Table 1.
Estimated associations between baseline predictor and change in health from appropriate use of MR analyses (i.e. in situations where RTM was assumed to occur), as shown in Figure 2
| 50% attrition rate | 70% attrition rate | |||||
|---|---|---|---|---|---|---|
| Dependency (b drop) | b pred (SE) | Coverage of 95% C.I. | b pred (SE) | Coverage of 95% C.I. | ||
| Pred. | V1 | V2 | ||||
| 0 | 0 | 0 | .10 (.04) | 95% | .10 (.05) | 95% |
| .3 | .3 | 0 | .10 (.04) | 95% | .10 (.05) | 93% |
| .3 | .3 | .1 | .09 (.04) | 91% | .08 (.05) | 93% |
| .3 | .3 | .3 | .04 (.04) | 67% | .03 (.05) | 73% |
Dependency is the magnitude (bdrop) of the regression of liability of dropping out on each of the three study variables. Pred = baseline predictor. V1 = the main variable at baseline (baseline health), V2 = the main variable at follow-up (follow-up health). SE = standard error. bpred = regression coefficient from predictor to change in health. Coverage of 95% C.I. = the percentage of the 500 samples with an estimated bpred with a 95% confidence interval containing the true population value. bpred and SE are average results over the 500 generated samples. N in the original samples was 1000. The first line shows results when attrition was completely random.
The true population value was bpred = .10.