Research Methodology

No. of cases in a given population at a designated time

The number of cases existing at a point in time

The number of cases existing over a specified time period

A defined population is surveyed and their disease status/exposure determined at a point in time. It can measure current disease burden, estimate risk, quick and inexpensive but nor suitable for rare disease, cannot measure incidence rate.

Take a random 1000 people and measure obesity. It just provides a snap shot and does not say if the disease incidence is increasing or decreasing

•Group of people followed through time. Incidence of an event/disease compared among those exposed and those unexposed to a risk factor. •Prospective best, but can be retrospective. •Advantages: can estimate incidence rate, risk and relative risk; can study multiple effects of a single exposure (including benefits), data on exposure prevents problems of recall bias, flexibility in choosing variables to systematically record •Disadvantages: large numbers needed, long duration, changes in exposure etc may make results irrelevant. (But nowadays we can use large datasets of real life medical care to analyse incidence and care).

Cumulative incidence is the number of new cases in a period of time divided by the number of people in the population at the beginning of the period

The null hypothesis is that one variable has no association with another vriable or set of variables or that 2 of more population distributions do not differ from each other.

The number of cars in brighton has no correlation to the season.

Advantages - can estimate the incidence rate, risk, relative risk and can study multiple effect of a single exposure Disadvantages - large numbers are needed, long duration, changes in exposure may make results irrelevant.

Case control studies look retrospectively using cases and controls. They look at the presence or absence of a risk factor. It compares the exposure in cases and controls.

Advantages - cheap, quick, few subjects needed, can use records and study multiple causes Disadvantages - cannot calculate prevalence or incidence rates, recall bias, validation difficult and suitable control group is hard to find

Patients identified who fit criteria for inclusion into study  Randomised to intervention group or control group  Each patient in the study receives the intervention/no-intervention or placebo and data is collected on specific outcome measures  The data from each group is compared to see if there is a difference between the outcome measures between intervention and control
A treatment or intervention is assessed by comparison with an alternative treatment or placebo. Random allocation to treatment minimises confounders and selection bias. Where possible both patient and person assessing the outcome should be blind to the treatment allocation = double blind.

Advantages - Control and treatment group are similar, minimises confounders and selection bias Disadvantages - expensive, time consuming, ethical problems

Descriptive studies are designed to describe the existing distributions of risk factors/ outcomes/ exposure. They do not test whether associations are causal