Bias in the study of prediction of change: a Monte Carlo simulation study of the effects of selective attrition and inappropriate modeling of regression toward the mean

Table 3 Estimated associations between baseline predictor and change in health from inappropriate use of MR analyses (i.e. in situations where RTM was assumed not to occur), as shown in Figure 3

Dependency is the magnitude (b_drop) 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. b_pred = regression coefficient from predictor to change in health. Coverage of 95% C.I. = the percentage of the 500 samples with an estimated b_pred with a 95% confidence interval containing the true population value. b_pred and SE are average results over the 500 generated samples. N in the original samples was 1000. Attrition rate was 50%. The first line shows results when attrition was completely random.
The true population value was b_pred = .10.

ISSN: 1471-2288