Introduction to epidemiological study design Study = basic tool in epidemiology Epidemiology = comparison —550 cases of stomach cancer Epidemiology = comparison —550 cases of stomach cancer in Hertfordshire in 2005 — Epidemiology = comparison —550 cases of stomach cancer in Hertfordshire in 2005 —Population 550,000 —Rate 100/100,000 Stomach cancer by age group, 2005, per 100,000 Stomach cancer in Hertfordshire, 1950-2005, per 100,000 Stomach cancer in SE England in 2005, per 100,000 Epidemiology = comparison —Type of comparison (= type of study) depends on purpose. —E.g. ◦Describe the disease / condition ◦Study (analyse) its determinants / causes ◦Study (analyse) prevention / treatment ◦ ◦ — Two primary criteria —Descriptive vs. analytical — —Observational vs. interventional Descriptive vs. analytical studies —describe a pattern of occurrence of a disease: descriptive studies (always observational). —to analyse the relationship between a disease and an exposure of interest: analytical studies (can be both observational and interventional) Descriptive studies —Describe patterns of disease occurrence — —Useful for: ◦health services planning ◦hypothesis formulation in research — —Usually based on existing data: ◦mortality ◦reporting of diseases (infections, STDs, cancers...) ◦hospital and medical records ◦Census — Descriptive studies 4 Ws : What? Who? Where? When? —What? health outcome / case / event — —Person (Who?) —Age, sex, …. — —Place (Where?) —Regions, countries, international comparisons — —Time (When?) —When events occurred: —● specific time period —● seasonal pattern (births, deaths, infections) — — Cross-sectional studies Cross-sectional studies —In a cross-sectional study, all information is collected at one point in time ◦Outcome ◦Exposures ◦Covariates —Sometimes called “survey” —Cross-sectional studies could be descriptive or analytical —Always observational —The unit of analysis is the individual — Cross-sectional study Time Survey – all measurements The only way to measure “exposures” and “outcomes” is - at the time of survey or - retrospectively Cross-sectional studies: Advantages —Relatively quick, do not require follow up —Provide a snapshot, e.g. prevalence of a disease or a risk factor in population —Allow examination of multiple diseases and multiple exposures —Can test or suggest hypotheses — Prevalence studies very important for planning/monitoring Doctor needs to know prevalence of a condition in diagnosis because rare conditions less likely Local Health Service Provider needs to know health needs of current population Cross-sectional studies: Limitations —Since both disease and exposures are measured at the same time, temporality is unclear —Difficult to estimate past exposure, especially if it occurred long time ago. Not ideal for studying exposures that change over time (e.g. diet). (but no problem with factors that are stable over time, e.g. genetic markers.) —Sensitive to reporting or recall bias if exposures are subjectively reported. —Sensitive to response rates and representativeness if used to estimate prevalence of a condition in population. — Ecological studies Ecological studies —The unit of analysis is a group (e.g. country, district, population etc) —Data cannot be disaggregated to the level of an individual. —Also sometimes called correlation studies or geographical studies —Include comparisons over time (time-series) —Usually cheap and quick Ecological fallacy —This is a logical fallacy in the interpretation of statistical data where inferences about the nature of individuals are deduced from inference for the group to which those individuals belong —Extrapolation from groups to individuals is conceptually inappropriate —Situation when individual-level and group-level (ecological) associations differ —Individual data are necessary to estimate the association at the level of the individual Ecological fallacy (1) Blood pressure Salt intake Ecological fallacy (2) Blood pressure Salt intake Ecological fallacy (3) Blood pressure Salt intake Ecological fallacy (4) Blood pressure Salt intake Example: The INTERSALT study —Ecological analysis ◦Increase in salt intake by 100 mmol/day was associated with increase in SBP by 7.1 mm Hg ◦ —Individual level analysis ◦ increase by 1.6 mm Hg of SBP — — From Elliott et al, BMJ 1996 Ecological studies: Advantages —Use existing (often routinely collected) data —Quick and cheap —Useful to general hypotheses —Differences in both exposure and outcome rates may be large, which increases the likelihood to find an association —Some exposures are difficult to measure in individuals and area-based measures are used instead (e.g. air pollution), and some exposures are inherently ecological (e.g. income inequality) Ecological studies: Disadvantages —Confounding: the groups, which are compared (e.g. countries) usually differ in many other factors than the exposure of interest. It is often impossible to reliably control for confounders. —There can be systematic differences in measurements of exposures and diseases (e.g. coding of causes of death) between populations. —Ecological fallacy: ecological studies compare groups but results are extrapolated to individuals. Cohort studies cohort time direction of enquiry Advantages of cohort study -Temporal sequence is clear (exposure before disease) -Less prone to ‘reverse causality’ -Allows calculation of disease incidence -Can examine many exposures simultaneously -Multiple outcomes can be examined 1. Not prone to reverse causality BUT -Might not be able to guarantee disease free at start (pre-clinical disease, undefined disease) -Not always simple one-way relationship between exposure and disease e.g. adult social deprivation  CHD but early life risk factors for CHD (e.g. childhood obesity) might influence schooling and adult social class so if you only follow them from mid life you may be missing part of the picture More on this in the health selection lectures of soc epi. 2. Can examine several exposures at once BUT -Should expect lag time to disease to be broadly similar -Sample is usually selected to be similar in many ways (e.g. civil servants) so might not get much variation in some exposures -Participant burden 3. Less bias cf case control because exposure measured before outcome Disease state should not be affecting reporting/recall of exposure Disadvantages of cohort study -Exposure may change over time -Some diseases take years/decades to develop so may not be suitable -Findings might not be relevant at end of study -High costs because large sample and long duration -Participant burden -Loss to follow-up usually depends on outcome of interest (selection bias) -Assessment of causality problematic in observational setting (although less problematic in cohort than other types of observational studies) NOTE: Exposure may change over time In diabetes-CHD example, simply looked at baseline glucose levels Statistical methods are available to cope with exposures varying over time. Essentially an individual contributes a years unexposed and b years exposed Some well-known cohort studies —British Birth Cohorts ◦Millennium Cohort Study ◦1970 British Cohort Study (BCS70) ◦1958 National Child Development Study ◦1946 National Survey of Health and Development —Studies of specific diseases (e.g. cardiovascular disease): ◦Whitehall II study ◦Framingham Study ◦HAPIEE (Health, Alcohol and Psychosocial Indicators in Eastern Europe) Birth cohorts: 1 week of birth Set up to study health and social circumstances of representative samples in Engl, Scotland and Wales Cardiovascular disease studies originally set up to study CVD but now broadened out to ageing studies Br Reg Heart Study set up to study factors responsible for variation in CHD and CHD biomarkers Studies sometimes set up to study effect of an exposure e.g. Dutch hunger winter families study Occupational studies (e.g. Wh2, Nurses Health Study) sample selected for convenience Summary of cohort studies —Exposure measured usually in healthy individuals —Follow up —Incidence —Time consuming & expensive —Temporality clear —Possibly the “best” observational design Case-control studies Cohort Start Unexposed Exposed All healthy Follow-up (wait) Disease assessment Controls Cases Start Look back Case-Control Case-control studies are —Ideal for rare diseases —Usually “retrospective” in design —Relatively quick —Relatively cheap Strengths of case-control studies —Quick (cases already exist, no need to wait) —Cheap (not necessary to examine large number of people) —Can examine many exposures —Suitable to study rare diseases —Suitable to study stable exposures (eg genetic markers) Weaknesses of case-control studies —Not suitable for rare exposure —Prone to misclassification of exposure —Prone to reverse causation (people with disease may have changed their behaviour) Intervention studies Basic features of intervention studies —An intervention study involves an intentional change in some aspect of environment or status of the subjects of the investigation. —Intervention studies differ from observational studies in that the researcher seeks to compare two or more groups that differ as a result of deliberate action rather than natural or found variation. — Everything except the intervention is (hoped to be) the same in the two groups Defined study sample Intervention group Control group Measure outcome Measure outcome Randomisation to two groups Key issues in RCTs —Careful entry criteria —Assessment (Pre- & Post-intervention) —Randomisation —Allocation Concealment —Blinding (Masking) The aim of randomisation is to… —create groups that are comparable with respect to known or unknown confounding factors — —There are two steps in the process 1.Generating an unpredictable allocation sequence e.g. tossing a coin, using a computer random number generator 2.Concealing the allocation sequence from the investigators —Not always possible Allocation concealment —… is making sure that neither investigator nor patient can predict group assignment — —Adequate methods —Off-site randomisation e.g. needing a phone call —Sequentially numbered, sealed, opaque envelopes Blinding —If participants or researchers know whether participant is receiving intervention then there is risk of: ◦Measurement error ◦Different investigations & care study group etc. ◦Acceptability bias (Researchers influence participants behaviour) —Different “levels” of blinding: can blind participants, researchers and/or statisticians or none Summary —Intervention studies are experiments —RCTs are the gold-standard design for assessing the effectiveness of interventions —Simple concept but many key features - need to carry out properly —Randomisation is the most important, but others —Not always applicable – PH interventions are usually more complex than a clear-cut simple experiment hierarchy of major study designs systematic review of RCTs RCT cohort case control interventional observational validity ecological cross-sectional Applications of different observational and analytical study designs Ecological Cross sectional Case control Cohort Investigation of rare disease ++++ - +++++ - Investigation of rare exposures ++ - - +++++ Examining multiple outcomes + ++ - +++++ Studying multiple exposures ++ ++ ++++ +++ Measurement of time relationships between expo and outcome + - + +++++ Direct measurement of incidence - - + +++++ Investigation of long latent period - - +++ +++