Validity of congenital malformation diagnoses in healthcare claims from a university hospital in Japan

Abstract

Purpose: This study aimed to assess the validity of diagnoses of congenital malformations (CMs) recorded in claims of a university hospital in Japan.Methods: Congenital malformations were identified according to Code Q00-Q89 of the International Classification of Diseases, 10th revision. All the children who had been diagnosed with CMs based on their claims in 2015 and within 1 year from their birth month were selected for this study. The infants' medical records were considered as a gold standard. Positive predictive values (PPVs) for CMs were calculated.Results: This study included 227 infants who had a CM diagnosis in their claims.Based on the algorithms established by the Quebec Pregnancy Cohort study group, the PPV for any CM was 90.7% and that for major CMs (MCMs) was 91.5%.Concerning MCMs of specific organ systems, those of the circulatory system (PPV 85.1%) were the most frequent, followed by cleft lip and cleft palate (PPV 100.0%), and other CMs of the digestive system (PPV 96.4%). Based on the EUROCAT classification, the PPV for any MCM was 88.5%. Specific MCMs reported in ≥20 infants were ventricular septal defect (PPV 96.0%), patent ductus arteriosus (PPV 72.7%) and cleft lip with or without cleft palate (PPV 100.0%). Conclusions:The PPVs for CMs in the Japanese administrative data were high enough to suggest that these data could be utilized for perinatal pharmacoepidemiological evaluations. The results were from a single center, and further validation studies are needed.

Validation studies of Japanese administrative health care data: A scoping review

“…For example, although values of validity measures differed, studies consistently showed lower accuracy of the diagnosis of ischemic stroke compared with hemorrhagic stroke 30,32,47,52 . We also found validation studies for rare conditions, such as type 1 diabetes, congenital malformations, and hemophilia A 31,35,37 . Another finding of this review was that most of the studies targeted one or several diseases.…”

“…Full‐text articles of 42 studies were retrieved, and six were excluded because they compared aggregate data, validated self‐report questionnaires using claims data as a reference standard, were review articles, were aimed at refining the cancer registry and not for research purposes, did not report case‐defining methods, or did not report a reference standard 12,19–23 . There were 36 eligible studies, including 29 studies 8,10,11,13,24–48 that validated administrative data using an external reference standard and seven studies 9,49–54 that validated administrative data using other data within the same database.…”

No association between major congenital malformations and exposure to Kampo medicines containing rhubarb rhizome: A Japanese database study

“…Methods for estimating GA in Japanese administrative data have been established, allowing the identification of early pregnancy prescriptions ( Ishikawa et al, 2018a ). Furthermore, it was shown that reliable teratogenic risk assessment is possible in Japanese administrative data ( Ishikawa et al, 2021 ). With the establishment of this method, the present study showed, based on a large database, that the association between KRR and the risk of MCMs in women prescribed MgO was comparable and that KRR could be a safe treatment option as a laxative during pregnancy.…”

“…Supplementary Table S3 shows the selected MCMs and their ICD-10 codes. Congenital minor malformations (ICD-10 codes Q00–Q89) were excluded, and diagnoses of MCMs given to newborns during the first year of life were extracted ( Ishikawa et al, 2021 ). In addition, the MCMs were listed separately individual organ system.…”

“…In addition, the MCMs were listed separately individual organ system. A study based on university hospital records demonstrated that the MCMs in the claims were validated against medical records, and the overall positive predictive value of the MCMs was about 90% ( Ishikawa et al, 2021 ).…”