Correlates of depressive symptoms in late middle-aged Taiwanese women: findings from the 2009 Taiwan National Health Interview Survey

The data for this cross-sectional study were obtained from the Taiwan National Health Interview Survey (NHIS) conducted in 2009. The NHIS is a nationally-representative survey of the total population of Taiwan conducted by the Health Promotion Administration of the Ministry of Health and Welfare once every 4 years.

Individuals in the NHIS were sampled using a multistage, stratified, systematic sampling design based on geographic location and level of urbanization. All selected individuals were interviewed by trained interviewers face-to-face using computer-assisted personal interviewing (CAPI). Further details regarding questionnaire content and sampling design can be obtained from the NHIS website (http://​nhis.​nhri.​org.​tw/). A total of 30,528 (1.33%) eligible individuals were sampled from the Taiwanese population of 22,942,706 as of December 31, 2008. A total of 25,632 individuals completed the survey (84.0% response). Of them, 2441 were females aged 50.0 to 65.0 years. A random sample of 638 respondents was further selected using the CAPI software to respond to a set of optional questions including those on depressive symptoms, which was the main outcome of interest in this study. Eighteen of these respondents had missing data for height or weight and 42 respondents had missing data on menopausal symptoms. As a result, a total of 578 women had complete data for all variables. After weights were applied to the raw data, a final sample of 533 women was available for analysis.

The study protocol was reviewed and approved by the institutional review board of the Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Taiwan (No. B10204005). Since the datafile contained only de-identified secondary data, the institutional review board waived the requirement for informed consent.

The outcome variable of this study was depressive symptoms in the week preceding the interview, measured by a 10-item Chinese version of the Center for Epidemiologic Studies Short Depression Scale (CES-D 10). The original 20-item version of the CES-D was developed in English in 1977 [22]. The 10-item brief version has been shown to adequately represent the full CES-D [23]. The psychometric properties of the Chinese version have been evaluated in elderly Chinese. Satisfactory content reliability and temporal reliability were attained with accuracy in classifying individuals with depressive symptoms and they were comparable to the original CES-D. A good internal consistency of Cronbach’s α = 0.78 and a moderate test-retest reliability at r = 0.44 over 3 years were observed [24]. The CES-D 10 comprises 10 questions with a four-point Likert scale graded as follows: rarely or none of the time = 0 (less than a day per week); some of the time = 1 (1 or 2 days per week); occasionally = 2 (3 or 4 days per week); and almost all the time = 3 (5 to 7 days per week), with a total score of 30. Two items (numbers 5 and 8) required reverse scoring. Respondents were classified as having depressive symptoms if they had a total score of 10 or above [25]. A confirmatory factor analysis of the Chinese CES-D 10 was conducted on a sample of 742 community-dwelling Chinese elderly aged 60 years and older. A two-factor model that divided items into “depressive affect/somatic retardation” and “positive affect” constructs were found to best fit the data. The reliability coefficients also revealed that most indicators were strongly related to their purported latent factors (range from 0.46–0.89) [26].

A total of 20 independent variables were evaluated in this study, including age, body mass index, educational level, marital status, nulliparity, living arrangements, religious belief, monthly personal income, work status, smoking habit, alcohol consumption, regular exercise, hypertension, hyperlipidemia, diabetes, perceived health status, use of emergency services in the past year, hospitalization in the past year, experience with any somatic climacteric symptoms, and experience with any psychological climacteric symptoms.

Age was categorized into 50.0–54.9, 55.0–59.9, and 60.0–65.0 years. Body mass index was categorized as normal, underweight, preobese, and obese, using the cut-off criteria set by the Ministry of Health and Welfare in Taiwan. Educational level was dichotomized into secondary school or above and elementary school or below. Marital status was dichotomized as married versus other, which included single, divorced, separated, and widowed. Work status was categorized into currently working, having worked in the past, and never having worked, with work defined as employment with paid compensation. Smoking habit was defined as either non-smokers or daily and occasional smokers. Alcohol consumption was defined as the consumption of alcoholic beverages in the past month. Regular exercise was defined as having engaged in any physical activity in the past 2 weeks. The presence of hypertension, diabetes, and hyperlipidemia were determined by two sequential questions: “Do you have this disease?” and “Were you told by a doctor or a nurse that you have this disease?” Only if the answers to both questions were affirmative, then the patient was classified as having the disease. The somatic climacteric symptoms and psychological climacteric symptoms were both dichotomous variables. A respondent was considered as having somatic climacteric symptoms if she had ever experienced any of the following 14 symptoms: hot flashes, night sweats, palpitations, chest tightness, dizziness, headache, vaginal dryness, loss of libido, dyspareunia, insomnia, muscle or joint ache, fatigue, dry skin, and other somatic changes. Moreover, a respondent was considered as having psychological climacteric symptoms if she had ever experienced any of the following 6 symptoms: anxiousness, dysphoria, panic attacks, feeling depressed, forgetfulness, and other psychological changes.