Care Consistency With Care Preferences in Nursing Homes: A Cluster-Randomized Study of the Effects of an Advance Care Planning Program (BEVOR)

Abstract

Background: In this study (NCT04333303), we investigated whether a complex advance care planning (ACP) intervention improves care consistency with care preferences in nursing home residents.

Methods: Forty-four German nursing homes were randomly assigned to an ACP intervention addressing the individual, institutional, and regional levels or to a control group (no intervention). The hospitalization rate over an observation period of 12 months (primary outcome) was analyzed as a surrogate for care consistency with care preferences at the nursing home level. Secondary outcomes comprised process-related and clinical parameters, including care consistency with care preferences (analysis level: residents/ nursing homes). Outcomes were evaluated by means of Poisson and logistic regression models with incidence rate ratios (IRR) and odds ratios (OR) as effect estimators in an intention-to-treat analysis.

Results: Of 44 nursing homes, 23 received the intervention. The hospitalization rate did not differ between the two groups (IRR 1.0; 95% CI: [0.97; 1.1]) but declined to a similar extent in both during the COVID-19 pandemic. The consistency of care with care preferences was similar in both groups as well (OR 0.9 [0.4; 1.9]). The predefined exploratory analysis suggests that care consistency with care preferences was more likely in the 6 out of 23 nursing homes that met predefined adherence criteria (OR 1.9 [0.7; 5.3]). Written emergency plans were significantly more common in the intervention group (IRR 11.6 [8.2; 16.4]), and even more so in adherent nursing homes (IRR 30.1 [15.7; 57.6]).

Conclusion: The intervention did not permeate sufficiently, especially due to the COVID-19 pandemic that may, in addition, have masked intrinsic shortcomings of the intervention. Thus, this trial does not allow a conclusive assessment of whether or not the intervention can promote care consistency with care preferences. However, exploratory analyses indicate that successful institutional implementation in conjunction with individual ACP conversations may increase care consistency with care preferences.

Figure 1

Consort flow diagram of study participants All nursing homes maintained group fidelity, i.e., the intervention group received and continued the intervention, the control group held back on implementation of advance care planning (ACP). In accordance with the study protocol, it is not possible to determine how many residents had ACP conversations. CG, control group; IG, intervention group; 12-mo. obs. period, 12-month observation period; n, number; NH, nursing homes; res, residents, who consented to individual data collection

Figure 2

Elements of the complex ACP program The intervention at individual level supports the residents in creating and documenting a relevant and valid advance care plan for future medical care. At institutional and regional levels, organization and system redesign and networking ensure that residents’ wishes are known and honored in the event of critical illness with decisional incapacity. First published in (24); for further details see additional files 1 (selected intervention details) and 2 (TIDeR checklist) in that publication. This figure is an adapted version of the figure titled “Overview of the elements of the regional ACP program,” originally published in BMC Trials (Götze et al.) (24); reproduced in accordance with CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).