Dp. Lubeck et al., Approaches to missing data inference results from CaPSURE - An observational study of patients with prostate cancer, PHARMACOECO, 15(2), 1999, pp. 197-208
Objective: There are multiple reasons for missing data in observational stu
dies; excluding patients with missing data can lead to significant bias. In
this study, we evaluated several methods for assigning missing Values to h
ealth service utilisation.
Design and setting: Cancer of the Prostate Strategic Urologic Research Ende
avor (CaPSURE) is a US national database of men with prostate cancer. Physi
cian visits and diagnostic tests for 342 patients newly diagnosed with pros
tate cancer were evaluated.
Patients and participants: Patients were followed for a full year (observed
data, n = 228) and patients with incomplete data (predicted data, n = 114)
were included.
Interventions: We used the following approaches for imputing missing data:
assigning the group mean, a time-specific mean, a patient-specific mean, a
stratified mean (by age, localised disease and insurance status) and carryi
ng the last observation forward and/or backward.
Main outcome measures and results: All prediction strategies resulted in hi
gher estimates (19.3 to 23.1) for annual physician visits than was observed
(17.1 +/- 15.5), and differences were statistically significant for both t
he last observation carried forward (23.1 +/- 15.5) and the patient's indiv
idual mean (22.7 +/- 36.1) when predicting physician visits. The same strat
egies had higher predicted values for x-rays (1.8 +/- 5.1 and 1.8 +/- 4.4 v
s 1.1 +/- 1.9 for the observed group), although the Last observation carrie
d forward was not statistically different from the observed value.
Conclusions: We were unable to identify a single optimal strategy. However,
imputation from individual means and the last observation carried forward
methods did not perform as well as the other strategies. While the differen
ces observed in this study were small, we anticipate that with increased le
ngth of follow-up and more dropouts, there would be greater differences amo
ng strategies.