Some Conceptual and Statistical Issues in Analysis of Longitudinal Psychiatric Data

TLDR

Longitudinal studies are common in psychiatry, yet most analyses discard much data or rely on restrictive repeated‑measure ANOVA methods that ignore missing data and serial correlation. The study examined the statistical and conceptual challenges inherent in longitudinal psychiatric data. Random regression models were employed to address missing data, serial correlation, time‑varying covariates, irregular measurements, and individual deviations, and were illustrated using the NIMH Treatment of Depression dataset. Both individual‑specific effects and serial correlation significantly influence longitudinal psychiatric outcomes, and omitting either leads to misleading uncertainty estimates.

Abstract

L studies have a prominent role in psychiatric research; however, statistical methods for analyzing these data are rarely commensurate with the effort involved in their acquisition. Frequently the majority of data are discarded and a simple end-point analysis is performed. In other cases, so called repeated-measures analysis of variance procedures are used with little regard to their restrictive and often unrealistic assumptions and the effect of missing data on the statistical properties of their estimates. We explored the unique features of longitudinal psychiatric data from both statistical and conceptual perspectives. We used a family of statistical models termed<i>random regression models</i>that provide a more realistic approach to analysis of longitudinal psychiatric data. Random regression models provide solutions to commonly observed problems of missing data, serial correlation, time-varying covariates, and irregular measurement occasions, and they accommodate systematic person-specific deviations from the average time trend. Properties of these models were compared with traditional approaches at a conceptual level. The approach was then illustrated in a new analysis of the National Institute of Mental Health Treatment of Depression Collaborative Research Program dataset, which investigated two forms of psychotherapy, pharmacotherapy with clinical management, and a placebo with clinical management control. Results indicated that both person-specific effects and serial correlation play major roles in the longitudinal psychiatric response process. Ignoring either of these effects produces misleading estimates of uncertainty that form the basis of statistical tests of hypotheses.