to the FDA, the regulatory agency and health care professionals in general risk losing control over the information on adverse events. The nature of public discourse involves many viewpoints of varying accuracy. This mixed-quality discourse may spread incorrect health information, which in turn could have negative consequences for public health such as noncompliance with medical recommendations. As a science-based agency, the FDA views the greater availability of quantitative data based on patient reports as a positive contribution overall. Despite the wider variance in the quality of patient reports, compared with other sources of information, increased reporting of adverse events and analyses of those reports will bring more data to bear on the discussion of drug safety already well under way in many places, including social networks and the Internet. Conclusion There is room to increase the rate at which patients report adverse events to the FDA, which in turn would increase the resulting benefits to public health. Patients reports may not be as reliable as those from health care providers or manufacturers, but they make different contributions to the early detection of adverse drug reactions. We have shown that the requirement of including FDA contact information for reporting adverse events in print direct-to-consumer advertisements for prescription drugs has had only a modest effect on reporting. However, additional publicity or education efforts could increase consumers awareness of this program and the attendant benefits. The authors thank John Quinn, Lynette Swartz, and Elaine Hu Cunningham of thefoodanddrugadministration s Center for Drug Evaluation and Research for their support. The authors have no financial interests in this article and received no funding to conduct this study. The opinions expressed in the article are those of the authors and are not intended to represent the opinions of the Food and Drug Administration. NOTES 1 Lazarou J, Pomeranz BH, Corey PN. Incidence of adverse drug reactions in hospitalized patients: a metaanalysis of prospective studies. JAMA. 1998;279(15): Field TS, Gilman BH, Subramanian S, Fuller JC, Bates DW, Gurwitz JH. The costs associated with adverse drug events among older adults in the ambulatory setting. Med Care. 2005;43(12): Moore TJ, Cohen MR, Furberg CD. Serious adverse drug events reported to the Food and Drug Administration, Arch Intern Med. 2007;167(16): Food and Drug Administration. Tollfree number for reporting adverse events on labeling for human drug products. Final rule. Fed Regist. 2008;73(209): US Code, sec Strauss A, Lawrence GA, Nelsen LM, Wiholm BE, Jones JK. Publicreleased version of the Adverse Event Reporting System (AERS) database. Clin Ther. 2005;27(3): Narayanan S, Desiraju R, Chintagunta PK. Return on investment implications for pharmaceutical promotional expenditures: the role of marketing-mix interactions. J Marketing. 2004;68(4): Ling DC, Berndt ER, Kyle MK. Deregulating direct-to-consumer marketing of prescription drugs: effects on prescription and over-thecounter product sales. J Law Econ. 2002;45(2): Iizuka T, Jin GZ. The effect of prescription drug advertising on doctor visits. J Econ Manage Strategy. 2005;14(3): Hilbe JM. Negative binomial regression. Cambridge (UK): Cambridge University Press; Poisson regression is a form of a generalized linear model in which the response variable is modeled as having a Poisson distribution. It assumes that the logarithm of the expected value of the response variable can be modeled by a linear combination of unknown parameters. A unique feature of Poisson distribution is that its mean is equal to its variance. When observed variance is greater than the mean (a situation known as overdispersion), the Poisson regression is not an appropriate model to use. In that case, the negative binomial regression a variant of Poisson regression is more appropriate. When there are excess zeros in the response variable, zeroinflated negative binomial regressions can be adapted to accommodate the excess zeros. 12 Olson MK. Pharmaceutical policy change and the safety of new drugs. J Law Econ. 2002;45(2): Weber J. Epidemiology of adverse reactions to nonsteroidal antiinflammatory drugs. Raven. 1984; (6): Campbell S. Promotional spending for prescription drugs [Internet]. Washington (DC): Congressional Budget Office; 2009 Dec 2 [cited 2012 Mar 27]. (Economic and Budget Issue Brief). Available from: files/cbofiles/ftpdocs/105xx/ doc10522/12-02-drugpromo_ brief.pdf 15 David G, Markowitz S, Richards S. The effects of pharmaceutical marketing and promotion on adverse drug events and regulation [Internet]. Cambridge (MA): National Bureau of Economic Research; 2009 Jan [cited 2012 Mar 27]. (NBER Working Paper No ). Available from: w14634.pdf 16 To access the Appendix, click on the Appendix link in the box to the right of the article online. 17 Ai CR, Norton EC. Interaction terms May :5 Health Affairs 1027

7 Adverse Drug Events in logit and probit models. Econ Letters. 2003;80(1): Warren JL, McBean AM, Hass SL, Babish JD. Hospitalizations with adverse events caused by digitalis therapy among elderly Medicare beneficiaries. Arch Intern Med. 1994;154(13): Government Accountability Office. Adverse drug events: the magnitude of the health risk is uncertain because of limited incidence data [Internet]. Washington (DC): GAO; 2000 Jan [cited 2010 Mar 2]. Available from: new.items/he00021.pdf 20 Food and Drug Administration. Reports received and reports entered into AERS by year [Internet]. Silver Spring (MD): FDA; [last updated 2011 Jul 15; cited 2012 Mar 27]. Available from: Drugs/GuidanceCompliance RegulatoryInformation/ Surveillance/AdverseDrugEffects/ ucm htm 21 Food and Drug Administration. Experimental study: toll-free number for consumer reporting of drug product side effects in direct-toconsumer television advertisements for prescription drugs. Fed Regist. 2009;74(158): Frosch DL, Krueger PM, Hornik RC, Cronholm PF, Barg FK. Creating demand for prescription drugs: a content analysis of television directto-consumer advertising. Ann Fam Med. 2010;5(1): Food and Drug Administration. Guidance for industry on consumerdirected broadcast advertisements; availability. Fed Regist. 1999; 64(152): Hammond IW, Rich DS, Gibbs TG. Effect of consumer reporting on signal detection: using disproportionality analysis. Expert Opin Drug Saf. 2007;6(6): Egberts TCG, Smulders M, dekoning FH, Meyboom RH, Leufkens HG. Can adverse drug reactions be detected earlier? A comparison of reports by patients and professionals. BMJ. 1996;313(7056): Aagaard L, Nielse LH, Hansen EH. Consumer reporting of adverse drug reactions: a retrospective analysis of the Danish adverse drug reaction database from 2004 to Drug Saf. 2009;32(11): Ekins-Daukes S, Irvine D, Wise L, Fiddes S. The yellow card scheme: evaluation of patient reporting of suspected adverse drug reactions. Pharmacoepidemiol Drug Saf. 2006;15: Danish Medicines Agency. One year with ADR consumer reports. Copenhagen: The Agency; 2004 Oct. 29 Mitchell AS, Henry DA, Sanson- Fisher R, O Connell DL. Patients as a direct source of information on adverse drug reactions. BMJ. 1988; 297(6653): ABOUT THE AUTHORS: DONGYI TONY DU, JOHN GOLDSMITH, KATHRYN J. AIKIN, WILLIAM E. ENCINOSA & CLARK NARDINELLI Dongyi Tony Du is an epidemiologist at the Food and Drug Administration. In this month s Health Affairs, Tony Du and his coauthors analyze reporting on adverse drug events to the Food and Drug Administration (FDA) in the wake of legislation enacted in That law required manufacturers to inform consumers via print advertising on how to make these adverse event reports, but actual reporting has remained low. The authors suggest additional measures, including educating consumers about the Adverse Event Reporting System. Du is an epidemiologist in the FDA s Center for Biologics Evaluation and Research. He is also an Oak Ridge Institute for Science and Education Fellow. He was in the inaugural class of FDA Commissioner s Fellows in This paper is part of his work as Commissioner s Fellow. Du received a medical degree from the Norman Bethune University of Medical Sciences; a doctorate in pharmacoeconomics from the University of Maryland at Baltimore; and a master s degreein immunology from the Changchun Institute of Biological Products, Chinese Ministry of Public Health. John Goldsmith is an economist at the FDA. John Goldsmith is an economist in the FDA s Office of Planning. He holds a doctorate in economics from the University of Maryland. Kathryn J. Aikin is a senior social science analyst at the FDA. Kathryn Aikin is a senior social science analyst in the FDA s Office of Prescription Drug Promotion. Her responsibilities include research; advisory and enforcement review of prescription drug promotion; and providing social science input into FDA policy, guidance, and regulation. Her research work focuses on direct-toconsumer (DTC) prescription drug advertising on such topics as consumers comprehension and perceptions of DTC advertising, national surveys of consumer and physician attitudes toward such advertising, and improvements to the consumer brief summary in DTC print ads. She is a member of the editorial board of the Journal of 1028 Health Affairs May :5

8 Public Policy and Marketing. She received a master s degreeanda doctorate in social psychology from Pennsylvania State University. William E. Encinosa is a senior economist at the Agency for Healthcare Research and Quality. William Encinosa is a senior economist in the Center for Delivery, Organization, and Markets at the Agency for Healthcare Research and Quality. He is also an adjunct associate professor at the Georgetown University Public Policy Institute. In 2009 Encinosa won the John M. Eisenberg Article-of-the-Year Award in Health Services Research from the Health Research and Educational Trust. He is a member of the editorial board of the American Journal of Managed Care. Encinosa has a doctorate in economics and a master s degree in mathematics from the University of Florida. Clark Nardinelli is director of the economics staff in the FDA Office of Policy, Planning, and Budget. Clark Nardinelli is the chief economist and director of the economics staff in the FDA Office of Policy, Planning, and Budget. His responsibilities include supervising and reviewing costbenefit and cost-effectiveness analyses of FDA regulations and other public health policies. Nardinelli earned a master s degree and a doctoral degree in economics from the University of Chicago. May :5 Health Affairs 1029