Validation of the Q87.11 ICD Code for Prader‐Willi Syndrome

BACKGROUND: Prader-Willi syndrome (PWS) is a rare genetic disorder that is identified by the ICD-10 code Q87.11. Although this code has been used in several large studies, its diagnostic validity has not been evaluated. This study assessed the accuracy of the Q87.11 code for identifying individuals with PWS. METHODS: The electronic health record (EHR) of a large health system was reviewed for patients with the Q87.11 code. Additionally, a 10% random sample of patients with free-text evidence of possible PWS but no code was reviewed. All charts were assessed for PWS using expert review. Diagnostic validity of the Q87.11 code was assessed at both the individual level (i.e., using a single diagnostic code at any point in the EHR to define the patient as having PWS) and encounter level (looking at coding behaviour for individual inpatient or outpatient episodes of care) using sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). RESULTS: Of 2 718 218 patients with ≥ 4 recorded encounters, 126 had at least one Q87.11 code; 122 were confirmed to have PWS by chart review. This yielded a PPV of 96.8% (95% CI: 92.0%-99.1%). Full-chart review of patients with text evidence of PWS but no code estimated a code sensitivity of 92.4% (95% CI: 68.9%-100%) and specificity and NPV of 100%. On a per-encounter basis, sensitivity was 79.4% (95% CI: 62.6%-91.1%) for inpatient and 46.0% (95% CI: 44.1%-47.9%) for outpatient visits. CONCLUSIONS: A single use of the Q87.11 diagnostic code produces excellent sensitivity, specificity, PPV and NPV for identifying patients with PWS, although sensitivity is lower for individual outpatient encounters compared with inpatient ones. The overall accuracy of the Q87.11 code supports the validity of using this code in the analysis of administrative claims datasets.