Cross Sectional Study K J Nathoo NECTAR/CERTC lunchtime lecture August 2011 Cross-sectional study OBJECTIVES Explain what is a cross sectional study. Its Uses Advantages and limitations Definitions Did investigator assign exposures? yes no Experimental study Observational study Observational studies Observational study comparative groups? yes no Analytical study Descriptive study Types of descriptive studies(1) Deal with individuals Case report, Case series report Cross sectional studies( prevalence studies) Surveillance( watchfulness over a community) Types of descriptive studies(2) Examine populations Ecological correlational studies Look for associations between exposures and outcomes in populations than in individuals inability to link exposures to outcomes in individuals and to control for confounding Observational:Analytical study Direction? Cohort: exposure- outcome Case-control: outcome- exposure Cross-sectional: exposure and outcome at the same time Cross-sectional study Observational Analytical( comparison group) Exposure and outcomes are assessed at the same time Does what? TIME Cross sectional studies are carried out at one point in time or over a short period SNAPSHOT of the outcome and characteristics associated with it, at a specific point in time Why carry out a cross-sectional study? Descriptive( survey) usually no hypothesis as such but AIM to describe the population with respect to an outcome and a set of risk factors Find the prevalence of the outcome of interest for the population or subpopulation at a given time point Cannot infer causality! Investigate associations between risk factors and the outcome of interest Limitation: carried out in one time point and give no indication of the sequence of events. Whether the event occurred before, after or during the onset of the disease outcome Associations? Other studies better for the purposes of understanding associations between exposure to risk factors and the outcomes of interest Xsectional studies indicate associations that MAY exist and therefore useful in generating hypothesis for future research Sampling issues Reference population Source population- external validity Sampling frame (made up of sampling units) sampling bias Sample Study subjects Sampling frame, sampling unit Sampling frame list of all the sampling units Listing of community, list of households Sampling unit This is a basic unit Eg person, household around which sampling procedure is planned SAMPLING Is the sample representative of the population? In a RANDOM sample every individual in the population has an equal probability of being selected “probability sampling” sampling Non-probability sampling Some individuals have a greater chance of being selected than others Eg convenience and availability sampling Sample selection bias Self selection: Can only include those who are available for inclusion Survivor bias: those who die are not available for the study may have different characteristics Response bias Also known as non-response bias Important in cross-sectional surveys Response rate Numerator( responded to the survey)/ denominator (who could have responded) Response rate70% significant bias Measurement issues standardised measure for the exposure standardised measure for the outcome of interest It forms the basis for future generalisations Measurement issues Random error: wrong result due to chance unknown source of variation distorts the sample in either direction Systematic error: wrong results due to bias Distorts the study findings in one direction Measurement issues Random and systematic errors can be components of 1. Sampling (reduced in probability sampling) 2. Measurement error Threatening the inference (conclusion)from the study Precision and Accuracy Strategies for minimising measurement error Precise= free from random error Accurate= free from systematic error PRECISION (RELIABLE) A very precise measurement is one that has nearly the same value each time it is measured A beam scale measure body wt with great precision Interview QAL of life vary form one occasion to another ACCURACY( VALID) Is the degree to which it actually represents what it is intended to represent Very important influence on the internal and external validity of the study The greater the error the less accurate the variable Advantages(1) Quick and inexpensive Provide prevalence info Control over selection of study subjects Advantages(2) Control over the measurements used Can study factors or outcome factors at the one time Early clues for hypothesis generation Disadvantages Not allow the true temporal sequence of exposure and outcome to be ascertained Potential Bias in measuring exposure Potential sampling and/ or survivor bias Does not yield incidence or true RR