Rationale A recent study into the patient perspective of patients with multiple chronic conditions in the Netherlands underlines the strain multimorbidity can put on people. Most patients would appreciate more coordination from and communication with their care providers. This call for better coordination of needs and preferences ties into the concept of Advance Care Planning (ACP). ACP is a structured process of communication in which patients and physicians discuss and, if applicable, document health preferences and goals of patients regarding their last phase in life. Most ACP studies have been performed amongst older, terminally ill patients with the main aim of establishing patients' preferences before they lose capacity. We want to investigate the potential of ACP to increase patient empowerment in a population of competent patients with multimorbidity, who are not necessarily in their last phase of life.

The distribution of healthcare expenditure among the population requiring care is skewed. In the Netherlands the top-10% most cost incurring patients account for 68% of expenditure. Many of these patients receive unnecessary or ineffective care, with a recent study estimating preventable spending at 10%. High-Need High-Cost patients comprise a very heterogeneous group, yet one common denominator explaining high cost is the high prevalence of multiple chronic conditions. Both overtreatment and conflicting treatment are legitimate concerns within this population. As multimorbidity and frailty increase with age, the older patient with multimorbidity is especially at risk. Targeted care programmes have been developed under the assumption that better coordination will lead to a reduction in healthcare utilization. However, although care might be identified as preventable or inefficient from a medical point of view, this is not necessarily the case from a patient perspective. We are interested how patients experience such care and thereby if better coordination would indeed lead to a reduction in utilization.

Because ACP supports patients in timely recognition and better expression of their needs and preferences, we hypothesize that care will address those needs and preferences more adequately, which will result in improved patient assessment of care. We further hypothesize that patient empowerment will enable better planning of care and decision making, which can result in less unwanted or preventable interventions. As a consequence healthcare utilization might decrease. However, another possibility is that rather than leading to a decrease, improved empowerment may lead to an increase in utilization because care which is deemed superfluous from a medical perspective might not be perceived as such by patients.

Objective The primary objective of our pilot study is to assess the feasibility of a formal Randomized Controlled Trial. Our secondary pilot objectives are to collect data on patient experience of healthcare, patient engagement, cost-effectiveness, and other data that might inform the design of a full-scale RCT.

Study design Randomized pilot study

Study population Patients over 65 years of age with polypharmacy, multimorbidity and multiple hospitalizations and/or ER admissions in the past year

Intervention One of the most well-researched ACP programmes is the Respecting Choices Programme. In this programme, a trained facilitator encourages patients to reflect on their goals, values and beliefs, to discuss and document their future choices, and to appoint a surrogate decision maker. The programme was translated to the Dutch context in previous studies in the nursing home setting and oncology care. Patients randomized to receive ACP will have two meetings with a trained facilitator within two months.

Main study parameters/endpoints Primary: trial-feasibility is defined as the successful inclusion of 50 patients in total, timely administration of the intervention in 25 patients, adherence to follow-up procedures and identification of problems or barriers during recruitment, inclusion, intervention administration and follow-up.

Secondary: main outcome for cost-effectiveness is total duration and number of hospital admissions, as a proxy for both costs and effects (iMCQ). In order to inform a future cost-effectiveness analysis (CEA), data on health-related quality of life (EQ5D-5L) will also be collected. Our outcomes for patient assessment of care and patient empowerment are the PACIC questionnaire, the ACP Engagement Survey and the appointment of a surrogate decision maker and/or the documentation of advance directives.

A recent study into the patient perspective of patients with multiple chronic conditions in the Netherlands underlines the strain multimorbidity can put on people. Conducting semi-structured, in-depth interviews it was established that besides the burden of disease itself, patients struggled with managing the logistics such as regular hospital visits. Another conclusion was that most patients would appreciate more coordination from and communication with their care providers. This call for better coordination of needs and preferences ties into the concept of Advance Care Planning (ACP). ACP is a structured process of communication in which patients and physicians discuss and, if applicable, document health preferences and goals of patients regarding their last phase in life. Most ACP studies have been performed amongst older, terminally ill patients with the main aim of establishing patients' preferences before they lose capacity. We are interested in potential of ACP to increase patient empowerment in a population of high-need competent patients with multimorbidity, who are not necessarily in their last phase of life.

When taking a closer look at healthcare expenditure, the distribution among the population requiring care is skewed. In the Netherlands, the top-1% most cost incurring patients are responsible for 24% of healthcare expenditure and the top-10% accounts for 68%. These numbers are in line with international data. Public healthcare expenditure has been rapidly increasing over the last decades, compromising the economic sustainability of healthcare systems around the world. In the Netherlands, healthcare expenditure was 40.3 billion euro (10,35% GDP) in 1998 and has risen to 96.1 billion euro (13,79% GDP) in 2016. The patients that account for this cost-concentration have become an increasing target for care management. Studies have shown these High-Need High-Cost patients receive unnecessary or ineffective care with a recent study estimating preventable spending at 10%. Reducing such care would not only reduce cost but has the potential to improve quality of care, patient assessment of care and their quality of life.

The High-Need High-Cost patient population is very heterogeneous, yet one common denominator explaining high cost is the high prevalence of multiple chronic conditions. Both overtreatment and conflicting treatment are legitimate concerns within this population. As multimorbidity and frailty increase with age, the older patient with multimorbidity is especially at risk. Targeted care programmes have been developed under the assumption that better coordination will lead to a reduction in healthcare utilization. But although care might be identified as preventable or inefficient from a medical point of view, this is not necessarily the case from a patient perspective. We are interested how patients experience such care and thereby if better coordination would indeed lead to a reduction in utilization.

In the United States, the Patient Self Determination Act (1991) embodied the use of advance directives to promote adequate communication. An advance directive (increasingly being replaced by the term advance decision) must relate to a refusal of specific medical treatment and can specify circumstances. It will come into effect when the individual has lost capacity to give or refuse consent to treatment. Careful assessment of the validity and applicability of an advance decision is essential before it is used in clinical practice. Valid advance decisions, which are refusals of treatment, are legally binding. However, studies demonstrated important limitations of advance directives such as the limited use of advance directives despite many promotional efforts and the absence of positive effects on the quality of care, quality of life (QOL) or satisfaction of patients and relatives. From the mid-nineties, advance directives evolved to ACP, a more comprehensive approach where filling in an advance directive or decision is only one part of a process of wider communication.

ACP is a formalized process of communication between patients, relatives and professional caregivers. It has been defined as "a voluntary process of discussion and review enabling individuals to express, and, if they wish, record views, values and specific treatment choices to inform their future care". ACP promotes the documentation of patients' preferences in their medical file, the communication of these preferences to family and friends, the periodic review of preferences as circumstances change and the nomination of a health care proxy. Modern ACP programmes aim at informing and empowering patients to express their preferences about their current and future treatment.

Studies addressing the effects of ACP programmes on end of life care are still rather scarce which is probably due to their challenging nature. The few studies that have been performed were mainly small scale with an observational design, often North American and predominantly performed in a geriatric context. A review of the literature shows that ACP programmes have the potential to improve communication between patients and healthcare professionals, increase the quality of life and well-being of patients and their relatives, reduce futile treatments and unnecessary hospitalizations, enhance provision of care consistent with patient goals and increase satisfaction with care. Other studies also show ACP has reduced costs. There are no studies to date that look at ACP in the wider context of multimorbidity alone.

The process of ACP can be formalized through a complex ACP intervention such as the Let Me Decide advance directive programme, the Respecting Choices Programme and the Making Advance Care Planning a Priority (MAPP) Programme. In considering these programmes we looked at (international) experience with the programme and their evidence-based effects. Let Me Decide had potential as this decreased emergency calls to the ambulance service and both hospitalisation and length of stay. However there was no difference in effect on patient satisfaction. MAPP showed favourable effects on in-hospital death and hospitalisation but no effect on hospital length of stay, hospice referral and days of hospice or palliative care. The Respecting Choices Programme has recently been studied in a large international trial and has been shown to increase compliance with end-of-life wishes, satisfaction with quality of death and patients' satisfaction while decreasing depression, anxiety and death in the ICU. A recent systematic review on interventions guiding ACP found that the main body of evidence regarding interventions that were evaluated in a controlled trial concerned Respecting Choices with six trials described in seven papers. This large body of evidence with demonstrated effects on both healthcare use and patient satisfaction combined with local experience with the programme led us to select the Respecting Choices Programme for this pilot study.

The programme was developed in the US and successfully trialled in a geriatric setting in Australia, showing that end of life care wishes were much more likely to be known and followed in the intervention group (86%) compared to the control group (30%). In this programme, a trained facilitator in collaboration with attending physicians encourages patients to reflect on their goals, values and beliefs, to discuss and document their future choices, and to appoint a surrogate decision maker.

Most ACP studies have been performed amongst older patients with the main aim of establishing patients' preferences before they lose capacity. We believe that ACP can also increase patient empowerment in a population of competent patients with multimorbidity, who are not necessarily in their last phase of life. Because ACP supports patients in timely recognition and better expression of their needs and preferences it will enable strategic and effective planning of care and decision making. Our hypothesis is that as a result care will more adequately address patients' needs and preferences resulting both in improved patient assessment of care and receiving less unwanted or preventable interventions. We hypothesize the latter is due to improved patient empowerment. However, another possible outcome is that improved empowerment will not lead to a decrease of healthcare utilization (and might even lead to an increase), as care which is deemed superfluous from a medical perspective might not be perceived as such by patients. The choice for the Respecting Choices Programme was made because it is one of the most evidence-based interventions available.

Because there is still some uncertainty about the feasibility of a formal RCT we have opted to conduct a randomized pilot study. In this pilot we will conduct the intended formal RCT on a smaller scale to draw lessons on the willingness to participate, response rates to questionnaires and follow-up rates as well as potential differences in these matters between the intervention and the control arm.

The objective of this pilot is to assess whether it is feasible to conduct a full-scale Randomized Controlled Trial studying the cost-effectiveness of ACP within the older population with multimorbidity. The primary research question of such a trial would be whether formalised ACP is cost-effective in achieving patient empowerment and improved patient assessment of care.

Inclusion Criteria:

• Polypharmacy (i.e. use of ≥ 5 medications for longer than six months)
• ≥ 2 hospitalizations, day clinic admissions and/or individual ER admissions in the last 12 months
• Charlson Comorbidity Index (CCI) of ≥ 5 (unadjusted for age)
• Written informed consent to participate

Exclusion Criteria:

• Unable to complete questionnaire or interview in Dutch
• Less than 6 months anticipated life expectancy
• Serious cognitive impairment (MMSE score < 16)