Development Of The Post-Stroke Early Depression Screening Tooland Its Preliminary Application

PART 1: DEVELOPMENT OF THE POST-STROKE EARLY DEPRESSION SCREENING TOOLObjective: We developed a new early symptom measurement for post-stroke early depression using the standard measurement development methods, and explored the factor structure, the reliability and validity of the new measurement.Methods: Based on stress theory on psychology, through a number of steps we establish the post-stroke early depression screening tool, we also established the norm of our measurement. Firstly, we constructed a 51-item,6-domain pool through systematic literature review(literature data,diagnostic criteria of major depression, qualitative study results of stroke patients with depression and patient interview). Secondly, we attained a35-item, 6-domian preliminary post-stroke early depression screening tool through research group discussion, interviewing of 74 healthcare professionals, Delphi consultation of 16 experts and pilot investigation of40 stroke patients. Thirdly, we interviewed 410 acute stroke patients using the post-stroke early depression screening tool(pre trial version). Critical ratio, correlation analysis, internal consistency and factor analysis were used to evaluate the tool’s item quality, 6 items were deleted due to unqualified results of the item analysis. They were: “I blame of my past bad living habits”, “I don’t want to communicate with others”, “I lose interest of everything”, “I pray for a miracle”, “I have a bad appetite” and “I feel I am losing weight”. Besides, 9 items were deleted due to unqualified construct validity though explaratory factor analysis and domain-total correlation. They were: “I feel I would never recover from the illness”, “I am very easy to get angry”, “I feel I am a burden of my family”, “I feel I was trapped by the illness”, “I feel it is my destiny to have the illness as a punishment of me”, “I feel I am useless”, “I always blame myself”, “I feel it is hard to do anything after stroke” and “I am easy to be fatigue than before”. The criteria validity of our measurement was good by comparing it with Hamilton Depression Rating Scale-24 version. The reliability of our measurement was also good based on the Cronbach α and test-retest reliability coefficient. We got the final version of our measurement including 20 items and 5 domains. Fourthly, we investigated 505 acute stroke patients using the final version of the post-stroke early depression screening tool. Confirmatory factor analysis was used to establish the stability of the factor structure. The reliability of our measurement was alsofurther established through internal consisitency(Cronbach α). The indexes of all confirmatory factor analysis indicators were all reached the expected criteria, and the domain and total measurement reliability were all exceeded the expected criteria of 0.70. Finally, we constituted the norm of our new measurement in Wenzhou city using a sample of 915 patients which were attained from the two big sample investigation of our study. The norms included the mean and standard deviation; the Z-score and T-score, and the cut-off of our measurement.Results: We finally established a new measurement of post-stroke early depression including 20 items and 5 domains. The domains of our new measurement are low, dull, emotional, nervous and wakefulness. The measurement is a self-rating scale using a 5 Likert scale ranging from 0 to4. The responses of the items are “no”, “rarely”, “sometimes”, “frequently”,“always” with each has a corresponding score from 0 to 4 respectively.Domain scores can be attained by adding up the scores in each domain,total score can be attained by adding up all the item scores of the measurement. The range of the total score is 0- 80, higher score indicates high level of early depression. Confirmatory factor analysis indicated that the measurement had a stabe factor structure including 5 domians; the reliability of the factors ranged from 0.775-0.908 and the reliability of the total scale was 0.888, which were all evidences of good reliability and validity. Besides, we established the norm of the post-stroke earlydepression screening tool including the mean(standard deviation) norm,the Z score and T-score norm and the cut-off value of the our measurement.The cut-off of the post-stroke early depression screening tool was defined as: scores < 11.5 = No PSD, scores 11.5-23.5 = Low PSD, scores 23.5-36.5= Moderate PSD, scores > 36.5 = High PSD.Conclusion: The post-stroke early depression screening tool showed stable factor structure and good reliability and validity. Our measurement can be used to detect early depressive symptoms of acute stroke patients and for the research on early depressive symptoms of acute stroke patient.PART 2: THE INCIDENCE AND INFLUENCING FACTORS OF EARLY DEPRESSION IN POST STROKE PATIENTSObjective: To explore the incidence and influencing factors of early depression in post-stroke patients and provide theoretical guideline of developing inventions related to early symptoms of post-stroke depression.Methods: A cross-sectional survey was carried on 304 acute stroke patients 7 to 30 days after stroke using the general questionnaire(the demographics questionnaire and the stroke related clinical questionnaire)and the post-stroke early depression screening tool. Descriptive statistics(percentage) was used to calculate the incidence of post-stroke early depression; t-test and chi-square test was employed to explore the influence factors of post-stroke early depression; we set significant factors in the t-test and chi-square test as independent variables and post-stroke early depression as dependent variables, and binary logistic regression analysis were used to attain the best predictors of post-stroke early depressive symptoms.Results: The incidence of early depressive symptoms of acute stroke patients was 76.0%, in which 26.3% were mild early depression, 29.3%were moderate early depression and 20.4% were severe early depression.Univariate analysis indicated that the influencing factors of post-strokeearly depression were average sleep time everyday, household income of the patient, living place of the patients, living rooms of the patients, left hemisphere stroke, right hemisphere stroke, stroke in frontal lobe, stroke in temporosphenoid lobe, stroke in parietal lobe, stroke in ventricle and Lacunar infarction. The binary logistic regression analysis showed that average sleep time everyday, household income of the patients, living rooms of the patients, left hemisphere stroke and right hemisphere stroke were the best predictors of post-stroke early depression. Average sleep time everday, household income and staying in 6-people ward were the protective factors of post-stroke early depression with odds ratio(OR)less than one; whereas living rooms of the patients, left hemisphere stroke and right hemisphere stroke were the risk factors of post-stroke early depression with OR more than one.Conclusion: The average sleep time everyday, household income of the patients, living rooms of the patients, left hemisphere stroke and right hemisphere stroke were the best predictors of post-stroke early depression.Though we can not change patients’ household income and stroke lesion location, improving patients’ sleep condition and change the ward rooms will be an effective solution to reduce the risk of post-stroke early depression in acute stroke patients.