Development And Psychometric Analysis O F The Multidimensional Pregnancy Psychosocial-stress Questionnaire

Objectives The purpose of the present study was to develop and evaluate the reliability,validity and responsiveness of the Multidimensional Pregnancy Psychosocial-stress Questionnaire(MPPQ), for the guidance of the research of the establishment and application of the pregnancy psychosocial-stress multidimensional evaluation method for pregnant women.Methods Theoretical concept was proposed base on the literature and reference psychosocial stress scales extensive review. The stress during pregnancy is divided into eight dimensionalities: communication barriers, family economic status, work pressure,unemployment, anxiety of fetus, anxiety of self, anxiety of childbirth, according to which the original scale was established. Then focus group discussion and expert advice to modify the questionnaire. The original questionnaire consists of two subscales including 50 items(life events 35 items, pregnancy-related anxiety 15 items).3 195 first trimester pregnant women were recruited in Hefei to collect psychosocial stress during pregnancy, 3 125 individuals were valid. The omission rate and response rate, two-sample t-test, correlation analysis and factor analysis were applied to analysis the items of the questionnaire. Eventually the Multidimensional Psychosocial-stress Questionnaire of Pregnancy(MPPQ) consisting of 23 items was established.Further Psychometric Analysis was carried out. 34 pregnant women were retested in 2 to 4 weeks, Pearson correlation coefficient were calculated to evaluate the reliability, and the Cronbach’s α coefficient were calculated to evaluate the internal consistency reliability of the questionnaire. Pregnant outcomes of 1 075 pregnant women were tracked, using Fisher-test to analysis the difference between pregnant women at different ages, education, family income etc. to evaluate the discriminantvalidity of the questionnaire, using χ2 test to evaluate the predictive validity.Results Through the analyzing of the response rate, 12 items <0.1 were deleted. One item was deleted using the t-test analysis. According to factor analysis, 3 items were deleted in life events questionnaire, while 5 items were deleted in pregnancy-related anxiety questionnaire. 4 items were deleted after Cronbach’s alpha coefficient analysis.According to three correlation coefficient indicators, items were deleted if indicators less 3, all items were selected. The final result shows that: after factor analysis(principal component analysis), internal consistency analysis and correlation analysis,27 items were deleted(20 of life events questionnaire, 7 of pregnancy-related anxiety questionnaire), therefore confirmed MPPQ were consisted of 23 items.Test-retest correlation coefficients of the two questionnaires were 0.679 and 0.615,Cronbach’s alpha coefficient were 0.637 and 0.785. MPPQ is divided into 8dimensions: communication problems, domestic economic problems, workload problems, work of husband, work of self, anxiety of fetus, anxiety of self, anxiety of childbirth, which consistent with the original theoretical conception. 5 common factors were extracted for life events questionnaire, the cumulative variance contribution rate is 55.363%; 3 common factors were extracted for pregnancy-related anxiety questionnaire, the cumulative variance contribution rate is 66.193%. Association between different maternal age, education level, the frequency of husband-smoking and MPPQ score was statistically significant(P <0.5), which confirmed the MPPQ has a certain discriminate validity. We did not find the association between MPPQ score and adverse pregnancy outcomes.Conclusion MPPQ is certain reliable and valid. It is can be used as the basic tool of psychosocial stress detecting in pregnant women during pregnancy.