Pilot study evaluating frailty-focused care for hospitalised patients with chronic obstructive pulmonary disease

Abstract

Background

Frailty is associated with morbidity and mortality among people with chronic obstructive pulmonary disease (COPD), yet there are no established care pathways to manage frail patients with COPD. To address this gap, we developed, implemented and assessed the feasibility of a new frailty-focused care model for patients hospitalised with exacerbations of COPD.

Methods

We conducted a prospective cohort study among hospitalised patients with acute exacerbations of COPD in an academic hospital in Canada over 18 months. We developed and implemented a frailty-focused care model using the degree of frailty to guide personalised assessments, education, and care interventions during and after hospital discharge. We assessed the feasibility of using frailty-focused care in a real-world setting with prespecified targets for recruitment rate, care model completion and collection of patient-reported outcomes including symptom burden, health confidence, health status and self-management scores. Patients were followed up at 3 months after hospitalisation to reassess patient-reported outcomes.

Results

87 patients used the frailty-focused care model during hospitalisation, 58 consented to participate in this study and 45 (78%) completed 3-month postdischarge follow-up. 47% (n=21) were at least mildly frail at baseline, with 35% (n=20) at risk of frailty. Target recruitment of 50 patients was achieved, and all participants completed core elements of frailty-focused care, including frailty assessment, personalised education and discharge planning. Patient-reported outcomes were measured in all participants in hospital, and in 78% (n=45) patients at 3-month follow-up. 23% (n=13) of patients initially reported feeling confident to manage their health condition. This improved to 62% (n=28) at 3-month follow-up.

Conclusions

This study provides a foundation to build innovative care models for frail individuals with COPD and shows it is feasible to use frailty-focused care in a real-world hospital setting. Future work requires strong patient engagement to better align frailty-focused care with patient-centred goals.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Frailty is prevalent in one out of five individuals with chronic obstructive pulmonary disease (COPD) and is strongly associated with poor outcomes including functional decline, poor quality of life, repeat hospitalisation and mortality. The European Respiratory Society identified frailty as a treatable trait in chronic lung disease, yet there are no practice recommendations on how to adapt respiratory care to the needs of individuals with frailty and chronic lung disease. To address this gap in care, we developed a frailty-focused care model for hospitalised patients with COPD and assessed whether such a model could be implemented in a real-world setting.

WHAT THIS STUDY ADDS

• This study demonstrated that it is feasible to implement frailty assessments into routine hospital care for COPD and execute specific lung health assessments, education and discharge planning tailored to the degree of frailty. The study found that almost one out of two patients hospitalised for COPD is already living with frailty, while another one out of three patients is vulnerable to frailty. After receiving a frailty-focused assessment and intervention in hospital, more patients reported feeling confident to manage their health condition.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This study provides a foundation on which to build frailty-focused care for individuals with chronic lung disease. Next steps will include meaningful patient and caregiver engagement to improve the model and ensure alignment with patient goals, while investigating whether frailty-focused care improves long-term patient-reported outcomes.

Background

Chronic obstructive pulmonary disease (COPD) is a progressive lung disease associated with high morbidity and poor quality of life.¹ Individuals with COPD experience non-pulmonary complications including peripheral muscle wasting and psychological comorbidity, which lead to progressive functional decline and frailty over time.²,⁴

Frailty results from deficits across multiple physiological systems, creating an increased vulnerability to stressors and subsequent negative health outcomes.^{5 6} People with COPD are twice as likely to experience frailty compared with those without COPD and are at higher risk of mortality.^{7 8} A recent systematic review reports that 20% of patients with COPD experience frailty and another 56% are deemed prefrail (at high risk of progressing to the frail state).³ Frailty predicts functional decline and rehospitalisation after acute exacerbations of COPD, reduced quality of life, increased symptom burden, longer length of hospital stay and increased mortality compared with non-frail people.⁸,¹² The prevalence and impact of frailty on people with chronic lung disease suggest that frailty is an important risk factor and may be used as a treatable trait in this population, meaning a clinically relevant, measurable and treatable characteristic to improve health outcomes.¹³,¹⁵ Recognising the importance of frailty in hospitalised patients, the National Health Service (NHS) mandated frailty screening for older adults in acute care.¹⁶ However, health providers lack guidance in adapting specific care for frail individuals with chronic lung disease.

We developed a unique frailty-focused COPD care model which streamlines clinical care to patients based on their degree of frailty. The care model begins at the onset of a hospitalisation and integrates with ongoing disease management services and programmes in the community after discharge (online supplemental appendix 1). We conducted this initial pilot study to evaluate whether implementing a complex personalised frailty-focused care model in the hospital setting is feasible, and to determine if relevant patient-reported outcomes could be assessed.

Methods

Study design, setting, and participants

We conducted a pilot prospective cohort study on the respiratory and medicine wards at an academic hospital for 18 months (November 2017 and June 2019). We enrolled consenting hospitalised patients with an admitting diagnosis of COPD exacerbation who used the frailty-focused COPD care model during the study period. A prior diagnosis of COPD was confirmed by prior lung function measurements in the medical chart showing airflow limitation (forced expiratory volume in 1 s/forced vital capacity ratio below 0.7). Participants were followed (in person) at 3 months after hospital discharge.

Patients and the public were not involved in the design, or conduct, or reporting, or dissemination plans of the research.

Frailty-focused care model

We developed a new frailty-focused care model for hospitalised patients with COPD, whereby care interventions were guided by the degree of frailty (online supplemental appendix 1). The model was developed by a collaborative multidisciplinary team of respirologists, geriatricians, nurses, physiotherapists, occupational therapists, social workers, dietitians and health administrators, who conducted an extensive review of regional practices in management of hospitalised patients with COPD. To ensure the appropriate personnel and care processes were operational, we implemented the care model as the standard of care for hospitalised patients with COPD exacerbation on the respiratory hospital ward for a 6-month period prior to study initiation.

Patients first received a frailty assessment on admission by hospital nursing staff. The frailty assessment was based on baseline status 2 weeks before hospitalisation, when the patient was clinically stable. Expected and suggested goals of care were then linked to the degree of frailty, to be reviewed by the team and discussed with the patient. Nursing staff and physicians placed consultation requests to allied health providers (social work, physiotherapy, occupational therapy, dietary services, home care) at the onset of hospitalisation based on the degree of frailty identified. A certified respiratory educator provided a detailed lung health assessment and education session tailored to each degree of frailty. This education involved information about COPD, inhaler device review/teaching, breathing techniques, coughing and positioning exercises to cope with dyspnoea and increased sputum production, assessment for community outreach programmes and pulmonary rehabilitation, assessment for need of community lung health resources (ongoing education, smoking cessation counselling), and review of supplemental oxygen use at home. The respiratory educator made recommendations to the physician team about referrals to outpatient community programmes and services that would be beneficial based on the degree of frailty. The care model was organised into three phases focused on ‘acute rescue’, ‘optimisation’ and ‘safe discharge’ from hospital. Figure 1 describes the care model process map and its roll-out during a hospitalisation.

Figure 1. Implementation of the frailty focused COPD care model on a hospital ward. ER, Emergency Room; CFS, Clinical Frailty Scale; COPD, chronic obstructive pulmonary disease.

Study procedures

Study personnel obtained informed written consent from eligible participants. The admitting nurse assessed and documented the degree of frailty using the validated Clinical Frailty Scale (CFS) and proceeded to follow the model of care (online supplemental appendix 1). The Clinical Frailty Scale was chosen to measure frailty because it is a validated instrument used in multiple healthcare settings internationally, is correlated with important long-term health outcomes and is scored based on information readily available from the patient or their caregiver.¹⁷

Study personnel collected sociodemographic information, comorbidities, smoking history and lung function from consenting patients and their electronic hospital chart. Patients completed questionnaires for dyspnoea (measured by the Modified Medical Research Council Scale, mMRC), and symptom burden (measured by the COPD Assessment Test, CAT), and validated Health Confidence Scores (HCSs), health status scores and chronic disease self-management questionnaire scores based on their status at the time of assessment.¹⁷,²¹ Participants were scheduled to follow-up 3 months after discharge to reassess patient-reported outcomes of dyspnoea, symptom burden, health confidence, health status and self-management scores.

Outcomes

The primary outcome measure was feasibility of implementation, defined a-priori by the recruitment of 50 patients to use the frailty-focused COPD care model over 18 months, with an 85% completion rate. We chose a sample size of convenience given the pilot nature of this study. Completion of the care model was defined as having a documented frailty assessment using the Clinical Frailty Scale and completion of a lung health assessment and education session with a certified respiratory educator during hospitalisation.

Secondary outcomes included a prespecified target of 80% completion of patient-reported outcome data collected during the index hospitalisation and 3 months postdischarge. We tested whether it was feasible to collect patient-reported outcome data in anticipated need for future studies of efficacy of the model against a control group.

Dyspnoea as measured by the mMRC score is a validated 5-point scale (0–4 points) with high inter-rater reliability and used to quantify functional disability due to shortness of breath among patients with respiratory diseases.¹⁸ A score of 0 indicates ‘dyspnoea only with strenuous exercise’ while the highest score of four indicates ‘too dyspnoeic to leave house or breathless when dressing’. A minimal clinically important difference (MCID) in the dyspnoea scale is 0.5–1 point.²² Symptom burden measured by the CAT score is a validated 40-point, 8-item questionnaire for patients with COPD.²³ Higher symptom burden scores indicate a large impact on the patients’ life and the MCID is2 points.²⁴

Patient-reported health confidence and health status were measured by validated questionnaires developed by R-Outcomes (online supplemental appendix 2).^{20 21} Patient-reported health confidence was measured using the HCS, a validated measure of a patients’ engagement in their own health.²⁰ The score was measured using a four-item survey, with each item allocated a 0–3 score; 3 (strongly agree), 0 (disagree).²⁰ Health status was measured using the validated howRu measure, also a four-item scaled survey with patients rating each item from 3 (none) to 0 (extreme), based on their experience 24 hours prior.²¹ The health confidence and health status scores were calculated as the sum of item scores, with transformation to a scale between 0 and 100. Higher scores represented greater health confidence and health status.²¹

Patient-reported self-management of chronic disease was assessed with the validated 6-item Stanford Self-Efficacy for Managing Chronic Disease Scale (Self-Management).¹⁹ This tool assessed patients’ confidence in managing their chronic condition.¹⁹ Patients rated each item from 1 (not at all confident) to 10 (totally confident), indicating the value that corresponds with their level of confidence to perform tasks related to disease management.¹⁹ A score of 7 for each item indicates an increased likelihood of achieving the behaviour.^{25 26} The overall score is the average score from all six items, with higher scores denoting greater confidence and ability for self-management.²⁶

Analysis

We used descriptive statistics (means using SD), proportions and median (IQR) to summarise demographics, frailty scores and baseline characteristics. We compared differences in patient-reported outcome measures between hospitalisation and 3 months postdischarge. Differences in categorical data were assessed by χ² test, while differences in continuous variables were assessed using the Kruskal-Wallis and analysis of variance testing.

Results

87 hospitalised patients for COPD exacerbation used the frailty-focused care model between November 2017 and June 2019. Of these, 58 (67%) patients agreed to participate in this pilot study. 45 of 58 patients (78%) completed the 3-month follow-up study visit and 13/58 (22%) patients were lost to follow-up. Reasons for loss to follow-up included death (five patients) and inability to contact after discharge (eight patients). Figure 2 describes recruitment of study participants.

Figure 2. Patient recruitment flow diagram. COPD, chronic obstructive pulmonary disease.

Baseline demographics and characteristics of the study cohort are described in table 1.

Table 1. Baseline demographics and characteristics among patients hospitalised for AECOPD, n=58 patients.

Variable	n=58
Age, median (IQR)	70 (63–75)
Female sex, n, %	36 (62)
Place of residence prior to hospitalisation, n, %
Home	40 (69)
Home with home care	13 (22)
Retirement home	4 (7)
Long term care facility	1 (2)
Comorbidities, n, %
Active malignancy	2 (3)
Diabetes	6 (10)
Heart disease	25 (43)
BMI (median, IQR)	25 (21–29)
BMI ≤21 kg/m2, n, %	16 (28)
Patient resuscitation preferences, n, %*
1 (full resuscitation, including cardiopulmonary resuscitation, intubation, intensive care unit admission)	29 (50)
2 (intensive care unit admission, intubation, but no cardiopulmonary resuscitation)	12 (20)
3 (no cardiopulmonary resuscitation, no intubation, no intensive care unit admission)	17 (29)
FEV1† (median, IQR)
FEV1 (litres)	0.86 (0.71–1.37)
FEV1 (% predicted)	40 (26–50)
6 min walk test distance‡ (median, IQR, metres)	261 (234–356)
Supplemental home oxygen use	20 (35)
Median (L/min)	3.0 (2.0–4.0)
Prior vaccination status, n,%
Pneumococcal vaccination	46 (79)
Influenza vaccination	49 (85)
Pulmonary rehabilitation completed, n, %	25 (43)
Patient sees a lung specialist, n, %	49 (85)
Prior intensive care unit admission (for COPD exacerbation), n, %	8 (14)
Prior Intubation, n, %	7 (12)
Current smoker (Tobacco), n, %	16 (28)
Pack/year history (median, IQR)	44 (36–60)
Frailty Score (Clinical Frailty Scale), n, %
Well or managing well	11 (19)
Vulnerable	20 (35)
Mildly Frail	11 (19)
Moderately or Severely Frail	16 (28)
Baseline Dyspnoea by mMRC score§ (median, IQR)	4 (3–34)
Baseline Dyspnoea by mMRC score ≥2, n, %	55 (95)
Baseline Symptom Burden by CAT score¶ (median, IQR)	27 (22–31)

^*Resuscitation wishes are categorised by this scheme at The Ottawa Hospital at the time of hospital admission for each patient.

^†FEV1 measures the volume of air expelled by the patient in 1 s during a forced exhalation manoeuvre.

^‡6MWT measures how far an individual can walk in 6 min, compared with a predicted distance.

^§mMRC was collected in 57 patients (1 patient did not complete questionnaires). Baseline refers to the measurements taken at the time of hospitalisation.

^¶CAT scores were collected in 57 patients (1 patient did not complete questionnaires).

AECOPD, acute exacerbations of COPD; BMI, body mass index; CAT, COPD Assessment Test; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 s; mMRC, Modified Medical Research Council; 6MWT, 6 min walk test.

62% of patients were female (36/58; 62%) and 53/58 (91%) patients were residing at home independently or with home care services prior to hospitalisation. 34% of patients were vulnerable (20/58, 34%), 11/58 (19%) were mildly frail, while 16/58 (28%) were moderately or severely frail at the time of hospitalisation. Most patients (55/58; 95%) reported that dyspnoea impacted their daily activity (mMRC score), and the median symptom burden score (CAT) was 27 (IQR 22–31), indicating a high burden of symptoms.

Feasibility outcomes

Feasibility of recruitment was achieved, having recruited 58 participants (defined by the ability to recruit 50 patients during the study period). The feasibility of care model completion was also achieved (defined as 85% of participants completed core elements of the care model), as all 58 patients received a frailty assessment using the Clinical Frailty Scale and completed the in-hospital lung health assessment with focused education.

Trends in patient-reported outcomes

Table 2 summarises the patient-reported outcomes measured at the time of hospitalisation and at 3 months after discharge.

Table 2. Description of patient-reported outcome measures during hospitalisation and at 3 months after discharge.

Variable	Baseline n=57*	3-month follow-up n=45	P value
Dyspnoea by mMRC score (median, IQR)	4 (3–4)	3 (2–3)	<0.001
Symptom Burden by CAT score (median, IQR)	27 (22–31)	21 (12–24)	<0.001
Health Confidence by HCS score† (mean, SD)	64.5±17.2	79.1±15.9	<0.001
Health Status by howRu score‡ (mean, SD)	43.1±19.7	67.8±20.8	<0.001
Self-Management Score§ (n≥7 points, %)
How confident do you feel that you can keep the fatigue caused by your disease from interfering with the things you want to do?	11 (19%)	19 (42%)	0.012
How confident do you feel that you can keep the physical discomfort or pain of your disease from interfering with the things you want to do?	13 (23%)	19 (42%)	0.036
How confident do you feel that you can keep the emotional distress caused by your disease from interfering with the things you want to do?	20 (35%)	28 (62%)	0.006
How confident do you feel that you can keep any other symptoms or health problems you have from interfering with the things you want to do?	14 (25%)	20 (44%)	0.099
How confident do you feel that you can do the different tasks and activities needed to manage your health condition so as to reduce your need to see a doctor?	13 (23%)	28 (62%)	<0.001
How confident do you feel that you can do things other than just taking medication to reduce how much your illness affects your everyday life?	15 (26%)	26 (58%)	0.001

^*One patient out of 58 did not complete the baseline patient-reported outcome questionnaires before discharge and was not included in this table. 13 patients were lost to follow-up, leaving 45/58 patients with 3-month patient-reported outcome data.

^†HCS is a patient-reported outcome measure on patient health and ability to make decisions regarding their health.

^‡howRu health status measure is a patient-reported outcome measure on how they feel and their ability to do thing.

^§Self-management score is measured by the 6-item Stanford Self-Efficacy for Chronic Disease scale. Proportion of patients who scored 7 or greater is presented in the table; a score of 7 or more indicates the behaviour has been met.

CAT, COPD Assessment Test; HCS, Health Confidence Score; mMRC, Modified Medical Research Council.

As 13 patients were lost to follow-up, posthospitalisation outcomes were assessed in 45/58 (78%) patients. This was just above the prespecified 20% loss to follow-up rate.

All patient-reported outcome measures improved from baseline to 3 months after hospital discharge. Dyspnoea (mMRC score) improved from a median score of 4 (IQR 3–4) at baseline to 3 (IQR 2–3) at 3 months postdischarge, while symptom burden (CAT score) improved from a median score of 27 (IQR 22–31) to 21 (IQR 12–24). Changes in both the dyspnoea and symptom burden scores exceeded the MCID threshold values.

The mean HCS improved from 64.5±17.2 at baseline to 79.1±15.9 at 3 months, and mean health status score improved from 43.1±19.7 to 67.8±20.8.

Patient scores for self-management of chronic disease were compared at hospitalisation and 3 months postdischarge. A larger proportion of patients reported improved confidence and self-management posthospitalisation; a greater proportion of patients felt that they could better manage their emotional distress (35.1% to 62.2%), pain (22.8% to 42.2%) and fatigue (19.3% to 42.2%). More patients indicated they knew how to modify their activities to better manage their health and reduce their need to see a doctor (22.8% to 62.2%) and felt confident in managing their condition with non-pharmacological means (26.3% to 57.8%) (table 2).

Discussion

In this pilot prospective cohort study, we demonstrated it is feasible to implement a new and complex frailty-focused care model among hospitalised patients with COPD exacerbation in a real-world hospital setting. Compliance was high in assessing frailty, completing lung health assessments/education tailored to degree of frailty and following care directives from healthcare staff. Collection of short-term patient-reported outcome data was suboptimal due to loss of follow-up at 3 months after hospital discharge (22% lost to follow-up at 3 months). Expected improvements were seen in health confidence, health status and chronic disease self-management scores 3 months after use of frailty-focused care in hospital, although this is likely related to ongoing recovery after hospitalisation and may not be solely due to frailty-focused care. However, the proportion of patients reporting they felt better equipped to manage their condition increased from 23% to 62% after frailty-focused care.

We observed that one out of two patients needing hospitalisation for COPD exacerbation was frail, with 47% of our cohort identified as mildly frail or greater using the Clinical Frailty Scale. This degree of pre-existing frailty is echoed in previous studies and places individuals with COPD at increased risk of adverse health outcomes such as rehospitalisation, ongoing functional decline and mortality.^{7 9 10} However, there are currently no established algorithms to administer frailty-focused care to patients living with COPD. Further, 35% of the cohort was assessed as vulnerable to frailty at the time of hospitalisation. This is an important stage to intervene as patients who are on the cusp of being frail are more likely to benefit from interventions to improve and prevent frailty.¹³ A large prospective cohort study found that after pulmonary rehabilitation, two-thirds of frail patients no longer met the frailty definition.¹³ This argues in favour of using frailty as a treatable trait to facilitate the development of individualised care plans that will re-direct health services and resources to benefit those with chronic lung disease at highest risk of poor outcomes.

Our team observed that the presence of a lung health educator who was dedicated to providing personalised support to patients helped to facilitate referrals to community lung health services after discharge from hospital, based on the degree of frailty identified. Busy clinical teaching units are not always aware of disease-specific community resources and this component of the care model provides a mechanism for improved health confidence and self-management in patients. Further investigation in larger studies with standard-care control groups is required to determine which aspects of the frailty-focused care model facilitated improvements in patient-reported outcomes. Additionally, the patient-reported outcomes used in this study warrant further use in larger studies for patients with chronic lung disease, as health confidence and self-efficacy scores, although valid, have not traditionally been used in lung health populations.

This pilot study of frailty-focused care has several strengths. Our comprehensive, innovative, frailty-focused care intervention was codeveloped with input from a team of interdisciplinary stakeholders and review of regional care practices. The model also included a dedicated lung health educator who enacted important referrals to community-based services according to patients’ degree of frailty. We observed the inclusion of a respiratory educator to be a vital component of this model in providing expert guidance to patients at a time when they are vulnerable to further deteriorations in health.

To our knowledge, this was the first study to assess the feasibility of using the frailty paradigm to guide care for hospitalised patients with COPD exacerbation.

Health systems are recognising the importance of risk assessment for frail patients and have already mandated frailty screening for older adults in acute care (NHSs in the UK).^{16 27} Even where policy to screen for frailty is in place, there is a gap in defining subsequent care steps, as identified by a recent study of UK hospitals.²⁷ Our frailty-focused care model for COPD mitigates this care gap, with clear definitions of care goals and directives based on the degree of frailty identified. Lastly, our frailty-focused care model was implemented in a real-world setting as the standard of care, supporting its generalisability to other large hospital settings.

We observed several opportunities for improvement while conducting this study. While our current care model was developed with a multidisciplinary panel of stakeholders, we did not capture patient or caregiver perspectives in the development of this model. Our frailty-focused model would benefit from meaningful patient engagement to align the model with patient goals and to integrate patient decision-support tools around care preferences (such as advanced care planning). Including datapoints on balance, frequency of falls and chest infections would also inform the impact of frailty-focused care. Further, we recognised that healthcare providers are not universally aware of the concept of frailty outside of geriatric medicine and are not formally educated about how to assess and measure frailty. To respond to this, we developed an educational intervention to improve the accuracy of frailty measurements by healthcare staff.²⁸ This study was conducted during a period of time when paper-charting was still in effect. To ensure efficiency in using this model, it must be built into comprehensive health records systems, which will improve ease of use by health providers, prompt providers to assess frailty, facilitate sharing of the model between health systems, and potentiate aggregate data collection. Following completion of this study, our team engaged hospital leadership to build the CFS measurement tool into a new hospital-wide electronic health record to facilitate future use. Lastly, this pilot study did not assess frailty at 3 months posthospitalisation or patient-reported outcomes against a control group who did not receive frailty-focused care. While health status and self-management score improvements are encouraging, they likely reflect the posthospitalisation recovery period in this study. Future studies with control groups are needed to understand which components of frailty-focused care can improve patient-reported outcomes. Lessons learnt from this study to improve and generalise the model will facilitate the next steps in assessing whether frailty-focused care can improve clinical and patient-reported outcomes.

Conclusions

We demonstrated that it is feasible to implement frailty-focused care for hospitalised patients with COPD exacerbation in a real world hospital setting. More research is needed to incorporate patient and caregiver perspectives to refine this care model, and future studies need to discern whether frailty-focused care can improve long-term patient-reported health outcomes compared with standard care. This study is an important first step towards the implementation of frailty-focused care for patients with chronic lung disease.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.