Validity of retrospectively-reported depressive episodes

Abstract

Background: Depression and other psychopathology are often assessed retrospectively. Few studies have evaluated the validity of these reports by comparing prospectively-assessed symptoms to retrospective reports during the same time period.

Methods: This study utilized a subset of participants (n = 68) from the Oregon Adolescent Depression Project who completed at least one mailer assessment of depressive symptoms during a retrospectively-reported depressive episode. Participants completed up to seven mailer assessments of depression and suicidal ideation and diagnostic assessments that included retrospectively-reported depressive episodes that coincided with the mailer assessments.

Results: Multilevel linear models examined differences in depressive symptoms and suicidal ideation during and between retrospectively-reported depressive episodes. Results showed that individuals reported significantly higher levels of depression and suicidal ideation for retrospectively-reported depressive episodes compared to when they were not in depressive episodes. In addition, the average level of depressive symptoms endorsed during retrospectively-reported depressive episodes reached established clinical cut-offs.

Limitations: Although we were able to determine whether symptoms during retrospectively-reported depressive episodes approached clinical cut-offs, we were unable to examine whether symptoms met criteria for depressive episodes. Additionally, we could not examine whether episode severity related to recall ability, and other forms of psychopathology were not assessed.

Conclusion: These findings provide critical evidence for the validity of retrospectively-reported depressive episodes. Future research should examine whether these findings generalize across varying recall periods and retrospective assessments for other psychopathology.

Keywords: Depression; Self-report; Validity.

Figure 1.

Oregon Adolescent Depression Project study timeline. ¹KSADS = version of the Schedule for Affective Disorders and Schizophrenia for School-Age Children that combined features of the Epidemiologic version (KSADS-E; Orvaschel, Puig-Antich, Chambers, Tabrizi, & Johnson, 1982) and included additional items to derive diagnoses of past and current psychiatric disorders in the revised DSM-III (American Psychiatric Association, 1987); ²HRSD = Hamilton Rating Scale for Depression (Hamilton, 1960); ³Battery of self-report measures included measures of stress, current depressive symptoms, other psychopathology (i.e., internalizing and externalizing), pessimism, attributions, self-consciousness, self-esteem, self-rated social competence, emotional reliance, future goals, coping skills, social support, interpersonal factors, physical health and illness, maturation, and other variables; ⁴LIFE = Longitudinal Interval Follow-Up Evaluation (Shapiro & Keller, 1979); ⁵CESD = Center for Epidemiologic Studies-Depression Scale (Radloff, 1977); ⁶Perceived Social Support Questionnaire (PSS; Procidano & Heller, 1983). *In the Oregon Adolescent Depression Project, annual mailer assessments spanned ages 18 to 32, thus starting before T3 and ending after T4. However, for the current study, only mailer assessments that occurred after T3 and before T4 were included in the study.