← All Studio courses Studio · HSCI 230

Evaluating Epidemiological Research

Decks for studio time, one for each lesson

Each deck guides studio time for its lesson: questions answered on students' devices, work in pairs and small groups, peer activities, and discussion as a class. A deck can also be presented without a room, worked through on your own (the recorded walkthroughs are being prepared, and until a deck's walkthrough is recorded its self-study format runs without the voice), or printed as a facilitator guide. All 12 decks are ready.

Lessons

12 of 12 decks ready
L · 01
Foundations of EpidemiologyOne river valley where families on wells near old smelter tailings are said to be sick more often: the turning points and phases of epidemiology, checking a claim of progress and Actor-Network Theory, biopower and the colonial origins of the discipline, the three pillars and five paradigms, the strengths and limits of quantitative science and other ways of knowing, misconduct, questionable practices, manufactured doubt and honest error, the replication crisis and its reforms, and OCAP® and EGAP.
L · 02
Systematic Reviews and Meta-AnalysisOne Newfoundland health authority deciding whether to fund a pharmacist-led blood pressure program: the evidence pyramid as a heuristic, DIKW, narrative and systematic reviews, the seven steps, PROSPERO and GRADE; MeSH, free text, Boolean operators, the search log and the AMSTAR 2 critical domains; inverse-variance weights, the fixed-effect estimate and the choice of model, worked by hand on four trials; the forest plot, Cochran's Q, I² and the methods for explaining heterogeneity; and publication bias, funnel plots, leave-one-out analysis and outcome scales.
L · 03
Introduction to Observational StudiesOne northwestern Ontario city's survey of shift work and type 2 diabetes: descriptive, experimental, quasi-experimental and observational studies, causal weight and the target trial; case reports, case series and surveys; the prevalence ratio and prevalence odds ratio worked by hand and the four design steps; duration, reverse causation and time-invariant exposures; and incidence from prevalence, repeated cross-sections compared with a cohort, and STROBE.
L · 04
Case-Control StudiesOne Saskatchewan health region's case-control study of new-onset asthma and grain handling: the study base and the odds ratio with its Woolf interval; incident cases, the four principles and the sources of controls; the rare-disease assumption, incidence density sampling, matching, STROBE and CASP; case-crossover and self-controlled case-series designs worked by hand; and case-case, case-only, case-cohort and two-stage designs.
L · 05
Cohort StudiesOne northwestern Ontario mill city whose health unit studies its workers and plans a trial with them: the cohort design compared with case-control and its variants; risks, the risk ratio and difference, person-time, rates and the rate ratio worked step by step; exposure scales, induction and lag, comparability, follow-up and losses; trial phases, equipoise, the design effect, allocation and masking; and intention to treat, Bonferroni, vaccine efficacy and CONSORT 2025.
L · 06
Ecological and Group-Level StudiesOne Alberta health zone weighing a well-testing program for its 24 communities: ecologic variants, rationales and designs; aggregate, environmental and global variables, the linear model and the group-level rate ratio worked by hand; the ecological and atomistic fallacies; misclassification, group-level confounding, effect modification and the modifiable areal unit problem; and multilevel and two-phase designs and group-level inference.
L · 07
Conceptualization, Measurement and Causal SpecificationOne central Alberta city whose health unit's adult study will guide a council health fund: constructs, validity evidence and the frameworks that choose what is measured; measurement invariance, self-rated health and ordinal steps worked by hand; reliability, four kinds of error and an attenuated slope; confounders, mediators and colliders, a restriction worked with numbers, and the obesity and birth weight paradoxes; and residual confounding, a lag analysis and simultaneity.
L · 08
Sampling, Selection Processes and External ValidityOne central Manitoba city with a regional hospital and a steel foundry: selection bias as a common effect and the selection bias factor; Berkson's bias in an in-patient table worked by hand; the healthy worker effect and four SMRs; differential attrition and a survey prevalence reweighted by age; Neyman bias through P = I × D and survivorship bias; and transportability by standardization and the three appraisal questions.
L · 09
Information Bias and Data QualityOne northern Alberta town's review of its data before its first community health report: sensitivity, specificity and observed odds ratios under nondifferential and differential error worked cell by cell; recall, social desirability, the mode of asking and the pulse oximeter; observer bias, overdiagnosis, lead time and surveillance bias worked by hand; and regression dilution, heaping and their corrections.
L · 10
Design-Specific and Temporal BiasesOne western Manitoba city whose evidence review committee must judge the studies behind its funding decisions: allocation concealment, blinding and the meta-epidemiologic evidence, with the odds ratio a flawed trial would report; placebo, nocebo and Hawthorne effects, the healthy adherer effect and contamination; immortal time bias, with person-time split at the first fill and a landmark design; lead-time bias, length-biased sampling and overdiagnosis; and time-window bias and age, period and cohort effects in repeated surveys.
L · 11
Confounding & Statistical InferenceOne New Brunswick river city weighing free strength-and-balance classes for older adults in every neighbourhood: the three conditions for a confounder, mediators and colliders, the hormone therapy reversal and a stratified risk ratio worked by hand; confounding by indication and its design solutions, time-varying confounding with inverse probability weights worked by hand, and the debate over adjusting for race and socioeconomic position; misspecification, multicollinearity and variance inflation, Type I and Type II errors, power, p-values and intervals; and Simpson's paradox worked by hand, the ecological and atomistic fallacies, the modifiable areal unit problem and the three missing data mechanisms with a complete-case prevalence.
L · 12
Integrated Appraisal of Epidemiological ResearchOne New Brunswick coastal city weighing a wood stove exchange rebate: the seven decisions of the inferential chain, the causal model step, STROBE, CONSORT 2025 and PRISMA 2020, and reporting compared with methodological quality; the five stages of appraisal, design alignment, selection, measurement and confounding, significance compared with importance and multiple testing worked by hand, and external validity; red flags, quality indicators and the E-value; and conflicting studies, Hill's viewpoints, inverse-variance pooling worked by hand, disciplined skepticism and the Wakefield case.
Students: join studio time. Open the join page and enter the four-letter code on the main screen, or scan the QR code. Join studio time