Introduction

High quality, person-centered medical care incorporates patient preferences, values, and goals for shared ownership in the decision-making process [1,2,3,4].

Patient preferences for information, involvement, diagnostics, treatment, self-management, care and end-of-life capture ethical considerations and a legal framework respecting patient autonomy [3].

Health care professionals (HCP) working in geriatric wards are trained to assess patient preferences and provide coordinated healthcare through a multidisciplinary team approach [2] and could, therefore, be considered more prone to address and follow patients’ preferences. Historically, the political and legal shift from paternalism to patient’s autonomy was introduced through the informed consent in health law, also in Norway. However, most laws in Europe including Norway still give the medical doctor (MD) the final say, challenging patient’s autonomy in medical decisions. Only a handful of countries outside the US recognize patient’ wishes as legally binding by the use of advance directives. During the past decade, attempts have been conducted to empower patients’ and families’ autonomy in medical decisions by the use of evidence-based communication tools to increase patients involvement, such as advance care planning (ACP) and shared decision-making (SDM) [5,6,7,8].

Still, the recent literature indicates ACP has severe implementation problems with barriers such as inconsistency in effect on goal-concordant care and less integration in wards for patients admitted with acute severe illness [5, 9]. In this study, prior to a randomized trial implementing ACP in geriatric wards, we aimed to identify to what extend HCP examine the patient and next-of kin preferences for information, involvement, and treatment and to synthesize the experience of clinical practice in accordance with these preferences. Finally, we assessed possible causes for discrepancies and HCP’s attitudes towards clinical decision-making.

Methods

We invited HCP working ≥20% in 12 geriatric wards in hospitals in the Southeastern region of Norway to a survey to explore HCP`s self-reported practice and attitudes for assessing patient preferences, prior to implementation of ACP. The participating hospitals had acute geriatric wards or internal medical wards with a small number of beds allocated to geriatric patients and with a geriatrician among the staff.

We conducted an anonymous, descriptive, cross-sectional web-based survey, inspired by others [10, 11] (Table 1 Supplementary). The recruitment period was four weeks from 18th of October 2022. The questionnaire consisted of four sections related to three different categories of patient preferences; information, involvement, and treatment. First, we examined whether HCP clarify the patient preferences. Second, we investigated the experiences of clinical practice in accordance with these preferences. Third, we asked for possible reasons for not informing or involving patients or next- of-kin. We also asked for HCP confidence in knowing the patient preferences for involvement (VAS scale 0–10). In the fourth section, we explored HCP’s attitudes towards who makes and who should make clinical decisions. We recorded HCP’s age, sex, profession, years in clinical practice and experience as a leader.

Table 1 Demographic and background characteristics of the 289 participants from 12 geriatric units in South-Eastern Norway Regional Health Authority
Full size table

Each survey participant gave written consent. Norwegian Centre for Research Data, the Data Protection Official for Research, approved the study (805491). The study is covered by the Act on medical and health research according to the Regional Ethical Committee and has been approved by local data protection officers at each trial site. The responses were stored at the University of Oslo’s Services for Sensitive Data.

Statistics

We report continuous variables as mean and standard deviation (SD) and categorical variables as frequencies and percentage. We dichotomized “clarify for patient preferences for treatment” into always/often versus sometimes/never and “experienced practice” into following (response “in line with”) versus not following the preferences (response not known, less than wanted and more than wanted) and compared the groups using Chi-square test. We used SPSS version 29.

Results

All 12 eligible hospitals agreed to participate. After inviting 470 HCP, 289 responded (response rate 61%), with mean age 37.8 years (SD 11.3), 235 (81.3%) women, 12.4 (SD 9.6) years of experience and 67 (23.2%) MDs (Table 1).

Clarification of preferences

When examining HCP, 192 (66.4%) sometimes/never clarified the patient preferences for information, 160 (55.4%) for involvement and 125 (43.3%) for treatment. HCP reported that 178 (61.6%) and 142 (49.1%) sometimes/never clarify preferences for information and involvement of next-of-kin (Table 2 and Table 2 Supplementary). The results did not change when we stratified for years of practice.

Table 2 Health care professionals’ assessments of patient preferences and attitudes to decision making, n (%)
Full size table

Inclusion of patient preferences in clinical practice

Of 289, 191 (66.1%) of HCP do not follow patient preferences for information either because patient preferences are unknown, or they choose to inform more or less than wanted. Further, 154 (53.9%) do not follow the preferences for involvement and 151 (52.2%) do not follow the preferences for treatment, and similar trends are reported towards next-of-kin (Table 2).

Discrepancy with norms and clinical practice

Reasons stated for not informing or involving patients were (1) not enough resources or time [180 (62.3%) for informing and 167 (57.8%) for involving], (2) patients being too sick [99 (34.3%) and 116 (40.1%)], and (3) the need for better knowledge and communication skills [97 (33.6%) and 105 (36.3%)], respectively. Other reasons were short hospital stays [72 (24.8%) and 66 (22.8%)] and limited access to single rooms [67 (23.2%) and 50 (17.3%)]. The median score for confidence in knowing the patient preferences for involvement was 6.0 (5–8) (Fig. 1 Supplementary display the distribution).

Only 115/164 HCP (70.1%) follow the patient preferences for treatment. Interestingly, 23/125 (18.4%) report conducting treatment in accordance with the patient preferences without clarifying them (Table 3), with significant differences in responses regarding clarifying preferences and experience of clinical practice (P < 0.01). In subgroup analyses with MD we found the same trends.

Table 3 The relationship between clarifying the patient preferences for treatment and experienced practice of following the preferences
Full size table

Attitudes to decision-making

In clinical practice, 61 (21.1%) report that HCP make the decisions alone, while only 13 (4.5%) think they should. Interestingly, 153 (52.9%) think that HCP, patients and next-of-kin should make decisions together, while only 74 (25.6%) report this to be integrated into their clinical practice (Table 2).

Discussion

We found that even in geriatric wards more prone to a person-centered approach, only half of HCP report clarifying patient preferences for information, involvement, and treatment. Further, more than half do not inform, involve, and treat in line with these preferences, and only 70% report they follow the clarified preferences for treatment. Lack of resources, time, knowledge, communication skills and confidence and too sick patients are some reported reasons for not discussing preferences with the patient. Few HCP think that they alone should make clinical decisions, while more than 50% report that HCP, patients and next-of-kin should make decisions together.

Considering the attempts to strengthen patient’s autonomy, the results indicate a lack of impact and that action is needed to empower patients and families in the health care system. This is of relevance for Norway with an even more paternalistic legal framework than neighboring countries, also for most countries since very few have legally binding obligations to patients.

When failing to clarify patient preferences, we may provide healthcare that is not goal-concordant and not complying with their rights for participation in decision-making processes [12]. In older adults, frailty and comorbidity frequently lead to a higher risk of complications, and individual patient priorities play a particularly important role in decision-making. Through conversations with patients about their preferences, we might ensure that patients understand their disease and prognosis, acknowledge their right to participate, and reduce both over-and under diagnosis and treatment. Involving next-of-kin allows us to ensure that patients` wishes are known [13].

HCP experience that they inform, involve, and treat in accordance with patient preferences, even without clarifying such preferences. This may reflect a persistent paternalistic and provider-orientated culture in the wards, or more focus on diagnosis and treatment than the (actual) problems experienced by the patient [4, 10, 14]. When one-third of the HCP report clarifying the patient preferences and still do not adapt the treatment accordingly, we can assume an experience of moral stress and a gap in patient and HCP’s expectations of treatment opportunities. If this is the case, there is an urgent need to develop a culture for educating and encouraging communication for assessing preferences, both for immediate and future care [15]. Proposed strategies for improvement at a clinical level are to normalize ACP conversations, team-discussion of prognosis and training in communication skills for quality and trust in clinician-patients conversations, especially in acute severe illness where the decisions have high impact [5, 6, 15, 16]. Further, there is a need for leadership and operational plans at organizational- and system-levels [9, 16].

Resources and time, severely ill patients and the need for more knowledge, communication skills and confidence are known also previously reported causes for not informing and involving according to wishes of the patient and next-of-kin [17]. Other possible explanations can be older patients with impaired cognitive function and complex and severe conditions with prognostic uncertainty [2]. Health literacy (the degree of having the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions,) is also known to be low among older patients [18].

HCP in our survey seems to favor shared decision-making between HCP, patients, and next-of-kin. Assessing preferences is valuable [7, 8], even in the context of acute hospitalization. Further, patients in geriatric wards suffer high morbidity and mortality, with frequent decisions about limit life-sustaining treatment, where next-of-kin often can inform about the patient preferences and priorities in life. Even if Norway and most European countries do not have advance directives, patient views should always be considered. This underscores the importance of having access to the patient’s views, where ACP could be an important source [19, 20].

One strength of the current study is the high number of respondents from different hospitals to elicit perspectives of important ethical and legal principles. The survey methodology has limitations with self-reported design, non-validated measures, lack of information of non-responders and other explanations for discrepancies, subjective experiences with personal opinions, not assessing goals and values or access to medical records.

In conclusion, our findings support a limited engagement in conversation and inclusion of patient preferences when providing healthcare to older patients in acute geriatric wards. There is an urgent need to develop a culture for respecting autonomy by involving patients and their next-of-kin into ethical healthcare discussions with a more personalized thinking. Clinicians may consider using ACP and develop communication skills to integrate patient’s autonomy when providing medical care.