Interventions to Promote End-of-Life Conversations: A Systematic Review and Meta-Analysis

CONTEXT

Although several interventions aimed to promote end-of-life conversations are available, it is unclear whether and how these affect delivery of end-of-life conversations. Measuring the processes associated with high-quality end-of-life care may trigger improvement.

OBJECTIVES

To estimate the effect of interventions aimed to promote end-of-life conversations in clinical encounters with patients with advanced chronic or terminal illness or their family, on process indicators of end-of-life conversations.

METHODS

Systematic review with meta-analysis (PROSPERO no. CRD42021289471). Four databases (PubMed, CINAHL, PsycINFO, and Scopus) were searched up to September 30, 2021. The primary outcomes were any process indicators of end-of-life conversations. Results of pairwise meta-analyses were presented as Risk Ratio (RR) for occurrence, standardized mean difference (SMD) for quality and ratio of means (ROM) for duration. Meta-analysis was not performed when fewer than four studies were available.

RESULTS

A total of 4,663 articles were scanned. Eighteen studies were included in the systematic review and 16 entered at least one meta-analysis: documented occurrence (n = 8), patient-reported occurrence (n = 4), patient-reported-quality (n = 4), duration (n = 4). There was significant variability in settings, patients' clinical conditions, and professionals. No significant effect of interventions on documented occurrence (RR 1.54, 95% CI 0.84-2.84; I2 91%), patient-reported occurrence (RR 1.52, 95% CI 0.80-2.91; I2 95%), patient-reported quality (SMD 0.83, 95% CI -1.06 to 2.71; I2 99%), or duration (ROM 1.20, 95% CI 0.95-1.51; I2 65%) of end-of-life conversations was found. Data on frequency were conflicting. Interventions targeting multiple stakeholders promoted earlier and more comprehensive conversations.

CONCLUSION

Heterogeneity was considerable, but findings suggest no significant effect of interventions on occurrence, patient-reported quality and duration of end-of-life conversations. Nevertheless, we found indications for interventions targeting multiple stakeholders to promote earlier and more comprehensive conversations.

[The nurse's role in the process of advance care planning]

The COVID-19 pandemic and its impact on older and frail people underlines the importance of advance care planning (ACP). ACP is a dynamic communication process involving patients, families and healthcare providers, which serves to discuss and document wishes and goals for future care. Currently, ACP practice is often suboptimal. This implies that important decisions about care and treatment may need to be made acutely in crises. Many factors contribute to suboptimal ACP practice. One such factor is ambiguity regarding roles and responsibilities of different disciplines in the ACP-process. The perception that having ACP conversations is primarily a physician's task is a misconception. Specific skills that could contribute to a holistic and person-centered ACP-process are largely lacking in nursing curricula and therefore, may be insufficient and under-utilized. For instance, nursing staff could involve persons in conversations about meaning, quality of life, loss and grief as a part of ACP. Moreover, they may communicate a patient's wishes to other healthcare providers including physicians. Acknowledgement of this potential role, by physicians as well as by nursing staff themselves, is needed for ACP to become a truly interprofessional process.

Consensus on treatment for residents in long-term care facilities: perspectives from relatives and care staff in the PACE cross-sectional study in 6 European countries

Background

In long-term care facilities often many care providers are involved, which could make it difficult to reach consensus in care. This may harm the relation between care providers and can complicate care. This study aimed to describe and compare in six European countries the degree of consensus among everyone involved in care decisions, from the perspective of relatives and care staff. Another aim was to assess which factors are associated with reporting that full consensus was reached, from the perspective of care staff and relatives.

Methods

In Belgium, England, Finland, Italy, the Netherlands and Poland a random sample of representative long-term care facilities reported all deaths of residents in the previous three months (n = 1707). This study included residents about whom care staff (n = 1284) and relatives (n = 790) indicated in questionnaires the degree of consensus among all involved in the decision or care process. To account for clustering on facility level, Generalized Estimating Equations were conducted to analyse the degree of consensus across countries and factors associated with full consensus.

Results

Relatives indicated full consensus in more than half of the residents in all countries (NL 57.9% - EN 68%), except in Finland (40.7%). Care staff reported full consensus in 59.5% of residents in Finland to 86.1% of residents in England. Relatives more likely reported full consensus when: the resident was more comfortable or talked about treatment preferences, a care provider explained what palliative care is, family-physician communication was well perceived, their relation to the resident was other than child (compared to spouse/partner) or if they lived in Poland or Belgium (compared to Finland). Care staff more often indicated full consensus when they rated a higher comfort level of the resident, or if they lived in Italy, the Netherland, Poland or England (compared to Finland).

Conclusions

In most countries the frequency of full consensus among all involved in care decisions was relatively high. Across countries care staff indicated full consensus more often and no consensus less often than relatives. Advance care planning, comfort and good communication between relatives and care professionals could play a role in achieving full consensus.

Electronic supplementary material

The online version of this article (10.1186/s12904-019-0459-9) contains supplementary material, which is available to authorized users.

[Effects of a stepwise approach to behavioural problems in dementia: a cluster randomised controlled trial]

OBJECTIVE

To investigate whether implementation of a stepwise multidisciplinary intervention ('STA OP!' ['STAND UP!']) is effective in reducing behavioural problems and depressive symptoms in nursing home residents with advanced dementia.

DESIGN

Cluster randomised controlled trial.

METHOD

We implemented the STA OP! protocol on the intervention units by training the entire multidisciplinary team. This team was trained in all 6 steps of the protocol during five 3-hour sessions. Professionals working on the control unit received training on general technical nursing skills, dementia management and pain, but then without the stepwise component. All elderly care physicians were given additional training in pain management in patients with dementia, based on the guidelines on pain in vulnerable older people. Measurements were taken at baseline, and after 3 and 6 months. We used longitudinal 'multilevel' techniques to correct for clustering of data (e.g. at unit level) for statistical analysis (Dutch Trial Register: NTR1967).

RESULTS

A total of 288 residents with dementia were included, from 12 nursing homes (21 units): 148 in the intervention group in 11 units and 140 in the control group in 10 units. On the units where the STA OP! protocol was used there was a significant decline in agitation, neuropsychiatric symptoms and depression compared with the control units at 6 months. Furthermore, use of anti-depressive medication was significantly lower on the intervention units (odds ratio: 0.32; 95% CI: 0.10-0.98).

CONCLUSION

This cluster RCT revealed that the stepwise multidisciplinary intervention STA OP! is effective in reducing behavioural problems and use of psycho-pharmaceuticals in nursing home residents with dementia.

European palliative care guidelines: how well do they meet the needs of people with impaired cognition?

OBJECTIVE

Numbers of people dying with cognitive impairment (intellectual disability (ID), dementia or delirium) are increasing. We aimed to examine a range of European national palliative care guidelines to determine if, and how well, pain detection and management for people dying with impaired cognition are covered.

METHODS

Questionnaires were sent to 14 country representatives of the European Pain and Impaired Cognition (PAIC) network who identified key national palliative care guidelines. Data was collected on guideline content: inclusion of advice on pain management, whether cognitively impaired populations were mentioned, assessment tools and management strategies recommended. Quality of guideline development was assessed with the Appraisal of Guidelines Research and Evaluation (AGREE) instrument.

RESULTS

11 countries identified palliative care guidelines, 10 of which mentioned pain management in general. Of these, seven mentioned cognitive impairment (3 dementia, 2 ID and 4 delirium). Half of guidelines recommended the use of pain tools for people with cognitive impairment; recommended tools were not all validated for the target populations. Guidelines from the UK, the Netherlands and Finland included most information on pain management and detection in impaired cognition. Guidelines from Iceland, Norway and Spain scored most highly on AGREE rating in terms of developmental quality.

CONCLUSIONS

European national palliative care guidelines may not meet the needs of the growing population of people dying with cognitive impairment. New guidelines should consider suggesting the use of observational pain tools for people with cognitive impairment. Better recognition of their needs in palliative care guidelines may drive improvements in care.

Verwijzen naar het hospice: verwijsgedrag van artsen en ervaren belemmeringen in Deventer en omgeving

PHYSICIANS' REFERRAL PATTERNS AND PERCEIVED BARRIERS IN THE DEVENTER REGION IN THE NETHERLANDS:

OBJECTIVE

To examine physicians' perceived referral patterns and barriers to referral of terminally ill patients to a hospice (institute).

DESIGN

Survey study among physicians practicing in hospital and other settings in the region of Deventer, The Netherlands, in 2011-2012.

METHOD

We translated two available American instruments into Dutch. The questionnaire assessed hospice referral, knowledge about hospice, attitudes and barriers and reasons not to refer. We queried physicians who had referred patients to the local hospice about expectations and suggested areas for improvement with two open-ended items.

RESULTS

In total, 240 physicians received the questionnaire. The response rate was 47%. The physicians were generally positive about hospice care. They indicated experiencing few barriers in hospice referrals, but 32% of the physicians (21% of those practicing in the hospital, and 39% in other settings), indicated the patient being unready as a strong barrier. Half of the physicians (51%) believed that hospice is being underutilized and 22% (35% and 14%, respectively) thought that they would refer more frequently if they had more knowledge about hospice care. Of the physicians, 35% answered all six knowledge questions correctly. Communication with the hospice may be improved.

CONCLUSION

Despite positive attitudes toward hospice care, it may be underutilized due to poor knowledge and communication with the hospice. Perhaps, this is also due to limited patient-physician communication on prognosis, which further research may address.

[Evaluation of a family booklet on comfort care in dementia by professional and family caregivers]

Families of nursing home residents with dementia are usually involved in care and treatment decisions. To this end, family needs to be informed on the course of the dementia and possible palliative care. Based on a Canadian booklet, we developed an adapted version for use in the Netherlands. Elderly care physicians (n = 30), nurses (n = 38), and bereaved families (n = 59) evaluated the booklet and possible implementation strategies. All respondents confirmed that in general, there is a need of an information brochure on comfort care and end-of-life issues for families. Most (93%) families believed they would have found the booklet useful when received earlier. Compared to the physicians, nurses more frequently found the booklet useful to most or all families (p = 0.04). Acceptance, as measured on an 8-item scale, was highest among families and lowest among physicians. Overall usefulness was often perceived as high (means 7.9 to 8.3; scale range 0-10; SD 0.9 to 1.4) and did not differ across groups of respondents (p = 0.29). All respondents agreed that professional caregivers should have a role in providing the booklet. Additionally, half (53%; no difference across groups) favoured availability of the booklet through families' own initiative; some already before admission. In conclusion, there is a high need for written information on palliative care. The booklet is highly appreciated. A further improved version may support professional and family caregivers in practice.

Ratings of symptoms and comfort in dementia patients at the end of life: comparison of nurses and families

After-death reporting by proxies on end-of-life outcomes is used in research and can also be used to target institutions directly to improve practice. We compared the scores of family caregivers and nurses on two End-of-Life in Dementia Scales (EOLD) scales: Symptom Management (SM; range 0-45) over the last 3 months of life and Comfort Assessment in Dying (CAD; range 14-42). Higher scores represent better outcomes. Four Dutch nursing homes retrospectively enrolled 48 decedents with dementia. Total mean scores for family caregivers and nurses were 28.7 (SD 9.6) versus 25.2 (SD 12.7) for the SM and 31.7 (SD 5.5) versus 32.8 (SD 8.2) for the CAD. Mean item scores also did not differ much. Concordance Correlation Coefficients were 0.42 (SM) and 0.04 (CAD). Mean evaluations of end of life with dementia corresponded reasonably well between family and professional caregivers, but correspondence of individual observations was poor to moderate, suggesting serious differences in individual ratings but little systematic difference.

[End-of-life with dementia in Dutch antroposofic and traditional nursing homes]

Every year more than 20,000 people with dementia die in Dutch nursing homes and this number steadily increases. Therefore, the importance of good end-of-life care for these patients including physical, psychosocial and spiritual care is evident. Although the training standards for Dutch nursing home physicians and nurses share a common standard, the philosophy of a nursing home may affect end-of-life care strategies for the residents. We compared end of life of nursing home residents with dementia in two anthroposophic and two traditional nursing homes in a retrospective study using the most specific instrument available: the End-of-Life in Dementia scales (EOLD). Family caregivers completed the EOLD questionnaire. There was no difference in mean Satisfaction With Care scale scores between both types of nursing homes: 32.9 (SD 4.3) and 31.6 (SD 4.9), respectively. The anthroposophic nursing homes had significant higher scores on the 'Symptom Management' ((32.9 (SD 7.5) versus 26.9 (SD 9.5)), and 'Comfort Assessment in Dying' scales (34.0 (SD 3.9) versus 30.8 (SD 5.8)) and on its subscale Well Being (7.7 (SD 1.2) versus 6.7 (SD 2.1)). Our results suggest that death with dementia was more favourable in anthroposophic nursing homes than in regular homes. The results inform further prospective studies on nursing homes how this and other philosophies are translated into daily nursing home practice, including decision making in multi-disciplinary teams, family consultation, and complementary non-pharmacological therapies.

[Pneumonia mortality risk in patients with dementia: nursing home physicians' use and evaluation of a prognostic score]

A validated prognostic score for mortality risk 14 days after antibiotics treatment of nursing home residents with dementia and pneumonia is available. Of the nursing homes contacted, 96% was prepared to participate in a clinical impact analysis to examine usefulness of the score in practice. After randomising nursing homes, physicians of 27 homes in the intervention group were asked to complete a questionnaire and use the score for the next case of pneumonia; the control group comprised physicians of the 27 other homes who only completed the questionnaire. The 38 respondents from the control group who all reported about a single patient did not differ from the respondents of the intervention group (31 physicians enrolled 34 patients). Only in 24 cases did physicians calculate the score. For 79% of those patients, the score was (at least somewhat) useful, but mostly to train prognostication competencies and for better documentation of prognosis; frequently treatment decisions had already been made. Of the total group of respondents, the majority was positive about the use of prognostic scores in general, but no-one in the participating homes had any experience with it. The prognostic score is potentially useful for an important group of patients with pneumonia, but further implementation research and inclusion of prognostic instruments in training curricula is needed.

Benefits and pitfalls of pooling datasets from comparable observational studies: combining US and Dutch nursing home studies

Different research groups sometimes carry out comparable studies. Combining the data can make it possible to address additional research questions, particularly for small observational studies such as those frequently seen in palliative care research. We present a systematic approach to pool individual subject data from observational studies that addresses differences in research design, illustrating the approach with two prospective observational studies on treatment and outcomes of lower respiratory tract infection in US and Dutch nursing home residents. Benefits of pooling individual subject data include enhanced statistical power, the ability to compare outcomes and validate models across sites or settings, and opportunities to develop new measures. In our pooled dataset, we were able to evaluate treatments and end-of-life decisions for comparable patients across settings, which suggested opportunities to improve care. In addition, greater variation in participants and treatments in the combined dataset allowed for subgroup analyses and interaction hypotheses, but required more complex analytic methods. Pitfalls included the large amount of time required for equating study procedures and variables and the need for additional funding.

[Dying with dementia: what do we know about it?]

Death with dementia is increasingly common, yet research on end of life with dementia and end-of-life care for such patients has been sparse. This article reviews recent studies in this area, most of which were done in US nursing homes. Research focused on five domains: prognosis, decision making, treatment, patient's health and suffering, and family's circumstances and satisfaction with care. Prognostication focused on developing risk scores for mortality within 6 months or a year, and while decision making was usually studied qualitatively, the other three domains were largely covered by a series of small, retrospective studies. Future direction in research is discussed, including the ongoing CASCADE project in Boston and the Dutch End of Life with Dementia Study (DEOLD). Both of these prospective studies in nursing home residents assess decision making, as well as factors associated with family's satisfaction and patient suffering. These studies will provide insight into interventions that are most likely to improve end of life care of patients with dementia in the respective countries and elsewhere.

Dementia severity, decline and improvement after a lower respiratory tract infection

OBJECTIVE

To assess decline and improvement in functional characteristics, cognition and restraint use after a lower respiratory tract infection (LRI) and describe variation by dementia severity.

DESIGN

Two prospective cohort studies.

SETTING

Nursing homes in the Netherlands and in Missouri, USA.

PARTICIPANTS

227 Dutch and 396 Missouri nursing home residents with dementia and LRI who were treated with antibiotics.

MEASUREMENTS

We compared functional characteristics (Activities of Daily Living [ADL], bedfast status, pressure ulcers, incontinence), cognition and restraint use 3 months after an LRI with status 1 to 2 weeks before diagnosis.

RESULTS

Residents with LRI frequently declined on all measures, but many also improved, including those with severe dementia. On the measures where residents could still decline further, residents with severe dementia showed higher variability than residents with less severe dementia. This was most obvious for bedfast status and restraint use.

CONCLUSIONS

Compared with less severely demented residents, residents with severe dementia showed more decline on measures where they still had room for change. However, on these measures, residents with severe dementia also improved more often. LRI does not necessarily lead to deterioration even in individuals with severe dementia.

[Treatment of pneumonia in nursing home residents with severe dementia: for residents with poor prognosis, a more reserved approach in The Netherlands and more active treatment in the United States]

OBJECTIVE

To describe differences in the treatment of pneumonia and in the association of treatment with prognosis in Dutch and American nursing home patients with late-stage dementia. Design. Prospective studies in The Netherlands and the American state of Missouri.

METHOD

In 61 Dutch nursing homes and 36 in Missouri, severely demented patients with pneumonia were included in the periods October 1996-July 1998 and August 1995-September 1998 respectively. Data was collected on their state of health, comorbidity, symptoms of pneumonia and treatment aspects such as antibiotic use, hospital admission and relief of symptoms. Comparisons were made between treatments in both countries and between groups of patients with a similar probability of mortality within 2 weeks.

RESULTS

A total of 328 Dutch and 280 American patients were selected. Antibiotics were more frequently withheld in The Netherlands (in 33% of patients) than in Missouri (24%). Differences in antibiotic use were more pronounced in patients with a poor prognosis (56% versus 15%). Dutch patients were more frequently dehydrated but were less likely to receive rehydration therapy than American patients, with a larger difference in patients with a poor prognosis (2% versus 63%). Treatments to relieve symptoms that were provided more often in patients with a poor prognosis (in 20-26%) were: oxygen (both countries), and in The Netherlands also opiates, and hypnotics, sedatives or anxiolytics.

CONCLUSION

In The Netherlands, curative treatment was frequently withheld in patients with severe dementia and pneumonia, and even more frequently when the prognosis was poorer. Conversely, treatment in Missouri was more active in patients with a poor prognosis. Despite more frequent palliative treatment goals in The Netherlands, treatments to relieve symptoms were provided infrequently and inconsistent with this approach. These insights may be helpful for decision-making in the treatment of pneumonia in patients with severe dementia.

[A cross-cultural study of physician treatment decisions in relation to demented nursing home patients who develop pneumonia]

This qualitative interview study in The Netherlands and North Carolina (US) found that physician treatment decisions are influenced by contextual differences in physician training and healthcare delivery in the US and The Netherlands. Dutch physicians treating nursing home residents with dementia and pneumonia assumed active, primary responsibility for treatment decisions while US physicians were more passive and deferential to family preferences, even in cases where they considered the families' wishes inappropriate. Dutch physicians knew their patients well and made treatment decisions based on what they perceived was in the best interest of the patient while US physicians reported limited knowledge of their nursing home patients due to a lack of contact time. Efforts to improve care for patients with poor quality of life who lack decision-making capacity must consider the context of societal values, physician training, and the processes by which physicians negotiate patient and family preferences.

[Quality of life and dementia. I. Model of assessment of wellbeing in dementia patients]

The introduction of cholinesterase inhibitors to improve the cognitive function and activities of daily living in patients with Alzheimer disease, raises the question whether these drugs also have the potential to improve the quality of life of these patients. In this article a model is presented to measure quality of life in patients with dementia, in which psychological well-being is chosen as the central measure. The presented model might be the starting point to develop instruments to measure quality of life in dementia. After a short introduction concerning the developments in quality of life research, the two most important characteristics of the concept--multidimensionality and subjectivity--are discussed against the background of the relevant literature on dementia. The dementia-specific dimensions--individual characteristics, psychological, social and physical dimension, and environment--and domains are presented, and put in a hierarchical model.

[Quality of life and dementia. II. Selection of a measurement instrument for wellbeing appropriate for the reference model]

Measuring demented patients' 'quality of life' is an important issue in the Netherlands. Due to diminished cognition and other characteristics of dementia, general instruments are not suitable. A conceptual model for quality of life in dementia should guide instrument choice. We used a recently developed model, in which 'well-being' is indicated as the standard for quality of life. Appropriateness of fit in the model is shown of three important instruments developed in the '90s. Eventually, the 'Dementia Quality of Life instrument' (DQoL or D-QoL) of Brod et al., 1999 was selected as exhibiting the best fit. The DQoL measures well-being as a balance of positive and negative aspects. Other instruments consider, for example, either positive or negative aspects or include determinants of well-being, such as the general health condition. In contrast to the other instruments, the DQoL requires only the patient's own opinion. We plan further validation of the DQoL in Dutch demented patients.

Decisions to treat or not to treat pneumonia in demented psychogeriatric nursing home patients: evaluation of a guideline

We evaluated a new guideline, in the form of a "checklist of considerations," to support end-of-life decision making in the treatment of demented patients with pneumonia. Questionnaires were sent to nursing home physicians (NHPs) in The Netherlands at three times: before implementation of the checklist (concerning 91 individual patients), during use of the checklist (concerning another 107 individual patients), and after data collection (concerning the targeted patient category of demented nursing home patients with pneumonia as a whole). In the last questionnaire, one NHP from each nursing home (n = 55 NHPs) gave his or her general opinion about the checklist. We measured the usefulness of the checklist in supporting decision making and its frequency of actual use. The NHPs accepted the contents of the checklist for use in the targeted patient category. It was used in 46% of the incident cases of pneumonia. The checklist was considered more useful in supporting decision making for the targeted patient category (85% of the NHPs) than for the individual patient (47%). Possible explanations for this discrepancy in "usefulness" include the difference in the nature of the outcome measures and the fact that the checklist was used more frequently for the "easier cases." Information on individual patient level, patient category level, and nursing home and NHP characteristics is used to suggest checklist improvements.

[The role of observer for the reliability of Dutch version of the Discomfort Scale-Dementia of Alzheimer Type (DS-DAT)]

The role of the observer in the reliability of the Dutch Discomfort Scale-Dementia of Alzheimer Type (DS-DAT). The Discomfort Scale of Dementia of the Alzheimer Type (DS-DAT) is an instrument to assess discomfort in severely demented patients. No data on the reliability of assessment using a Dutch translation were available. In this paper, we analyse the role of the observer in the reliability of rating. This is of importance for studies in which many physicians perform multiple assessments. Twenty-eight nursing home physicians in training rated the DS-DAT in five nursing home patients with dementia presented on videotape. This was repeated after five months. All the physicians were previously trained in the use of the instrument. The results were statistically analysed using random effects analysis of variance. The Intra-class Correlation Coefficient (ICC) was 0.74 for inter-observer reliability and 0.97 for intra-observer reliability. Variance between subsequent assessments was small, but physicians appeared to differ somewhat among themselves in the way they rated the videotaped patients. A future complete reliability assessment of rating the DS-DAT in clinical practice would involve patient variation as well, scoring patients in clinical practice.

Decisions to treat or not to treat pneumonia in demented psychogeriatric nursing home patients: development of a guideline

Non-treatment decisions concerning demented patients are complex: in addition to issues concerning the health of patients, ethical and legal issues are involved. This paper describes a method for the development of a guideline that clarifies the steps to be taken in the decision making process whether to forgo curative treatment of pneumonia in psychogeriatric nursing home patients. The method of development consisted of seven steps. Step 1 was a literature study from which ethical, juridical and medical factors concerning the patient's health and prognosis were identified. In step 2, a questionnaire was sent to 26 nursing home physicians to determine the relative importance of these factors in clinical practice. In a meeting of nine experienced physicians (step 3), the factors identified in step 2 were confirmed by most of these professionals. To prevent the final guideline being too directive, a concept guideline that included ethical and legal aspects was designed in the form of a "checklist of considerations" (step 4). Experts in the fields of nursing home medicine, ethics and law reviewed and commented on the concept guideline (step 5). The accordingly adapted "checklist of considerations" was tested in a pilot study (step 6), after which all experts endorsed the checklist (step 7). The resulting "checklist of considerations" structures the decision making process according to three primary domains: medical aspects, patient's autonomy, and patient's best interest (see annex at end of paper).