Validity of participant-reported diagnoses in an online patient registry: A report from the NF1 Patient Registry Initiative☆

Abstract Background With increased internet accessibility worldwide, it is now possible to assemble individuals with rare diseases through web-based patient registries. However, the validity of participant-reported medical diagnoses is unknown. The objective of this study was to evaluate the accuracy of participant-reported Neurofibromatosis Type 1 (NF1) diagnoses among participants in the NF1 Patient Registry Initiative (NPRI). Methods Subjects enrolled in the NPRI from 5/17/2011 to 7/7/2014 were included. Medical records (MRs) were obtained for participants who returned medical record release forms (MRRFs) during the study period. Participants were classified as having definite, probable, suspected, or no NF1 diagnosis based on MR information. To assess whether a returned MRRF served as a reliable marker of MR-documented NF1, we calculated the positive predictive value (PPV) as the proportion of individuals with MR-documented NF1 among those from whom MRs were obtained. We further examined whether a returned MRRF predicted the number of reported NF1 clinical signs in multivariable linear regression analyses. Results A total of 1456 individuals were included in the analyses. Of 416 individuals who returned MRRFs, 205 MRs were reviewed within the study period. The PPV ranged from 72.0 to 98.5% when including definite or definite/probable/suspected cases, respectively. The mean number of reported NF1 clinical signs was similar between those who returned (mean = 3.3 ± 1.2) and did not return (mean = 3.2 ± 1.3) their MRRFs. MRRF return was not a significant predictor of the number of NF1 clinical signs after adjusting for covariates. Conclusion These data strongly suggest that individuals enrolling in the NPRI accurately report their NF1 diagnosis.