BackgroundSeafarers have reported impaired health and health-related quality of life (HRQOL). Social support might increase HRQOL, but little is known about this association among Chinese seafarers. The aim of this study was to describe social support and explore its association with HRQOL among Chinese seafarers.MethodsA cross-sectional survey was conducted in the ports of Nantong and Rugao, China, from April to December 2013. A total of 917 Chinese seafarers were interviewed on social support, mental distress, perceived occupational stress, and HRQOL using the following self-administered questionnaires: The Social Support Rating Scale, Self-rating Depression Scale, Occupational Stress Questionnaire, and the World Health Organization Quality of Life-BREF (WHOQOL-BREF) questionnaire. Hierarchical linear regression modelling was used to analyze the association between seafarers’ subjective level of social support and their HRQOL.ResultsOf the 917 male Chinese seafarers included in the study, 40.7% perceived high levels of social support, and 39.1% were highly satisfied with their overall quality of life (QOL). Hierarchical regression analysis showed significant associations between level of social support and all health dimensions in the WHOQOL-BREF, even after adjusting for depressive symptoms, occupational stress, occupational activities, sleep duration, and other relevant covariates. Compared with the medium or low level social support group, seafarers with a high level of social support had better QOL scores in the general facet health and QOL (β = 2.43, p<0.05), and the physical health (β = 3.23, p<0.001), psychological health (β = 5.56, p<0.001), social relation (β = 6.07, p<0.001), and environment domains (β = 4.27, p<0.001). In addition, depression, occupational stress, occupational activities, and sleep duration were found to be determinants of seafarers’ HRQOL.ConclusionsChinese seafarers have poorer HRQOL than the general population, but social support has a significant positive effect on their HRQOL. Efforts to improve social support should be undertaken.

The prevalence of metabolic syndrome (MS) varies worldwide due to genetic and environmental factors. A population-based cross-sectional study, with 37,582 participants recruited in Nantong, China in 2007-2008 (stage I) and 2013 (stage II). Socio-demographic, lifestyle factors, disease history and fasting blood sample were collected. The prevalence of MS was much higher in 2013 (42.6%) than that in 2007-2008 (21.6%), which was significantly higher in older people in both stages. Participants with two or more familial history of diseases were associated with a higher MS prevalence compared to those who didn’t have familial history of diseases. Total physical activity (PA) was associated with 24 and 19% decreased risk of MS in men and women. Occupational PA in moderate and vigorous intensity was associated with a 25%-45% decreased risk of MS. Female smokers who smoked more than 10 cigarettes per day or over 25 years were associated with 96 and 74% increased MS risk, respectively. The highest quartile of rice wine consumption was associated with a lower risk of MS with OR of 0.63 in women, compared with female non-drinkers. These findings indicated that PA and rice wine are two protective factors in MS prevention in rural areas of East China.

BackgroundIt is challenging to deal with mixture models when missing values occur in clustering datasets.Methods and ResultsWe propose a dynamic clustering algorithm based on a multivariate Gaussian mixture model that efficiently imputes missing values to generate a “pseudo-complete” dataset. Parameters from different clusters and missing values are estimated according to the maximum likelihood implemented with an expectation-maximization algorithm, and multivariate individuals are clustered with Bayesian posterior probability. A simulation showed that our proposed method has a fast convergence speed and it accurately estimates missing values. Our proposed algorithm was further validated with Fisher’s Iris dataset, the Yeast Cell-cycle Gene-expression dataset, and the CIFAR-10 images dataset. The results indicate that our algorithm offers highly accurate clustering, comparable to that using a complete dataset without missing values. Furthermore, our algorithm resulted in a lower misjudgment rate than both clustering algorithms with missing data deleted and with missing-value imputation by mean replacement.ConclusionWe demonstrate that our missing-value imputation clustering algorithm is feasible and superior to both of these other clustering algorithms in certain situations.

We evaluated how metabolic disorders affected antihypertension therapy. 2,912 rural Chinese patients with hypertension who provided blood samples, demographic and clinical data at baseline and after 1 year of antihypertension therapy were evaluated. At baseline, 1,515 patients (52.0%) were already receiving drug therapy and 11.4% of them had controlled blood pressure (BP). After 1 year, all 2,912 patients were receiving antihypertension therapy that was administered by community physicians, and 59.42% of them had controlled BP. Central obesity and abnormal triglyceride, high-density lipoprotein cholesterol, and glucose were associated with 15–70% higher risks of uncontrolled hypertension. Metabolic syndrome using the JIS criteria was associated with poor BP control (odds ratio: 1.71 and 1.54 for the baseline and follow-up datasets, respectively). The risk of uncontrolled hypertension increased with the number of metabolic disorders (p for trend <0.01). The presence of ≥3 metabolic disorder factors was associated with higher risks of poor BP control. The associations of metabolic factors and uncontrolled hypertension were stronger for the standard and modified ATP III criteria, compared to the IDF and JIS criteria. Metabolic factors were associated with less effective antihypertension therapy, and all definitions of metabolic syndrome helped identify patients with elevated risks of uncontrolled hypertension.

The prevalence of metabolic syndrome increases rapidly worldwide, and its association with physical activity (PA) varies with race and lifestyles. Little is known about the association in rural China. The Nantong Metabolic Syndrome Study recruited 13,505 female and 6997 male participants in 2007 and 2008. Socio-demographic characteristics, and physiological and behavioural data were collected. Logistic regression model was applied to estimate associations of metabolic syndrome and its components with different PAs. The overall metabolic syndrome prevalence was 21.6% in current study. Increasing total PA or moderate-to-vigorous-intensity occupational PA was associated with decreasing 5%-60% risk of having metabolic syndrome and abnormal metabolic syndrome components in both genders. An association between leisure-time PA and blood pressure was found in men, but no associations between leisure-time PA and metabolic syndrome components were found in women. Commuting PA, such as walking and taking bus, by bicycle and walking only, was associated with decrease of 20%-45% risk of several abnormal metabolic syndrome components in women. This study provides information for future investigation into the nature of these associations so that recommendations can be developed to reduce the prevalence of metabolic syndrome and its components among rural population in China.