Frailty trajectories and their associations with short and long-term clinical outcomes in patients with heart failure: an analysis of over 400,000 individuals using a primary care database in the UK

Abstract Introduction Frailty is an important comorbidity in patients with heart failure (HF). It is characterised by a gradual deterioration in body systems and loss of inbuilt reserves. There is strong evidence that frailty at baseline relates to poor prognosis. However, frailty is a dynamic health state that changes over time. Little work has been done to study frailty trajectories in HF patients and how this impacts on clinical outcomes. Methods Electronic health records of patients registered with primary care practices in the UK included in the Clinical Practice Research Datalink (CPRD): GOLD and AURUM databases with linkage to national hospital admissions and death data were utilised. We identified adult patients with a new diagnosis of HF. We used the secondary care administrative records frailty (SCARF) index to assess frailty status. Patients were classified into 4 frailty groups: SCARF index 0-0.05 (non-frail), 0.06-0.11 (mildly frail), 0.12-0.18 (moderately frail), ≥0.19 (severely frail). Frailty trajectories (baseline vs 6 months) and their associations with short (1-year) and long-term (5-year) all-cause mortality and hospitalisation was evaluated using cox proportional hazard models. Results This study evaluated a total of 412,173 patients with HF median (IQR) age: 78 (69-85) years, 47% female, 92% White. About 70% of individuals were frail at baseline: 92,353 (22%); 81,668 (20%) and 110,270 (27%) were mildly, moderately and severely frail respectively. Predictors of worse frailty status included increasing age, female sex, lower socioeconomic status (SES), higher burden of comorbidities including anaemia and renal dysfunction (all p0.001). In cox regression models adjusted for age, sex, ethnicity, year of HF diagnosis, place of diagnosis (hospital or community) and medications, worsening frailty status at baseline was related to increasing risk of all-cause hospitalisations and death at both 1 year and 5 years. Majority of patients (N=332,896, 81%) had valid follow up data at 6 months. About 80% had no change in frailty status (N=260,732, 78%); 10,046 (3%) and 2291 (0.1%) patients had frailty status improved by 1 and ≥2 levels respectively; 42,650 (13%), 12,886 (4%) and 4,291 (1.3%) patients had frailty status worsened by 1,2 or 3 levels respectively. (Figure 1) In cox regression models adjusted for age, sex, ethnicity, year and place of HF diagnosis, medications and baseline frailty status, improvement in frailty status at 6 months was associated with lower, while deterioration in frailty status was associated with higher risk of 1-year and 5-year all-cause hospitalisation and mortality. (Table 1) Conclusion Majority (70%) of patients with a new diagnosis of HF in primary care were frail. One in 5 patients experienced a deterioration in frailty status at 6 months. The degree of deterioration in frailty status was related to a progressive increased risk of death and hospitalisation at 1 and 5 years.