Compare Data Quality Between ERPO And PRO

A Patient Reported Outcome (PRO) is a questionnaire used in a clinical trial, where the responses are collected directly from the patient. It reflects patients’ health related quality of life and plays an important role in clinical assessment."Guidance for Industry, Patient-Reported Outcome Measures:Use in Medical Product Development to Support Labeling Claims", launched by FDA, define PRO as:A PRO is any report of the status of a patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else.PRO is widely used in modern clinical trials. With the development of modern computers, electronic systems, known as EDC, are also popular and widely accepted by doctors and patients. As the result, electronic patient reported outcomes (EPRO) are used more frequently. We often hold the point that EPRO can get accurate data with high data quality in time, with its convenience for use. These conclusions were mostly built on dummy cross-over study. What the real picture is in real clinical trial should be explored. With this kind of study, we can find the true story about PRO and EPRO and apply it in the future.ObjectivesPicking up questionnaires with both PRO and EPRO strategies which are widely used in Pfizer clinical database, we aim to compare the data quality between PRO and EPRO from the perspective of clinical data management and give guidance on future use of PRO and EPROMethodsBased on real clinical data in Pfizer clinical database, we picked two questionnaires, The25-ltem National Eye Institute Visual Functioning Questionnaire (NEI-VFQ-25, VFQ) and Western Ontario and McMaster Universities Arthritis Index(WOMAC),which both have PRO and EPRO strategies. The EPRO method for VFQ is telephone from call center and the EPRO methods for WOMAC are Interactive Web Response System (IWRS) and Personal Digital Assistant (PDA). The other data which were used in the analysis are collection date of vital signs, dosage information, and subject summary. The main tool used are Microsoft excel and SAS9.2.Resultsl.For VFQ, these two data collection methods:PRO and EPRO(1) On these data points, PRO data quality is significantly better than EPRO:①patients with no baseline in PRO group are less than in EPRO(p<0.001)②patients without questionnaire data in PRO group are less than in EPRO(p<0.001)⑤patients with extra questionnaires in PRO group areless than in EPRO(p<0.001)⑧patients in PRO group with questionnaire collected later than last visit date are less than in EPRO(p<0.001).(2) On these data points, EPRO data quality is significantly better than PRO:④questionnaires with missing value in EPRO group are less than in PRO(p<0.001)⑥questionnaires with discrepancy in EPRO group are less than in PRO(p<0.001)(3) On these data points, there is no significant difference between PRO and EPRO⑦questionnaires not within visit window③missing questionnaires2. For WOMAC, two EPRO methods were used:eDiary(EPRO1) and IWRS(EPRO2).For EPRO1,(1) On these data points, PRO data quality is significantly better than EPRO:⑤patients with extra questionnaire in PRO group are less than in EPRO(p<0.05)(2) On these data points, EPRO data quality is significantly better than PRO:③missing questionnaires in PRO group are more than in EPRO(p<0.05)④questionnaires with missing value in PRO group are more than in EPRO(p<0.001)⑦questionnaires not within visit window in PRO group are more than in EPRO(p<0.001)(3) On these data points, there is no significant difference between PRO and EPRO:①patients with no baseline②patients without questionnaire data⑧patients with questionnaire collected later than last visit date.For EPRO2,(1) On these data points, PRO data quality is significantly better than EPRO:④questionnaires with missing value in PRO group are less than in EPRO(p<0.001)(2) On these data points, EPRO data quality is significantly better than PRO:⑦questionnaires not within visit window in PRO group are more than in EPRO(p<0.001)(3) On these data points, there is no significant difference between PRO and EPRO:①patients with no baseline②patients without questionnaire data⑤patients with extra questionnaires⑧patients with questionnaire collected later than last visit date.ConclusionIn general, EPRO strategy performs better on reducing missing values in the questionnaire, PRO strategy performs better on reducing extra questionnaires. On other data point, there is no uniform rule among three kinds of EPRO strategies. Different EPRO strategies perform differently on data points. EPRO with telephone gets more data issues about data quality. IWRS and eDiary perform better than telephone.