Comparing two models of outpatient specialised palliative care

Patients with life-threatening illness often suffer from reduced quality of life (QOL) and multiple symptoms, such as pain, fatigue, and dyspnoea, related to their illness and/or its treatment [1, 2]. Ideally, these patients should receive palliative care that addresses their specific needs. The many well-defined clinical pathways, i.e. the overarching guidelines for medical treatment, care and follow-up, in general ensure that this care is made available, but they may not always leave room for a person-centred and individual approach in respect to symptom control, psychosocial and spiritual support, along with practical issues [3, 4].

In 2019, 10,160 patients in Denmark suffering from cancer and 1133 with other diagnoses were referred to specialist palliative care (SPC). The median survival from first contact with the team was only 38 days for the cancer patients and 37 days for patients suffering from other diagnoses [5]. Referral to SPC thus occurs late in life and, for some patients, this leaves health professionals with little time to arrange and evaluate care, which is why it is important to learn from other patients’ and carers’ experiences when developing practice. The Danish Health Authority defines palliative care in line with the World Health Organization’s definition [6], as described in Table 1.

In Denmark, the hospitals are responsible for SPC. The care may be organised as either inpatient hospitalisation or at outpatient clinics. Inpatient care in this context refers to care provided to hospitalised patients by an SPC consultation service. Outpatient refers to care provided to non-hospitalised patients who either visit a clinic at the hospital or receive home visits by the SPC consultant team from the hospital SPC consultation service.

Two recently published meta-analyses, including both inpatient and outpatient SPC, showed that interventions were associated with improvements in patient QOL [7, 8], while only one found a positive result for symptom management [8]. Of the 43 included studies in the latter meta-analysis, satisfaction with care was assessed in 11 trials and, of those, seven reported a significant improvement in satisfaction among SPC recipients [8]. It is not clear whether the type of intervention – inpatient or outpatient – was of importance.

From the patient perspective, personalised care provided by inter-professional teams and communication are key elements of the experience of quality of care and satisfaction [9‐11]. Depending on their illness, however, patients may have divergent palliative care needs, which the SPC team should be in a position to address. Personalised palliative care should be tailored to provide the right level of intervention for the right patient in the right setting at the right time [11].

In an interview study, Wentlandt et al. (2016) identified six themes of importance in relation to quality of care and satisfaction in palliative care units. Both patients and families emphasized the importance of an interdisciplinary team for the management of SPC, i.e. doctor, nurse, physiotherapist, nutritional therapist, social worker, psychologist and priest; that the care is attentive and personalised; and that it is family-centred. In addition, the authors found that SPC should be allocated appropriate resources and adequate staffing to provide consistent care, and that the setting should be non-institutionalized, thereby allowing for both privacy and socialization [10]. Communication was the most prevalent theme, which encompassed the following: the SPC team should address expectations and explain care goals, keep patients and families informed about the patient’s condition, listen actively to validate patients’ concerns and individual needs, and provide a safe space for conversations about death and dying [10]. These results are consistent with the qualitative results of a systematic review of patients’, families’, and staff’s satisfaction with end-of-life care [12]. In regard to the intervention studies included in the review, Dy et al. (2008) found that palliative care and hospice teams improve patient and family satisfaction. However, it was also found that many studies does not include satisfaction as an outcome, and that more focus on satisfaction-based elements might improve the effectiveness of end-of-life care interventions and their evaluation [12].

While many researchers have studied in-hospital SPC, much fewer have investigated outpatient SPC in connection with home visits [13, 14]. As a result of this, many published studies does not include standardized outpatient SPC- models. For this reason, it can be problematic to compare interventions, organization or clinical results [13]. In a review study, Hui et al. (2018) found that studies involving interdisciplinary SPC teams were more likely to have a positive outcome than those involving solely nurse-led palliative care, though no single study directly compared interdisciplinary teams with single-practitioner-led models [11]. A review from 2013 found only four well-designed RCTs, but a growing body of non-randomized studies that indicated that outpatient SPC may: improve patient satisfaction, symptom control and QOL; reduce health care utilization; and lengthen survival in a lung cancer population [13]. In addition, only four studies concerning outpatient SPC were found by Zimmermann et al. (2008). Three of those showed significantly improved satisfaction when receiving SPC [15].

This seems to show that both inpatient and outpatient SPC improves patient satisfaction, but it is not clear how the outpatient SPC were organized, as the studies do not describe a standardized outpatient SPC model. However, it is clear that core elements of quality of care and satisfaction are personalised care by inter-professional teams and the teams’ ability to communicate open-minded [9, 10, 16]. This is an individual approach to symptom control, psychosocial and spiritual support, together with practical issues.

In Denmark, most patients are very satisfied with the SPC they receive, however the factors of age, marital status and place of care were found to have some impact, in a nationwide survey from 2012 [17]. In other studies that examined the effect on early SPC no difference were found between intervention and standard care [18, 19].

Although several issues may be common to patients and families, studies show that factors important to quality at the end of life differ by role and by individual [17, 20, 21]. Models may vary in relation to different national health and welfare systems, geography, culture and traditions, and the literature does not clearly prescribe a model for outpatient SPC [22]. Thus, when deciding how to organize outpatient SPC in the future, it is relevant to study outpatient SPC and include both patients’ and families’ perspectives.

Thus, the overall aim of this study was to compare two different models for outpatient SPC, i.e. first contact between patient, next-of-kin and doctor/nurse in the form of a home visit, and the first visit taking place in the hospital setting. This was with a view to gaining new knowledge about satisfaction, symptom control and QOL in relation to outpatient SPC and to include this knowledge in future service planning and organization.

A total of 190 participants were enrolled in the study (67 from R/K and 123 from N/S/N). During 2017, 1038 were registered to the SPC teams (371 to R/K and 667 to N/S/N). A total of 220 eligible patients were listed in the study while 113 were not. The SPC team nurse who was responsible for informing patients about the study and receiving potential participants’ acceptance form was challenged for time. For this reason, including participants for research studies could not be prioritised. A total of 705 were invited to participate and, of those, 190 accepted the invitation. The flow of participants and their reasons for leaving the study is shown in Fig. 1. Of the 190 invited, 102 returned the first questionnaire and were thus included in the study; 55 returned the second questionnaire and 31 the third. Of the 102 the age range was 38 to 88, with a mean age of 68 (SD 9.47). Patients enrolled but not included were the 88 who did not return the first questionnaire. Of those, 80 died within 110 days and the mean residual life was 13 days. We found no difference between participants and those not included, with respect to sex, mean age or distribution between the hospitals.

At baseline, the only significant difference in demographic characteristics between the participants from N/R and N/S/N were their level of education, basis for income and distance to hospital, which was expected, given the hospitals’ geographic location (Table 3).

At baseline, patients from hospitals R/K had a significantly higher overall mean score for satisfaction with care (65.91 vs. 55.83; p = 0.03), measured by FAMCARE-P16. Based on the individual score of > 64 as generally satisfied, patients from R/N were satisfied, while patients from N/S/N were not satisfied, although in comparing each individual question, we found no significant differences between the two groups. However, more patients were satisfied with the availability of nurses as compared to that of doctors (Table 4).

Satisfaction with communication, measured by CPWQ-com, was equal in the two groups at baseline. More patients were satisfied with nurses’ abilities to listen and communicate than doctors. In terms of QOL, measured by EORTC QLQ-C15-PAL, we found that patients from hospitals R/K had a significantly higher mean score in relation to physical function (66.99 vs. 58.01; p = 0.01). No other differences were found (Table 5).

Comparing differences in satisfaction from baseline to one and 3 months, we found no differences in development of either FAMCARE-P16 or CPWQ-com between the two hospitals.

The development in overall QOL did not differ between the two hospitals at one or 3 months. However, in relation to the symptom insomnia, the mean difference in scores were significantly higher p = 0.01 at 1 month follow-up for patients at hospital R/K (mean change − 28.81; SE 11.05). In addition, the development in the symptom constipation was significant (p = 0.04). The mean change was higher for patients from R/K at 3 months follow-up (mean change − 23.09; SE 11.25) (Table 6 and Fig. 2).

In this study, we found no differences over time between the two models of SPC consultations in terms of overall satisfaction, satisfaction with communication or overall QOL. We found a significant difference in the development of the symptom insomnia at 1 month and in constipation at 3 months. At baseline, there were no significant differences in the symptom insomnia between the two groups that scored in mean 26 (R/K) and 34 (N/S/N). But, at the 1 month follow-up, patients from both groups reported much higher mean scores, i.e. 85 (R/K) and 64 (N/S/N), reflecting a worsening of their insomnia, which was more severe at hospital R/K. In comparison, Nordic studies found a mean score of 24 in a nationally representative sample of advanced cancer patients in Denmark [2] and 39-41in a regional sample of cancer patients in Norway admitted to palliative care [33]. SPC involves the sub-group of patients who are most burdened by symptoms and problems; and as SPC is initiated at the end-stage of life, progression of the disease and symptoms is often rapid [2]. Patients may thus experience a worsening of symptoms, which may explain the higher mean score for such palliative care patients. Johnsen et al. found insomnia to be one of the most frequent symptoms in advanced cancer patients [2], and Rabow et al. found that sleep quality improved with outpatient palliative medicine consultations [13]. However, this was not the case in our study. On the contrary, it seems that patients sleep quality worsened and more so at the hospital R/K. An important difference between the included hospitals were the location and duration of the initial visit. The SPC team from N/S/N did home visits that often lasted more than 1 hour, which may well have helped patients to establish a closer, more trusting relationship with the team during the first consultation. This resulted in a more comprehensive symptom assessment. Getting a chance to know the team and establish a relationship along with being involved are some of the fundamental perspectives of open and effective communication, and represent two of the core themes in a person-centred approach to care [34]. Context is the third theme found by Kitson et al. (2013) [34]. Even though Wentlandt et al. (2016) found that home visits represent a non-institutionalized setting that allows for both privacy and socialization [10], sufficient time resource allocation and a qualified SPC team who have chosen to work within the palliative care setting and who are dedicated to offering a person-centred approach seem to be at least as important as the setting in which the consultation takes place.

In relation to communication, between 95.8 and 97.9% of the patients found, the nurses to be good at both listening and communicating, and more patients were satisfied with nurses’ abilities to listen and communicate than they were with doctors’ abilities; which was also found by Ross [35]. In our study, the nurses conducted the follow-up telephone consultations and were first in line to answer phone calls from the patients. This division of labour may also reflect the reason why nurses were perceived as available and therefore able to answer questions to a larger degree than were doctors. The very high number of patients who were satisfied with the nurses’ communication indicates that the nurses succeeded in not only open communication, but also in possessing the personal qualities such as being polite, respectful, sensitive and welcoming [34]. All those abilities should be present in establishing a therapeutic nurse/patient relationship. This appears to be crucial to ensuring and identifying patients’ holistic needs, and to their involvement in decision-making. A holistic approach is a factor that is important to patients, when health professionals provide emotional support and support patients’ lives [36].

We know from previous studies that SPC improves satisfaction with care [13, 37, 38]. In this study, we found a significant difference in satisfaction at baseline between the two groups – patients from R/K scored significantly higher, with a mean score of 65.91, as opposed to N/S/N55.83 at, p = 0.03. In examining the individual items, the ones representing the biggest difference are those concerning attention and information from doctors. This probably reflects the fact that the patients at N/S/N only see a doctor during their first consultation, and even though it is a home visit, follow-up is primarily conducted by the patients’ GPs and community home care nurses. In comparison, patients from R/K have follow-up consultations that are conducted by a doctor and nurse from the SPC team every 3 weeks and, in the meantime, if needed, follow-up phone calls with SPC nurses. In addition, the SPC team at N/S/N only had one doctor to cover all home visits, which meant there was very little time to respond to patients’ unscheduled enquiries. The difference in satisfaction concerning attention and information from doctors might have been more related to pressure of work rather than to personal ability. The level of staffing is critical and, as shown in Table 2, R/N had a broader multidisciplinary team than did N/S/N. To access non-medical, non-nursing members of the team, the patients needed a referral from the team doctor or nurse, which may also have meant a restriction on access to the multidisciplinary team members. The shortage of doctors was also reflected in the longer waiting time at N/S/N, compared to R/K [5]. This represents inequality in access to SPC and may be another explanation as to why patients were less satisfied at N/S/N. For those patients, the inequality is exacerbated by living in a rural area, which is characterised by a shortage of GPs as well as a lower level of income and education [39]. Other studies have shown that rural residents may be less likely to receive palliative care [40, 41]. Besides, those who do not receive palliative care are more likely to receive potentially aggressive end-of-life care, spend more days in hospital in the last part of life and die in an acute care hospital [41, 42]. One reason why the difference in satisfaction is not great may be that the SPC team is made up of professionals who have actively chosen palliation as their profession and that they are dedicated to and focused on providing a person-centred approach.

A strength of the study was that patients were included from a broad, heterogeneous area and the inclusion period proceeded for 1 year. In addition, data were collected not only at baseline but also one and 3 months later. This gave us a picture of the development over time and allowed us to reveal consequences of inequity in the SPC.

On the other hand, even though we included participants over a year, only 102 of the 705 possible participants returned the first questionnaire. This may have introduced a selection bias, as the mean residual life of included patients was longer than for those who did not return the first questionnaire. Further, both symptoms and QOL seemed to worsen in the late palliative care phase, which we may not have captured. We used questionnaires to collect data; a limitation to this approach may be that, despite being validated, answers are limited to the questions asked. We would have liked to elicit more detailed and nuanced patient perspectives on what it was in the SPC consultations that was important to them. We therefore conducted interviews in a later study.

Specialist palliative care is in request for many patients during the late phase of their disease. Outpatient SPC may be organized using a wide variety of models. In this article, we describe two such models – one of which involves the first encounter in the form of a home visit, while, in the other, the first encounter takes place in a hospital-based outpatient clinic. We analysed the two models using a questionnaire design and found no significant difference in satisfaction with care, communication with health professionals or in overall quality of life. In fact, it seems more important that patients have access to SPC than which model is used. Furthermore, patient participation and involvement in care seem to be important to level of satisfaction, and it may that the time available matters more than the context. Thus, SPC must leave room for a person-centred and individual approach concerning symptom control, psychosocial and spiritual support, as well as for practical issues.