THE IMPACT OF PHYSICAL FUNCTION ON HEALTHCARE RESOURCE USE AND CLINICAL OUTCOMES IN PATIENTS WITH HEART FAILURE: A UK BIOBANK STUDY
Author(s)
Walter Abhayaratna, PhD1, Foroogh Shamsi, PhD2, Joe Hollinghurst, PhD3, Julie T. Mortensen, MSc, PhD4, Lærke Marie Sidenius Nelson, PhD2, Matthew Turner, PhD3, Nia Jenkins, PhD3, René Lindholm Cordtz, PhD2, Andrew Thompson, PhD5;
1Australian Government, Canberra, Australia, 2Novo Nordisk A/S, Copenhagen, Denmark, 3Health Economics and Outcomes Research Ltd, Cardiff, United Kingdom, 4Novo Nordisk, Søborg, Denmark, 5Lane Clark & Peacock LLP, London, United Kingdom
1Australian Government, Canberra, Australia, 2Novo Nordisk A/S, Copenhagen, Denmark, 3Health Economics and Outcomes Research Ltd, Cardiff, United Kingdom, 4Novo Nordisk, Søborg, Denmark, 5Lane Clark & Peacock LLP, London, United Kingdom
OBJECTIVES: Patients with heart failure (HF) often experience limitations in physical function. This study investigated the effect of physical function on clinical outcomes and healthcare resource utilisation (HCRU) in patients with HF.
METHODS: UK Biobank participants aged 40-69 with HF, linked primary and secondary care data and no recent cancer were eligible. Self-reported physical function was quantified using excess metabolic equivalent of task (MET) score, measured as excess MET-hours per week. World Health Organisation recommends ≥7.5 excess MET-hours per week, hence physical function categories were low (<3); moderate (≥ 3 & <7.5); recommended (≥7.5 & <15); and high (≥15). Associations between MET-score and hospitalisation for HF (HHF) with competing risk of all-cause mortality (ACM) were quantified using multivariate cause-specific Cox models. Associations between MET-score and HCRU (hospitalisation (for any reason) and length of stay (LOS)) in the five years following recruitment were quantified using multivariate negative-binomial hurdle models. Hazard (HR) and odds (OR) ratios [95% CI] were reported for each MET-score category vs high.
RESULTS: 1,661 participants with HF were included. Compared with high MET-scores, low MET-scores were associated with increased hazard of HHF (1.35 [1.08-1.68]) and ACM (1.20 [0.80, 1.81]); moderate MET-scores also had increased hazard of HHF (1.29 [1.04-1.58]) and ACM (1.35 [0.90-2.04]). Similarly, low and moderate MET-scores were associated with increased odds of hospitalisation (1.84 [1.26-2.68] and 1.47 [1.01-2.12] respectively) and LOS >0 days (1.85 [1.27-2.69] and 1.45 [1.01-2.09], respectively) compared with high MET-scores. Recommended MET-scores were associated with similar hazard and odds of all outcomes compared with high MET-scores.
CONCLUSIONS: Patients with HF with low and moderate physical function are at increased risk of HHF and ACM and accrue excess HCRU compared with those with high physical function levels. Management strategies to improve physical function in patients with HF could improve clinical outcomes and reduce healthcare burden in this population.
METHODS: UK Biobank participants aged 40-69 with HF, linked primary and secondary care data and no recent cancer were eligible. Self-reported physical function was quantified using excess metabolic equivalent of task (MET) score, measured as excess MET-hours per week. World Health Organisation recommends ≥7.5 excess MET-hours per week, hence physical function categories were low (<3); moderate (≥ 3 & <7.5); recommended (≥7.5 & <15); and high (≥15). Associations between MET-score and hospitalisation for HF (HHF) with competing risk of all-cause mortality (ACM) were quantified using multivariate cause-specific Cox models. Associations between MET-score and HCRU (hospitalisation (for any reason) and length of stay (LOS)) in the five years following recruitment were quantified using multivariate negative-binomial hurdle models. Hazard (HR) and odds (OR) ratios [95% CI] were reported for each MET-score category vs high.
RESULTS: 1,661 participants with HF were included. Compared with high MET-scores, low MET-scores were associated with increased hazard of HHF (1.35 [1.08-1.68]) and ACM (1.20 [0.80, 1.81]); moderate MET-scores also had increased hazard of HHF (1.29 [1.04-1.58]) and ACM (1.35 [0.90-2.04]). Similarly, low and moderate MET-scores were associated with increased odds of hospitalisation (1.84 [1.26-2.68] and 1.47 [1.01-2.12] respectively) and LOS >0 days (1.85 [1.27-2.69] and 1.45 [1.01-2.09], respectively) compared with high MET-scores. Recommended MET-scores were associated with similar hazard and odds of all outcomes compared with high MET-scores.
CONCLUSIONS: Patients with HF with low and moderate physical function are at increased risk of HHF and ACM and accrue excess HCRU compared with those with high physical function levels. Management strategies to improve physical function in patients with HF could improve clinical outcomes and reduce healthcare burden in this population.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
CO6
Topic
Clinical Outcomes
Topic Subcategory
Clinical Outcomes Assessment
Disease
SDC: Cardiovascular Disorders (including MI, Stroke, Circulatory)