The Pocket Guide to Critical Appraisal. Iain K. Crombie

whether lung cancer occurs more frequently in smokers than in non‐smokers.

      Complications

      Some cohort studies identify the group from some time in the past and follow them up to the present. These studies might at first appear to be looking backwards in time, but they are not. Historical records are used, and time flows forwards from the point at which the individuals are identified. This type of design is usually called a retrospective cohort study. (Note the term historical cohort study is used when the exposed group is recruited from one period and the comparison group comes from an earlier period.)

      Terms of identification

      The terms prospective and longitudinal suggest a cohort study, although these terms are also used in clinical trials. The term retrospective is used for one type of cohort study, but is also used with case–control studies.

Case–control studies

Case–control studies ask what makes a group of individuals different from a control or comparison group. Often the study group will have a disease, which the control group do not have. For example, the cases could be women with breast cancer and the controls could be women of similar age who do not have the disease. The study compares the characteristics of the cases and controls to identify the factors from the past that might have caused the disease. The control group is often selected to be similar to the cases in factors such as age, gender, and life circumstances. This type of study can also explore why some people behave in a particular way. For example, a study could explore why some patients miss clinic appointments, by comparing those who fail to attend with those who do attend.

      Essential features

      Case–control studies look backwards in time from some event (e.g. the diagnosis of breast cancer or failing to attend a clinic) to try to identify factors in the past that might explain why that event occurred. The direction of time is crucial for distinguishing between cohort studies and case–control studies: cohort studies look forwards; case–control studies backwards.

      Complications

      The use of a control group is not a defining characteristic, since RCTs always have one, and cohort studies often do.

      Terms of identification

      Several terms are used for this type of study including: case–control, case‐referent, case‐comparator, and case‐comparison. Because the method looks backwards in time it is sometimes called a retrospective study, but this term can be used with cohort studies.

RCTs

      The RCT should be the easiest method to identify. This design is used to test whether one health care intervention is superior to another. RCTs are most often used to test drugs, but they can be used to investigate many different types of health care interventions: surgery, vaccination, anti‐pressure sore mattresses, and health education. RCTs often compare a new treatment against the currently accepted best treatment. If there is no existing treatment, the new one is compared against a placebo (an inert substance or a dummy procedure).

      Essential features

RCTs are always concerned with effectiveness. The key element of this design is that patients are randomly allocated to receive a new treatment or to the conventional one (or a placebo). The outcome of the new treatment is compared with that of the conventional one, identifying which is superior. RCTs are also concerned with the side effects of treatments. Most RCTs compare two treatments, but sometimes more than two treatments can be investigated. This adds complexity to the conduct and analysis of the study, although the resulting study is a valid RCT.

      Complications

      The phrase clinical trial is often used as shorthand for an RCT. However, clinical trial can refer to studies in which patients are allocated treatment in a non‐random way (non‐randomised studies). For example, cohort studies can be used to assess effectiveness: a group of treated patients and an independent control group are followed up to see which group gains most benefit.

      Terms of identification

      The term randomisation almost always identifies an RCT, as does the equivalent phrase random allocation. (Note that the term random selection may refer to a survey.) The terms blinding, placebo, effectiveness, efficacy, and evaluation, or phrases like assess the value of or improve the outcome, can be used in both RCTs and non‐randomised studies.

Cohort studies that evaluate the effectiveness of interventions

An increasingly common research design is a special type of cohort study that evaluates the effectiveness of treatments. It compares the outcomes among a group of patients given one treatment with outcomes in other patients who have the same disease but were given a different treatment. The difference from a conventional RCT is that patients are not randomised to treatments. Instead the clinician responsible for the patient decides which treatment should be given. This type of study evaluates treatments as they are used in routine clinical practice. A common form of this design uses electronic health records to identify patients with a defined disease and the treatments that they were given. The effects of the treatments on the outcome measures are also obtained from electronic data. When the two groups are recruited and followed up in the same time period, this design is called a concurrent cohort study. In a different form of this design, recently diagnosed patients are given the new treatment and are compared with patients from the past who were given a different treatment. These studies are called historical cohort studies.

      Essential features

      The essential feature of this design is that the treatment patients receive is selected by the doctors who are responsible for their care, based on an assessment of their specific clinical circumstances. Rather than being randomised to treatment groups, clinical judgement is used to decide which treatment will be best for each patient. As a result, the patients given one treatment may differ systematically from those given the other treatment. This is the major weakness of the research design, as the two groups of patients will be different at the start of the study (i.e. before they receive their treatment).

      Complications

A special type of cohort treatment study is the quasi‐randomised study. In this design patients are assigned by the researchers to treatment groups using a non‐random method such as date of clinic appointment (assignment by alternate days) or date of birth (assignment by odd or even dates). Because the assignment process is predictable, it could be tampered with, creating a risk of bias. This design is best viewed as similar to an RCT, but one at high risk of bias due to poor concealment of treatment allocation.

      Terms of identification

      There are no specific terms which identify this research design. The terms concurrent cohort study and historical cohort study could refer to this design, but they could also refer to conventional cohort studies (those which do not evaluate the effectiveness of treatments). The term non‐randomised study is often used for this type of study, although that label refers to a group of research designs which include case–control studies, surveys, interrupted time series, and case series.

Systematic reviews

      Systematic reviews seek to identify all the papers published on a specific health topic to obtain a summary of the findings from