Assessing the impact of cardiovascular events on health-related quality of life outcomes in DAPA-HF

Background Heart failure (HF) patients are at increased risk of cardiovascular (CV) events, including hospitalisation for HF (hHF), myocardial infarction (MI) and stroke, imposing a significant burden on health related quality of life (HRQoL). DAPA-HF was a multinational clinical trial (NCT03036124) investigating the efficacy and safety of dapagliflozin for the treatment of HF with reduced ejection fraction. Patient reported outcomes were collected. The objective of this study was to estimate the impact of CV events on patient HRQoL over time, as assessed through EQ-5D-5L and Kansas City Cardiomyopathy Questionnaire (KCCQ) total symptom score (TSS) and clinical symptom score (CSS). Methods Mixed effects regression models were developed based on pooled individual patient data from DAPA-HF to estimate the impact of hHF, MI and stroke on patient utility (EQ-5D-5L questionnaire responses weighted according to the societal value placed on given health states), and KCCQ TSS score. Utility was estimated using UK-specific tariffs after mapping EQ-5D-5L to EQ-5D-3L values in line with NICE guidance. A subject-specific intercept was incorporated, and estimates were adjusted for the incidence of events occurring within one month prior, two to four months prior, and 4 to 12 months prior to questionnaire completion. Results Mean patient baseline utility was 0.716 (95% CI: 0.711, 0.722), with KCCQ TSS 73.6 (73.0, 74.2). The incidence of CV events was consistently associated with reduced patient HRQoL, assessed through either EQ-5D or KCCQ TSS. In the first month following the event, hHF was associated with a 0.083 (0.06, 0.107) reduction in patient utility, and 16.9 (14.5, 19.4) reduction in KCCQ TSS (Fig. 1). Comparing measures, the disease specific measure KCCQ appeared more sensitive than EQ-5D to changes in HRQoL following hHF events and less sensitive to changes following MI and stroke events. Comparing events using the generic EQ-5D measure, at two months post-event, patients with MI and stroke returned to baseline utility; patients with hHF remained below baseline utility at each assessment point for 12 months (Fig. 2); where patients had a mean reduction of 0.02 (0.005, 0.035) utility and 0.5 (−1.1, 2.1) KCCQ-TSS compared to those without an hHF event. Conclusion The incidence of cardiovascular events imposes a considerable burden on HRQoL in patients with HFrEF. HF specific events may be better characterised with a disease specific tool, whereas for wider CV events a generic tool may be preferable. The impact of hHF on HRQoL was noteworthy in its persistence across the measures used up to one year. Interventions that reduce the risk of these events have the potential to significantly improve patient quality of life. FUNDunding Acknowledgement Type of funding sources: Private company. Main funding source(s): AstraZeneca

Evaluating the key predictors of health-related quality of life in patients with heart failure and reduced ejection fraction: results from the DAPA-HF trial

Background The DAPA-HF trial demonstrated that dapagliflozin was superior to placebo at preventing cardiovascular death and hospitalisation for heart failure (hHF) events in patients with chronic heart failure with reduced ejection fraction (HFrEF). The trial also demonstrated a clinically important benefit of dapagliflozin on health-related quality of life (HRQoL). However, key predictors of HRQoL in HFrEF patients remain uncertain. The objective of this study was to determine, using DAPA-HF trial data, the patient characteristics and disease-related events associated with patient HRQoL, measured by health state utility values. Methods Mixed effects regression models were developed based on pooled individual patient data from DAPA-HF to determine patient utility estimated from responses to the EQ-5D-5L questionnaire, incorporating a subject specific random intercept. In line with NICE guidance, utility estimates were derived using UK-specific utility tariffs after mapping EQ-5D-5L data to EQ-5D-3L values. Univariable analysis was first undertaken to assess candidate predictors of utility; followed by a multivariable model including statistically significant predictors, e.g. Kansas City Cardiomyopathy Questionnaire Total Symptom Score (KCCQ-TSS) and the incidence hHF events, and controlling for differences in baseline characteristics. All variables were included in a single model to provide independent (adjusted) estimates for each covariable. Results 19,983 EQ-5D-5L questionnaires from 4,744 patients were included. Mean patient utility at baseline was 0.716 (95% CI: 0.711, 0.722). Univariable analysis demonstrated NYHA, KCCQ-TSS, T2DM, BMI, age, geographic location, non-ischaemic/unknown aetiology and atrial fibrillation were statistically significant in their association with patient utility while prior hHF, race, eGFR and left ventricular ejection fraction were not. Multivariable analysis results are summarised in Fig. 1. The baseline characteristic with the greatest impact on EQ-5D was KCCQ-TSS quartile, with EQ-5D increasing with KCCQ-TSS and the difference in utility between patients in quartile 1 (lowest score) and quartile 4 (highest score) estimated at 0.233 (0.226, 0.240). When controlled for baseline characteristics, being post-event was significantly associated with HRQoL; patients who experienced hospitalisation for HF had 0.036 (0.014, 0.058) lower utility on average within one month of the event and 0.025 (0.011, 0.039) lower utility up to one-year after the event. For patients who had stroke or myocardial infarction events there were reductions in utility of 0.206 (0.141, 0.272) and 0.108 (0.039, 0.177) respectively at 1 month. Conclusion HF symptoms, measured by the KCCQ, were strongly associated with patient health utility. Therapeutic interventions that can improve HF symptoms have the potential to improve HRQoL and reduce the burden of HF. FUNDunding Acknowledgement Type of funding sources: Private company. Main funding source(s): AstraZeneca