INTRODUCTION
Hong Kong is one of the cities with the longest life expectancy. The proportion of elderly aged 65 or above is projected to increase from 20.5% in 2021 to 36.0% in 2046.1) Frailty, which comes along with aging, is associated with multiple hospitalizations and tremendous healthcare burden.2,3) According to the 2024–2025 Budget of the Hong Kong Government,4) elderly patients aged 65 or above share 54.5% of the total public medical service cost. The cost for in-patient services (both acute and convalescent) in 2024–2025 is estimated to be HKD 7,140 per day, while the cost per Accident and Emergency Department (AED) attendance is estimated to be HKD 2,080.
Frail elderly often present with complicated medical problems such as recurrent sepsis, feeding problems, pressure injuries, etc.,5-7) and are in need of intensive medical and nursing care. Traditionally the majority of such elderly are managed in residential care homes, yet with the arising concept of aging and dying in place,8) a growing number of such being cared at home has been observed.
Residential care homes have more readily available nursing staff for various kinds of nursing care procedures. In Hong Kong, they are often supported by the regional hospital’s Community Geriatric Assessment Team for chronic disease management, and by visiting doctors who manage patients’ episodic illnesses. On the other hand, a service gap exists for frail older adults residing in community. Often the complexity of their medical problems lays beyond the scope of primary care, yet follow-ups in specialty clinics are relatively infrequent and such patients usually have difficulty traveling to hospitals. Community nursing services can provide task-based nursing procedures, yet unable to manage patients’ complex medical needs. As a result, this gap leads to low threshold for AED attendance, frequent hospitalizations and discharge problems.
The New Territories West Cluster of Hong Kong provides medical services to a population of 1.1 million people with Tuen Mun Hospital being its major regional hospital. In view of such a service gap and that a new service model is deemed necessary to cater the rising demand, the geriatric team of Tuen Mun Hospital has developed the Integrated Care at Home (ICAH) program since 2018, aiming to establish an on-site, need-based, integrated model to facilitate frail elderly care in the community, so as to support aging in place. We also target to reduce AED attendance and hospital stay, empower caregivers, and relieve carer stress.
MATERIALS AND METHODS
Program Design and Services
ICAH program is a geriatrician-led, multidisciplinary, case-manager based program, that works under close collaboration with community nursing services. Cases are recruited through proactive screening. Inclusion criteria include community-dwelling older adults with advanced frailty under the care of Tuen Mun Hospital geriatric team, who are chair- to bedridden and functionally dependent with a Clinical Frailty Scale of at least 7,9) in need of intensive medical and nursing care, with dedicated caregivers, and preferably with 2 or more admissions within the past 6 months, or being hospitalized for more than 1 month.
Under ICAH program, a designated community nurse will be assigned as the case manager for each patient. Regular nursing visits and on-site follow-up by geriatricians are provided. General nursing services include wound care, medication management, support on medical devices, etc., while specialized wound nurse visits could be arranged for complicated wounds. Medical follow-up includes chronic disease management, medication reconciliation and symptom control, etc. Nursing visits could be as frequent as daily if required, otherwise typically once every 2–4 weeks. Medical visits are individualized, but typically once every 2–3 months. Ad-hoc nursing phone consultations are available within office hours, further supported by ad-hoc medical on-site or tele-consultations if needed, which could usually be arranged within 1–3 days. Community allied health services would be referred if required after comprehensive geriatric assessment. More complex interventions could be carried out at home, such as subcutaneous fluids infusion, intravenous medications, and more complex wound management. With adequate support and education, caregivers are empowered to perform procedures like oropharyngeal suction or use of narcotics in palliative care.
Clinical admissions to designated geriatric wards could be arranged if necessary. Admitted cases are co-managed with the hospital team, and early post-discharge support is provided. Collaborated services with the Integrated Discharge Support Program for elderly (IDSP) for transitional care could be provided, which is a pre-existing multidisciplinary post-discharge program but with the lack of direct medical support. Community support would be further tailored by need via liaison with medical-social collaboration services, through referrals to medical social workers and non-government organizations that provide home care support services. Case conferences are held regularly to review the progress of each case. Advanced care planning is proactively discussed with all cases who are ready for discussion, and peri-bereavement and dying in place support are provided to families in need. Recruited cases are followed until patients either pass away or move to residential care homes, or when they no long require intensive medical support.
Study Design and Data Collection
In this retrospective observational study, data of all patients recruited into ICAH program from February 1, 2018 till August 31, 2023 were reviewed, and data up to February 29, 2024 were retrieved. Patients’ demographic information, consultation records and admission records were obtained from the Hospital Authority’s electronic patient records. Patients’ clinical outcomes, including institutionalization, change of feeding mode, mortality, place of death, and any resuscitation received, would be reviewed.
Advanced dementia was defined as Reisberg Functional Assessment Screening Tool (FAST) staging of least 7a.10) Advanced parkinsonism was defined as stage 5 disease by Hoehn & Yahr.11) Advanced heart failure was defined as class IV of the New York Heart Association classification. Severity of pressure injury was classified according to the National Pressure Ulcer Advisory Panel staging system.12) Recurrent sepsis referred to more than 1 episode of infection that required systemic antibiotic use. Regarding complaints leading to ad-hoc consultations or clinical admissions, chest symptoms included cough, shortness of breath or sputum retention. Urinary symptoms included dysuria, hematuria, turbid urine, retention of urine, or urinary catheter-related problems. Skin conditions referred to skin rash, pruritus or cellulitis. Wound conditions referred to new occurrences, worsening or infection of wounds. Gastrointestinal symptoms included vomiting, diarrhea or constipation. Eye conditions referred to eye discharge, redness or pain. Musculoskeletal conditions referred to musculoskeletal pain or swelling. Suboptimal glucose control referred to either hyperglycemia or hypoglycemia. Post-discharge review referred to any proactive or by-request review shortly after a hospitalization episode. This study was approved by the Hospital Authority Central Institutional Review Board (Ref No. CIRB-2024-120-1). Informed consent was waived for this retrospective study.
Outcomes
Primary outcomes were patients’ 180-day AED attendance and hospitalization days before and after program recruitment, which was obtained by dividing the total outcome counts by the total number of surviving observation days, then multiplied by 1,000 patient days (PD) and were compared by incidence rates. The total observation days for AED attendance excluded the count of hospitalization days as patients would not be at risk for AED attendance while being hospitalized. Secondary outcome was the major caregivers’ stress burden as assessed by Relative Stress Scale (RSS) at baseline and at 3-month post recruitment, which is a 15-item questionnaire focusing on different aspects of perceived carer stress.13)
Statistical Analyses
Statistical analyses were performed using SPSS (Window version 22.0; IBM Corp., Armonk, NY, USA). All continuous variables in this study had skewed distribution using the Kolmogorov-Smirnov test and were expressed as medians with interquartile ranges (IQR), while categorical variables were expressed as numbers with percentages (%). Incidence rates were presented with 95% confidence interval (CI) by Poisson distribution, and were compared using chi-square test. Mann-Whitney U test was used to compare non-parametric continuous data. All statistical tests were two-sided and p<0.05 was considered to be statistically significant.
RESULTS
Study Population and Baseline Characteristics
From February 1, 2018 to August 31, 2023, 76 cases were recruited with a median age of 90 and a median Clinical Frailty Scale of 8, which indicates very severe frailty (Table 1). Seventy of the recruited patients (92.1%) had advanced dementia, 39 (51.3%) had a history of stroke, 17 (22.4%) had a history of fracture hip, 33 (43.4%) had recurrent sepsis within the past year, and 23 (30.3%) had deep pressure injuries of stage 3 or above. The median body weight was 41 kg. Thirty patients (39.5%) were on feeding tubes, 9 (11.8%) were on indwelling urinary catheters, and 4 (5.3%) were on long-term oxygen therapy. Twenty-five patients (32.9%) were living in public housing estates, and 19 (25%) were on the Comprehensive Social Security Assistance scheme, which is a welfare program provided by the Social Welfare Department of Hong Kong for local residents who cannot support themselves financially to meet their basic needs. Patients in general had 2 (IQR 1–2) caregivers looking after them. Foreign domestic helpers were involved in the care of 48 patients (63.1%).
Program-Related Service Provision and Clinical Outcomes
As of February 29, 2024, a median of 0.4 on-site medical visits and 3.7 nursing visits were paid to each patient per month; 0.4 ad-hoc nursing phone consultations and 0.1 ad-hoc medical consultations were delivered per patient per month (Table 2). Subcutaneous fluid was administered to one patient. Narcotics were used in six patients for symptom relief. Conservative sharp wound debridement of pressure injuries was performed on four patients. After advanced care planning, 33 (43.4%) patients were signed in the Do Not Attempt Cardiopulmonary Resuscitation (DNACPR) Form for non-hospitalized patients.
A total of 165 ad-hoc medical consultations and 35 clinical admissions were provided, with reasons shown in Tables 3 and 4, respectively. The most common complaints for ad-hoc consultations were chest symptoms (16.4%), skin conditions (14.5%), wound conditions (10.3%), and urinary symptoms (9.7%), etc. The most common reasons for clinical admissions were chest infection (25.7%), decreased general condition and appetite (22.9%), and wound conditions (14.3%), etc. Within the program, two patients were weaned off artificial feeding tubes; one patient switched from oral to tube feeding, and one patient was put on temporary tube feeding during his admission for an infected pressure injury, within which he passed away.
As of February 29, 2024, 66 patients were discharged from the ICAH program, among whom four had moved to residential care homes, two were discharged as they no longer required intensive medical and nursing support, and continued routine follow-up in our specialty clinic. The remaining 60 patients passed away, of which 30 died within 180 days after recruitment. The median survival among those who died was 5.6 months (IQR 2–15.7) after recruitment.
Among the 60 patients who passed away, 31 (51.7%) died in a geriatric ward, while 9 (15%) died in non-geriatric wards; 17 (28.3%) died in AED, and 2 (3.3%) died at home. Dying in place support was provided to 3 patients (3.9%) during their last journey of life. The most common cause of death was pneumonia, which occurred in 54 patients (90%); 38 (63.3%) did not receive any cardiopulmonary resuscitation (CPR) or intubation, while 22 (36.7%) received invasive resuscitation, most of which occurred on ambulance or in AED. Eleven patients with a valid DNACPR Form for non-hospitalized patients received CPR on ambulance, after which nine of them were terminated after arrival to AED.
Study Outcomes
The rates of AED attendance and hospitalization days 180 days pre- and post-program were shown in Table 5. AED attendance rates decreased from 15.3 per 1,000 PD (95% CI, 12.9–17.9) to 3.2 per 1,000 PD (95% CI, 2.1–4.5) (p<0.001). Rates of hospitalization days decreased from 266.4 per 1,000 PD (95% CI, 257.8–275.2) to 29.7 per 1,000 PD (95% CI, 26.5–33.3) (p<0.001). The median Relative Stress Scale decreased from a baseline of 24.5 (IQR 14–34) to 16 (IQR 11–28) at 3-month post recruitment (p=0.001).
DISCUSSION
In this retrospective before-and-after study we recruited 76 patients who are amongst the oldest and frailest with intensive medical needs, with an average of one AED visit and 38 hospitalization days over the past 6 months. After recruitment into ICAH program, there was a significant reduction in AED attendance and hospital stay, and their caregivers had a lower RSS score. This is a relatively small-scale study but is also one of the first published data exploring the efficacy of home-based medical care models designated for severely frail older adults in Hong Kong.
In Hong Kong, about 90% of inpatient services are managed by public hospitals14) as public tertiary healthcare is easily accessible and very affordable. Without adequate support in community, frail older patients have a low threshold to visit AED for acute medical problems. Local analysis reported substantial utilization of hospital services by patients in their last year of life in terms of AED attendances, hospital admissions and patient days, with a rise starting in the last 6 months and a surge in the final two months of life.14) The median age of our patients was 90, and their median Clinical Frailty Scale was 8, which refers to patients being “completely dependent on personal care and approaching end of life,”9) and 39.5% patients died within 6 months after recruitment into ICAH. Even with the lack of control group for comparison, the significant decrease in AED attendance and hospital stay after ICAH is still impressive in view of their anticipated need for intensive hospital services while approaching end-of-life and considering the local health-seeking behaviour.
Various international studies have demonstrated the efficacy and cost-effectiveness of integrated home-based medical services, including home-based primary care and palliative care programs, on reducing hospital attendance and medical costs.15,16) A local study also showed that a virtual ward program for post-discharge older patients was able to reduce unplanned emergency hospital readmissions.17) Such programs share similar service provision with ICAH, such as home visits by physicians, telephone consultations, community resource networks, etc., yet they have different healthcare goals and are designed for different patient groups,18) thus comparison with such programs requires careful interpretation.
A home-based palliative care program for older adults in the United States resulted in a reduction of hospitalization days from 6.59 to 1.46 days, and a non-significant reduction of AED visits from 0.23 to 0.11 in 6 months.19) The number of AED visits and hospital days before and after this program were significantly less than those of ours, which could possibly be explained by the differences in patient characteristics and needs. Patients in the United States study were younger than ours, with less dementia while significantly more cancer and organ failure cases. Our patients with advanced frailty share a different end-of-life trajectory which involves a more slowly progressive functional deterioration, compared to a more abrupt decline over the last few months as in malignancy or organ failure.20,21) Besides, unlike general palliative patients whose major symptoms are cancer pain and breathlessness, our study found that patients with advanced frailty usually present with recurrent systemic infections, wound problems and decreased general conditions. Therefore, conventional palliative care models may not fit our group of very frail patients,20) and that a specific geriatric-targeted care model is deemed necessary.
Similarly, even though many home-based primary care programs resulted in reduced hospital attendance, their baseline AED visits and hospital stay are still significantly less than those of this study, as their patients are younger and with better functional status.22,23) The degree of reduction of hospital attendance is also not as obvious as ours. A retrospective cohort in Canada demonstrated a non-significant decrease of AED visit rates from 4.1 to 3.7 visits per 1,000 PD and hospital stay rates from 41.8 to 39.5 days per 1,000 PD after patients received an integrated home-based primary care program, while those who received standard care continued to rise.24) Our program is unique as ICAH provides service beyond the boundary of either primary or palliative care, instead we look after patients throughout community, in-patient, transitional, and even end-of-life phases, where patients are largely managed by the same medical team. This continuity of care undoubtedly contributes to the substantial reduction in both AED visits and hospital stay, to even less than those of a primary care program.
Our study provides valuable insights on how dying in place is a challenge in Hong Kong. A local study revealed that up to 31% patients would opt to die at home,8) yet in reality 90% of deaths in Hong Kong occur in public hospitals.14) This is highly affected by local social and cultural contexts, such as potential social taboo, depreciation of property value and the lack of medical support associated with dying at home.25) Current public healthcare does not cover death certification at home, while private services and post-mortem logistic arrangements can be quite costly. Without a proper Medical Certificate of the Cause of Death,26) families may need to go through police investigations and coroner referral, which may lead to unnecessary stress while going through grief. Therefore many of our caregivers chose to send patients to AED at their very last moments, with 28.3% of our deceased patients being certified in AED despite ICAH support. Even though this longstanding social culture may not be changed within a short time, we were still able to keep patients out of hospital for as long as possible, and to create a platform to facilitate those who opted to die in place.
There was also unfortunately a significant number of patients who had a valid DNACPR Form for non-hospitalized patients, but were still given CPR on ambulance under the constraint of Fire Services Ordinance of Hong Kong, which specifically indicates the duty of Fire Services Department to assist any person who needs immediate medical attention by resuscitating or sustaining one’s life.27) This figure corresponds to that of another local study, within which 87.5% of patients who just died at home in a cancer-predominant palliative program received CPR by emergency rescue personnel despite the presentation of valid advance directives or DNACPR forms.28) This undoubtedly imposes unnecessary suffering to patients with known terminal illnesses. Currently the legislation of the Advance Decision on Life-Sustaining Treatment Bill is in progress,29) and hopefully by passing the Bill and its relevant legislative amendments, patients’ end-of-life journey in the community could be better protected.
The baseline RSS score among our caregivers was 24.5, higher than the cutoff of 23, indicating an increased risk of clinically significant psychological distress.30) This corresponds to previous studies which showed that caregiver burden is particularly high amongst family caregivers of frail older people with multiple comorbidities.31) Some commonly observed source of caregiver stress among our patients includes recurrent or distressing symptoms, reluctance for hospitalization but not being aware of alternative options, and fear of imposing complications or delaying treatments due to the lack of caring skills. Such observations clearly indicate the need and value of this program by providing carer education and empowerment. A systemic review also suggested that education, counselling and psychological support can help relieve stress, depression and role strain of caretakers,32) which could in turn delay institutionalization of frail older adults.33) As a result, we managed to keep most (94.7%) patients in place during the study period with only a minority being relocated to residential care homes. A significantly decreased RSS score also reflected the effectiveness of ICAH program by relieving carers’ stress burden.
One major limitation of our study is that it was retrospective and there was no control group nor randomization, so we could not prove the causal relationship between ICAH program and the reduction in hospital attendance. The sample size was small and it only covered patients in a territorial hospital, so results may not be generalizable to other parts of Hong Kong. In fact, the study population was confined to a highly selected group where there was good family support and dedicated caregivers to start with. The small sample size also limits the possibility of a comprehensive cost-effective analysis. Moreover, RSS has not been validated in Chinese, and RSS focuses on psychological stress but no other aspects such as financial burden or perceived health status. Qualitative assessment of caregiver burden has recently been switched to a validated Chinese version of the Zarit Caregiver Burden Interview,34) but the data collected is not yet adequate for analysis.
ICAH is still an ongoing program and we have been modifying and optimizing our services constantly. At the moment we are trying to utilize more telemedicine to minimize unnecessary travelling time, and implementing more use of gerontechnology to facilitate remote patient care. We hope that by running and expanding the program in a sustainable way we can facilitate aging in place for more frail older adults.
In conclusion, ICAH program demonstrated positive effects in reducing AED visits and hospital stay for very frail older patients while alleviating caregiver burden. A larger scale, prospective study with a control group may be considered in future to further establish the efficacy and cost-effectiveness of the program.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
