1.1 subjects
Postmenopausal women who underwent medical examinations in one health-checkup agency in Tianjin from May 2015 to October 2015 were enrolled in this study.
Inclusion criteria: 1) patients had menopause for more than one year; 2) patients did not have medical history of cardiovascular disease, such as coronary heart disease, ischemic stroke, etc.; 3) patients was able to complete the questionnaire independently.
Exclusion criteria: 1) patients was unable to participate in daily physical activity; 2) patients had a medical history of psychosis; 3) patients was unwilling to participate in this study; 4) patients did not complete the questionnaire for various reasons.
1.2 Calculation of sample size
According to the requirements of logistic regression analysis, the sample size is generally 20 to 30 times the number of variables. The number of variables in our present study was 12. Considering the drop out of the samples and the invalid questionnaire, we set the drop out rate as 10% to obtain the sample size greater than 264. Thus, we finally determined to include 300 cases in this study.
1.3 Theoretical framework
SET is designed to explore the relationship between the social ecological environment and human behavior, and how they interact and influence each other. This theory focuses on individuals in the social ecological system, and believes that interaction and mutual-influence of individuals and the environment are dynamic and coadapted, emphasizes that the social ecological environment is of great significance for the exploration and interpretation of human behavior, and to a certain extent, provides a more comprehensive theoretical basis for social work.
This theoretical model consists of micro-level system, meso-level system and macro-level system. The micro-level system refers to individuals who exist independently in the social ecological environment, with physical, psychological, and social attributes. The meso-level system refers to small-scale organizations that have an impact on individuals, including families, colleagues, or other social groups. The macro-level system refers to a social system that is larger than the meso-level system, mainly including environment, culture, community, etc. These three-level systems are always interacting and influencing each other. Therefore, in order to understand and solve problems more comprehensively, we should explore the underlying causes of human behavior from multiple levels and perspectives.
1.4 Research tools
1.4.1 basic questionnaire on general data
A self-designed basic questionnaire for general information, such as age, working condition, education, marriage status, (body mass index) BMI, medical history, blood pressure, total cholesterol and triglycerides, was used.
1.4.2 Ischemic Cardiovascular Disease (ICVD) 10-year Risk Assessment Form
The "Ischemia Cardiovascular Disease (ICVD) 10-year Risk Assessment Form" (14) developed by the research team of the national "Tenth Five-Year Plan" research project was used to do the risk assessment of ICVD in postmenopausal women, including 6 risk factors: age, systolic blood pressure, total cholesterol, body mass index (BMI), diabetes mellitus (DM) and smoking. Each risk factor was given different risk scores according to different conditions of cases, and finally the sum of risk scores of 6 factors was calculated for each case. According to the absolute risk, the samples were divided into two groups: the low-risk group (absolute risk <10%) and the high-risk group (absolute risk ≥10%).
1.4.3 Self-rating depression scale
The self-rating depression scale (15) developed by W.K.Zung in 1965 was used to assess the subjective feelings, including 20 items, 10 of which were scored in reverse. A 4-level scoring of 1 to 4 points was used. Finally, the sum of the scores of 20 items was calculated to obtain the original total score. Standard score = original total score × 1.25.
1.4.4 D-type personality scale
The D-type personality scale (DS14) with good reliability and validity (16) was developed by Denollet J, a Dutch scholar, and translated and revised by Yu XN and Zhang JX from the Chinese Academy of Sciences. The scale consists of 2 dimensions of social inhibition and negative emotions. The questionnaire used a 5-level scoring of 0 to 4 points, of which 2 items were scored in reverse. The score of one dimension was the sum of the scores of the items in this dimension. The total score of the scale was the sum of the scores of all items in this scale. The social inhibition score and the negative emotions score were both greater than 10 points at the same time, which can be diagnosed as Type D personality.
1.4.5 Social Support Rating Scale
Social Support Rating Scale (SSRS) with good reliability and validity (17) was developed led by Xiao SY in 1986. Cronbach 'a of the the total score is 0.92, and Cronbach 'a of each item score is 0.89~0.94. The scale has 10 items and consists of 3 dimensions, including objective support, subjective support and the rate of social support utilization. The score of one dimension is the sum of the scores of the items in this dimension, and the total score of social support is the sum of the scores of the three dimensions. The higher the total score is, the more social support is and the less social isolation is.
1.4.6 Abbreviated – Neighborhood Environment Walkability Scale
The Abbreviated – Neighborhood Environment Walkability Scale (ANEWS) (18) amended by Zhou Rena from the Fudan university in 2011 consists of 5 dimensions, including 17 items. The questionnaire uses a 5-level scoring from 1 to 5 points. The score of one dimension is the sum of scores of items in this dimension, and the total score of the scale is the sum of the scores of all items. The higher the total score is, the better the walking environment is. Cronbach’s α of the scale is 0.807, the intra-group correlation coefficient (ICC) is 0.945, and the Spearman correlation coefficient is 0.721.
1.4.7 Women’s Health Initiative Physical Activity Questionnaire
The Women’s Health Initiative Physical Activity Questionnaire (WHI-PAQ) (19) developed by Anne-Marie Meyer, a American scholar, and translated and revised by Yang CJ, mainly consists of two parts, including physical exercise and housework, with a total of 8 items, such as walking, light, moderate and heavy physical exercise, housework, garden activities, and sitting and sleeping time. The formula is: total energy consumption = (weekly activity frequency × minutes per time × corresponding metabolic equivalent) / ( 60 minutes/hour). WHI-PAQ classifies the intensity of exercise into 3 levels according to total energy consumption, including low-intensity exercise (<500 MET × min × wk (-1)), medium-intensity exercise (500~1200 MET × min × wk (-1)) and High-intensity exercise (>1200 MET × min × wk (-1)). The test-retest reliability of the scale is 0.950, and each dimension is 0.957-0.669. The Spearman rank correlation coefficient of overall standard validity is 0.884, the content validity index (CVI) is 0.923, and the content validity ratio (CVR) is 0.846, with good reliability and validity.
1.5 Methods for performing the questionnaire survey
The researcher conducted a questionnaire survey for postmenopausal women who met the inclusion criteria. Before filling out the questionnaire, based on the guidelines, how to fill out and what to note was introduced by the researchers. After completing the questionnaires, the researchers checked the submitted questionnaires on the spot. For the questionnaires with the missing items, remedial measures were took in time, and then the researchers took them back without any missing items after checking again.
1.6 Statistical analysis
SPSS 18.0 statistical software was used for data entry and processing, and double entry and validation were used. If the quantitative data conformed to the normal distribution, the mean±standard deviation (SD) was used, and the difference between groups was analyzed by t-test. If quantitative data did not conform to a normal distribution, the median and inter-quartile range were used to describe the data, and the difference between groups was analyzed by rank-sum test. For qualitative data, frequency and percentage was used to describe the data, and the difference between groups was analyzed by Chi-square test. Binary Logistic regression analysis was used to analyze and explore the factors affecting the occurrence of ICVD in postmenopausal women, and P<0.05 in two sides was considered statistically significant.