| ObjectiveTesting the reliability,validity and the screening function of DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure—Adult(DCSM-A).Methods103 patients with mental disorders and 104 patients with psychosomatic disorders,who aged between 18 and 65 and meet the inclusion criteria were selected.At the same tine,102 healthy people whose demographic characteristics are similar to the above two groups as normal controls were selected.First,all the individuals need to fill in their general demographic information.Then they completed DCSM-A,SF-36 and WHODAS2.0.After that,we retest reliability of 20 samples in mental disorder group according to their scores of DCSM-A.And we retest reliability of 20 samples in normal control group,whose demographic characteristics are similar to mental disorder group.SPSS20.0 was used to analyze the three groups' general information,psychological evaluation results and the split-half reliability coefficients,internal consistency reliability,item analysis,test-retest reliability and test-retest differences of DCSM-A.We analyzed the correlation between SF-36,WHODAS2.0 with DCSM-A to test the concurrent validity of DCSM-A Results1.The difference between the groups: the DCSM-A total score in the control group(8.67±8.64),psychosomatic diseases group(7.88±5.51)and mental disorders group(13.13±11.78)showed statistically significant differences(F=10.18,P=0.000).The score in the mental disorders group was higher than that in the control group and psychosomatic diseases group(P<0.05).The differences between the control group(42.24±10.29),psychosomatic disease group(42.43±7.35)and mental disorders group(59.46±15.60)were statistically significant(F=74.95,P=0.000).In all dimensions of SF-36(except body function),the difference between the other dimensions was statistically significant(F=11.14?171.59,P<0.0001),except for the body pain psychosomatic disease group,the score of the other dimension mental disorders group was lower than that of the control group and the psychosomatic disease group(P<0.05).2.Reliability analysis: the fractional half-reliability and Cronbach's coefficients of DCSM-A were 0.82 and 0.84 for the total sample,0.84 and 0.90 for the control group,0.81 and 0.85 for the mental disorders group,and 0.58 and 0.64 for the psychosomatic diseases group,respectively.The retest reliability of the total DCSM-A score in the total retest sample is 0.89,and that of the field retest is 0.50-0.94.The total retest reliability of the control group was 0.90,and the field was divided into0.32-0.97.The total retest reliability of the mental disorders group was 0.88,and the field score was between 0.54 and 0.98.In the mental disorders group,only the rescheduling scores in the substance dependence field were statistically significant(P=0.02),while the rescheduling scores in other fields were not statistically significant(P<0.05).3.Validity analysis: the correlation between total DCSM-A score and total WHODAS2.0 score(r=0.59)and field score(r=0.26?0.59)was statistically significant(P<0.01),the correlation(r=0.17?0.46)between the total score of WHODAS2.0 and DCSM-A(excluding substance dependence)was statistically significant(P<0.01);Most correlation values(r=0.15?0.48)were statistically significant(P<0.01).The correlation coefficient between DCSM-A total score and SF-36 life vitality(VT)was the highest(r=-0.43),and the correlation coefficient with other SF-36 subscales ranged from-0.20 to-0.39(P<0.01);In the domain correlation,the correlation coefficient between DCSM-A somatic symptoms and SF-36 somatic pain(BP)was high(r=-0.54),and most of the correlation coefficients between other domains were statistically significant(P<0.01).4.Screening efficacy analysis: the sensitivity of DCSM-A total score to distinguish mental disorders from normal control was 78.6%,and the specificity was46.1%.The sensitivity to distinguish psychosomatic diseases from normal control was 71.2%,and the specificity was 41.2%.When any field is divided into screening indexes,the optimal demarcation value of each symptom field is different in different disease groups as the demarcation value increases,the sensitivity decreases and the specificity increases.The sensitivity,specificity,positive predictive value and accuracy of screening for psychosomatic diseases were 87.4%,63.7%,70.9%,83.3%,71.2%,62.7%,66.1% and 67.0% respectively.Conclusions1.The Chinese version of DCSM-A is the same as the original version.Both the total score and the field score have good retargetry reliability.2.DCSM-A can distinguish normal control,psychosomatic diseases and mental disorders,and is significantly correlated with WHODAS2.0 and SF-36 scores,suggesting that the Chinese version of DCSM-A has good empirical validity and calibration validity.3.The total score of DCSM-A is not suitable as A screening index,and the American demarcation value standard has high sensitivity and low specificity,which is suitable for screening.The sensitivity and specificity of the best demarcation value standard are moderate,which is suitable for the initial diagnosis. |
PDF Full Text Request