Background. Psychiatric epidemiological surveys in China have repeatedly found much lower prevalence estimates than in most other parts of the world.Method. Face-to-face household interviews of 5201 subjects (2633 in Beijing and 2568 in Shanghai respectively) were conducted from November 2001 to February 2002 using a multistage household probability sampling method. A Chinese version of the World Health Organization Composite International Diagnostic Interview (CIDI) was used for assessment.Results. Twelve-month prevalence of any DSM-IV mental disorder in metropolitan China is estimated to be 7·0%, with major depressive disorder (2·0%), specific phobia (1·9%), and intermittent explosive disorder (1·7%) the most common disorders. Of these, 13·9% are classified as serious, 32·6% moderate, and 53·5% mild. Only 3·4% of respondents with any disorder sought treatment within the previous 12 months.Conclusions. Although the general pattern of disorders, risk factors, and unmet need for treatment are similar to those in other countries, a low prevalence of mental disorders is found in metropolitan China. Resolving methodological problems that cause downward bias in estimates, such as stigma-related under-reporting and diagnostic incongruity with a somatopsychic mode of symptom presentation may lead to more accurate and probably higher prevalence estimates in future epidemiological studies. As a low prevalence still translates into an enormous number of people in China, measures are urgently needed to address the huge unmet need for treatment of mental disorders.

Previous studies in Western countries have consistently documented positive associations of smoking and heavy drinking with depressive symptoms but a prospective analysis of these relationships among middle-aged and elderly community members in China have not previously been reported. Using data from the China Health and Retirement Longitudinal Study, a two-wave nationally representative survey conducted in 15,628 adults 45 years of age and older, we estimated the prospective association between depressive symptoms and an array of smoking and drinking behaviors. Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D) short form. Inverse associations were the dominant pattern of association. For the population as a whole, individuals with baseline depressive symptoms were less likely to start drinking (OR=0.7, 95% CI=0.5, 0.9) or smoking (OR=0.6, 95% CI=0.4, 0.8). Similarly, baseline drinkers and smokers were less likely to develop depressive symptoms (ORdrinkers=0.6, 95% CI=0.5, 0.7; ORsmokers=0.7, 95% CJ=0.6, 0.9). No evidence was found for an increased incidence or persistence of depressive symptoms among high-frequency drinkers or heavy smokers or vice versa. Males who had never smoked prior to the onset of depressive symptoms tended to have more rapid onset of tobacco dependence compared to those without such symptoms. Males and females had different association patterns. The study is observational in nature and provides limited evidence for causality. The results are inconsistent with previous findings in Western countries, throwing into question the presumed universality of the association between alcohol drinking or tobacco use and depression among middle-aged and elderly adults.

The secondary analysis of existing data has become an increasingly popular method of enhancing the overall efficiency of the health research enterprise. But this effort depends on governments, funding agencies, and researchers making the data collected in primary research studies and in health-related registry systems available to qualified researchers who were not involved in the original research or in the creation and maintenance of the registry systems. The benefits of doing this are clear but the barriers are many, so the effort of increasing access to such material has been slow, particularly in low- and middleincome countries. This article introduces the rationale and concept of the secondary analysis of existing data, describes several sources of publicly available datasets, provides general guidelines for conducting secondary analyses of existing data, and discusses the advantages and disadvantages of analyzing existing data.