1

EPID 765 METHODS AND ISSUES IN PHARMACOEPIDEMIOLOGY

3 credits

*PRE*-requisites: EPID 600 and BIOS 600 or equivalents

Lead-Instructor: Til Stürmer; sturmer@unc.edu; 919 966 7433

Co-Instructor: Michele Jonsson Funk; mfunk@unc.edu; 919-843-0384

January 10 – April 25, 2019

Tuesday, Thursday, 2:00 – 3:15

Room: Rm MGG 2301 Pharmacoepidemiology: Application of epidemiologic knowledge, methodology, and reasoning to the study of the effects (beneficial and adverse) and uses of drugs in human populations

Pharmacoepidemiology is a public health discipline that mainly relies on non-experimental (epidemiologic) methods to assess intended and unintended drug effects to support decision-makers in the absence of specific evidence from experimental studies (randomized controlled trials). This course is for clinicians, pharmacists, epidemiologists and scientists from related fields in academia, industry and regulatory agencies. It will provide an introduction and overview of pharmacoepidemiologic topics, methods, databases, and review examples of current research. The course will look at specific aspects and potential pitfalls of epidemiologic study designs when applied to the study of drug effects and provide an overview of ways to limit the potential for bias.

Course objectives: introduce participants to most important issues and career options in pharmacoepidemiology; acquire basic understanding of how non-experimental studies on drugs can draw from standard epidemiologic techniques and unique research challenges and opportunities. Provide the tools necessary to evaluate published pharmacoepidemiologic studies and to design and implement (note: additional courses/skills required) pharmacoepidemiologic studies using state-of-the-art methodology to limit the potential for bias.

Structure: Journal club discussion, lectures, case studies, invited speakers. The course is organized as a sequence of relevant topics. Most lessons will start with a 30 minute discussion of the topic followed by a 45 minute lecture. Discussions are either student led (journal club) or led by the instructors. Preparation and active participation in the discussion is expected from all.

2

Grasp of challenges and concepts is encouraged over knowledge about solutions. Readings are from a draft PE textbook by the lead instructor, from the Textbook of Pharmacoepidemiology (see below), and from the literature.

Expectations:

• All students are expected to read required materials (all provided on Sakai) before class and to participate actively in class discussions. Suggested readings are intended for future reference.

• All students will lead a journal club discussion. I will ask for volunteers but may need to assign students

• All students are expected to write a term paper on a pharmaco-epidemiologic topic (see below).

Grading: 30%: Class participation 30%: Presentation and discussion of journal club article 40%: Term paper Each will be graded on a 4 point scale

- 4: fully acceptable by professional colleague - 3: evidence of a colleague in training - 2: some merit but insufficient for scientific interchange - 1: unacceptable or incomplete

An overall grade of at least 2.5 is required for a pass; students with a grade of 3.5 or higher will receive an honor grade. Journal Club presentation:

During the last 20 minutes of class. The student will provide a brief summary of the paper (3 minutes, absolutely not more than 5!) and then lead the discussion. The discussion should cover important aspects of methods, results, and conclusions. Do not try to cover everything, but rather focus on specific aspects of these. Are the conclusions supported by the data presented? Would you change clinical practice based on the data presented?

Term paper:

The term paper should provide an overview of a chosen drug-outcome association. This is not to imply that the field of pharmacoepidemiology is restricted to this kind of study but rather to acknowledge that many challenges when addressing drug-outcome associations using non-experimental methods are unique to the field.

The paper should develop the history of the evidence ideally from case-report (or any other form of signal) to the current state of knowledge. The paper should indicate an understanding of the advantages and disadvantages of specific studies based on their design and analytic methodology, make a non-formal summary of the evidence taking design and analytic methodology into account, discuss limitations of the existing evidence and whether and how

3

these can be addressed, and finally propose possibilities to overcome evidence gaps based on existing or new data.

The overview should be based on the published literature and not cover a topic that has been recently reviewed (2013 or later) because that leaves little room for additional work and interpretation. For those taking the database class we encourage choosing the same topic but this is not necessary. A maximum of 4 students can team up for the term paper. For group term papers, we expect a qualitatively more substantial contribution and that students within a group evaluate their respective contributions.

Each student should submit a very brief (1 paragraph, max. 250 words) proposal for a topic indicating why this topic was chosen and whether it is suited to address the above mentioned points. These brief proposals are due 2/27 to allow discussion of problems/alternatives in the week before the spring break. Each student will need an agreed upon topic before the spring break (by 3/7). Each student will present their paper in class at the end of the course. We will have approximately 10 minutes per student/project including discussion (the final time will be a function of the number of projects/teams). The presentation will be limited to 5 minutes and 6 slides, sharp. The term paper is due 4/8 at midnight.

The term paper should conform to the instructions for authors of Pharmacoepidemiology and Drug Safety (review category, maximum 3,000 words). In addition, there should be at least one figure and not more than 3 tables. All facts presented should be referenced and any plagiarism avoided. Some of these papers may lead to publications while others will not.

The term papers will be graded based on the understanding of the difficulties to make decisions about the benefit and harm of a specific drug/drug class in the presence of less than ideal data, limitations of the existing evidence, and the proposal to overcome these. The latter should be based on an understanding of real life constraints (rather than “we propose a RCT enrolling 100k people to be treated over 20 years”). General statements (“could be biased”) should be avoided. Instead, an assessment of the direction and magnitude of potential biases of a specific study in relation to other studies should be made [e.g., “Study A did not control for SES. Low SES (defined as) vs. high SES (defined as) has been shown to be a risk factor for Y (RR;95%CI)[reference] and to be associated with barriers to receiving treatment B vs. C (OR;95%CI) [reference]; thus not controlling for SES would tend to bias the RR of B vs. C on Y in the direction of X and based on the strength of associations described above, the magnitude of bias would likely be sufficient to explain the observed result/discrepancies”].

4

Course Policies and Resources

Recognizing,

Valuing and

Encouraging

Inclusion and

Diversity in the

Classroom

We share the School`s commitment to diversity. We are committed to ensuring that the School is a diverse, inclusive, civil and welcoming community. Diversity and inclusion are central to our mission — to improve public health, promote individual well-being and eliminate health inequities across North Carolina and around the world. Diversity and inclusion are assets that contribute to our strength, excellence and individual and institutional success. We welcome, value and learn from individual differences and perspectives. These include but are not limited to: cultural and racial/ethnic background; country of origin; gender; age; socioeconomic status; physical and learning abilities; physical appearance; religion; political perspective; sexual identity and veteran status. Diversity, inclusiveness and civility are core values we hold, as well as characteristics of the School that we intend to strengthen.

We are committed to expanding diversity and inclusiveness across the School—among faculty, staff, students, on advisory groups, and in our curricula, leadership, policies and practices. We measure diversity and inclusion not only in numbers, but also by the extent to which students, alumni, faculty and staff members perceive the School’s environment as welcoming, valuing all individuals and supporting their development.”

In this class, we practice these commitments in the following ways:

• Develop classroom participation approaches that acknowledge the diversity of ways of contributing in the classroom and foster participation and engagement of all students.

• Structure assessment approaches that acknowledge different methods for acquiring knowledge and demonstrating proficiency.

• Encourage and solicit feedback from students to continually improve inclusive practices.

As a student in the class, you are also expected to understand and uphold the following UNC policies:

• Diversity and Inclusion at the Gillings School of Global Public Health:

http://sph.unc.edu/resource-pages/diversity/

• UNC Non-Discrimination Policies:

http://policy.sites.unc.edu/files/2013/04/nondiscrim.pdf • Prohibited Discrimination, Harassment, and Related Misconduct at UNC:

https://deanofstudents.unc.edu/incident-reporting/prohibited-harassmentsexual-misconduct

Accessibility UNC-CH supports all reasonable accommodations, including resources and services, for students with disabilities, chronic medical conditions, a temporary disability, or a pregnancy complication resulting in difficulties with accessing learning opportunities. All accommodations are coordinated through the UNC Office of Accessibility Resources &

5

Services (ARS), https://ars.unc.edu/; phone 919-962-8300; email

ars@unc.edu. Students must document/register their need for accommodations with ARS before accommodations can be implemented.

Counseling and

Psychological

Services

CAPS is strongly committed to addressing the mental health needs of a diverse student body through timely access to consultation and connection to clinically appropriate services, whether for short or long-

term needs. Go to their website: https://caps.unc.edu or visit their facilities on the third floor of the Campus Health Services building for a walk-in evaluation to learn more.

UNC Honor

Code

As a student at UNC-Chapel Hill, you are bound by the university’s

Honor Code, through which UNC maintains standards of academic excellence and community values. It is your responsibility to learn about and abide by the code. All written assignments or presentations (including team projects) should be completed in a manner that demonstrates academic integrity and excellence. Work should be completed in your own words, but your ideas should be supported with well-cited evidence and theory. To ensure effective functioning of the

Honor System at UNC, students are expected to:

a. Conduct all academic work within the letter and spirit of the Honor Code, which prohibits the giving or receiving of unauthorized aid in all academic processes.

b. Learn the recognized techniques of proper attribution of sources used in written work; and to identify allowable resource materials or aids to be used during completion of any graded work.

c. Sign a pledge on all graded academic work certifying that no unauthorized assistance has been received or given in the completion of the work.

d. Report any instance in which reasonable grounds exist to believe that a fellow student has violated the Honor Code. Instructors are required to report suspected violations of the Honor Code, including inappropriate collaborative work or problematic use of secondary materials, to the Honor Court. Honor Court sanctions can include receiving a zero for the assignment, failing the course and/or

suspension from the university. If you have any questions about your rights and responsibilities, please consult the Office of Student

Conduct at https://studentconduct.unc.edu/, or consult these other resources:

• Honor system module.

• UNC library’s plagiarism tutorial. • UNC Writing Center handout on plagiarism.

Recommended parallel course: Because pharmacoepidemiology relies heavily on the use of large automated healthcare databases, participants are encouraged to also take EPID 766 “Epidemiologic Research with

6

Healthcare Databases” taught in spring 2019 (Lead-Instructor: Dr. Jennifer Lund). Students are also encouraged to come to the Pharmacoepidemiology Seminar on Mondays, 3:30-4:30.

Textbook: Textbook of Pharmacoepidemiology. 2nd edition. Brian L Strom, Stephen E Kimmel (eds). 480 pages, paperback, ISBN 978-1-118-34486-6, Wiley-Blackwell, 2013. Available in the library for this course. This is a reasonably priced textbook providing a good overview, including overview of often used healthcare databases. You can get one for <$20.

Selection of and some comments on other pharmacoepidemiology textbooks:

Pharmacoepidemiology, 5th edition. Brian L Strom, Stephen E Kimmel, and Sean Hennessy (eds). 976 pages, hardcover, ISBN: 978-0-470-65475-0, Wiley-Blackwell 2012. The “bible” in its 5th edition. "The book is essential reading for anyone interested in pharmacoepidemiology." (IJE); “…an excellent textbook and a comprehensive reference which belongs in the library of every pharmaceutical manufacturer and regulator." (Eur J Public Health). ~$340 (e-book: ~$300) Full text e-book available online at UNC. http://search.lib.unc.edu/search?R=UNCb8080048

Pharmacoepidemiology and Therapeutic Risk Management. Abraham G Hartzema, Hugh H Tilson, K Arnold Chang, (eds). 1050 pages, softcover, ISBN: 978-0929375304, 2008. Much extended scope compared with prior versions (see below). Good examples. PERM from now on. ~$200, but I saw one on sale for $99. Currently revised for next edition.

Pharmacoepidemiology: An Introduction, 3rd edition. Abraham G Hartzema, Miquel Porta, Hugh H Tilson, 670 pages, softcover, ISBN 0-929375-18-1, 1998. Provides the reader with an overview of pharmacoepidemiology, as well as the epidemiology of specific disease states. Includes an annotated bibliography of pharmacoepidemiologic studies as of 20 years ago. A little old, ~$25

Powerful Medicines: The Benefits, Risks, and Costs of Prescription Drugs. Jerry Avorn. Hardcover, 464 pages, ISBN: 978-0-375-41483-1 (0-375-41483-5), 2004. No textbook but comprehensive behind-the-scenes look at issues that affect everyone: our shortage of data comparing the worth of similar drugs for the same condition; alarming lapses in the detection of lethal side effects; the underuse of life-saving medications; lavish marketing campaigns that influence what doctors prescribe; and the resulting upward spiral of costs that places vital drugs beyond the reach of many Americans. Hardcover ~$20, Paperback ~$10

Selection of epidemiology textbooks

Modern Epidemiology, 3rd edition. Kenneth J Rothman, Sander Greenland, Timothy L Lash. 851 pages, hardcover, ISBN: 978-1451190052, Lippincott Williams & Wilkins, 2008 (mid-cycle revision 2012). The bible in its 3rd edition. Still the best and cheaper than many others. A must have for anyone interested in non-experimental population research. ~$80

Epidemiology. An introduction. 2nd edition. Kenneth J Rothman. hard/paperback, ISBN: 978-0199754557. Oxford University Press, 2012. A simple (but neither simplistic nor outdated as so many others) overview of the concepts that are the underpinnings of epidemiology, so that a coherent picture of epidemiology thinking emerges. The emphasis is not on statistics, formulas, or computation, but on epidemiologic principles and concepts". ~$40

7

Lessons

# Day Date Contents Instructor

1 Thu 1/10 What is pharmacoepidemiology? TS

2 Tue 1/15 Sources of data for PE Mitch Conover

3 Thu 1/17 Drug, outcome, and comorbidity data TS

4 Tue 1/22 Methodologic Challenges in PE

• Confounding (by indication, frailty)

• Selection bias (healthy initiator, sick stopper, healthy user)

TS

5 Thu 1/24 Study Design Solutions

• New user design

• Active comparators

TS

6 Tue 1/29 Risk periods

• First treatment carried forward

• As treated

• Induction, carry-over, lag periods

TS

7 Thu 1/31 Propensity scores TS

8 Tue 2/5 Disease risk scores TS

9 Thu 2/7 Non-uniform treatment effects TS

10 Tue 2/12 Instrumental variables TS

11 Thu 2/14 Validation Studies TS

12 Tue 2/19 Crystal ball PE, immortal time bias, immeasurable time bias

TS

13 Thu 2/21 Adherence & persistence TS

14 Tue 2/26 Patients treated contrary to prediction TS

We 2/27 Term paper proposal

15 Thu 2/28 Variability in treatments & variable selection (including hdPS)

TS

16 Tue 3/5 Potentially inappropriate prescribing TS

17 Thu 3/7 The opioid epidemic Nab Dasgupta

Thu 3/7 Agreed upon term paper topic

Spring Break (3/9 – 3/17)

8

18 Tue 3/19 Methods review TS

19 Thu 3/21 Practical lessons for a successful career in pharmacoepidemiology

Nancy Dreyer

20 Tue 3/26 Paper Presentation I

21 Thu 3/28 Paper Presentations II

22 Tue 4/2 Medications in pregnancy Sara Ephross

23 Thu 4/4 Methodology Richard Wyss

Mo 4/8 Term paper due

24 Tue 4/9 Pragmatic trials Michael Kappelman

25 Thu 4/11 Propensity score trimming Robert Glynn

26 Tue 4/16 Self-controlled designs Jesper Hallas

27 Thu 4/18 Single arm studies with external comparator

Christina Mack

28 Tue 4/23 Antidepressants and suicide Matt Miller

29 Thu 4/25 Wrap-up TS

9

Lesson 1: What is pharmacoepidemiology?

Required readings:

• None

Suggested readings:

• Weed DL, Hursting SD. Biologic plausibility in causal inference: Current method and practice. Am J Epidemiol 1998;147:415-25.

• Evans SJW. Pharmacoepidemiology. Br J Clin Pharmacol 2012; 73(6):973–8.

• Venulet J. From experimental to social pharmacology. Int J Clin Pharmacol 1974;10(3):203-5.

Lesson 2: Sources of data for PE

Required readings:

• Strom BL. Overview of automated databases in pharmacoepidemiology. Pharmacoepidemiology, 5th edition, chapter 11

• Journal club article: Characterizing RWD Quality and Relevancy for Regulatory Purposes (Page 4-13 of the PDF). https://healthpolicy.duke.edu/sites/default/files/atoms/files/characterizing_rwd.pdf)

Suggested readings:

• Toh S et al. Examples of automated databases 123. Textbook of Pharmacoepidemiology. 2nd edition, chapter 9

• Eichler H-G, Abadie E, Breckenridge A, Leufkens H, Rasi G (2012) Open Clinical Trial Data for All? A View from Regulators. PLoS Med 9(4): e1001202. doi:10.1371/journal.pmed.1001202

• Johannesdottir SA, Erzsebet HP, Ehrenstein V, Schmidt M, Pedersen L, Sørensen HT.. Existing data sources for clinical epidemiology: The Danish National Database of Reimbused Prescriptions. Clinical Epidemiology 2012; 4: 303-313.

• Corrigan-Curay J, Sacks L, Woodcock J. real-world evidence and real-

world data for evaluating drug safety and effectiveness. JAMA.

September 4, 2018. Volume 320, Number 9.

• Sherman RE, Anderson SA, Dal Pan GJ, et al. Real-World Evidence –

What Is It and What Can it Tell Us? N Eng J Med 2016; 375;23

Lesson 3: Drug, outcome, and comorbidity data

Required readings:

10

• Suzanne L. West, Brian L. Strom, Charles Poole. Validity of Pharmacoepidemiologic Drug and Diagnosis Data. Pharmacoepidemiology, 5th edition, chapter 41

• Journal club article: Lin KJ, Glynn RJ, Singer DE, Murphy SN, Lii J, Schneeweiss S. Out-of-system Care and Recording of Patient Characteristics Critical for Comparative Effectiveness Research. Epidemiology. 2018 May;29(3):356-363.

Suggested readings:

• Dickstein C. Health Care Data and the SAS System Course Notes. Chapter 3.5 “Pharmacy Codes”. SAS Institute Inc. 2009; ISBN 978-1-60764-401-9

• Stürmer T, Jonsson Funk M, Poole C, Brookhart MA. Nonexperimental Comparative Effectiveness Research Using Linked Healthcare Databases [invited commentary for The Changing Face of Epidemiology]. Epidemiology 2011;22(3):298-301.

Lesson 4: Methodologic challenges in PE

• Confounding (by indication, frailty)

• Selection bias

Required readings:

• Glynn RJ, Knight EL, Levin R, Avorn J. Paradoxical relations of drug treatment with mortality in older persons. Epidemiology 2001;12:682-9.

• Journal club article: Emilsson L, García-Albéniz X, Logan RW, Caniglia EC, Kalager M, Hernán MA. Examining Bias in Studies of Statin Treatment and Survival in Patients With Cancer. JAMA Oncol. 2018 Jan 1;4(1):63-70.

Suggested readings:

• Shrank WH, Patrick AR, Brookhart MA. Healthy User and Related Biases in Observational Studies of Preventive Interventions: A Primer for Physicians. Journal of General Internal Medicine. May 2011, Volume 26, Issue 5, pp 546-550

• Simpson SH, Eurich DT, Majumdar SR, Padwal RS, Tsuyuki RT, Varney J, Johnson JA. A meta-analysis of the association between adherence to drug therapy and mortality. BMJ 2006;333:15.

• Brookhart MA, Patrick AR, Dormuth C, Avorn J, Shrank W, Cadarette SM, Solomon DH. Adherence to lipid-lowering therapy and the use of preventive health services: an investigation of the healthy user effect. Am J Epidemiol 2007;166:348-54.

• Schneeweiss S, Patrick AR, Sturmer T, et al. Increasing levels of restriction in pharmacoepidemiologic database studies of elderly and

11

comparison with randomized trial results. Medical Care. 2007 45: 10(2):131-142.

• Dormuth CR, Patrick AR, Shrank WH, Wright JM, Glynn RJ, Sutherland J, Brookhart MA. Statin adherence and risk of accidents: a cautionary tale. Circulation 2009;119:2051-7.

• Rothman KJ. Six Persistent Research Misconceptions. J Gen Intern Med 2014;29(7):1060–4.

• Pearce N. The use of beta agonists and the risk of death and near death from asthma. J Clin Epidemiol 2009;62:582-7.

• Miettinen OS. The need for randomization in the study of intended effects. Stat Med 1983;2(2):267-71.

• Zhang HT, McGrath LJ, Wyss R, Ellis AR, Stürmer T. Controlling confounding by frailty when estimating influenza vaccine effectiveness using predictors of dependency in activities of daily living. Pharmacoepidemiol Drug Saf. 2017 Dec;26(12):1500-1506.

Lesson 5: Study Design Solutions

• New user design

• Active comparators

Required readings:

• Wayne A. Ray. Evaluating Medication Effects Outside of Clinical Trials: New-User Designs. Am J Epidemiol 2003;158:915-20.

• Journal club article: Chang HY, Singh S, Mansour O, Baksh S, Alexander GC. Association Between Sodium-Glucose Cotransporter 2 Inhibitors and Lower ExtremityAmputation Among Patients With Type 2 Diabetes.JAMA Intern Med. 2018 Sep 1;178(9):1190-1198.

Suggested readings:

• Lund JL, Richardson DB, Sturmer T. The Active Comparator, New User Study Design in Pharmacoepidemiology: Historical Foundations and Contemporary Application. Curr Epidemiol Rep 2015;2:221–8.

• Kramer, Lane, Hutchinson. Analgesic use, blood dyscrasias, and case-control pharmacoepidemiology. A critique of the International Agranulocytosis and Aplastic Anemia Study. J Chron Dis 1987;40:1073-85.

Lesson 6: Risk periods

o First treatment carried forward o As treated o Induction, carry-over, lag periods

Required readings:

12

• Rothman KJ. Induction and latent periods. Am. J. Epidemiol. (1981) 114 (2): 253-259

• Journal club article: Azoulay L, Suissa S. Sulfonylureas and the Risks of Cardiovascular Events and Death: A Methodological Meta-Regression Analysis of the Observational Studies. Diabetes Care 2017. 2017 May;40(5):706-714.

Suggested readings:

• Hernán MA, Alonso A, Logan R, et al. Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease. Epidemiology

(Cambridge, Mass.) 2008;19(6):766-779.

• Stampfer MJ. ITT for observational data: worst of both worlds? Epidemiology. 2008 Nov;19(6):783-4

Lesson 7: Propensity scores

Required readings:

• Stürmer T, Schneeweiss S, Brookhart MA, Rothman KJ, Avorn J, Glynn RJ. Analytic strategies to adjust confounding using exposure propensity scores and disease risk scores: Nonsteroidal antiinflammatory drugs and short-term mortality in the elderly. Am J Epidemiol 2005;161:891-8.

• Journal club article: Ripollone JE, Huybrechts KF, Rothman KJ, Ferguson RE, Franklin JM. Implications of the Propensity Score Matching Paradox in Pharmacoepidemiology. Am J Epidemiol. 2018 Sep 1;187(9):1951-1961.

Suggested readings:

• Parsons LS. Reducing bias in a propensity score matched-pair sample using greedy matching techniques, 2001. (http://www2.sas.com/proceedings/sugi26/p214-26.pdf)

• Brookhart MA, Schneeweiss S, Rothman KJ,.Glynn RJ, Avorn J, Stürmer T. Variable selection for propensity score models. Am J Epidemiol 2006;163:1149-1156.

• Stürmer T, Wyss R; Glynn RJ; Brookhart MA. Propensity scores for confounder adjustment when assessing the effects of medical interventions using nonexperimental study designs. Journal of Internal Medicine 2014;275(6):570-80.

• Stürmer T, Joshi M, Glynn RJ, Avorn J, Rothman KJ, Schneeweiss S. A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially

13

different estimates compared with conventional multivariable methods. Journal of Clinical Epidemiology 2006;59:437-47.

Lesson 8: Propensity scores cont.’d & disease risk scores

Required readings:

• Glynn RJ, Gagne JJ, Schneeweiss S. Role of disease risk scores in comparative effectiveness research with emerging therapies. Pharmacoepidemiology and Drug Safety 2012; 21(S2): 138-147.

• Journal club article: Desai RJ, Wyss R, Jin Y, Bohn J, Toh S, Cosgrove A, Kennedy A, Kim J, Kim C, Ouellet-Hellstrom R, Karami S, Major JM, Niman A, Wang SV, Gagne JJ. Extension of Disease Risk Score-Based Confounding Adjustments for Multiple Outcomes of Interest: An Empirical Evaluation. Am J Epidemiol. 2018 Nov 1;187(11):2439-2448.

Suggested readings:

• Arbogast PG, Kaltenbach L, Ding H, Ray WA. Adjustment for multiple cardiovascular risk factors using a summary risk score. Epidemiology 2008; 19: 320-37.

• Arbogast PG, Ray WA. Use of disease risk scores in pharmacoepidemiologic studies. Statistical Methods in Medical Research 2009; 18: 67-80.

• Arbogast PG, Ray WA. Performance of disease risk scores, propensity scores, and traditional multivariable outcome regression in the presence of multiple confounders. AJE 2011.

• Hansen BB. The prognostic analogue of the propensity score. Biometrika 2008; 95(2): 481-488.

• Wyss R, Hansen BB, Ellis AR, Gagne JJ, Desai RJ, Glynn RJ, Stürmer T. The "Dry-Run" Analysis: A Method for Evaluating Risk Scores for Confounding Control. Am J Epidemiol. 2017 May 1;185(9):842-852.

Lesson 9: Non-uniform treatment effects

Required readings:

• Sturmer T, Rothman KJ, Glynn RJ. Insights into different results from different causal contrasts in the presence of effect-measure modification. Pharmcoepidemiol Drug Saf. 2006; 15: 698-709.

• Journal club article: Yoshida K, Hernández-Díaz Sonia, Solomon DH, Jackson JW, Gagne JJ, Glynn RJ, Franklin JM., Jessica M. Matching Weights to Simultaneously Compare Three Treatment Groups:

14

Comparison to Three-way Matching. Epidemiology. 2017 May;28(3):387-395.

Suggested readings:

• Lunt M, Solomon D, Rothman KJ, Glynn R, Hyrich K, Symmons DP, Stürmer T. Different methods of balancing covariates leading to different effect estimates in the presence of effect modification. AJE 2009; 169:909-917

Lesson 10: Instrumental variables

Required readings:

• Brookhart MA, Wang PS, Solomon DH, Schneeweiss S. Evaluating short-term drug effects using a physician-specific prescribing preference as an instrumental variable. Epidemiology 2006;17:268–75.

• Journal club article: Desai, RJ, Mahesri M, Abdia Y, Barberio J, Tong A, Zhang D, Mavros P, Kim SC, Franklin JM. Association of Osteoporosis Medication Use After Hip Fracture With Prevention of Subsequent Nonvertebral Fractures. JAMA Network Open. 2018;1(3):e180826. doi:10.1001/jamanetworkopen.2018.0826

Suggested readings:

• Brookhart MA, Schneeweiss S. Preference-Based Instrumental Variable Methods for the Estimation of Treatment Effects: Assessing Validity and Interpreting Results. Int J Biostat 2007;3:Article 14.

• Rubin DB. The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials. Stat Med 2007;26:20-36.

• Garabedian LF, Chu P, Toh S, Zaslavsky AM, Soumerai SB. Potential Bias of Instrumental Variable Analyses for Observational Comparative Effectiveness Research. Ann Intern Med. 2014;161:131-138. doi:10.7326/M13-1887

• Swanson SA, Hernán MA. Commentary: how to report instrumental variable analyses (suggestions welcome). Epidemiology 2013;24(3):370-4.

• Swanson SA, Hernán MA. Think globally, act globally: An epidemiologist's perspective on instrumental variable estimation. Stat Sci 2014;29(3):371-374.

• Swanson SA, Miller M, Robins JM, Hernán MA. Definition and evaluation of the monotonicity condition for preference-based instruments. Epidemiology 2015;26(3):414-20.

Lesson 11: Validation studies

15

Required readings:

• Setoguchi S, Solomon DH, Glynn RJ, Cook EF, Levin R, Schneeweiss S. Agreement of diagnosis and its date for hematologic malignancies and solid tumors between medicare claims and cancer registry data. Cancer Causes Control 2007;18:561-9.

• Journal club article: Cuthbertson, CC, Kucharska-Newton A, Faurot KR, Stürmer T, Jonsson Funk M, Palta P, Windham B, Thai S, Lund JL. Controlling for Frailty in Pharmacoepidemiologic Studies of Older Adults: Validation of an Existing Medicare Claims-based Algorithm. Epidemiology. 2018. 2018 Jul;29(4):556-561.

Suggested readings:

• Stürmer T, Marquis MA, Zhou H, Meigs JB, Lim S, Blonde L, MacDonald E, Wang R, LaVange LM, Pate V, Buse JB. Cancer incidence among those initiating insulin therapy with glargine versus human NPH insulin. Diabetes Care 2013;36:3517-25.

• Brouwer ES, Napravnik S, Eron JJ, Simpson RJ, Brookhart MA, Stalzer B, Vinikoor M, Moore MF, Stürmer T. Validation of Medicaid claims-based diagnosis of myocardial infarction using an HIV clinical cohort. Medical Care 2014; doi: 10.1097/MLR.0b013e318287d6fd. PMID: 23604043. [Epub ahead of print]

• Wang T, Hong J-L, Gower EW, Pate V, Garg S, Buse JB, Stürmer T. Incretin-based therapies and diabetic retinopathy: real-world evidence in older U.S. adults. Diabetes Care 2018 Sep;41(9):1998-2009.

• Warren JL, Harlan LC, Fahey A, Virnig BA, Freeman JL, Klabunde CN, Cooper GS, Knopf KB. Utility of the SEER-Medicare data to identify chemotherapy use. Med Care 2002;40(8 Suppl):IV-55-61.

• Stürmer T, Glynn RJ, Rothman KJ, Avorn J, Schneeweiss S. Adjustments for unmeasured confounders in pharmacoepidemiologic database studies using external information. Med Care 2007;45(10 Supl 2):S158-65.

• Kiyota Y, Schneeweiss S, Glynn RJ, Cannuscio CC, Avorn J, Solomon DH. Accuracy of Medicare claims-based diagnosis of acute myocardial infarction: estimating positive predictive value on the basis of review of hospital records. Am Heart J 2004;148:99-104.

• Winkelmayer WC, Schneeweiss S, Mogun H, Patrick AR, Avorn J, Solomon DH. Identification of individuals with CKD from Medicare claims data: a validation study. Am J Kidney Dis 2005;46:225-32.

• Gillian CH. Validation of death and suicide recording on the THIN UK primary care database. Pharmacoepidemiol Drug Saf 2008

• Chubak J, Pocabelli G, Weiss NS. Tradeoffs between accuracy measures for electronic health care data algorithms. Journal of Clinical Epidemiology 2012; 65: 343-349.

• Brunelli SM, Gagne JJ, Huybrechts KF, Wang SV, Patrick AR, Rothman KJ, Seeger JD. Estimation using all available covariate

16

information versus a fixed look-back window for dichotomous covariates. Pharmacoepidemiol Drug Saf 2013;22(5):542-50.

• Young JC, Stürmer T, Lund JL, Jonsson Funk M. Predictors of prevalent statin use among older adults identified as statin initiators based on Medicare claims data. Pharmacoepidemiol Drug Saf 2016;25(7):836-43.

• Li X, Stürmer T, Brookhart MA. Evidence of sample use among new users of statins: implications for pharmacoepidemiology. Med Care 2014;52(9):773-80.

Lesson 12: Crystal ball PE, immortal time bias, immeasurable time bias

Required readings:

• Suissa S. Immortal time bias in pharmaco-epidemiology. Am J Epidemiol. 2008 Feb 15;167(4):492-9.

• Journal club article: Suissa S. Lower Risk of Death With SGLT2 Inhibitors in Observational Studies: Real or Bias. Diabetes Care 2018;41:6–10

Suggested readings:

• Suissa S. Immeasurable time bias in observational studies of drug effects on mortality. Am J Epidemiol. 2008 Aug 1;168(3):329-35.

• Suissa S, Azoulay L. Metformin and the Risk of Cancer: Time-related biases in observational studies. Diabetes Care. 2012 December; 35(12): 2665–2673.

• Shintani AK, Girard TD, Arbogast PG, Moons KGM, Ely EW. Immortal time bias in critical care research: application of time-varying Cox regression for observational cohort studies. Crit Care Med. 2009 November; 37(11): 2939–2945

Lesson 13: Adherence and Persistence

Required readings:

• Benner JS, Glynn RJ, Mogun H, Neumann PJ, Weinstein MC, Avorn J. Long-term persistence in use of statin therapy in elderly patients. JAMA. 2002 Jul 24-31;288(4):455-61.

• Journal club article: Brookhart MA, Reams D, Dluzniewski PJ, Kshirsagar A, Walsh L, Bradbury BD. Estimating the Effect of Preventable Treatment Discontinuation on Health Outcomes. Epidemiology. 2018 Jan;29(1):134-141.

17

Suggested readings:

• Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005 Aug 4;353(5):487-97.

• Brookhart MA, Patrick AR, Schneeweiss S, Avorn J, Dormuth C, Shrank W, van Wijk BL, Cadarette SM, Canning CF, Solomon DH. Physician follow-up and provider continuity are associated with long-term medication adherence: a study of the dynamics of statin use. Arch Intern Med. 2007 Apr 23;167(8):847-52.

• Sattler EL, Lee JS, Perri M 3rd. Medication (re)fill adherence measures derived from pharmacy claims data in older Americans: a review of the literature. Drugs Aging. 2013 Jun;30(6):383-99. doi: 10.1007/s40266-013-0074-z.

Lesson 14: Patients treated contrary to prediction

Required readings:

• Stürmer T, Rothman KJ, Avorn J, Glynn RJ. Treatment effects in the presence of unmeasured confounding: Dealing with observations in the tails of the propensity score distribution – a simulation study. American Journal of Epidemiology 2010;172:843-854.

• Journal club article: Li F, Thomas LE, Li F. Addressing Extreme Propensity Scores via the Overlap Weights. Am J Epidemiol. 2019 Jan 1;188(1):250-257.

Suggested Readings:

• Rothman KJ, Gallacher JEJ, Hatch EE. Why representativeness should be avoided. Int J Epidemiol 2013;42:1012–4.

Lesson 15: Variability in treatments & variable selection (including hdPS)

Required readings:

• Myers JA, Rassen JA, Gagne JJ, Huybrechts KF, Schneeweiss S, Rothman KJ, Joffe MM, Glynn RJ. Effects of Adjusting for Instrumental Variables on Bias and Precision of Effect Estimates. Am J Epidemiol 2011;174(11): 1213–1222.

• Journal club article: Karim ME, Pang M, Platt RW. Can We Train Machine Learning Methods to Outperform the High-dimensional Propensity Score Algorithm? Epidemiology. 2018 Mar; 2018 Mar;29(2):191-198.

Suggested readings:

18

• Goodwin JS, Goodwin JM. The tomato effect – rejection of highly efficacious therapies. JAMA 1984;251(18):2388-90.

• Schneeweiss S, Rassen JA, Glynn RJ, Avorn J, Mogun H, Brookhart MA. High-dimensional propensity score adjustment in studies of treatment effects using health care claims data. Epidemiology 2009;20(4):512-522.

• Rassen JA, Glynn RJ, Brookhart MA, Schneeweiss S. Covariate Selection in High-Dimensional Propensity Score Analyses of Treatment Effects in Small Samples. American Journal of Epidemiology 2011;173(12):1404-1413.

Lesson 16: Potentially inappropriate prescribing

Required readings:

• Jirón M, Pate V, Hanson LC, Lund JL, Jonsson Funk M, Stürmer T. Trends in Prevalence and Determinants of Potentially Inappropriate Prescribing in the US 2007 – 2012. J Am Geriat Soc 2016;64(4):788-97.

• Journal club article: Mantri S, Fullard M, Gray SL, Weintraub D, Hubbard RA, Hennessy S, Willis AW. Patterns of Dementia Treatment and Frank Prescribing Errors in Older Adults With Parkinson Disease. JAMA Neurol. 2018 Oct 1. doi: 10.1001/jamaneurol.2018.2820

Suggested readings:

• Hill-Taylor B, Sketris I, Hayden J, Byrne S, O'Sullivan D, Christie R. Application of the STOPP/START criteria: a systematic review of the prevalence of potentially inappropriate prescribing in older adults, and evidence of clinical, humanistic and economic impact. J Clin Pharm Ther. 2013 Oct;38(5):360-72.

• Bradley MC, Motterlini N, Padmanabhan S, Cahir C, Williams T, Fahey T, Hughes CM. Potentially inappropriate prescribing among older people in the United Kingdom. BMC Geriatrics 2014, 14:72

• Jirón M, Pate V, Lund JL, Hanson LC, Jonsson Funk M, Stürmer T. Potentially Inappropriate Medication Use in US older adults 2008 – 2014: Both too much and too little. Manuscript under internal reviews (confidential – do not circulate)

Lesson 17 The opiod epidemic

Required readings:

19

• Journal club article: Larochelle MR, Bernson D, Land T, et al. Medication for Opioid Use Disorder After Nonfatal Opioid Overdose and Association With Mortality A Cohort Study. Ann Intern Med. 2018;169:137-145. doi:10.7326/M17-3107

Lesson 18 Methods review

Required readings:

• Journal club article: Raaschou P, Söderling J, Turesson C, Askling J; ARTIS Study Group.Tumor Necrosis Factor Inhibitors and Cancer Recurrence in Swedish Patients With Rheumatoid Arthritis: A Nationwide Population-Based Cohort Study. Ann Intern Med. 2018 Sep 4;169(5):291-299.

Lesson 19 Practical lessons for a successful career in pharmacoepidemiology

Required readings:

• Journal club article: Wadtröm H, Frisell T, Askling J, ARTIS Study Group.Malignant Neoplasms in Patients With Rheumatoid Arthritis Treated With Tumor Necrosis Factor Inhibitors, Tocilizumab, Abatacept, or Rituximab in Clinical Practice: A Nationwide Cohort Study From Sweden. JAMA Intern Med. 2017 Nov 1;177(11):1605-1612.

Lesson 20 Paper Presention I

Lesson 21 Paper Presention II

Lesson 22 Medications in Pregnancy

Required readings:

• Journal club article: Bateman BT, Heide-Jørgensen U, Einarsdóttir K, et al. β-Blocker Use in Pregnancy and the Risk for Congenital Malformations: An International Cohort Study. Ann Intern Med. 2018 Nov 20;169(10):665-673.

Lesson 23 Methodology

Required readings:

• Journal club article: Richardson DB, Keil AP, Kinlaw AC, Cole SR. Marginal Structural Models for Risk or Prevalence Ratios for a Point

20

Exposure Using a Disease Risk Score. Am J Epidemiol. 2019 Feb 6. doi: 10.1093/aje/kwz025.

Lesson 24 Pragmatic trials

Required readings:

• Journal club article: Hernán MA, Robins JM. Per-Protocol Analyses of Pragmatic Trials. N Engl J Med. 2017 Oct 5;377(14):1391-1398.

Suggested readings:

• Ford I, Norrie J. Pragmatic Trials. N Engl J Med 2016;375:454-63. DOI: 10.1056/NEJMra1510059

Lesson 25 Propensity score trimming

Required readings:

• Journal club article: Yoshida K, Solomon DH, Haneuse S, Kim SC, Patorno E, Tedeschi SK, Lyu H, Franklin JM4, Stürmer T, Hernández-Díaz S, Glynn RJ. Multinomial Extension of Propensity Score Trimming Methods: A Simulation Study. Am J Epidemiol. 2019 Mar 1;188(3):609-616.

Lesson 26 Self-controlled designs

Required readings:

• Journal club article: Bykov K, Mittleman MA, Glynn RJ, Schneeweiss S, Gagne JJ. The Case-Crossover Design for Drug-Drug Interactions: Considerations for Implementation. Epidemiology. 2019 Mar;30(2):204-211.

Lesson 27 Single arm studies with external comparator

Required readings:

• Journal club article:

Lesson 28 Antidepressants and suicide

21

Required readings:

• Journal club article: Brännström J, Lövheim H, Gustafson Y, Nordström P. Association Between Antidepressant Drug Use and Hip Fracture in Older People Before and After Treatment Initiation. JAMA Psychiatry. 2019 Feb 1;76(2):172-179.

Lesson 29 Wrap-up

Required readings:

• Journal club article: Schneeweiss S, Rassen JA, Brown JS, et al. Graphical Depiction of Longitudinal Study Designs in Health Care Databases. Ann Intern Med. 2019 Mar 12. doi: 10.7326/M18-3079.