Health Care Use of Frail Older People: Follow‐Up Outcome Evaluation of an Intensive Primary Care Programme at 24 and 36 Months

As populations age worldwide, there is growing interest in supporting older people to safely remain at home for as long as possible. Primary care can play a crucial role in systematically identifying and managing older patients with complex needs 1. However, preventive intervention trials to improve the practice and organization of primary care have had mixed success in reducing aged residential care (ARC) placement 2-8. The Kare programme is an intensive primary care intervention at the Waitematā District Health Board in New Zealand. We previously reported that at 12 months the Kare programme delayed ARC placement and mortality, decreased emergency department (ED) attendance, and increased use of support services 9. The aim of the current study was to compare the outcomes of Kare patients with usual care patients at 24 and 36 months. The methods of Kare have been previously described in detail 9. Briefly, Kare was a nurse-led pragmatic intervention trial that aimed to delay ARC placement and reduce unplanned hospitalizations for frail older people (see Supporting Information Methods 1). Practice nurses did comprehensive assessments, care planning, and proactive follow-up visits of Kare patients for 12 months and annual reviews thereafter. Practice nurses participated in gerontology workshops, trained in Kare processes, and were supported by gerontology nurse specialists to implement this model of care. Funds were provided to facilitate care planning meetings and input from the usual general practitioner (GP). We examined the health services outcomes of 1177 older primary care patients aged 75+ (65+ for Māori or Pacific Peoples) enrolled in nine participating general practices and 3681 propensity-score matched patients enrolled in 67 practices providing usual care (n = 4858). Usual care received by controls may have included referrals to physicians in various specialties, gerontology nurse specialists, and other services deemed necessary by their GP. We used linked administrative data to determine older patients' health services outcomes. The Northern A Health and Disability Ethics Committee approved an ethics application waiver for the Kare programme and its evaluation. We fitted logistic mixed models to model the probability of residential care placement, mortality, and Needs Assessment and Service Coordination (NASC) assessment for home support at 24 and 36 months. Negative binomial mixed models were fitted to model rates of hospital admissions, ED attendance, outpatient consults with specialist physicians, and use of community nursing and allied health services at 24 and 36 months. All models specified were adjusted for the same patient and practice characteristics as the original 12-month evaluation. Table 1 summarizes differences in 24-month and 36-month outcomes of Kare patients and their matched controls after fully adjusting for patient and practice characteristics. The odds of ARC placement were not different in Kare patients compared to matched controls at either follow-up period (OR 0.98, CI 0.74–1.30 at 24 months; OR 0.96, CI 0.76–1.22 at 36 months). There was no association between Kare and rates of acute admissions at either follow-up period (IRR 1.02, CI 0.92–1.13 at 24 months; IRR 1.02, CI 0.92–1.14 at 36 months). However, we found over 20% higher rates of elective admissions in Kare patients at both follow-up periods (IRR 1.23, CI 1.08–1.40 at 24 months; IRR 1.26, CI 1.12–1.42 at 36 months). Rates of total visits to the ED were 44% lower at 24 months (IRR 0.56, CI 0.45–0.72) and 43% lower at 36 months (IRR 0.57, CI 0.46–0.71) in Kare patients compared to controls. Higher rates of outpatient specialist services, gerontology nurse services, and allied health therapist sessions, and higher odds of needs assessment were observed in Kare patients at both follow-up periods. The odds of death at 24 or 36 months were similar in Kare patients and controls (OR 0.81, CI 0.65–1.02 at 24 months; OR 1.02, CI 0.81–1.28 at 36 months). We tested the effectiveness of Kare in improving older patients' health services outcomes at 24 and 36 months. Although ARC placement and acute hospital admissions were similar between Kare patients and matched controls, differences in other health care use suggest that Kare may be tipping the balance of health care use towards planned care. Kare funded time for nurses to facilitate comprehensive assessment, care planning, and proactive follow-up visits throughout the study. These intervention components may have contributed to Kare patients' increased use of non-acute, proactive, and supportive care (including outpatient specialist services and elective admissions for cardiovascular disease, cancer, etc.) and lower ED attendance rates, which were still observed at 36 months. The long-term impacts observed in this follow-up evaluation despite nominal intervention costs (practices were paid USD415 per patient per annum for additional hours dedicated to Kare and USD3600 per practice per annum for workforce development) warrant further study in larger and more ethnically diverse practice populations. Intervention concept and design by D.N., M.D., M.L.B.; design of follow-up evaluation by L.P., N.K., A.C.; data acquisition by J.W., R.A.F., D.L.R.; and data analysis by L.P., N.K., A.C., J.W. All authors contributed to the interpretation of findings, drafting/revising the manuscript, and provided approval of the final version. This work was supported by Waitematā District Health Board. The authors declare no conflicts of interest. Data S1: Supporting Information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.