Lecture 8 Epidemiology GPH-GU 2106 Fall 2024 PDF

Lecture 8 Epidemiology GPH-GU 2106 - Fall 2024 Lecture 8 – RCTs 1 Lecture 8 Agenda To describe the different types of RCTs. To understand the key components of RCTs. 2 1 ...

Lecture 8 Epidemiology GPH-GU 2106 - Fall 2024 Lecture 8 – RCTs 1 Lecture 8 Agenda To describe the different types of RCTs. To understand the key components of RCTs. 2 1  Lecture 8 Epidemiologic Study Design Typology l Based on design l Based on Unit of Analysis l Observational l Individuals l Cross-Sectional l Groups/Ecologic Studies l Cohort l Case-Control l Experimental l Clinical Trials l Field Trials 3 What is a randomized trial? l A planned experiment among humans where: l the investigator randomly assigns participants or groups to either an intervention or control condition l Intervention efficacy is evaluated by comparing outcomes among those who receive the intervention to those receiving a control therapy/intervention 4 2  Lecture 8 Types of randomized trials 1. Natural 2. Community 3. Cluster experiment trials randomized trials l levels of exposure to a l experimental study l clusters (e.g. individuals presumed causative where one group of in communities, agent differ in the community/ies receives hospitals, or other population in a way that an intervention and aggregates of individuals) is relatively unaffected another community/ies are randomized and all by other, extraneous does not consenting persons factors such that the enrolled situation resembles a planned trial l Classic example: Water l Classic example: l Classic example: John fluoridation trials influenza vaccination in Snow – Cholera and the some communities and comparing dental caries Broad Street Pump not in others to assess in Grand Rapids, MI (intervention) vs. herd immunity. Muskegon, MI (control) 5 When to choose a community or cluster randomized trial l Nature of the intervention – is it a mass media campaign? Or is it a population-level interventions? l Acceptability and reduced stigma – can everyone gets the same treatment within a cluster? l Are there enough sub-groups within clusters – can you get data on these subgroups for extra analysis? l Despite the above, community and cluster trials allow for assessment of population-level effects (Population Attributable Fractions!!) 6 3  Lecture 8 Types of randomized trials (cont.) 4. Individual-level randomization – randomize eligible individuals to an intervention (treatment) or a control/placebo/standard of care condition l Key characteristics: l Randomization l Blinding l Control/placebo group vs. “controlled” trial 7 FDA/WHO classification of RCTs l Phase I: Initial studies to determine metabolism, pharmacologic actions, and safety of drug in humans l Phase II: Controlled clinical studies evaluating preliminary efficacy of drug in patients with disease , determine common short-term side effects l Phase III: Expanded trials for gathering additional information on overall benefit-risk relationship of the drug [Needed for FDA approval] l Phase IV: post-marketing trials in general population 8 4  Lecture 8 1. Randomization l Method of assigning participants to different trial arms l Chance should be the only factor that determines group assignment l Neither participant or investigator know in advance which arm participant will be randomized to l Want to avoid bias in assignment to trial arm that may be introduced by the investigator l Ensures that intervention and control groups “look alike” with respect to all other factors except for the treatment at the time of enrollment l Some examples of randomization (aka random allocation): l Random number table; computer generated programs; sealed envelopes with randomization info l Some examples of nonrandom allocation: l Alternate assignment of treatments; Assignment by day of the week 9 Random sampling vs Randomization Random sampling Study sample Population Randomization Experimental group Control group Study participants Population l Random sampling ensures generalizability of survey results. l Randomization in experiments ensures comparability of the experimental group and the control group when the pool of study participants is large. 10 5  Lecture 8 Table 2: Baseline sociodemographic, drug use and health care characteristics of STRIVE participants by trial arm Intervention Control arm arm Characteristic Total (n=418) PMI (n=222) VDI (n=196) p-value % (n) % (n) % (n) Site 47% (195) 50% (111) 43% (84) 0.343 29% (119) 27% (59) 31% (60) 25% (104) 23% (52) 27% (52) Age; Mean (SD) 26.6 (3.9) 26.7 (3.9) 26.4 (4.0) 0.434 Gender Male 76% (316) 76%(167) 76% (149) 0.971 Female 24% (98) 24% (52) 24% (46) Race/Ethnicity Black or African American 7% (28) 7% (15) 7% (13) 0.843 Hispanic/Latino 27% (111) 25% (56) 28% (55) White 57% (237) 57% (126) 57% (111) Other (Asian, American Indian, 10% (42) 11% (25) 9% (17) Mixed, Other) Frequency of alcohol use (L3M) None 30% (125) 31% (68) 29% (57) 0.874 Daily (everyday) 10% (40) 11% (23) 9% (17) Weekly (1 – 6 days/week) 32% (132) 31% (67) 34% (65) Monthly (1 – 3 days/month) 28% (116 28% (61) 28% (55) 11 When is it unethical to randomize? l An effective treatment already exists: l e.g., in trials of therapies to prevent mother-to-child HIV transmission, cannot randomize mothers to a placebo treatment – need to provide standard of care l Personal choice - cannot randomize to different interventions l e.g., trials of different types of contraceptives (pill vs IUD), are ethically questionable because women have the right to select a method of their choice. (Can randomize within method type e.g., pill A vs pill B) l Risks of new treatment likely to exceed risks of existing treatment…we’ll come to this again when we discuss equipoise 12 6  Lecture 8 2. Blinding l Randomization to intervention or control condition is not apparent to everyone involved in the study during the course of the trial l Blinding is used to: l avoid bias in (1) enrollment, (2) during trial, (3) follow-up l 3 levels of blinding: l Single blind - participants are blinded but investigators are aware of who is intervention and control arm l Double blind - neither participants nor investigators know who is receiving the intervention l Triple blind – participants and investigators don’t know intervention assignment. And also, data analyses are conducted in a manner that is removed from the investigators 13 3. “Control” condition l The control condition provides a comparison arm by which the investigator gauges the effect of the treatment l Controls may receive no treatment (e.g., placebo) if no standard of care is currently available l If there is an established standard of care it would be unethical to withhold this from controls l In these cases, standard of care becomes the reference control l E.g. MIRA trial – “…women [in comparison arm] receiving a state-of- the-art HIV prevention package (including risk reduction counseling, voluntary counseling and testing, and diagnosis and treatment of STIs)…” (Padian et al, 2007) 14 7  Lecture 8 4. “Controlled” trial l RCT = randomized CONTROLLED trial not randomized control trial. “Controlled” refers to: a. Pre-specified hypotheses b. Established inclusion/exclusion criteria c. Baseline assessments d. Primary and secondary endpoints (e.g., behavioral change, HIV/HCV incidence) to evaluate hypotheses e. Carefully detailed study protocols (enrollment, treatment delivery and follow up) f. Rigorous monitoring of treatment and outcomes g. Monitoring retention & loss to follow-up h. Ethical considerations i. Analysis plans and stopping rules 15 4a. Pre-specified hypotheses l All trials are based on a large body of epidemiological, clinical, behavioral, evidence supporting the need for therapy/intervention l Questions to consider prior to trial: l What effect is expected from the intervention/therapy? l How ”much” of an effect is expected? l What are the adverse events anticipated? l How much time needed to detect outcomes in trial arms? l What is the required sample size at onset of trial, taking into consideration potential loss to follow up during observation period? 16 8  Lecture 8 4b. Inclusion/exclusion criteria lWant a sample most efficient for answering the clinical question l Eligibility is pre-established as follows: l Ensure participants meet criteria for intervention l By some sociodemographic characteristics and other health-related events (absence of contraindications etc.) l Exclude those with difficulty in complying l Exclusions made to control error l These criteria will impact sample size required to detect an association between intervention and outcomes of interest l Note, the more strict the eligibility criteria, the less generalizable the results 17 4c. Baseline assessments l Done to characterize the study cohort l Identifying information (name, address, ID#) l Demographics (age, race, gender, etc.) l Clinical factors of relevance l In addition, the first table of a final report of any RCT typically compares the baseline characteristics between the two study groups (intervention and control condition) to ensure that randomization worked! 18 9  Lecture 8 4d. Trial outcomes or endpoints l Outcomes or endpoints: l Primary endpoint – morbidity or mortality specific endpoint l Secondary endpoint – disease indicator, health behavior, etc. Selection of the “best” endpoint is often complicated so investigators may sometimes choose surrogate endpoints E.g., the primary endpoint for most current clinical trials of HIV disease therapy is CD4 cell count or HIV viral load; secondary endpoints may be an AIDS defining event or death. 19 4e. Study protocols l Ensure all trial steps are fully documented and adhered to by staff over entire study period l Adherence to the intervention protocol: l Measuring compliance of once a day vs. complex medication schedules via self-report, pill counts or urinary metabolite levels (e.g. MEMS caps) l Behavioral intervention should be well tolerated – are participants attending all sessions? 20 10  Lecture 8 4f. Rigorous monitoring l Rigorous monitoring and assessments of all endpoints at all follow visits: l Surveys, medical exams, tests, etc. to assess for outcomes and factors associated with the intervention and the trial outcomes! l Measuring endpoints: l Time at risk measured via person-time 21 4g. Retention and Loss to follow-up l Trial retention: l Call participants the day before clinical visits l Provide reimbursement l Frequency and duration of follow up depend upon: l Type of endpoint (e.g., response to treatment, development of new disease, progression of disease, behavioral change, sustainability of change) l The level of risk (higher the risk, more frequent the follow up) 22 11  Lecture 8 4g. Retention and Loss to follow-up (cont.) l Losses to follow up (LTF) must be minimized because: l Losses are often selective (e.g., high risk persons drop out of trials) and this introduces bias l Losses to follow up should be comparable in the intervention and control arms to avoid biased comparisons l Losses to follow up reduce study power by reducing the person-time of observation 23 4h. Ethical considerations l Equipoise – There must be genuine doubt about efficacy of treatment BUT sufficient belief it may work l Informed Consent – must be obtained l Stopping rules - Based on interim analyses, before the trial is over, to determine if the intervention is 1. beneficial (and should not be withheld from placebo group) or is 2. harmful (and trial should be stopped) or 3. the evidence is inconclusive (and trial should continue) l Data Safety and Monitoring Boards (DSMB) l Many trials have been interrupted prior to completion 24 12  Lecture 8 Source: Grant RM, et al. N Engl J Med. 2010;363:2587-99. 25 Source: Grant RM, et al. N Engl J Med. 2010;363:2587-99. 26 13  Lecture 8 4h. Ethical considerations (cont.) l Analyzing by intention to treat l A procedure in the conduct and analysis of RCTs l All participants allocated to a given arm of the trial are analyzed together as representing that trial arm irrespective of whether they received or completed the prescribed treatment l If the analysis is not based on original assignment, you are breaking the randomization and the groups may not be comparable anymore l Results can be invalidated 27 4i. Assessing associations in RCTs l Outcomes at fixed points in time l What is proportion with outcome at each follow up l Logistic regression (Odds ratio) l Events measured in person-time l Rate of outcomes per 100 person years l Poisson regression, incidence rate ratio (IRR) l Time-to-event l Time from enrollment until outcome l Cox proportional hazard regression, hazards ratio (HR) l Kaplan-Meier survival analyses, log rank test 28 14  Lecture 8 Reporting trial results l The CONSORT guidelines: http://www.consort- statement.org/ 29 CIDUS-III/DUIT Baseline Visit (N=1949) Not Eligible (n=1120) HIV-/HCV- (n=1049); Eligible (n=829) HIV+/HCV- (n=17); HIV+/HCV+ (n=37); Not All HCV+/HIV- tested (n=17) Did not Return Returned for CIDUS-III/DUIT Results & for Test results Strive Enrollment Visit (n=207) (n=622) Reasons for not consenting to STRIVE (n=91): Enrolled in STRIVE via Not Interested/No Time (n=12); Study not started or ended (n=9); CIDUS-III/DUIT (n=531) Plan on moving (n=4); Other Enrolled in STRIVE from (n=14); No reason given (n=52) other studies (n=123): STRIVE Results Visits (n=507) Did not Return for STRIVE Results Visit Total Enrolled (n=24) (n=630) Enrolled in STRIVE but missed Randomized (n=418) randomization (n=212) VDI PMI n=196 n=222 30 15  Lecture 8 RCTs – Advantages & Disadvantages Advantages Disadvantages l Can demonstrate cause-effect l Only ethically appropriate approach relationships for some research questions l May be faster & cheaper than l More resource intensive cohort studies (i.e. observing & whether smoking cessation l Many interventions not suitable programs reduce smoking) for blinding l Allow investigators to control l Tested interventions may be exposure levels as needed different from common practice l Limited generalizability due to the use of volunteers, eligibility criteria, and loss to follow-up 31 16