HRM 751 OBSERVATIONAL RESEARCH METHODS

2016 Course Coordinator: Andrew Mente Instructors: Andrew Mente, John Oudyk, Alexandra J. Mayhew, Sohail Mulla, & Binod Neupane Time: Monday mornings 9:00 am to 12:00 pm Locations: Large group lecture: MDCL 2232 Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
DATE FORMAT TIME LECTURER GR1 GR2 GR3 GR4 CLASS TOPIC Sept. 12 Lecture Tutorial 9-10 10-12 AM AJM JO BN SM 1. Course introduction. Overview of Observational Studies Sept. 19 Lecture Tutorial 9-10 10-12 AM AJM JO BN SM 2. The Research Question & Theoretical Frameworks Sept. 26 Lecture Tutorial 9-10 10-12 AM AJM JO BN SM 3. Study designs for epidemiology ? I Oct. 3 Lecture Tutorial 9-10 10-12 AM AJM JO BN SM 4. Study designs for epidemiology – II Oct. 10 ** Thanksgiving: NO CLASS** Oct. 17 Lecture Tutorial 9-10 10-12 AJM AJM JO BN SM 5. Survey methods, questionnaire design and implementation PROTOCOL ABSTRACT DUE Oct. 24 Lecture Tutorial 9-10 10-12 AJM JO BN SM 6. Conceptualization and measurement Oct. 31 Lecture Tutorial 9-10 10-12 AJM JO BN SM 7. Sampling, sample size and power Nov. 7 STUDENTS PRESENT 9-12 —- AJM JO BN SM 8. Student Protocol Presentations Nov. 14 Lecture Tutorial 9-10 10-12 AJM JO BN SM 9. Analysis of discrete (categorical) outcomes Nov. 21 Lecture Tutorial 9-10 10-12 JO AJM JO BN SM 10. Analysis of continuous outcomes Nov. 28 TEST Tutorial 9-10 10-12 AM AJM JO BN SM 11. In class test and protocol development Dec. 5 Tutorial 9-12 —- AJM JO BN SM 12. Final protocol critiques and reviews Dec. 12 — — — — — — — 13. FINAL COURSE PAPER DUE
OBJECTIVES The course introduces students to basic concepts and methods used in observational (non- experimental) studies to conduct needs assessments (e.g., prevalence of disease or disorder), to understand the determinants of health (e.g., association between independent/exposure variables and dependent/outcome variables in analytic research) and to assess the impact of interventions implemented to improve health or alter life quality (e.g., non-experimental program evaluations). The topics will focus on three broad areas: (1) the formulation of research questions and use of theory to explicate the relationships among the core variables of interest; (2) the basic elements and options for research studies: sampling, measurement and analysis; and (3) the identification and control of error.
2
COURSE CONTENT The readings for the course draw from primary literature and electronic text books available through the McMaster library subscriptions. References and links to the weekly required readings are provided on Avenue. Each week, supplementary readings that provide additional explanations of concepts, or expand on topics in the required readings will be listed under the heading ?Optional Readings.?
EDUCATIONAL METHODS (1) There will be 10 lectures that present core concepts and issues to all students. The lectures will be about 50 minutes each, at fixed times throughout the course in a central location. Some time will be set aside for large-group questions/discussions.
(2) There will be small group tutorials to expand on/explore the material presented during the lectures. Students are expected to read the book chapters/articles identified for each unit and to identify questions and issues for discussion. It is anticipated that from 25-50% of tutorial time will be taken up with student-initiated material. Each tutorial will also have 1-3 assignments. It is anticipated that the presentation and discussion of each assignment should take about 20-30 minutes each.
(3) Each student will develop a research protocol in the general form of a written grant application to conduct an observational (non-experimental) study. The study will address a research question of interest to the student and include: a problem statement/formulation; brief review of relevant literature with special emphasis on theoretical framework and presumed mechanisms of effect; a description of the research design, including sampling, measurement, data collection and analysis; discussion of the threats to validity, strategies to control error, risks to subjects and ethics. By Class #5 (October 17, 2016), students will choose the question they wish to address and e-mail a half-page (not longer) description of the question to their tutor for clearance. Class #8 (Nov. 7, 2016) will be devoted to student presentations and discussions of their protocols. The idea is to present the protocol in point form (5-10 minutes), allowing 10-15 minutes discussion to identify and resolve methodological issues associated with the proposal. One student selected at random will serve as respondent and lead the discussion. This same student, in point form, will write a one page critical review of the project, summarizing methodological and substantive issues raised during the discussion. This critical review will be given to both the tutor and student presenter the following week (Class #9, November 14, 2016).
EVALUATION OF STUDENTS The evaluation of students (grades) will be divided into 4 components:
(1) Written protocol. Described in the previous paragraph, the written protocol will constitute 60% of the grade and will be marked by your tutor. The protocol will be from 15-20 (maximum) pages, double spaced, excluding references, figures, tables, and should take the general form of an operating grant application to CIHR. Please use Times Roman 12 point or Arial 11 point, with 2 cm margins.
(2) Tutorial: There are two ratings: i) participation/contribution (seeks clarification, asks good questions, facilitates participation of others, etc.) and ii) comprehension (understanding of key methodologic principles and concepts). Both components are rated on a 7-point likert scale. We feel that we will obtain a better assessment of your participation/contribution and comprehension by having each faculty member rate your weekly participation. The tutors will rate your participation each week. The in-class participation/contribution and comprehension ratings will constitute 10% of the final HRM 751 grade.
3
Scoring Scheme: The two components (Participation/Contribution and Comprehension) will carry equal weight. The first rating bears on your attendance and (appropriate) level of contributions to class discussion. The second rating bears on levels of understanding and knowledge inferred by the tutor from your verbal contribution to class discussions. Each component is scored out of 7, so your highest participating rating in a given week is 14. For example, if you receive a 6 (out of 7) for participation/contribution and a 7 (out of 7) for comprehension, your mark for that week would be 13. The same rating system is used every week. Your final participation mark is simply the sum of all your weekly participation marks divided by the highest possible participation marks one can achieve in those weeks. For example, a student attends the 11 weekly tutorials and gets the following weekly ratings (each out of 14): 13, 12, 13, 14, 10, 11, 14, 9, 13, 12, and 14. The sum of these ratings is 135. The highest possible rating is 14 marks x 11 wks= 154. Thus, this student’s participation is 135 / 154 = 87.7%. Because participation accounts for 10% of your grade, this student’s final participation mark is 8.8 out of 10.
(3) Critical written review in point form and no more than one page in length single spaced of student protocol presented in Class #8. This review will be worth 10% of the grade. This review is due during Class #9.
(4) In-class test. There will be one in-class test that will consist of multiple choice and/or short response questions. This will take place during Class #12. The test will constitute 20% of the grade.
EVALUATION OF COURSE AND TUTORS This course has been revised substantially from earlier years, and it will be necessary for students and tutors to evaluate the course and unit objectives, along with the usefulness of the lectures, readings, assignments and tests for achieving these objectives. At the end of each class, 5-10 minutes will be set aside for students to make verbal recommendations for changing the unit. Evaluation forms (multiple choice ratings, with a space for comments) will be completed on avenue. These will be completed anonymously, collated/analysed at the end of the course.
4
SCHEDULING AND TOPICS
Class Date Topic Class 1 Sept 12, 2016 Course introduction, Overview of Observational studies Class 2 Sept 19, 2016 The Research Question & Theoretical Frameworks Class 3 Sept 26, 2016 Study designs for epidemiology?I Cohort and case-control studies Class 4 Oct 3, 2016 Study designs for epidemiology?II Cross-sectional surveys, ecological & hybrid studies Class 5 Oct 17, 2016 Survey methods, questionnaire design and implementation Protocol abstract due Class 6 Oct 24, 2016 Conceptualization and measurement Class 7 Oct 31, 2016 Sampling, sample size and power Class 8 Nov 7, 2016 Student protocol presentations Class 9 Nov 14, 2016 Analysis of discrete (categorical) outcomes Class 10 Nov 21, 2016 Analysis of continuous outcomes Class 11 Nov 28, 2016 Protocol development and critical review of a research proposal In-class test Class 12 Dec 5, 2016 Final protocol critiques and reviews Protocol critique due No class Dec 12, 2016 Final protocol due
5
CLASS 1: Sept 12, 2016, INTRODUCTION: OVERVIEW OF OBSERVATIONAL STUDIES & CONCEPTS
Lecture: 9:00 am – 10:00 am ? MDCL 2232 COURSE INTRODUCTION, LOGIC OF INQUIRY: THEORY, CAUSATION, MEDIATORS AND MODERATORS
To Learn About 1. Objectives for observational studies 2. The attributes of answerable questions/hypotheses 3. Concepts of causality and causal relationships 4. Basic types of observational studies 5. Briefly review confounding and bias
Key Concepts Descriptive studies Analytic studies Study question
Hypothesis testing Independent variables Dependent variables Confounding variables
Research design Correlation vs. causation Validity Bias
Required Readings
1. Ahrens W, Kirckberg K, Pigeot I. (2005). An Introduction to Epidemiology (pages 1-6). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
2. Mann CJ. Observational research methods. Research design II: cohort, cross-sectional, and case-control studies. Emerg Med J. 2003;20:45-60. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ 12533370
3. Grimes, D.A., & Schulz, K.F. (2002). Bias and causal associations in observational research. The Lancet, 359, January 19, 248-252. http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/?term=11812579
4. Course handout ?Confounders and Mediators example?
Optional Readings Course handout ?Common Priority Areas for Funding Agencies?
Wild P. (2005). Design and Planning of Epidemiological Studies (pages 465-469). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
Pearce N, Greenland S. Confounding and Interaction (pages 371-398). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
Kestenbaum A. (2009). Chapter 11: Effect Modification (pages 113-120). In Adeney KL, Weiss NS (Eds),
6
Epidemiology and Biostatistics: An introduction to Clinical Research. Seattle: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-0-387-88433-2
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
1. Class discussion: issues/questions arising from the readings
2. Please think of a question which can be answered using observational methods, and which you are contemplating for this course. To help focus your question, it may be helpful to review the following two websites:
http://guides.mclibrary.duke.edu/ebmtutorial
http://www.uic.edu/depts/lib/lhsp/resources/pico.shtml
Note that the above approach works best for phrasing a clinical question, and that you are free to choose a non-clinical question (as long as it is relevant to health). Be prepared to identify the patient/population, intervention/exposure, comparison, and outcome components of your question (eg, apply the PICOT method).
3. Consider which observational study design would be best suited for your question. In doing so, provide specific reasons why alternative designs are not suitable.
4. Consider in what circumstances observational studies are done in preference to randomized clinical trials.
7
CLASS 2: Sept. 19, 2016, THE RESEARCH QUESTION & THEORETICAL FRAMEWORKS
To Learn About 1. Identifying and framing a good research question. 2. Various theoretical frameworks and mechanisms of effect. 3. The use of theoretical frameworks to help guide data analysis.
Key Concepts Framing a good research question Sufficient cause Component cause
Strength of effects Interaction Proportion of disease due to specific causes Induction period
Philosophy of science Bradford Hill criteria for causation
Lecture: 9:00 am -10:00 am ? MDCL 2232 THEORETICAL FRAMEWORKS
Required Readings
1. Thabane L, Thomas T, Ye C, Paul J. Posing the research question: not so simple. Can J Anaesth 2009;56:71-79. http://libaccess.mcmaster.ca/login?url=http://www.springerlink.com/content/57642r8jt635x243/
2. Brazil K, Ozer E, Cloutier MM, et al. From theory to practice: Improving the impact of health services research. BMC Health Serv. Res. 2005;5(1):1. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ 15638931
3. Rothman KJ, Greenland S. (2005). Chapters 1.1 and 1.2 (pages 45-57). Basic Concepts. In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
Optional Readings
Brown AF, Ettner SL, Piette J, Weinberger M, Gregg E, Shapiro MF, Karter AJ, Safford M, Waitzfelder B, Prata PA, Beckles GL. Socioeconomic position and health among persons with diabetes mellitus: a conceptual framework and review of the literature. Epidemiol Rev 2004;26:63-77. (Sample article showing development of a theoretical framework) http://libaccess.mcmaster.ca/login?url=http://epirev.oxfordjournals.org/cgi/reprint/26/1/63
Course handout. Causal criteria summary. (Nice brief summary table of the Bradford Hill guidelines for assessing causation)
8
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
1. Class discussion: issues/questions arising from the lecture and readings. 2. Review the question you developed for session 1 and consider whether it adheres to the FINER guildeline. 3. Develop a theoretical framework related to your protocol question. Prepare a draft section explaining your theoretical framework for inclusion in your final project.
9
CLASS 3: Sept 26, 2016, STUDY DESIGNS FOR EPIDEMIOLOGY ? I
To Learn About 1. The design of cohort and case-control studies 2. Advantages and disadvantages of each study design 3. Types of research questions that are suitable for each design
Key Concepts Temporality Study population
Exposure status Outcome status Confounding
Matching
Lecture: 9:00 am -10:00 am ? MDCL 2232 COHORT AND CASE-CONTROL STUDIES
Required Readings
1.Miller AB, Goff DC, Bammann K, Wild P. (2005) Cohort Studies (pages 253-286). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
2. Kestenbaum A. (2009). Chapter 6: Case-control studies (pages 45-57). In Adeney KL, Weiss NS (Eds), Epidemiology and Biostatistics: An introduction to Clinical Research. Seattle: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-0-387-88433-2
3. Bland JM, Altman DG. Matching. British Medical Journal. 1994;309(6962):1128 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2541945
Optional Readings
Grimes DA, Schulz KF. Cohort studies: marching towards outcomes. Lancet. 2002;359(9303):341-5. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ ?term=cohort+marching+towards+outcomes
Schulz KF, Grimes DA. Case-control studies: research in reverse. Lancet. 2002;359(9304):431-4. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ 11844534
Grimes DA, Schulz KF. Compared to what? Finding controls for case-control studies. Lancet. 2005;365(9468):1429-33. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ 15836892
Grimes DA, Schulz KF. Making sense of odds and odds ratios. Obstet Gynecol. 2008;111 (2 Pt 1):423-6. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ 18238982
Mamdani M, Sykora K, Li P et al. Reader?s guide to critical appraisal of cohort studies: 2. Assessing
10
potential for confounding. BMJ. 2005; 330; 960-2. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ ?term=Reader%E2%80%99s+guide+to+critical+appraisal+of+cohort+studies%3A+2.+Assessing+potential+ for+confounding
Kestenbaum A. (2009). Chapter 10: Methods to Control for Confounding (pages 101-111). In Adeney KL, Weiss NS (Eds), Epidemiology and Biostatistics: An introduction to Clinical Research. Seattle: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-0-387-88433-2
Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, McQueen M, Budaj A, Pais P, Varigos J, Lisheng L; INTERHEART Study Investigators. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004;364:937- 52. (NOTE: Sample article of a famous case-control study) http://libaccess.mcmaster.ca/login?url=http://www.thelancet.com/journals/lancet/article/PIIS0140- 6736%2804%2917018-9/fulltext
Kannel, W.B., Dawber, T.R., Kagan, A., Revotskie, N. & Stokes, J. (1961). Factors of risk in the development of coronary heart disease ? six-year follow-up Experience. The Framingham Study. Annals of Internal Medicine, 55(1), 33-50. (NOTE: Sample article of a famous cohort study). http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ ?term=Factors+of+risk+in+the+development+of+coronary+heart+disease+%E2%80%93+six-year+follow- up+Experience.+The+Framingham+Study
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
1. Class discussion: issues/questions arising from the readings.
2. Consider your protocol question (or a different question of interest) and develop a list of advantages and disadvantages associated with developing i) a cohort study and ii) a case-control study to answer this question. Consider the issues of feasibility, cost, prevalence versus incidence, and ethics.
3. Consider your protocol question and develop a list of potential confounding variables. Can you measure these variables? What methods would you consider using to control for these potential confounders?
4. If you are proposing a cohort study for your protocol, start drafting a section for your final paper that considers inclusion criteria for your cohort and how you will measure of exposure and outcome status. Give some thought to how you can maximize complete follow-up.
5. If you are proposing a case-control study for your protocol, start drafting a section for your final paper that considers how you will identify your cases and what population will serve as appropriate controls. What ratio of cases to controls will you use?
11
CLASS 4: Oct. 3, 2016, STUDY DESIGNS FOR EPIDEMIOLOGY ? II
Lecture: 9:00 am -10:00 am ? MDCL 2232 CROSS-SECTIONAL SURVEYS, ECOLOGICAL AND HYBRID STUDIES
Required Readings
1. Kestenbaum A. (2009). Chapter 4. Cross-sectional Studies (pages 29-31). In Adeney KL, Weiss NS (Eds), Epidemiology and Biostatistics: An introduction to Clinical Research. Seattle: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-0-387-88433-2
2. Kass PH, Gold EB. (2005) Modern Epidemiologic Study Designs (pages 321-333). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
3. Parkin DM, Bray FI. (2005) Ecological Studies (pages 176-184). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
4. London School of Economics. FAQ 8: When is it good to use a longitudinal design? Available at: http://www.lse.ac.uk/media@lse/research/EUKidsOnline/BestPracticeGuide/FAQ/FAQ-8.pdf
5. Choi, B.C., & Noseworthy, A.L. (1992). Classification, direction, and prevention of bias in epidemiologic research. Journal of Occupational Medicine. 34(3), 265-271. http://circ.ahajournals.org/cgi/content/extract/108/17/2154
Optional Readings
Cook PF. Scientific inquiry. Study designs for program evaluation: how do we know what works? J Spec Pediatr Nurs. 2009;14(1):70-72. http://onlinelibrary.wiley.com.libaccess.lib.mcmaster.ca/doi/10.1111/j.1744- 6155.2008.00177.x/abstract;jsessionid=C7E26E2ED696D2C570C83A2381CFD4AC.d02t03
World Health Organization. Module 4: How to evaluate the programme. Available at http://www.who.int/roadsafety/projects/manuals/alcohol/4-How%20to.pdf
Redelmeier, D.A., & Tibshirani, R.J. (1997). Association between cellular-telephone calls and motor vehicle collisions. New England Journal of Medicine, 336, (7), 453-458.) http://content.nejm.org/cgi/content/short/336/7/453 (NOTE: Sample article of a case-crossover design)
Martinez-Sanchez JM, Blanch C, Fu M, et al. Do smoke-free policies in work and public places increase smoking in private venues? Tob Control. 2013; Apr 25 [Epub ahead of print].[NOTE: Sample article of an ecological study] http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1136/tobaccocontrol- 2012-050877
Anand SS, Yusuf S, Jacobs R, Davis AD, Yi Q, Gerstein H, Montague PA, Lonn E. Risk factors, atherosclerosis, and cardiovascular disease among Aboriginal people in Canada: the Study of Health
12
Assessment and Risk Evaluation in Aboriginal Peoples (SHARE-AP).Lancet. 2001 ;358(9288):1147-53. (NOTE: Sample article of a famous cross-sectional study) http://libaccess.mcmaster.ca/login?url=http://www.thelancet.com/journals/lancet/article/PIIS0140- 6736%2801%2906255-9/fulltext
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
1. Class discussion: issues/questions arising from the readings.
2. Consider the types of bias outlined by Choi (1992). Which types of bias are you concerned about for your study? What effect would do you expect they would have on your results? Consider how you could prevent or minimize these biases, or how you could control for them in your analysis. Draft a section for your final paper that addresses these issues.
13
CLASS 5: Oct 17, 2016, SURVEY METHODS, QUESTIONNAIRE DESIGN AND IMPLEMENTATION
To Learn About 1. Advantages, disadvantages and trade-offs among various survey methods. 2. Questionnaire design and implementation. 3. Approaches to maximize response rates. 4. Approaches to minimize bias.
Key Concepts Questionnaire design Types of interviews (eg.
Face-to-face, telephone, etc.) Question development and design
Steps of survey implementation Questionnaire bias
Lecture: 9:00 am -10:00 am ? MDCL 2232 SURVEY METHODS
Required Readings
1. Van den Broeck J, Chhagan M, Kauchali S.(2013). Chapter 18. Questionnaires (pages 357-378). In Van den Broeck J, Brestoff JR (Eds) Epidemiology: Principles and Practical Guidelines. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-94-007-5989-3
2. Cordier S, Steward PA. (2005) Exposure Assessment (pages 438-445). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
Optional Readings
Choi, B.C., & Pak, A.W. (2005). A catalog of biases in questionnaires. Preventing Chronic Diseases, 2(1), A13. http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1323316&blobtype=pdf (Note: Not mandatory; Read only if using a questionnaire in your study)
Boynton, P.M., & Greenhalgh, T. (2004). Selecting, designing, and developing your questionnaire. BMJ, 328: 1312-5 http://www.bmj.com/cgi/content/full/328/7451/1312 (Note: Nice reference; read if using a questionnaire)
Boynton, P.M. (2004). Administering, analysing, and reporting your questionnaire. BMJ, 328: 1372-5. http://www.bmj.com/cgi/content/full/328/7452/1372 (Note: Nice reference; read if using a questionnaire)
Dillman, D.A.(2006). Chapter 7. Implementation Procedures (pages 149-193). In D.A. Dillman, , Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method. (2nd ed, ). New York: John Wiley & Sons, Inc. http://libaccess.mcmaster.ca/login?url=http://www.MCMU.eblib.com/EBLWeb/patron?target=patron&e xtendedid=P_281571_0&
Dillman, D.A.(2006). Internet and Interactive Voice Response Surveys ( pages 352-412). In D.A. Dillman, Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method. (2nd ed). New York: John Wiley &
14
Sons, Inc. http://libaccess.mcmaster.ca/login?url=http://www.MCMU.eblib.com/EBLWeb/patron?target=patron&e xtendedid=P_281571_0& (Note: Not mandatory; Read only if using an internet or interactive voice response survey in your study)
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
** Protocol Abstract Due**
1.Class discussion: issues/questions arising from the readings.
2. Consider whether you will need to use any form of a questionnaire for your study. If so, think about whether there is an existing questionnaire that is suitable or if you?ll need to create a questionnaire. Draft a section for your final paper that addresses these issues, as well as details of survey administration and data collection. Be prepared to discuss the strengths and limitations of various survey methods (i.e. face-to-face, telephone, internet, and postal surveys) and defend your choice.
15
CLASS 6: Oct 24, 2016, CONCEPTUALIZATION & MEASUREMENT
To Learn About 1. Levels of measurement 2. Types, sources, and he impact of measurement error on statistical estimates and associations 3. Assessing reliability 4. Assessing validity 5. Relating measurement objectives to measurement adequacy
Key Concepts Levels of measurement Objectives of measurement Measurement error
Types of reliability (eg. Test- retest) Responsiveness Sensitivity to change
Types of validity (eg. Content) Statistics used for reliability and validity
Lecture: 9:00 am -10:00 am ? MDCL 2232 SELECTING CONCEPTS, CHOOSING/DEVELOPING MEASURES, RELIABILITY AND VALIDITY
Required Readings
1. Streiner, D.L., & Norman, G.R. (2008). Chapter 8. Reliability. In: D.L. Streiner, G.R. Norman, Health measurement scales: a practical guide to their development and use (4th ed,). Oxford: Oxford University Press. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1093/acprof:oso/9780199231881.001.0001
2. Streiner, D.L., & Norman, G.R. (2008). Chapter 10. Validity. In: D.L. Streiner, G.R. Norman, Health measurement scales: a practical guide to their development and use (4th ed,). Oxford: Oxford University Press. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1093/acprof:oso/9780199231881.001.0001
3. Coggon D, Rose G, Barker DJP. 4. Measurement error and bias. In Epidemiology for the uninitiated. (4th ed). BMJ. http://www.bmj.com/about-bmj/resources-readers/publications/epidemiology-uninitiated/4- measurement-error-and-bias Optional Readings
Taylor, R., Reeves, B., Mears, R., Keast, J., Binns, S., Ewings, P., & Khan, K. (2001). Development and validation of a questionnaire to evaluate the effectiveness of evidence-based practice teaching. Medical Education. 35(6), 544-547. http://www.ncbi.nlm.nih.gov/pubmed/11380856
16
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
1. Class discussion: issues/questions arising from the readings.
2. You have a 12-item rating scale to measure depressed mood in adults. Each item (e.g., how much of the time do you feel down in the dumps or blue?) has 5 response options (all of the time, most of the time, some of the time, a little of the time, and none of the time). You would like to use the scale for multiple purposes: estimation, measuring change, screening. Discuss some approaches to evaluating it for these purposes.
3. Consider the scales and questionnaires that will be used in your study. Has their reliability and validity been established in a population similar to yours? If not, think through what might be required to establish reliability and validity. Refer to the article: Development and validation of a questionnaire to evaluate the effectiveness of evidence-based practice teaching by R. Taylor and colleagues, Medical Education, 35(6), 544-547, 2001 for an example of the establishing validity and reliability
17
CLASS 7: Oct 31, 2016, SAMPLING, SAMPLE SIZE AND POWER
To Learn About 1. Basic approaches and concepts applicable to sampling research subjects 2. Advantages, disadvantages and trade-offs among sampling strategies in cost, complexity and accuracy 3. Choosing appropriate sampling strategies for research questions and study designs 4. Issues underlying and approaches to sampling elusive populations 5. Evaluating and adjusting for nonresponse and subject attrition in research studies 6. Methods for estimating the number of subjects needed to meet statistical power requirements to answering research questions 7. The effects on statistical power of sample loss, matching, adjustment for confounding and the assessment of effect modification
Key Concepts Sampling unit Sampling frame Types of sampling strategies (eg. Probability, quota, etc.)
Matching Precision of estimate Stratified/ matched sampling Correlation Risk
Prevalence Incidence Sample size and statistical power
Lecture: 9:00 am -10:00 am ? MDCL 2232 SAMPLING: BASIC CONCEPTS
Required Readings 1. Van den Broeck J, Fossgard Sandoy I, Brestoff, JR. (2013).Chapter 9 The Recruitment, Sampling, and Enrollment Plan (pages 171-196). In Van den Broeck J, Brestoff JR (Eds) Epidemiology: Principles and Practical Guidelines. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-94-007-5989- 3
2. Brestoff JR, Van den Broeck. (2013). Chapter 7 Study Size Planning (pages 137-155). In Van den Broeck J, Brestoff JR (Eds) Epidemiology: Principles and Practical Guidelines. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/978-94-007-5989-3
3. Norman G. Sample size calculations: should the emperor?s clothes be off the peg or made to measure? BMJ. 2012;345:e5278. http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1136/bmj.e5278
Optional Readings Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159. http://journals2.scholarsportal.info/details.xqy?uri=/00332909/v112i0001/155_app.xml
Elashoff JD, Lemeshow S. (2005) Sample Size Determination in Epidemiologic Studies (pages 559-594). In Ahrens W, Pigeot I (Eds.) Handbook of Epidemiology. Germany: Springer. [NOTE: Excellent working examples of calculations specific to study designs and analyses] http://libaccess.mcmaster.ca/login?url=http://dx.doi.org/10.1007/b137839
Hoenig JM, Heisey DM. The Abuse of Power: The Pervasive Fallacy of Power Calculation for Data Analysis. Am Stat. 2001;55(1):19-24.[NOTE: Interesting article about the pitfalls of post hoc power analyses] http://libaccess.mcmaster.ca/login?url=http://journals2.scholarsportal.info/details.xqy?uri=/00031305/v
18
55i0001/19_taoptpopcfda.xml
Kamangar F, Islami F. Sample Size Calculation for Epidemiologic Studies: Principles and Methods. Arch Iran Med. 2013;16(5):295-300.[NOTE: Nice overview of sample size methodology] http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov/pubmed/23641744
Naing L, Winn T, Rusli BN. Practical Issues in Calculating the Sample Size for Prevalence Studies. Arch Orofac Sci. 2006;1:9-14. http://libaccess.mcmaster.ca/login?url=http://www.dental.usm.my/aos/index.html
Sample size calculations printout, from St. George?s Hospital Medical School. (NOTE: Very useful and practical guide for doing your sample size/power calculations) http://www- users.york.ac.uk/~mb55/guide/size.htm
Schafer JL. Multiple Imputation: a primer. Stat Methods Med Res. 1999;8(1):3-15. http://libaccess.mcmaster.ca/login?url=http://www.ncbi.nlm.nih.gov.libaccess.lib.mcmaster.ca/pubmed/ 10347857
Useful websites and online statistical calculators
Department of Biostatistics. Vanderbilt University. PS: Power and Sample Size Calculation. http://biostat.mc.vanderbilt.edu/wiki/Main/PowerSampleSize
Lenth, RV. (2006-9). Java Applets for Power and Sample Size. http://homepage.stat.uiowa.edu/~rlenth/Power/
www.statpages.org/
For the tutorial: 10:00 am – 12:00 pm Tutorial groups: Group 1: HSC 1J10; Group 2: HSC 1J7; Group 3: HSC 1J8; Group 4: HSC 4N55A
1. Class discussion: issues/questions arising from the readings.
2. Draft a section for your final paper that addresses your sampling strategy and come prepared to discuss this during the tutorial. Consider how you will define the sampling frame, and the sampling unit. How will you recruit your study participants? Identify the strengths and weaknesses of your approach and consider what types of biases might be introduced.
3. In evaluating the effectiveness of some programs, it is often difficult, if not impossible, to use an experimental design. Discuss briefly the general approach to sampling (and design) that you might take to evaluate ?modern? neonatal intensive care for sick newborns. How would you deal with issues of generalizability?different newborns, places and times.
4. Draft a section for your final paper that describes how you estimated the sample size requirements for your study and be prepared to discuss in class. To help guide you, consider the type of comparison you are proposing and whether your outcome will be measured as a continuous, dichotomous or categorical variable. Think about where you will find the parameters necessary for your calculation and whether or not these sources are appropriate to inform your sample size calculation. If you are