Interpreting SF-12 mental component score: an investigation of its convergent validity with CESD-10

• Published in: Quality of life research Vol. 24; no. 9; pp. 2209 - 2217
• Main Authors: Yu, Doris S. F., Yan, Elsie C. W., Chow, Choi Kai
• Format: Journal Article
• Language: English
• Published: Cham Springer 01.09.2015; Springer International Publishing; Springer Nature B.V
• Subjects: Adolescent; Adult; Age groups; Aged; Aged, 80 and over; Asian Continental Ancestry Group; Depression - diagnosis; Depression - psychology; Depressive Disorder - diagnosis; Depressive Disorder - psychology; Disease; Female; Health Surveys; Hong Kong; Households; Humans; INSTRUMENT DEVELOPMENT; Interviews; Male; Medicine; Medicine & Public Health; Mental depression; Mental disorders; Mental health; Middle age; Middle Aged; Nursing schools; Performance evaluation; Polls & surveys; Population; Public Health; Quality of Life - psychology; Quality of Life Research; Sensitivity and Specificity; Sociology; Surveillance; Validity; Young Adult; Hong Kong; China; Depression; Adult Chinese population; Short-Form 12 Health Survey; Mental health
• DOI: 10.1007/s11136-015-0959-x
• ISSN: 0962-9343

Objective To examine the convergent validity of Mental Component Scale of the Short-Form 12 (SF-12 MCS) with the Center for Epidemiologie Studies Depression Scale (CESD-10). The CESD-10 is a screening tool for probably clinically significant depression in the Chinese population. Methods Data were obtained from a household survey carried out in Hong Kong. A two-stage stratified sampling method successfully interviewed 1795 adult subjects from 1239 households. Data on SF-12 MCS and the CESD-10 were extracted. Receiver operating characteristics (ROC) analyses were performed to examine the convergent validity of SF-12 MCS against the CESD-10 threshold for probably clinically significant depression for the younger to middle-aged, late middle-aged and older population cohorts. Results ROC analysis indicated the excellent convergent validity of SF-12 MCS with the CESD-10 threshold for identifying probably clinically significant depression, with the area under curve ranged from 0.81 to 0.85. The optimal cutoff scores for depression among the younger to middle age group, late middle age group and older age group were 48.1, 50.2 and 50.2, respectively, with sensitivities ranged from 77 to 83 % and specificities ranged from 73 to 78 %. Bootstrapping estimates of the mean difference indicated no significant difference in the optimal cutoff scores between these age cohorts. Conclusion SF-12 is a widely adopted measure to capture the health profile of Chinese population. The study findings indicated the satisfactory performance of the SF-12 MCS in identifying probably clinical depression. Future study is warrant to examine the diagnostic validity of the SF-12 MCS by using gold standard to assess clinical depression.