Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Effects of facilitated family case conferencing for advanced dementia: A cluster randomised clinical trial

  • Meera Agar ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

    Affiliations Faculty of Health, University of Technology Sydney, Ultimo, New South Wales (NSW), Australia, South Western Sydney Clinical School, University of New South Wales, Liverpool, NSW, Australia, Ingham Institute for Applied Medical Research, Liverpool, NSW, Australia, Improving Palliative Care through Clinical Trials (ImPaCCT), Sydney, NSW, Australia

  • Tim Luckett ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Data curation, Funding acquisition, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

    tim.luckett@uts.edu.au

    Affiliation Faculty of Health, University of Technology Sydney, Ultimo, New South Wales (NSW), Australia

  • Georgina Luscombe ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Data curation, Formal analysis, Funding acquisition, Methodology, Writing – original draft, Writing – review & editing

    Affiliation Sydney Medical School, The University of Sydney, Ultimo, NSW, Australia

  • Jane Phillips ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Funding acquisition, Methodology, Writing – review & editing

    Affiliation Faculty of Health, University of Technology Sydney, Ultimo, New South Wales (NSW), Australia

  • Elizabeth Beattie ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Funding acquisition, Methodology, Project administration, Supervision, Writing – review & editing

    Affiliation School of Nursing, Queensland University of Technology, Herston, Queensland (QLD), Australia

  • Dimity Pond ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Funding acquisition, Methodology, Writing – review & editing

    Affiliation School of Medicine and Public Health, The University of Newcastle, Newcastle, NSW, Australia

  • Geoffrey Mitchell ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Funding acquisition, Methodology, Writing – review & editing

    Affiliation Faculty of Medicine, The University of Queensland, St Lucia, QLD, Australia

  • Patricia M. Davidson ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Funding acquisition, Methodology, Project administration, Supervision, Writing – review & editing

    Affiliations Faculty of Health, University of Technology Sydney, Ultimo, New South Wales (NSW), Australia, School of Nursing, Johns Hopkins University, Baltimore, Maryland, United States of America

  • Janet Cook ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Data curation, Project administration, Supervision, Writing – review & editing

    Affiliation Faculty of Health, University of Technology Sydney, Ultimo, New South Wales (NSW), Australia

  • Deborah Brooks ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Data curation, Project administration, Supervision, Writing – review & editing

    Affiliation School of Nursing, Queensland University of Technology, Herston, Queensland (QLD), Australia

  • Jennifer Houltram ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Data curation, Formal analysis, Supervision, Writing – review & editing

    Affiliation Centre for Health Research and Evaluation (CHERE), Faculty of Business, UTS, Haymarket, NSW, Australia

  • Stephen Goodall ,

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Funding acquisition, Methodology, Supervision, Writing – review & editing

    Affiliation Centre for Health Research and Evaluation (CHERE), Faculty of Business, UTS, Haymarket, NSW, Australia

  • Lynnette Chenoweth

    Contributed equally to this work with: Meera Agar, Tim Luckett, Georgina Luscombe, Jane Phillips, Elizabeth Beattie, Dimity Pond, Geoffrey Mitchell, Patricia M. Davidson, Janet Cook, Deborah Brooks, Jennifer Houltram, Stephen Goodall, Lynnette Chenoweth

    Roles Conceptualization, Funding acquisition, Methodology, Writing – review & editing

    Affiliation Centre for Healthy Brain Ageing, University of New South Wales, Randwick, NSW, Australia

Effects of facilitated family case conferencing for advanced dementia: A cluster randomised clinical trial

  • Meera Agar, 
  • Tim Luckett, 
  • Georgina Luscombe, 
  • Jane Phillips, 
  • Elizabeth Beattie, 
  • Dimity Pond, 
  • Geoffrey Mitchell, 
  • Patricia M. Davidson, 
  • Janet Cook, 
  • Deborah Brooks
PLOS
x

Abstract

Background

Palliative care planning for nursing home residents with advanced dementia is often suboptimal. This study compared effects of facilitated case conferencing (FCC) with usual care (UC) on end-of-life care.

Methods

A two arm parallel cluster randomised controlled trial was conducted. The sample included people with advanced dementia from 20 Australian nursing homes and their families and professional caregivers. In each intervention nursing home (n = 10), Palliative Care Planning Coordinators (PCPCs) facilitated family case conferences and trained staff in person-centred palliative care for 16 hours per week over 18 months. The primary outcome was family-rated quality of end-of-life care (End-of-Life Dementia [EOLD] Scales). Secondary outcomes included nurse-rated EOLD scales, resident quality of life (Quality of Life in Late-stage Dementia [QUALID]) and quality of care over the last month of life (pharmacological/non-pharmacological palliative strategies, hospitalization or inappropriate interventions).

Results

Two-hundred-eighty-six people with advanced dementia took part but only 131 died (64 in UC and 67 in FCC which was fewer than anticipated), rendering the primary analysis under-powered with no group effect seen in EOLD scales. Significant differences in pharmacological (P < 0.01) and non-pharmacological (P < 0.05) palliative management in last month of life were seen. Intercurrent illness was associated with lower family-rated EOLD Satisfaction with Care (coefficient 2.97, P < 0.05) and lower staff-rated EOLD Comfort Assessment with Dying (coefficient 4.37, P < 0.01). Per protocol analyses showed positive relationships between EOLD and staff hours to bed ratios, proportion of residents with dementia and staff attitudes.

Conclusion

FCC facilitates a palliative approach to care. Future trials of case conferencing should consider outcomes and processes regarding decision making and planning for anticipated events and acute illness.

Trial registration

Australian New Zealand Clinical Trial Registry ACTRN12612001164886

Background

Care in advanced dementia requires a palliative approach focused on quality of life (QOL) [14]. Nursing home residents with advanced dementia often receive suboptimal palliation due to limited staff awareness and training [2, 514]. Communication between staff, health services and families (including surrogate decision-makers) addressing palliative care needs is often poor [6, 7]. Decisions include whether hospital transfers and acute interventions such as intravenous antibiotics will offer net benefit versus detriment to QOL [8, 9, 12, 1518]. Surrogates should be involved in symptom management decisions and care planning, and provide the personal context [7, 10, 14].

Case conferencing brings together health professionals and other surrogate decision-makers to build consensus on goals of care and appropriate steps for current management and advance care planning. Case conferencing shows promise for improving symptom management in people with dementia living in nursing homes but systematic reviews have highlighted methodological weaknesses in the evidence base [19, 20]. Evidence from randomised controlled trials (RCTs) in the community palliative care setting is more compelling with better maintenance of physical and mental health [21] and decreased hospitalization seen, [22] provided the process is facilitated by appropriate training and organizational support [2325].

The current study compared the efficacy of facilitated case conferencing versus usual care in improving end of life (EOL) care for persons with advanced dementia living in nursing homes.

Methods

Study design

A cluster RCT design was used because the intervention aimed to change the approach to dementia palliative care across nursing homes and randomizing individual residents or staff members would have resulted in contamination. The parallel cluster RCT design is described in a published protocol [26] (see also S1 File), and was conducted over a predefined study period of 18 months between February 2013 to December 2014. Human Research Ethics Committees of the University of NSW, University of Technology Sydney, Queensland University of Technology approved the trial. This paper focuses on the study’s first aim, namely to compare the efficacy of a facilitated approach to family case conferencing with usual care. As such, it focuses on data relevant to evaluating quality of end of life (EOL) care for participants who died during the study period. Data relevant to understanding processes influencing implementation and evaluating cost-effectiveness will be reported in other papers. Details are reported according to the Consolidated Standards of Reporting Trials Checklist for Cluster Randomised Controlled Trials (see S1 Table).

Randomization and blinding

The study’s statistician (GL) generated a block randomization schedule using a computer-generated allocation sequence, and allocated nursing homes to each arm after collection of institutional-level baseline data, stratifying by organizational affiliation. Due to the system-level nature of the intervention, participating investigators, project managers and nursing home managers could not be blinded to the evaluative aim of the research or to nursing home allocation. Staff, residents and families at each nursing home were blinded to the evaluative aim of the study, but those in nursing homes allocated to the intervention arm were aware of associated changes to practice (see below). Research staff were blinded to the evaluative aim of the study and collected data from nursing homes in only one arm to reduce the likelihood they would notice systematic differences in practice.

Sites

Sites were 20 nursing homes in two major Australian cities meeting the following criteria: 1) ≥100 beds, 2) ≥50% people with dementia, and 3) designated as facility providing intensive level of nursing home care. Nursing homes were identified from an Australian government website list [27] and approached in alphabetical order to minimise selection bias.

Participants

The MORECare initiative[28] informed target participants, namely people with advanced dementia living in residential care (‘residents’) where surrogate decision-maker involvement for palliative care planning is needed. Potentially eligible residents were identified by nursing home managers and screened by the study team. Residents needed to have dementia documented in nursing home records and advanced dementia as determined by scores on the: 1) Functional Assessment Staging Tool (FAST)[29] in dementia (≥6a, stable for 1 month), and 2) Australia–modified Karnofsky Performance Status (AKPS) [30] ≤50. These criteria were chosen because a FAST stage 7c combined with functional dependency (measured here by the AKPS) is predictive of an average survival of <6 months, and the study’s primary endpoint focused on end of life care [31]. A family member with legal authority provided written informed consent on behalf of the resident. Either the same or another family member involved in making decisions about the resident’s care also gave written informed consent for their own participation in the study. Written informed consent was collected after randomization of nursing homes.

All permanent registered and enrolled nurses and care assistants of participating nursing homes were invited to join the study via staff meeting presentations and email circulars and gave informed consent after randomization of nursing homes.

Intervention

Facilitated case conferencing (FCC) was compared with usual care (UC) as follows.

Facilitated case conferencing (FCC).

Theoretical frameworks underpinning FCC included the expected trajectory of advanced dementia [2] and evidence-based strategies for organizational culture change (clinical leadership [32] and train-the-trainer[33]). A Registered Nurse was trained as a Palliative Care Planning Coordinator (PCPC) in each nursing home working for 2 days per week or equivalent to: 1) identify residents with advanced dementia likely to benefit from a case conference; 2) organise, set an agenda, chair and document case conferences with optimal participation by family, multi-disciplinary nursing home staff and external health professionals (e.g. General Practitioner’s (GP’s)); 3) develop and oversee implementation of palliative care plans; and 4) train nursing and direct care staff in person-centred palliative care. The key features of the case conference model were: use of pre-defined specific clinical triggers for a case conference; used a shared agenda setting model where the resident, their family and all multidisciplinary staff could specify a priori areas for discussion; required attendance of the resident and/or their substitute decision maker or family member(s); was facilitated by the PCPC to ensure optimal participation of attendees; and was followed by a communication strategy to summarise actions and plan arising from the case conference. Discussion topics were not limited and were individualised to what was seen as important for the resident; and could include care planning, current and future treatment decision making, information sharing, meeting resident preferences or needs and advance care planning.

Usual care (UC).

In nursing homes randomised to UC, no staff education, training or support was provided. No restrictions were placed on nursing homes’ education programme, or approach to care planning and decision-making.

Data collection

Details of baseline, process and outcome data are provided in the published protocol [26]. Resident-level measures used validated measures in advanced dementia which perform well with surrogate report. Data on resident and nursing home variables likely to influence EOL care [34], and FCC fidelity was also collected. All nurse and family-rated measures were collected via face-to-face or telephone interview with the research team.

The primary outcome was family-rated EOL care (End of Life in Dementia (EOLD) Scales)[35] optimally assessed four to six weeks following the residents death: a) symptom-related comfort during the last 7 days of life (Comfort Assessment in Dying with Dementia; CAD–EOLD); b) symptom management in the last 90 days of life (Symptom Management at the End of Life in Dementia; SM–EOLD); and c) family or caregiver’s satisfaction with care during the last 90 days of life (Satisfaction With Care at the End of Life in Dementia; SWC–EOLD). Resident-level secondary outcomes were nurse-rated CAD-EOLD and SM-EOLD ratings as soon as possible following the resident’s death; nurse-rated QOL three monthly (Quality of Life in Late-stage Dementia (QUALID) Scale [36]), and symptoms and care in the last month of life extracted from nursing home and medical records (symptom control medication [e.g. analgesics] versus diagnosis oriented [e.g. antibiotics, anti-epileptics, anxiolytics, steroids] commencement, cessation and dose alteration; non-pharmacological strategies [e.g. family attendance, body positioning]; symptom assessment frequency; acute care episodes and potentially inappropriate non-palliative interventions [ventilation, resuscitation, enteral feeding, intravenous antibiotics and fluids, dialysis, transfusion, oxygen and surgery). Comorbidities were collected from the resident record classified as life threatening (e.g. cancer, organ failure, neurological), disabling (e.g. hearing/vision, musculoskeletal) or inter-current acute (e.g. infections).

Nursing home variables were: staff attitudes to, knowledge of and confidence in providing palliative care to people with advanced dementia (Palliative Care for Advanced Dementia [qPAD] [37]); ratio of nursing staff hours to beds; and proportion of residents with dementia.

Fidelity to protocol (intervention ‘dose’) at the resident level was collected for use in per protocol (PP) analyses, but could not be measured as planned [26] as many UC nursing homes did not routinely collect detailed information about case conferences (e.g. triggers, attendance, issues discussed), and encouraging this data collection may have led to contamination. A simpler dose measure was used, namely whether or not participating residents received a case conference during their time in the study. Dose at the nursing home level consisted of four indicators concerning the extent to which PCPCs: 1) were able to work 2 days per week, 2) were supported by managers, 3) fulfilled expectations outlined in training, and 4) diffused their role among other staff. Each indicator was scored 0, 1 or 2, with 0 representing a lesser extent, 1 a moderate extent and 2 a large extent.

Sample size

Assuming an intracluster correlation coefficient (ICC) of 0.05 (estimated from unpublished data sourced from Dutch nursing homes), a sample size of 8 clusters per arm with 15 residents (who died during the study period and for whom EOL outcomes would be available) per cluster (i.e. N = 240 in total), was considered adequate to identify a between-arm difference of 0.5 standard deviation (SD) on the EOLD scale with a two-sided 5% significance level and power of 80%. We conservatively anticipated a 10% resident dropout rate (e.g. withdrawal of consent to participate in the study; resident moved to another nursing home). Allowing for this, a recruitment target of 272 people with advanced dementia (17 per site) was set. This calculation incorporated an estimate that almost all people (98%) meeting the inclusion criteria would die (and so yield data on EOL care) within the study period (<18 months) based on review of dementia- specific mortality data from local nursing homes and evidence from the literature relating to prognostic variables referred to above [31, 38]. In other words, with a 10% withdrawal rate and 2% survival rate, of 272 participants we would expect 27 to withdraw, and of the remaining 245, five to survive to the end of the study period, resulting in a total sample size of 240 available for analysis.

Analysis

Between-arm differences at baseline in resident and nursing home characteristics and secondary outcomes were analysed using two-tailed t-tests for continuous variables where the distribution was normal or the Mann-Whitney U z-test where distribution was significantly skewed. Chi-square tests were used to determine whether differences were significant on categorical variables.

Analyses of primary and secondary outcomes were on an intention to treat (ITT) basis. Generalised linear mixed models (GLMMs) were used to analyse the impact of the intervention on EOL care (EOLD) and quality of life (QUALID; last measurement during the 90 days prior to resident death). GLMMs were run without adjustment to the dependent variables, and following an evaluation of model best fit using information criteria, were conducted assuming normal distributions with identity links. Residual plots were visually inspected to ensure there were no substantial deviations from normality or homoscedasticity. GLMMs allow for the inclusion of fixed and random effects in the model. These models account for nested sources of variability in data, such as when residents with different characteristics are clustered within nursing homes. In these models, the nursing home was included as a random effect, accounting for intra cluster-correlations in the sample and consequently producing better fixed-effect estimates. While some nursing home level variables were included in the models, the inclusion of a random effect allows for other unmeasured nursing home level effects. In all models, there were resident-level control variables (comorbidities: presence or absence of a disabling condition, an inter-current acute condition, or a life-threatening condition) and nursing home-level control variables (proportion of residents with dementia; baseline ratio of nursing staff hours to bed number; and median staff qPAD Knowledge Test and Attitude Scale scores). These models were rerun in a series of Per Protocol (PP) analyses, excluding nursing homes from the FCC arm that did not implement the intervention to any degree planned (i.e. had a score of ‘0’ on all four nursing home dose variables), and including the resident-level dose variable of whether or not a case conference was received.

Statistical analyses were conducted using IBM SPSS Statistics version 22 (IBM Corp., Armonk, NY). Alpha, the significance level threshold, was set at 0.05 for all statistical tests.

Results

Forty-nine nursing homes were approached to recruit the 20 needed (Fig 1). Reasons for declining to participate included a lack of interest in research, other research projects or confidence in their case conferencing and palliative care programs. Due to difficulties recruiting facilities, three facilities were included who had less than 100 beds (two UC with 46 and 88 beds; one FCC with 75 beds). Three hundred and nineteen residents were assessed for eligibility, with 25 not eligible. Informed consent was obtained for 294 residents, and four died prior to the intervention period commencing and four withdrew. The baseline sample comprised 130 residents (UC facilities) and 156 residents (FCC facilities). During the study, 131 (46%) participants died, 64 in UC and 67 in FCC. Median time to death was 7 months (inter-quartile range [IQR] 9), with no difference between the UC and FCC arms (P = 0.27).

thumbnail
Fig 1. Study flowchart for IDEAL study residents included in EOLD analysis (N = 131).

† Original nursing home eligibility criteria were amended, allowing three facilities with bed counts <100 to be included.

https://doi.org/10.1371/journal.pone.0181020.g001

Baseline characteristics of the 20 nursing homes and 131 people who died during the study are in Table 1. Participants in FCC and UC arms had similar baseline characteristics, though FCC participants were more likely to be born in Australia and were visited less frequently. FCC and UC nursing homes were similar in proportion of people with dementia during the previous 3 months and nursing staff hours to bed ratios. However, nurses and care assistants in the FCC arm had significantly higher baseline qPAD knowledge scores, indicating a greater knowledge of palliative care for advanced dementia.

thumbnail
Table 1. Baseline characteristics of nursing homes (N = 20) and participants who died (N = 131).

https://doi.org/10.1371/journal.pone.0181020.t001

End of life outcomes

Family-reported EOL outcomes were available for approximately three-quarters of the participants who died (76%, 99/131), with no systematic differences in demographic characteristics for people with or without a family EOLD report. Family-rated EOLD measures were completed at a median 7.3 weeks (IQR 5.7; range 2.7 to 24 weeks) and nurse reports at a median 2.4 weeks (IQR 4.0; range 0 to 22 weeks); Table 2 outlines summary statistics. There was no significant group effect on the EOLD scales in any of the models (i.e. family-rated CAD-EOLD, SM-EOLD and SWC-EOLD or nurse-rated CAD-EOLD and SM-EOLD).

thumbnail
Table 2. EOLD scores of nursing homes participants who died (N = 131).

https://doi.org/10.1371/journal.pone.0181020.t002

ICCs for the CAD-EOLD and SWC-EOLD were less than assumed (0.008 and negative respectively) whereas for SM-EOLD the ICC was greater than assumed (0.124). Staff-reported EOLD were missing for four people. There was no significant group effect on the EOLD scales in any of the models. In the ITT models, presence of inter-current acute comorbidity was associated with lower family rated satisfaction with care (SWC-EOLD) and poorer staff rated comfort level (CAD-EOLD) (see Table 3).

thumbnail
Table 3. Generalised linear mixed model predictors of quality of end of life care in dementia (EOLD): Intention to treat models.

https://doi.org/10.1371/journal.pone.0181020.t003

PP analyses indicated positive associations between family-rated EOLD satisfaction with care and better staff attitudes, increased ratios of nursing staff hours to beds, and lower proportions of people with dementia in the nursing home, in addition to absence of inter-current acute comorbidities (see Table 4). Symptom management was rated as poorer by staff for those participants who had a case conference, regardless of treatment arm. In the FCC arm, 69% (n = 46/67) of residents who died received at least one case conference compared with 44% in the UC arm (n = 28/64) (P = 0.004). Staff ratings of CAD-EOLD were higher in nursing homes with higher proportions of people with dementia. CAD-EOLD was also rated as lower in people with a comorbid disability or with an inter-current acute condition.

thumbnail
Table 4. Generalised linear mixed model predictors of quality of end of life care in dementia (EOLD): Per protocol models with resident level dose.

https://doi.org/10.1371/journal.pone.0181020.t004

Ratings of the person’s quality of life (QUALID) during the 90 days prior to death were significantly associated with staff qPAD knowledge in the ITT model, with better quality of life being associated with nursing homes with higher staff knowledge scores (coefficient -1.04, P = 0.036). This effect was not observed in the PP model.

Symptoms and care during the last month of life

The prevalence of symptoms observed to be documented in nursing progress notes during the month prior to death differed between the arms, with the FCC arm significantly more likely than UC to have documentation of pain/discomfort (P = 0.04), restlessness (P = 0.02), constipation (P = 0.002) skin tears (P = 0.005) and other symptoms (including being resistive to care, agitation, pressure areas, wounds, respiratory infection and chest rattles) (P < 0.001). Symptoms not observed more commonly in one arm versus the other included difficulty swallowing/eating/drinking, drowsiness, breathlessness, coughing, choking/gurgling, vomiting, fear or anxiety, diarrhoea and depression (P > 0.05).

Equal numbers (47%) in FCC (31/66) and UC (n = 29/62) had formal pain assessments, but more documentation of pain assessment frequency was seen in the FCC arm (97% (30/31)) than UC (79% (23/29)). Pain assessments were daily or more often in 40% of the FCC participants (12/30) compared with 17% (n = 4/23) of those in UC (P = 0.08).

The majority of participants died in the nursing home in both arms. Pharmacological and non-pharmacological changes to management, and physician input during the month prior to death were significantly more common in the FCC arm (see Table 5). Medication initiations in FCC were more frequently symptom-oriented (e.g. analgesics, laxatives) than diagnosis-oriented (e.g. antibiotics, anti-epileptics, anxiolytics, steroids).

Discussion

This RCT evaluated EOL outcomes of facilitated family case conferencing in advanced dementia from perspectives of EOL care received (satisfaction) and resident outcomes (comfort, quality of life and quality of dying) [39]. While the study recruited sufficient participants overall, a lower than estimated mortality rate meant the participants with primary outcome data (after death) did not meet target sample size. Although no significant intervention effects were observed on EOL outcomes or QOL, management was more consistent with a palliative approach in FCC.

Higher rates of nurse-documented pain and restlessness, and more frequent pain assessment, which are less readily observed and under-reported symptoms[14], suggest increased awareness of symptoms as a result of FCC. Staff in both arms of the study tended to perceive symptom management as poorer for residents who had a case conference than those who didn’t, suggests case conferences foster proactive identification of symptoms.

Resident and nursing home variables are also informative in understanding factors associated with better EOL care. Inter-current acute comorbidities associated with lower family rating of satisfaction with care and lower nurse-ratings of comfort, highlights the need to plan care not only for expected gradual decline and the last hours to days of life, but also for anticipated acute illness, comorbidities and “crisis events” (e.g. aspiration pneumonia). Case conferences have also been used in nursing homes for other decision-making (e.g. behavioral and psychological symptoms of dementia) [40], and integration of these models is important. Quality of EOL care was associated with proportion of people with dementia, higher nursing and direct care staff hours and attitudes towards advanced dementia, consistent with previous evidence that being cared for in a dementia unit reduces likelihood of hospitalisation [41]. However, it is less clear why these relationships were observed only in the PP analyses or why staff knowledge about advanced dementia was not predictive of EOL care given its association with QOL in the last 90 days of life.

Our study has a number of limitations. The inclusion criteria aimed to capture a group of residents with advanced dementia with a high mortality rate. However, even with a participant population where nearly 40% were totally or completely bedfast (compared to 9% in the largest-scale study to date, the CASCADE [Choices, Attitudes, and Strategies for Care of Advanced Dementia at the End-of-Life] Study [38]) and an 18 -month study period (compared with 12 months in the CASCADE Study), the mortality rate was only 46% (compared with 41% in the CASCADE Study). The local Australian data we further relied on to estimate mortality identified a 50% mortality rate in high level care residents with all-stage dementia, but was not able to distinguish mortality due to dementia versus other comorbid illnesses or limit to those with advanced stage dementia only. We recommend that future studies should use more conservative estimates utilizing published mortality rates. We followed previous trials of case conferencing [42, 43] in taking a pragmatic approach aimed at improving external validity by reflecting the diversity of current practice [44, 45]. While staff, residents and families at each nursing home were blinded to the evaluative aim of the study, those in nursing homes in the intervention arm were aware of the introduction of the PCPC role and changes in case conferencing and staff education and so may have been more inclined to report favourably on the quality of palliative care offered as a result. Whilst we introduced measures to limit sampling bias, it seems likely that participating nursing homes were more proactive in palliative care and quality improvement than the sector average. With the exception of one US [46] and one Dutch [47] study, mean scores for family and nurse ratings on the SWC-EOLD, SM-EOLD and CAD-EOLD across both arms were either comparable with, or higher than, those reported in the international literature [4854], and hospitalization and potentially non-palliative interventions use were low [12, 43]. It is also possible that we did not control for important confounder, mediator or suppressor variables amongst the myriad resident, family and nursing home factors that may have been influential [55]. Whilst the EOLD Scales have demonstrated responsiveness in previous longitudinal research [56], their focus on end of life care may have lacked sensitivity to complex interactions with nursing home, staff, and family characteristics; missed benefits earlier in the person’s care, or failed to detect the surrogate decision-maker’s perceived support and engagement in decisions and care choices. Also, timing of EOLD scale administration occurred substantially later than the optimal window in many cases, risking a recall bias of unknown direction and similar across both arms. Measurement of family experience, such as decisional conflict and bereavement, may also have been informative. Finally, dose measurement failed to incorporate case conference details, PCPC motivation and adherence to care plans, all of which may have affected EOL care. UC nursing homes can use more informal means other than case conferencing to share decisions with families (i.e. one-to-one conversations between nursing staff and families); such interactions where not captured in the dose measure but may have positively influenced EOL care and family satisfaction. It should also be noted that diagnosis-oriented medications are sometimes used with intent of symptom alleviation (e.g. corticosteroids for breathlessness from respiratory conditions) which may have led to underestimation of palliative pharmacological strategies.

Conclusion

This study is one of the few RCTs of palliative care interventions in nursing homes worldwide, and appears to be the first to test efficacy of facilitated family case conferencing for people with advanced dementia. The study’s primary endpoint of quality of EOL care was under-powered and did not show evidence of effect. In spite of these limitations, a systematic approach to facilitating a palliative approach and skills enhancement drove improvements in care. Given the growing burden of dementia globally, these data will be formative in interventions aimed to improving palliative care in nursing homes in the future.

Supporting information

S1 Table. CONSORT (Consolidated Standards of Reporting Trials) checklist.

https://doi.org/10.1371/journal.pone.0181020.s002

(DOCX)

Acknowledgments

We thank the people with dementia and their families, the residential aged care facilities and their staff who participated in this study.

References

  1. 1. van der Steen JT, Radbruch L, Hertogh CM, de Boer ME, Hughes JC, Larkin P, et al. White paper defining optimal palliative care in older people with dementia: a Delphi study and recommendations from the European Association for Palliative Care. Palliat Med. 2014;28(3):197–209. http://dx.doi.org/10.1177/0269216313493685. pmid:23828874.
  2. 2. Mitchell SL, Teno JM, Kiely DK, Shaffer ML, Jones RN, Prigerson HG, et al. The clinical course of advanced dementia. New England Journal of Medicine. 2009;361(16):1529–38. pmid:19828530.
  3. 3. Bosek MS, Lowry E, Lindeman DA, Burck JR, Gwyther LP. Promoting a good death for persons with dementia in nursing facilities: family caregivers' perspectives. JONA'S healthcare law, ethics and regulation. 2003;5 (2):34–41. pmid:12789031.
  4. 4. Mitchell SL. CLINICAL PRACTICE. Advanced Dementia. N Engl J Med. 2015;372(26):2533–40. Epub 2015/06/25. pmid:26107053; PubMed Central PMCID: PMCPMC4539157.
  5. 5. Teno JM. Now is the time to embrace nursing homes as a place of care for dying persons. J Palliat Med. 2003;6(2):293–6. pmid:12854949.
  6. 6. Birch D, Draper J. A critical literature review exploring the challenges of delivering effective palliative care to older people with dementia. J Clin Nurs. 2008;17(9):1144–63. pmid:18416791.
  7. 7. Chang E, Daly J, Johnson A, Harrison K, Easterbrook S, Bidewell J, et al. Challenges for Professional Care of Advanced Dementia. Int J Nurs Pract. 2009;15:41–7. pmid:19187168
  8. 8. Di Giulio P, Toscani F, Villani D, Bruneeli C, Gentile S, Spadin P. Dying with Advanced Dementia in Long-Term Care Geriatric Institutions: A Retrospective Study. J Palliat Med. 2008;11(7):1023–8. pmid:18788965
  9. 9. Engel SE, Kiely DK, Mitchell SL. Satisfaction with End-of-Life Care for Nursing Home Residents with Advanced Dementia. J Am Geriatr Soc. 2006;54(10):1567–72. pmid:17038076
  10. 10. Bayer A. Death with dementia—the need for better care. Age Ageing. 2006;35(2):101–2. pmid:16414963.
  11. 11. Hertogh CMPM. Advanced Care Planning and the Relevance of a Palliative Care Approach in Dementia. Age and Ageing. 2006;35:553–5. pmid:17047005
  12. 12. Mitchell SL, Kiely DK, Hamel MB. Dying with advanced dementia in the nursing home. Arch Intern Med. 2004;164(3):321–6. pmid:14769629.
  13. 13. Rurup ML, Onwuteaka-Philipsen BD, Pasman HRW, Ribbe MW, van der Wal G. Attitudes of physicians, nurses and relatives towards end-of-life decisions concerning nursing home patients with dementia. Patient Educ Couns. 2006;61(3):372–80. pmid:15975757.
  14. 14. McAuliffe L, Nay R, O'Donnell M, Featherstonehaugh D. Pain Assessment in Older People with Dementia: Literature Review. J Adv Nurs. 2009;65(1):2–10. pmid:19016920
  15. 15. Sampson EL, Gould V, Lee D, Blanchard MR. Differences in care received by patients with and without dementia who died during acute hospital admission: a retrospective case note study. Age Ageing. 2006;35(2):187–9. pmid:16407434.
  16. 16. Givens JL, Jones RN, Shaffer ML, Kiely DK, Mitchell SL, Givens JL, et al. Survival and comfort after treatment of pneumonia in advanced dementia. Arch Intern Med. 2010;170(13):1102–7. pmid:20625013.
  17. 17. Meier DE, Ahronheim JC, Morris J, Baskin-Lyons S, Morrison RS. High short-term mortality in hospitalized patients with advanced dementia: lack of benefit of tube feeding. Arch Intern Med. 2001;161(4):594–9. pmid:11252121.
  18. 18. Sorrell JM. Use of feeding tubes in patients with advanced dementia: Are we doing harm? Journal of Psychosocial Nursing and Mental Health Services. 2010;48 (5):15–8. http://dx.doi.org/10.3928/02793695-20100331-02. pmid:20415291.
  19. 19. Reuther S, Dichter MN, Buscher I, Vollmar HC, Holle D, Bartholomeyczik S, et al. Case conferences as interventions dealing with the challenging behavior of people with dementia in nursing homes: a systematic review. Int Psychogeriatr. 2012;24(12):1891–903. http://dx.doi.org/10.1017/S1041610212001342. pmid:22883019.
  20. 20. Phillips JL, West PA, Davidson PM, Agar M. Does case conferencing for people with advanced dementia living in nursing homes improve care outcomes: evidence from an integrative review. Int J Nurs Stud. 2013;50(8):1122–35. http://dx.doi.org/10.1016/j.ijnurstu.2012.11.001. pmid:23200128.
  21. 21. Mitchell GK, Del Mar CB, O'Rourke PK, Clavarino AM. Do case conferences between general practitioners and specialist palliative care services improve quality of life? A randomised controlled trial (ISRCTN 52269003). Palliat Med. 2008;22(8):904–12. pmid:18772208.
  22. 22. Abernethy AP, Currow DC, Shelby-James T, Rowett D, May F, Samsa GP, et al. Delivery strategies to optimize resource utilization and performance status for patients with advanced life-limiting illness: results from the "palliative care trial" [ISRCTN 81117481]. J Pain Symptom Manage. 2013;45(3):488–505. Epub 2012/10/30. pmid:23102711.
  23. 23. Abernethy AP. Individualised multi-disciplinary case conferences: can they be practically added to specialist palliative care? 8th Australian Palliative Care Conference; Sydney2007.
  24. 24. Mitchell G, Cherry M, Kennedy R, Weeden K, Burridge L, Clavarino A, et al. General practitioner, specialist providers case conferences in palliative care—lessons learned from 56 case conferences. Aust Fam Physician. 2005;34(5):389–92. pmid:15887946.
  25. 25. Mitchell GK, De Jong IC, Del Mar CB, Clavarino AM, Kennedy R. General practitioner attitudes to case conferences: how can we increase participation and effectiveness? Medical Journal of Australia. 2002;177(2):95–7. pmid:12098350.
  26. 26. Agar M, Beattie E, Luckett T, Phillips J, Luscombe G, Goodall S, et al. Pragmatic cluster randomised controlled trial of facilitated family case conferencing compared with usual care for improving end of life care and outcomes in nursing home residents with advanced dementia and their families: the IDEAL study protocol. BMC Palliative Care. 2015;14(1):63. http://dx.doi.org/10.1186/s12904-015-0061-8. pmid:26589957; PubMed Central PMCID: PMCPMC4654825.
  27. 27. Australian Government Department of Social Services. myagedcare: Find a service 2012 [cited 2013 12th November]. Available from: http://www.myagedcare.gov.au/service-finders#block-finder-aged-care-home-finder-agedcarehome.
  28. 28. Higginson IJ, Evans CJ, Grande G, Preston N, Morgan M, McCrone P, et al. Evaluating complex interventions in end of life care: the MORECare statement on good practice generated by a synthesis of transparent expert consultations and systematic reviews. BMC Med. 2013;11:111. Epub 2013/04/27. pmid:23618406; PubMed Central PMCID: PMCPMC3635872.
  29. 29. Reisberg B. Functional assessment staging (FAST). Psychopharmacol Bull. 1988;24(4):653–9. pmid:3249767.
  30. 30. Abernethy A, Shelby-James T, Fazekas B, Woods D, Currow D. The Australia-modified Karnofsky Performance Status (AKPS) scale: a revised scale for contemporary palliative care clinical practice. BMC Palliative Care. 2005;4(7):http://www.biomedcentral.com/content/pdf/1472-684X-4-7.pdf.
  31. 31. Coventry PA, Grande GE, Richards DA, Todd CJ. Prediction of appropriate timing of palliative care for older adults with non-malignant life-threatening disease: a systematic review. Age Ageing. 2005;34(3):218–27. pmid:15863407.
  32. 32. Clay-Williams R, Nosrati H, Cunningham FC, Hillman K, Braithwaite J. Do large-scale hospital- and system-wide interventions improve patient outcomes: a systematic review. BMC Health Serv Res. 2014;14:369. Epub 2014/09/05. pmid:25187292; PubMed Central PMCID: PMCPMC4282191.
  33. 33. Ersek M, Kraybukk B, Hansen N. Evaluation of a train-the-trainer program to enhance hospice and palliative care in nursing homes. J Hospice Palliat Nurs. 2006;8:42–9.
  34. 34. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M, et al. Developing and evaluating complex interventions: the new Medical Research Council guidance. Bmj. 2008;337:a1655. Also available open-source online at www.mrc.ac.uk/complexinterventionsguidance. http://dx.doi.org/10.1136/bmj.a1655. pmid:18824488; PubMed Central PMCID: PMCSource: NLM. PMC2769032.
  35. 35. Volicer L, Hurley AC, Blasi ZV. Scales for evaluation of End-of-Life Care in Dementia. Alzheimer Dis Assoc Disord. 2001;15(4):194–200. pmid:11723370.
  36. 36. Weiner MF, Martin-Cook K, Svetlik DA, Saine K, Foster B, Fontaine CS. The Quality of Life in Late-stage Dementia (QUALID) Scale. J Am Med Dir Assoc. 2000;1:114–6. pmid:12818023
  37. 37. Long CO, Sowell EJ, Hess RK, Alonzo TR. Development of the questionnaire on palliative care for advanced dementia (qPAD). Am J Alzheimers Dis Other Demen. 2012;27(7):537–43. http://dx.doi.org/10.1177/1533317512459793. pmid:23002199.
  38. 38. Mitchell SL, Miller SC, Teno JM, Davis RB, Shaffer ML. The Advanced Dementia Prognostic Tool: A risk score to estimate survival in nursing home residents with advanced dementia. Journal of Pain and Symptom Management. 2010;40(5): pmid:20621437
  39. 39. Stewart AL, Teno J, Patrick DL, Lynn J. The concept of quality of life of dying persons in the context of health care. J Pain Symptom Manage. 1999;17(2):93–108. Epub 1999/03/09. pmid:10069149.
  40. 40. Reuther S, Holle D, Buscher I, Dortmann O, Muller R, Bartholomeyczik S, et al. Effect evaluation of two types of dementia-specific case conferences in German nursing homes (FallDem) using a stepped-wedge design: study protocol for a randomized controlled trial. Trials. 2014;15:319. pmid:25118091.
  41. 41. Mitchell SL, Teno JM, Intrator O, Feng Z, Mor V. Decisions to forgo hospitalization in advanced dementia: a nationwide study. J Am Geriatr Soc. 2007;55(3):432–8. pmid:17341248.
  42. 42. Abernethy AP, Currow DC, Hunt R, Williams H, Roder-Allen G, Rowett D, et al. A pragmatic 2 x 2 x 2 factorial cluster randomized controlled trial of educational outreach visiting and case conferencing in palliative care-methodology of the Palliative Care Trial [ISRCTN 81117481]. Contemp Clin Trials. 2006;27(1):83–100. pmid:16290094.
  43. 43. Gozalo P, Teno JM, Mitchell SL, Skinner J, Bynum J, Tyler D, et al. End-of-life transitions among nursing home residents with cognitive issues. N Engl J Med. 2011;365(13):1212–21. Epub 2011/10/14. pmid:21991894; PubMed Central PMCID: PMCPMC3236369.
  44. 44. Roland M, Torgerson DJ. What are pragmatic trials? Bmj. 1998;316(7127):285. Epub 1998/02/24. pmid:9472515; PubMed Central PMCID: PMCPMC2665488.
  45. 45. Lurie JD, Morgan TS. Pros and cons of pragmatic clinical trials. Journal of comparative effectiveness research. 2013;2(1):53–8. Epub 2013/11/19. pmid:24236521.
  46. 46. Zimmerman S, Cohen L, van der Steen JT, Reed D, van Soest-Poortvliet MC, Hanson LC, et al. Measuring end-of-life care and outcomes in residential care/assisted living and nursing homes. J Pain Symptom Manage. 2015;49(4):666–79. Epub 2014/09/11. pmid:25205231.
  47. 47. van Soest-Poortvliet MC, van der Steen JT, de Vet HC, Hertogh CM, Deliens L, Onwuteaka-Philipsen BD. Comfort goal of care and end-of-life outcomes in dementia: A prospective study. Palliat Med. 2015;29(6):538–46. Epub 2015/02/19. pmid:25690602.
  48. 48. Kiely DK, Volicer L, Teno J, Jones RN, Prigerson HG, Mitchell SL. The validity and reliability of scales for the evaluation of end-of-life care in advanced dementia. Alzheimer Dis Assoc Disord. 2006;20(3):176–81. pmid:16917188; PubMed Central PMCID: PMCNIHMS101664 PMC2671933.
  49. 49. van der Steen JT, Gijsberts MJ, Muller MT, Deliens L, Volicer L. Evaluations of end of life with dementia by families in Dutch and U.S. nursing homes. Int Psychogeriatr. 2009;21(2):321–9. pmid:19193253
  50. 50. van der Steen JT, Deliens L, Ribbe MW, Onwuteaka-Philipsen BD. Selection Bias in Family Reports on End of Life with Dementia in Nursing Homes. J Palliat Med. 2012;15(12):1292–6. PMC3509357. pmid:23153076
  51. 51. Kiely DK, Shaffer ML, Mitchell SL. Scales for the Evaluation of End-of-Life Care in Advanced Dementia: Sensitivity to Change. Alzheimer disease and associated disorders. 2012;26(4):358–63. PMC3424399. pmid:22273800
  52. 52. van der Steen JT, Ribbe MW, Deliens L, Gutschow G, Onwuteaka-Philipsen BD. Retrospective and prospective data collection compared in the Dutch End Of Life in Dementia (DEOLD) study. Alzheimer Dis Assoc Disord. 2014;28(1):88–94. Epub 2013/05/02. pmid:23632265.
  53. 53. van Soest-Poortvliet MC, van der Steen JT, Zimmerman S, Cohen LW, Klapwijk MS, Bezemer M, et al. Psychometric properties of instruments to measure the quality of end-of-life care and dying for long-term care residents with dementia. Qual Life Res. 2012;21(4):671–84. Epub 2011/08/05. pmid:21814875; PubMed Central PMCID: PMCPMC3323818.
  54. 54. Abbey J, Douglas C, Courtney M, Parker D, Yates P. Develop, trial and evaluate a model of multidisciplinary case conferencing for residents with end-stage dementia. Report to the Prince Charles Hospital Foundation, 2008.
  55. 55. MacKinnon DP, Krull JL, Lockwood CM. Equivalence of the mediation, confounding and suppression effect. Prev Sci. 2000;1(4):173–81. Epub 2001/08/29. pmid:11523746; PubMed Central PMCID: PMCPMC2819361.
  56. 56. Kiely DK, Shaffer ML, Mitchell SL. Scales for the Evaluation of End-of-Life Care in Advanced Dementia: Sensitivity to Change. Alzheimers Disease and Associated Disorders. 2012;Epub ahead of print.