Development And Evaluation Of Patient-reported Outcomes Instrument For Stroke

Objective:This study is based on the US Food and Drug Administration definition of patient-reported Outcome Instruments, to develop a patient-feeling centered stroke disease PRO scale that suit our own culture, and evaluate the performance and function of the scale, so that this scale is applicable to the clinical treatment evaluation and clinical trial of new medicines.Method:The PRO instrument development standard and procedures are followed during the development of this scale; the framing theory and the draft scale are formed after significant information of stroke disease was gained through reading previous document and literature, and through consulting doctors and patients. Then the primary scale was formed by patients and experts’judgment of the relevance between the item and dimension in the expected structure, and the correction of the frame item pool. Extract samples from a small area for pre-examination, select the items according to classic measure theory and item response theory, so as to eliminate the unqualified items, and start formal examination with revised scale. Select the items once again using the same method after the scale copies are collected, so that the final Stroke Patient PRO Scale is formed. Check the reliability, validity, and feasibility of the scale, and analyze the possible differential item functioning caused by gender.Results:Final "Stroke Patient PRO Scale" consists of46items,10dimensions, and4domains, and is formed after three items correction of content validity degree, pre-examination and formal examination.20items are included in the physical domain, which belong to four dimensions:Somatic symptoms, cognitive ability, language communication and self-care ability;14items are included in the psychological domain, which belong to three dimensions:anxiety, depression, and avoidance;7items are included in the social domain, which belong to two dimensions:social communication and family support;5items are included in the treatment domain, which belong to satisfactory dimension. Then the assessment of the reliability, validity, fairness, and feasibility is conducted. Reliability:reliability is verified by Cronbach’s α coefficient with0.905of total scale.Validity:structural validity result shows that the multi dimensional measurement of the scale fulfilled the expectation. Fairness:DIF showed that only PHD9and PSD3have cross gender DIF. Feasibility:among the sample survey of the stroke patients, the scale copy acceptance rate is99.2%, completion rate is98.9%, and the average filling time is8.9minutes.Conclusion:The stroke patient PRO scale has good reliability, validity, fairness, and feasibility, thus can be considered as the clinical treatment evaluation instrument for stroke patients.