OBJECTIVES: To describe the impact of hypertension (HTN) on hospitalisations due to cardiovascular diseases (CVDs) in individuals with different blood pressure (BP) levels using data from routine health check-ups reported in a Japanese claims database.
METHODS: Using data from the Japan Medical Data Center (JMDC) database, a retrospective longitudinal cohort study was conducted on adults aged 40-64 years with both systolic and diastolic pressure readings between January 2008 and January 2015 (first check-up defined as index date). Individuals were categorised as naïve to HTN therapy (naïve) or having prior HTN therapy (experienced), depending on the prescription of HTN drugs within a 6-month look-back period. Time from index date to hospitalisation due to CVDs was compared among individuals with different BP levels at index date (optimal, normal, high-normal, Grade1, Grade2 and Grade3) using Cox proportional hazards model. Models with time-dependent variables reflecting the BP variation over follow-up were also used.
RESULTS: Of the 740,784 naïve individuals included, 61.8% were male, the mean age was 47.9 years and 46.9% were classified into the optimal group. The probability of hospitalisation increased with HTN severity at index date, with HRs of 1.95, 2.85 and 6.44 for Grade1, Grade2 and Grade3 respectively (p<0.0001, reference: optimal). Naïve individuals with hypertension at last check-up had a significantly higher probability of hospitalisation compared to individuals without hypertension (HR: 1.75, p<0.0001). Of the 72,828 experienced individuals included, 74.2% were male, the mean age was 53.6 years and 14.6% were classified into the optimal group. There was a non-statistically significant trend towards increased probability of hospitalisation in patients with more severe HTN at index date. CONCLUSION: This study adds quantitative evidence about the impact of hypertension on the risk of hospitalisations for CVD in Japan. Routine health check-ups are useful to identify patients at risk of CVD hospitalisation.