Ms. Sullivan and Dr. Payne Reply

To The Editor: Drs. Gupta and Sharma point out limitations with the Seasonal Pattern Assessment Questionnaire in detecting individuals experiencing seasonal affective disorder. As a self-report measure, the Seasonal Pattern Assessment Questionnaire has been used with college samples yielding high retest reliability over seasons (1) . We do, however, recognize that the Seasonal Pattern Assessment Questionnaire may be overinclusive (2) . Thus, for further validation of the seasonal affective disorder status, we also administered the Beck Depression Inventory–II. Seasonal affective symptoms likely exist on a continuum. Although scores on both the Seasonal Pattern Assessment Questionnaire and Beck Depression Inventory–II were correlated in our study, the majority of individuals with seasonal affective disorder based on the questionnaire did not qualify for a diagnosis of major depressive disorder. However, scores on the Beck Depression Inventory–II were still higher for subjects in the seasonal affective disorder group (excluding those subjects with seasonal affective disorder/major depressive disorder) relative to those subjects not reporting symptoms of seasonal affective disorder.

To address concerns that the Seasonal Pattern Assessment Questionnaire is overinclusive, we more closely examined the group identified as seasonal affective disorder. As Drs. Gupta and Sharma suggest, there were individuals in the seasonal affective disorder group who could be considered subsyndromal seasonal affective. Although these subjects would not be identified by the Seasonal Pattern Assessment Questionnaire as having subsyndromal seasonal affective disorder, the Beck Depression Inventory–II better describes the severity of disordered mood in these subjects. Using the major depressive disorder criterion of the Beck Depression Inventory–II as an additional indicator of seasonal affective disorder (instead of the Seasonal Pattern Assessment Questionnaire as a single indicator), we observed that the majority of subjects in the seasonal affective disorder group would be considered subsyndromal seasonal affective, while only a few met the criterion for seasonal affective disorder as illness. The few remaining participants with seasonal affective disorder/major depressive disorder were not group outliers in terms of Beck Depression Inventory scores, since there were other participants whose scores were borderline major depressive disorder. Additionally, many participants reported December as the month in which they experience the full extent of atypical seasonal changes. Taken together, these data support the hypothesis that some participants who appeared to experience subsyndromal seasonal affective disorder may have been on the verge of experiencing full seasonal affective disorder as the Winter season progressed. However, cognition may have already been impacted. Consistent with our primary finding, participants who qualified for the diagnosis of subsyndromal seasonal affective disorder based on both the Seasonal Pattern Assessment Questionnaire and Beck Depression Inventory reported cognitive difficulties equivalent to those participants with major depressive disorder. Even with an adjusted identification criterion, cognitive failures were higher in the subsyndromal seasonal affective disorder group relative to individuals with no disordered mood. It is possible that high cognitive failures early in the season are identifying characteristics of seasonal affective disorder. Cognitive failures for the subsyndromal seasonal affective disorder group could be linked to ruminative cognitive styles that may predict vulnerability to seasonal depression (3) .