统计问题第74问：巢式病例对照研究

Question

The association between suicide and the use of calcium channel blockers in people with hypertension was investigated using a nested case-control study design. Data were taken from the general practice research database, a computerised database of anonymised longitudinal medical records collected prospectively in primary care. The cohort of people with records between January 1991 and August 1998 was used.

At the end of data collection, 38 people who had committed suicide had received a prescription for an antihypertensive drug within six months of their death. One hundred and forty controls—matched for age, sex, general practice, and index date—were selected from the same cohort. The index date was the date of suicide, with controls being treated with an antihypertensive drug on the same date. The study found that calcium channel blockers were not associated with an increased risk of suicide compared with other antihypertensive drugs.

Which of the following statements, if any, are true?

a) Cases are people who committed suicide

b) The nested case-control study was retrospective in design

c) The choice of controls was prone to selection bias

d) Data collection was subject to recall bias

提示：这是一道多选题。

Answer

Answers a and b are true; c and d are false.

The nested case-cohort study is an observational design that incorporates the case-control approach within an established cohort. The design overcomes some of the disadvantages associated with case-control studies while incorporating some of the advantages of a cohort study. The general practice research database provided a prospective cohort. After data collection had finished, a case-control approach was adopted for the investigation of calcium channel blockers as a risk factor for suicide in people with hypertension. Those people who committed suicide and received a prescription for an antihypertensive drug within six months of death were the cases (a is true). Controls were selected from the same cohort as the cases. Cases and controls were matched for age, sex, general practice, and index date. Case-control studies and the importance of matching, plus cohort studies, have been described in previous questions.

Both cases and controls were identified after data had been collected for the cohort. Therefore, the above nested case-control study was retrospective in design (b is true).

Both selection and recall bias are of particular concern in case-control studies. Selection bias can occur if the controls are selected from a different population to the cases. This may result in systematic differences between cases and controls, including differences in social background and general health status. In the above study, controls were selected from the same population as the cases so selection bias would not have occurred (c is false).

Case-control studies are retrospective in design, with people recalling information about past exposure to risk factors. Recall bias is caused by systematic differences between cases and controls in the accuracy of information recall. Generally, cases recall information more accurately than controls, possibly because of an association with the disease or outcome of interest. Although the cases and controls were identified retrospectively, the data were collected prospectively so were not prone to recall bias (d is false).

所以答案是选择 a b

每天学习一点，你会更强大！