The association between continuity of care and the risks of cardiovascular diseases and mortality in primary care patients with diabetes mellitus

Abstract

Continuity of care—providing a consistent source of care, traditionally with a designated physician—has been suggested to be beneficial to diabetes management. However, high patient volume and physician mobility make its implementation in public clinics in Hong Kong challenging. Continuity of care at the clinic or team level, which was evaluated by the patient’s attendance to the same clinic or to a small group of physicians continuously, might offer more suitable alternatives. This thesis examined the associations between cardiovascular diseases (CVD) or all-cause mortality risks of type 2 diabetes mellitus (T2DM) patients in general out-patient clinics (GOPC) in Hong Kong with clinic-based and team-based continuity of care. This retrospective cohort study used data from the Clinical Management System of the Hospital Authority of Hong Kong. The study included T2DM patients aged 18 or older who had no CVD on or before baseline, which was the date of the earliest GOPC attendance between 1 January 2008 and 31 December 2018. Both team-based and clinic-based continuity of care were evaluated with the usual provider continuity index (UPCI), calculated from the GOPC attendances within two years before baseline. In clinic-based continuity of care, patients were categorised into four groups by UPCI—<0.50, 0.50-0.75, 0.76-0.99 and 1.0— while in team-based continuity of care, patients were divided into quartiles with cut-offs of <0.50, 0.50-0.74, 0.75-0.91 and 0.92-1.0. The group with the lowest UPCI was the reference for relative risk assessments. Propensity score fine stratification weights were applied to improve balance among groups. CVD and all-cause mortality hazard ratios were assessed with multivariable Cox regression adjusted with baseline characteristics. For clinic-based continuity of care, 311,839 patients were included after weighting. The median follow-up periods for CVD of the four patient groups were 66.5, 72.5, 80.5 and 78.5 months, with incidence rates of 29.9, 27.3, 27.0 and 24.4 cases/1000 person-year respectively. Patients with a clinic-based UPCI of 0.50-0.75, 0.75-0.99 and 1.0 had a 10%, 17% and 24% lower CVD risk and 16%, 24% and 25% lower all-cause mortality risk relative to those with a UPCI <0.50. Being younger and having fewer comorbidities were correlated with a larger reduction in CVD risk in those with a higher UPCI. For team-based continuity of care, 312,068 patients were included after weighting. The median follow-up periods for CVD in the 1st-4th quartiles were 77.5, 77.5, 77.5 and 78.5 months, while the incidence rates were 26.1, 25.4, 24.6 and 23.6 cases/1000 person-year respectively. Patients in the 2nd, 3rd and 4th quartiles had a 5%, 8% and 13% lower CVD risk and 5%, 7% and 6% lower all-cause mortality risk, respectively, relative to those in the 1st quartile. Reduction in CVD risk was more prominent in those who were male, younger and with fewer comorbidities. This study demonstrated that CVD and all-cause mortality risks were lower in T2DM patients with higher clinic-based and team-based continuity of care. These approaches might serve as alternatives to individual-based continuity of care in public clinics to enhance the quality of care for T2DM patients.