Heart failure (HF) is a major public health problem that is associated with high hospital readmission rates, mortality, and health care expenditures [
1]. Despite treatment advances, mortality remains high with approximately 50% of HF patients dying within 5 years of their diagnosis [
1]. Overall, 1 in 8 deaths list HF as a contributing cause [
2]. While all-cause mortality and the mortality attributable to cardiovascular disease (CVD) have declined over the past decades, declines in rates of CVD mortality began slowing in 2011 [
3‐
9]. Mortality attributable to HF, in contrast, began increasing around this time [
4,
10]. Explanations for the increase in HF mortality may include the aging population, the increasing prevalence of comorbidities, and the growing prevalence of HF with preserved ejection fraction, which lacks specific evidence-based treatments [
11‐
14]. Examining temporal trends in HF mortality is important as this may reflect changes in individual, or institutional factors, and can inform healthcare policy and practice [
15]. Integrated healthcare delivery systems are associated with better adherence to evidence-based care guidelines, survival rates, and reduced racial disparities [
16‐
18]. Kaiser Permanente Southern California (KPSC) serves approximately 4.6 million enrollees and largely reflects the diverse population of southern California, making it ideal to study mortality trends [
19]. Therefore the purpose of the current study was to examine secular trends in HF mortality among adults ≥ 45 years old in KPSC, California, and the US, from 2001 to 2017.
Discussion
In the current study, HF mortality rates in KPSC were similar to California but lower than the US, with rates increasing from 2001 to 2017. This was observed in KPSC men but not among women. Trends among KPSC members ≥ 65 years old increased as in the overall population, while no change in HF mortality rates were observed in KPSC members 45–64 years old. Mortality with HF as a contributing cause decreased significantly among KPSC men, but not women. For HF mortality as both an underlying and contributing cause, smaller changes in KPSC compared to California and the US were observed between older (2001–2011) and contemporary (2011–2017) time periods.
HF mortality rates were lower in KPSC compared to the US, and the magnitude of differences in mortality rates between KPSC and the US decreased between 2001 and 2017. Overall, US rates in the current study fluctuated between 48 and 58 per 100,000 PY for all adults ≥ 45 years old, consistent with previous studies. An analysis of data from the National Center for Health Statistics (NCHS) reported 2015 age-adjusted mortality rates with HF as the underlying cause of death ranging from 1.7 per 100,000 PY among US adults 40–49 years old to 441.1 per 100,000 PY among those ≥ 80 years old [
27]. While the age compositions of populations in each study may account for the variability of rates, the current study results are similar in magnitude to observed US rates. Reasons for lower HF mortality rates in KPSC, particularly in 2001, may require additional investigation. However, this may be partially attributable to better treatment and management of other preventable causes of death; a previous KPSC study noted that HD mortality decreased between 2001 and 2016 [
15]. Lower rates of HF mortality compared to the US may also be attributable to the KPSC integrated healthcare delivery model. KPSC has previously noted improvements in HF risk factors, including blood pressure and lipids [
28,
29]. Additionally, while KPSC is representative of the Southern California population, some differences in mortality rates may be explained by the sociodemographic differences between KPSC, California, and the US.
HD mortality, and more specifically, mortality attributable to HF began slowing around 2011, and even increasing more recently [
4,
6,
10]. Sidney and colleagues examined US mortality trends attributable to all HD from 2000 to 2015, noting decreasing rates of HF mortality between 2000 and 2011 but annual increases between 2011 and 2015 by 3.73% (95% CI 3.21%, 4.26%) [
10] An additional study by Sidney reported this trend continued through 2017, and represented a 38% increase in deaths with HF listed as an underlying cause between 2011 and 2017 [
14]. Similar trends were reported using NCHS data, where age-adjusted HF mortality decreased between 2000 and 2012 but increased between 2012 and 2014 [
4]. In KPSC, we noted that mortality with HF as an underlying cause increased consistently between 2001 and 2017, which contrasts with the larger AAPCs in California and the US between 2011 and 2017. We did not examine factors contributing to these increases. However, increased awareness of HF as a diagnosis may contribute to the increased designation of HF as the underlying cause of death.
Sex-based differences in HF mortality were noted among adults ≥ 65 years old in KPSC, where women compared with men had a larger AAPC. This is consistent with data from California but differs from that of the US. One explanation could be the higher rates of HF mortality in the US compared to KPSC and California at the beginning of the study period; although it is important to note that an increasing AAPC was noted for HF mortality among older adults from 2011–2017. We also note some differences between KPSC, California, and the US with respect to race/ethnicity. However, we interpret these results cautiously due to small sample sizes within subgroups. Similar to this study, Sidney and colleagues noted differences in annual declines of HF mortality in 2000–2011 and 2011–2015 among sex and race/ethnicity groups [
10]. In contrast, an analysis of the National Inpatient Sample between 1991 and 2015 reported HF mortality rates varied by race/ethnicity but not sex [
27].
Previous studies using national data have reported declines in mortality with HF as a contributing cause. Glynn and colleagues reported a decline through 2011–2012, but an increase through 2012–2017, while a separate study reported a decline through 2014 [
4,
23]. Reasons why declines in HF mortality are observed through 2011 may include the emphasis of blood pressure and cholesterol management, and increased smoking cessation. Sustained increases in comorbidities such as obesity and diabetes, reduced mortality from other forms of CVD, and the growth of the population ≥ 65 years old may explain more recent increases in HF mortality [
10,
14,
15,
30,
31]. Emphasis on treating and managing HF and accompanying risk factors may help reduce these increasing rates of mortality.
We acknowledge some limitations. HF is considered a mediator between various disease states and mortality, and HF listed as an underlying cause of death may be attributable to other conditions [
1]. Thus, ICD-10 coding guidelines suggest listing other plausible heart conditions as the underlying cause of death and listing HF when the etiology cannot be determined. Regardless, we observed similar patterns in mortality with HF listed as a contributing cause of death. Additionally, deaths among KPSC enrollees occurring outside of California may not be completely captured. After 2011, a law was established that prevented the SSA from disclosing state death records received through its contracts. However, it was assumed that most deaths outside of the state of California were reported to KPSC by family, employers, caregivers, doctors outside of the state, and law enforcement. Additionally, cause of death was not available on all enrollees, which may underestimate HF-specific mortality. In the current study, we noted small numbers of HF mortality among stratified subgroups by younger age (45–64 years) and race/ethnicity, which limits the interpretability of our findings. Rates of mortality in the KPSC population may not be generalizable to uninsured populations. Additionally, race/ethnicity was not available for all members. Finally, factors associated with HF mortality were not examined, and this warrants further investigation. Prior work has shown that patient self-care (e.g., weight monitoring, medication adherence, healthy lifestyle behaviors, and comorbidity management) and other care-related factors (e.g., patient education and connection with care) are important in the prognosis of HF patients. [
32‐
34]
In conclusion, lower rates of HF mortality were observed in KPSC compared to the US. Given the aging of the US population and increasing prevalence of HF, it will be important to examine individual and care-related factors driving susceptibility to HF mortality.
Acknowledgements
Parts of this study were presented as an abstract at the American Heart Association Quality of Care and Outcomes Research virtual conference, May 15-16, 2020. The abstract was published online: Mefford, M., Zhuang, Z., Liang, Z., Chen, W., Watson, H., Koyama, S.Y., Lee, M., Sidney, S., & Reynolds, K. (2020). Abstract 270: Temporal Trends in Heart Failure Mortality in an Integrated Healthcare Delivery System, California and the US, 2001-2017. Circulation: Cardiovascular Quality and Outcomes, 13.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.