Advance care planning for adults with heart failure.

BACKGROUND

People with heart failure report various symptoms and show a trajectory of periodic exacerbations and recoveries, where each exacerbation event may lead to death. Current clinical practice guidelines indicate the importance of discussing future care strategies with people with heart failure. Advance care planning (ACP) is the process of discussing an individual's future care plan according to their values and preferences, and involves the person with heart failure, their family members or surrogate decision-makers, and healthcare providers. Although it is shown that ACP may improve discussion about end-of-life care and documentation of an individual's preferences, the effects of ACP for people with heart failure are uncertain.

OBJECTIVES

To assess the effects of advance care planning (ACP) in people with heart failure compared to usual care strategies that do not have any components promoting ACP.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, CINAHL, Social Work Abstracts, and two clinical trials registers in October 2019. We checked the reference lists of included studies. There were no restrictions on language or publication status.

SELECTION CRITERIA

We included randomised controlled trials (RCTs) that compared ACP with usual care in people with heart failure. Trials could have parallel group, cluster-randomised, or cross-over designs. We included interventions that implemented ACP, such as discussing and considering values, wishes, life goals, and preferences for future medical care. The study participants comprised adults (18 years of age or older) with heart failure.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted outcome data from the included studies, and assessed their risk of bias. We contacted trial authors when we needed to obtain missing information.

MAIN RESULTS

We included nine RCTs (1242 participants and 426 surrogate decision-makers) in this review. The meta-analysis included seven studies (876 participants). Participants' mean ages ranged from 62 to 82 years, and 53% to 100% of the studies' participants were men. All included studies took place in the US or the UK. Only one study reported concordance between participants' preferences and end-of-life care, and it enrolled people with heart failure or renal disease. Owing to one study with small sample size, the effects of ACP on concordance between participants' preferences and end-of-life care were uncertain (risk ratio (RR) 1.19, 95% confidence interval (CI) 0.91 to 1.55; participants = 110; studies = 1; very low-quality evidence). It corresponded to an assumed risk of 625 per 1000 participants receiving usual care and a corresponding risk of 744 per 1000 (95% CI 569 to 969) for ACP. There was no evidence of a difference in quality of life between groups (standardised mean difference (SMD) 0.06, 95% CI -0.26 to 0.38; participants = 156; studies = 3; low-quality evidence). However, one study, which was not included in the meta-analysis, showed that the quality of life score improved by 14.86 points in the ACP group compared with 11.80 points in the usual care group. Completion of documentation by medical staff regarding discussions with participants about ACP processes may have increased (RR 1.68. 95% CI 1.23 to 2.29; participants = 92; studies = 2; low-quality evidence). This corresponded to an assumed risk of 489 per 1000 participants with usual care and a corresponding risk of 822 per 1000 (95% CI 602 to 1000) for ACP. One study, which was not included in the meta-analysis, also showed that ACP helped to improve documentation of the ACP process (hazard ratio (HR) 2.87, 95% CI 1.09 to 7.59; participants = 232). Three studies reported that implementation of ACP led to an improvement of participants' depression (SMD -0.58, 95% CI -0.82 to -0.34; participants = 278; studies = 3; low-quality evidence). We were uncertain about the effects of ACP on the quality of communication when compared to the usual care group (MD -0.40, 95% CI -1.61 to 0.81; participants = 9; studies = 1; very low-quality evidence). We also noted an increase in all-cause mortality in the ACP group (RR 1.32, 95% CI 1.04 to 1.67; participants = 795; studies = 5). The studies did not report participants' satisfaction with care/treatment and caregivers' satisfaction with care/treatment.

AUTHORS' CONCLUSIONS: ACP may help to increase documentation by medical staff regarding discussions with participants about ACP processes, and may improve an individual's depression. However, the quality of the evidence about these outcomes was low. The quality of the evidence for each outcome was low to very low due to the small number of studies and participants included in this review. Additionally, the follow-up periods and types of ACP intervention were varied. Therefore, further studies are needed to explore the effects of ACP that consider these differences carefully.