作者: Lucía Cea Soriano , Montse Soriano‐Gabarró , Luis A. García Rodríguez
DOI: 10.1002/PDS.3877
关键词: Medical record 、 Colorectal cancer 、 Pharmacoepidemiology 、 Read codes 、 Response rate (survey) 、 Internal medicine 、 Medical diagnosis 、 Referral 、 Fast track 、 Medicine
摘要: Purpose To validate the recorded diagnoses of colorectal cancer (CRC) and identify false negatives in The Health Improvement Network (THIN) primary care database. Methods We conducted a validation study incident CRC cases THIN among patients aged 40–89 years from 2000–2011. Read code entries (N = 3805) were verified by manual review patients' electronic medical records (EMRs) including free-text comments. Incident ascertained following validated against two data sources deemed gold standards: (i) questionnaires sent to practitioners (PCPs; for random sample 100 potential cases), (ii) Hospital Episode Statistics (HES) linked practices. False identified searching International Classification Diseases-10 codes related HES. Results Of 3805 via codes, 3033 (80.0%) considered definite after EMRs. positive predictive value (PPV) was 86.0% removing ‘fast track referral suspected CRC’. response rate PCPs 87.0% (n = 87), PPV 100% based on PCP questionnaires. Using HES, 97.9% (556/568), negative 6.1% (36/592). Conclusions CRC diagnostic have high PPV, which is increased further low. © 2015 Authors. Pharmacoepidemiology Drug Safety published John Wiley & Sons Ltd.