Question
Researchers used a case-control study to assess the risk of thrombosis associated with the use of oral contraceptives. Participants were premenopausal women aged less than 50 years, not pregnant, not within four weeks postpartum, and not using a hormone excreting intrauterine device or depot contraceptive.
Women were identified as a case if they had a first objectively diagnosed episode of deep venous thrombosis or pulmonary embolism. A total of 1524 cases and 1760 otherwise healthy controls were identified. The women were then asked about use of oral contraceptives in the past year. The odds ratio of venous thrombosis for current users of oral contraceptives compared with non-users was 5.0 (95% CI 4.2 to 5.8).
Which of the following, if any, are true?
a) It was possible to estimate the population at risk in this case-control study.
b) The odds ratio is an estimate of the population relative risk.
c) Recent users of oral contraceptives were five times as likely to experience deep venous thrombosis or pulmonary embolism than non-users.
d) The result was statistically significant at the 5% critical level of significance.
提示：这是一道多选题。
Answer
Answers b, c, and d are true, whereas a is false.
Case-control studies have been described in a previous question. The women in this study were selected on the basis of their disease status. Cases were those with a first objectively diagnosed deep venous thrombosis or pulmonary embolism. Controls were otherwise healthy women. The cases and controls were asked about past exposure to the risk factor—oral contraceptives. The aim of the study was to establish whether recent oral contraceptive use increased the risk of deep venous thrombosis or pulmonary embolism.
Case-control studies are retrospective in design and so cannot be used to estimate the population at risk (a is false). It would have been possible to estimate the population at risk if women had been followed prospectively. In this instance, the proportion of women exposed to the risk factor that subsequently experienced the disease would have estimated the risk for the population. In this case-control study, women were selected on the basis of their disease status; therefore, the proportion with venous thrombosis was dependent on the ratio of cases to controls. The proportion of women using oral contraceptives that experienced venous thrombosis would, therefore, be dictated by the ratio of cases to controls.
Given that the population at risk cannot be estimated using a case-control study, risks and relative risks cannot be calculated. It is possible, however, to derive an odds ratio for a case-control study, which estimates the population odds ratio and, in turn, the population relative risk (b is true). Since the odds ratio is an estimate of the population relative risk it is often, although incorrectly, referred to as a relative risk.
To calculate the odds ratio, the odds of venous thrombosis if using oral contraceptives were divided by the odds of venous thrombosis if not using oral contraceptives. Odds are an alternative way of expressing probability. The odds of venous thrombosis if using oral contraceptives are equal to the number of women with venous thrombosis that were using oral contraceptives divided by the number of healthy controls using oral contraceptives. The odds of venous thrombosis if not using taking oral contraceptives were calculated in a similar fashion.
The odds ratio for venous thrombosis for recent users of oral contraceptives compared with non-users were 5.0. Therefore, the odds of venous thrombosis for women using oral contraceptives were five times those of non-users of oral contraceptives (c is true). Women using oral contraceptives were five times as likely—namely four times more likely—to develop venous thrombosis compared with non-users.
其实我觉得OR这么解释疾病的风险，是不对的。因为郑老师我也错了哈哈。
Last week’s question described the relationship between the 95% confidence interval for the population relative risk and the P value for the hypothesis test of statistical significance. If the 95% confidence interval does not include unity—the state of equipoise—then the P value will be less than 0.05. The same relationship holds true for the population odds ratio. The 95% confidence interval for the population odds ratio in this study was 4.2 to 5.8, and hence did not include unity. Therefore, the P value for the hypothesis test of statistical significance for the population odds ratio was less than 0.05 (that is, less than 5%). In this study, the null hypothesis would be rejected in favour of the alternative at the 5% level of significance; therefore, the odds ratio was significantly different from unity (d is true).
所以答案是选择 b c d
每天学习一点，你会更强大！