The implication of selection bias in clinical studies of late life depression: an empirical approach

Anxiety and the risk of death in older men and women

“…These participants gave their informed consent and agreed to be interviewed in their homes. Of the non-responders, 126 had died before approach; 44 could not be contacted, 134 were too ill or cognitively impaired to be interviewed, and 394 were unwilling to participate (Van Exel et al , 2000).…”

Setting and registry characteristics affect the prevalence and nature of multimorbidity in the elderly

“…It is especially known that psychiatric morbidity was not well recognized in general practice [14,15]. Moreover, the underdiagnosis of psychiatric conditions was even more pronounced among somatically ill patients [16].…”

Challenges in Identifying Individualized Brain Biomarkers of Late Life Depression

“…In a highly heterogeneous patient group, estimating patient-control differences can result in either a reduced or null signal due to the combination of subgroups with inconsistent effects (especially in high sample size studies), or can result in a spuriously strong signal due to overrepresentation of a subgroup (more likely in small sample sizes). This type of sampling bias, which occurs when some members of the LLD population are systematically more likely to be selected in a study sample, may also be driven by oversampling of community-dwelling versus inpatient individuals [ 72 , 73 ] and may particularly impact underrepresented communities [ 74 ]. Beyond patient-control group comparisons, regression-style analyses are still fitted using data from all patients and therefore also do not account for heterogeneity.…”