Validation of hospital discharge diagnoses for hypertensive disorders of pregnancy

INTRODUCTION: A correct diagnosis of preeclampsia and gestational hypertension is important for treatment and epidemiological studies. Changes in diagnostic criteria and underreporting in certain subsets of patients may hamper validity of the diagnoses.

MATERIALS AND METHODS: We validated the discharge diagnoses of preeclampsia and gestational hypertension, which are reported to the Danish National Patient Registry, in a cohort of 2163 pregnant women by retrospective evaluation of electronic hospital data.

RESULTS: A preeclampsia discharge diagnosis was found in 113 (5.2%) of the participants. After validation, significantly more patients fulfilled criteria for diagnosis of preeclampsia (n = 163, 7.5%, p = 0.002); more had severe preeclampsia, 14 (0.6%) vs. 70 (3.2%), p < 0.001 and gestational hypertension, 62 (2.9%) vs. 46 (2.1%), p = 0.12. The diagnostic sensitivity for preeclampsia by discharge diagnosis was 55.8%; severe preeclampsia 18.6%; gestational hypertension 39.1%. Corresponding positive predictive values were 80.5, 92.9 and 29.0%. Misclassification occurred in 4.3, 2.7 and 3.3%, respectively. Misclassification was more prevalent in obese compared to lean women (10% vs. 3.6%, p < 0.0001).

CONCLUSIONS: Discharge diagnoses substantially underestimated the prevalence of preeclampsia, especially severe preeclampsia. Misclassification was most common in obese preeclamptic women. These findings depict the limitations associated with the direct use of discharge diagnoses of hypertensive disorders in pregnancy for research purposes.