Multimorbidity and complex multimorbidity, their prevalence, and associated factors on a remote island in Japan: a cross-sectional study

BACKGROUND: Evidence is still limited on the prevalence of multimorbidity and complex multimorbidity in Japan, as well as their associated factors. Additionally, no studies regarding multimorbidity and complex multimorbidity in Japan have used patient data obtained directly from medical records. The primary objective was to clarify the prevalence of multimorbidity and complex multimorbidity using patient data obtained directly from medical records. The secondary objective was to elucidate factors associated with multimorbidity and complex multimorbidity. METHODS: We conducted a cross-sectional study at Tarama Clinic on Tarama Island, a remote island in Okinawa, Japan. Among patients who visited Tarama Clinic from April 1 to June 30, 2018, those who were ≥ 20 years of age, lived on Tarama Island, and had capacity to give consent were eligible for study inclusion. We collected the following data using medical records, medical expense receipts, and self-administered questionnaires: multimorbidity and complex multimorbidity, participants’ characteristics, and potentially associated factors. Multimorbidity was defined as having ≥ 2 among 17 chronic health conditions, in accordance with previous studies conducted in Japan, and having ≥ 2 among 16 chapters of the International Classification of Primary Care, 2nd edition (ICPC-2). Complex multimorbidity was defined as having ≥ 3 among 16 chapters of the ICPC-2. RESULTS: We included 355 study participants. Descriptive analysis showed multimorbidity prevalence measured using the 17 chronic health conditions in this area was 30.0% (age ≥ 20 years) and 57.5% (age ≥ 65 years). Multimorbidity prevalence measured using the ICPC-2 was 32.9% (age ≥ 20 years) and 60.3% (age ≥ 65 years). We also found a 20.9% (age ≥ 20 years) and 45.1% (age ≥ 65 years) prevalence of complex multimorbidity, respectively. Poisson regression with robust error variance showed that age ≥ 60 years was associated with multimorbidity. Age ≥ 60 years (adjusted prevalence ratio [aPR] 1.887 and 1.831 for ages 60–79 and ≥ 80 years, respectively) and being unemployed (aPR 1.251) were associated with complex multimorbidity. However, having hazardous drinking or more was inversely associated with complex multimorbidity (aPR 0.745). CONCLUSIONS: The population-based prevalence of multimorbidity and its upward trend with increasing age on a remote island in Japan was consistent with previous reports in the country. Multimorbidity was associated with age 60 years or older, and complex multimorbidity was associated with age 60 years or older, being unemployed, and not having hazardous drinking or more. Our study findings indicated a possible association between the coexistence of diseases and social determinants of health (SDH) in Japan. To improve care for patients with multimorbidity and complex multimorbidity, more research that takes SDH into account is warranted, and evidence-based policymaking is essential for Japan. The present study can provide a foundation for accumulating such evidence.