Scholarly article on topic 'Delivery of institutional long-term care under two social insurances: Lessons from the Korean experience'

Delivery of institutional long-term care under two social insurances: Lessons from the Korean experience Academic research paper on "Economics and business"

CC BY-NC-ND
0
0
Share paper
Academic journal
Health Policy
OECD Field of science
Keywords
{"Long-term care policy" / "Coordination of care" / "Social insurance" / "Older people"}

Abstract of research paper on Economics and business, author of scientific article — Hongsoo Kim, Yong-Il Jung, Soonman Kwon

Abstract Little is known about health and social care provision for people with long-term care (LTC) needs under multiple insurances. The aim of this study is to compare the profile, case-mix, and service provision to older people at long-term care hospitals (LTCHs) covered by the national health insurance (NHI) with those of older people at long-term care facilities (LTCFs) covered by the public long-term care insurance (LTCI) in Korea. A national LTC survey using common functional measures and a case-mix classification system was conducted with a nationally representative sample of older people at LTCFs and LTCHs in 2013. The majority of older people in both settings were female and frail, with complex chronic diseases. About one fourth were a low-income population with Medical-Aid. The key functional status was similar between the two groups. As for case-mix, more than half of the LTCH population were categorized as having lower medical care needs, while more than one fourth of the LTCF residents had moderate or higher medical care needs. Those with high medical care needs at LTCFs were significantly more likely to be admitted to acute-care hospitals than their counterparts at LTCHs. The current delivery of institutional LTC under the two insurances in Korea is not coordinated well. It is necessary to redefine the roles of LTCHs and strengthen health care in LTCFs. A systems approach is critical to establish person-centered, integrated LTC delivery across different financial sources.

Academic research paper on topic "Delivery of institutional long-term care under two social insurances: Lessons from the Korean experience"

Health Policy xxx (2015) xxx-xxx

Contents lists available at ScienceDirect

Health Policy

journal homepage www.elsevier.com/locate/healthpol

Delivery of institutional long-term care under two social insurances: Lessons from the Korean experience

Hongsoo Kim *, Yong-Il Jung, Soonman Kwon

Graduate School of Public Health and Institute of Health and Environment, Seoul National University, Gwanak-Ro 1, Gwanak-Gu, Seoul 151-742, Republic of Korea

ARTICLE INFO

ABSTRACT

Article history:

Received 3 March 2015

Received in revised form 24 July 2015

Accepted 27 July 2015

Keywords:

Long-term care policy Coordination of care Social insurance Older people

Little is known about health and social care provision for people with long-term care (LTC) needs under multiple insurances. The aim of this study is to compare the profile, case-mix, and service provision to older people at long-term care hospitals (LTCHs) covered by the national health insurance (NHI) with those of older people at long-term care facilities (LTCFs) covered by the public long-term care insurance (LTCI) in Korea. A national LTC survey using common functional measures and a case-mix classification system was conducted with a nationally representative sample of older people at LTCFs and LTCHs in 2013. The majority of older people in both settings were female and frail, with complex chronic diseases. About one fourth were a low-income population with Medical-Aid. The key functional status was similar between the two groups. As for case-mix, more than half of the LTCH population were categorized as having lower medical care needs, while more than one fourth of the LTCF residents had moderate or higher medical care needs. Those with high medical care needs at LTCFs were significantly more likely to be admitted to acute-care hospitals than their counterparts at LTCHs. The current delivery of institutional LTC under the two insurances in Korea is not coordinated well. It is necessary to redefine the roles of LTCHs and strengthen health care in LTCFs. A systems approach is critical to establish person-centered, integrated LTC delivery across different financial sources.

© 2015 Published by Elsevier Ireland Ltd.

1. Introduction

Building sustainable long-term care (LTC) systems and quality LTC provision are a shared health policy agenda in many developed countries experiencing an aging population. Institutional LTC is a key component of the continuum of LTC in most developed countries, and aims to maintain the health and well-being of the frailest older population [1]. Institutional LTC is the most expensive form of LTC, so LTC reforms often target deinstitutionalization and promote community-based LTC, but institutional care is still

* Corresponding author. Tel.: +822 880 2723; fax: +822 762 9105. E-mail address: hk65@snu.ac.kr (H. Kim).

http://dx.doi.org/10.1016lj.healthpol.2015.07.009 0168-8510/© 2015 Published by Elsevier Ireland Ltd.

an essential service for older people with complex medical conditions and severe functional limitations [1,2]. The most common institutional LTC settings are long-term care hospitals (LTCHs) and long-term care facilities (LTCFs), but their roles and the coordination of care across the two settings vary across countries, according to health and LTC delivery models and financial schemes [1,2].

In Korea, which has the most rapidly aging population in the world, the provision of LTCH and LTCF services is financed by two distinct social insurances: LTCH services are covered by the National Health Insurance (NHI), and LTCF services are covered by Long-Term Care Insurance for the Elderly (LTCI) [3,4]. The separation of the two types of service funded by the respective insurances is rooted in a broader health and social care context. First, LTCHs under

H. Kim et al. / Health Policy xxx (2015) xxx-xxx

NHI were introduced to control the provision of acute-care hospital beds. Korea has the second-highest number of acute-care hospital beds among OECD countries; more than 90% of the beds are supplied by private institutions [5,6], and policies to control bed size have not been successful. LTCHs, first introduced in 1993 [7], are an alternative type of hospital for acute-care services. In the early 2000s, small-and medium-size hospitals were oversupplied, many of which had financial struggles due to their lower competitiveness compared with larger hospitals with more skilled medical staff and more advanced medical technology. To address these conditions, the Korean government encouraged less competitive, non-general hospitals to switch from acute care hospitals to LTCHs [8,9], demand for which was expected to increase due to the rapid population aging. Some financial support was provided to hospitals electing to switch; barriers to enter the LTCH market were set low; and the workforce and facility requirements for being certified as a LTCH were less strict than those for being an acute-care hospital [8].

The populations to be served by the LTCHs were broadly defined by medical law as people who mainly needed care for geriatric or chronic diseases, or those in a recovery period after surgery or injury [10]. The number of LTCHs doubled over five years, from 639 in 2008 to 1356 in 2013 [11], and the number of beds increased by about 20% on average per year, from 66,727 in 2007 to 161,054 in 2012 [6]. Before the introduction of LTCI, social admissions in LTCHs were inevitable because there was no public financing available once older people were discharged from LTCHs.

In contrast, LTCF care in Korea targets the beneficiaries of the public LTCI implemented as a separate social insurance scheme in July 2008 [1], partially due to path dependency created by Korea's running the NHI as a social insurance for 30 years. Similar to the NHI, LTCI was operationalized by the National Health Insurance Services (NHIS), the centralized, single insurer of the public LTCI [4,12]; but financially, the two social insurances were designed to be separate, and the benefits under the two insurances were designed not to overlap. Because LTCFs were established under the welfare act for the aged, LTCF service has several unique characteristics [4,12,13]. First, the majority of LTCF residents covered by LTCI are people aged 65 or older who have passed a certain threshold of functional limitation set by the standardized, national care-need certification system. Second, LTCFs are not health care organizations but entities providing social care services; LTCFs primarily offer non-medical care, mainly support for the daily living of older people with functional limitations. Thus, more than 70% of the current workforce in LTC institutions are also personal care assistants, and the nursing staff requirement is only 1 per 25 residents. Nursing staff do not need to be registered nurses (RNs), and no in-house medical staff is mandated. Rather, a community doctor with an LTCF contract is supposed to visit the facility once every two weeks for general check-ups and to update prescriptions, etc. LTCF residents are supposed to visit outpatient clinics or be transferred to hospitals when they have health and medical care needs beyond a general check-up.

There were several advantages to introducing LTCI and LTCFs separately from existing LTCH services under the NHI. First, it could help prevent the medicalization of LTC. Adding LTC services to the existing health care benefits package of the NHI could have resulted in a rapid increase in health care utilization by older people with complex health and social care needs. Politically, introducing a brand-new social insurance including new benefits to support the frail elderly and decrease family burden was likely to be more attractive to the public, who would have to pay more for another mandatory social insurance. In addition, administratively, it would be easier to make and manage a financial account for LTCI separate from the NHI; financial sustainability was at the top of the agenda in designing LTCI.

Seven years has passed since LTCI was introduced in 2008, and it has had several early successes. About 6.1% (n = 378,493) of Korean people aged 65 or older with the most severe functional limitations were the beneficiaries of LTCI at the end of 2013, almost 1.5 times higher than the 4.2% at the end of 2008 [14,15]. About 71.0% of the public is aware of LTCI, and more than 88.6% is willing to use the service, showing high acceptance of the new insurance by the public [16]. The family members of LTCI beneficiaries have reported satisfaction with the services their older relatives have received and the decreased burden of family caregiving [16]. As for the number of institutions, more than 4648 LTCFs provide institutional LTC services that are reimbursed by the public LTCI [14].

In contrast to such success, the health- and care-service delivery across LTCHs and LTCFs under the two insurances is not organized well. LTCH patients and their length of stay (LOS) rapidly increased between 2008 and 2013: the number of patients rose by 78.9% (from 185,464 to 331,919 patients) and the average LOS rose by 28.9% (from 127.8 to 164.7 days) [17]. Social readmissions with a long LOS are still a policy concern, although the government in 2009 implemented a policy to increase copayments for light care from 20% to 40% [18]. A recent study reported one third of older LTCH patients had a low need for medical and nursing treatment, although it used a relatively small and convenient sample [3]. The same study also reported complex conditions and unmet health care needs in older LTCF residents.

Some overlap between LTCH and LTCF services is inevitable and even maybe necessary, but the current mixed roles of LTCHs and LTCFs under the two insurances could make health and LTC systems in Korea inefficient, ineffective, and unsafe. Policy interventions are necessary, but no empirical evidence except Roh et al.'s study [3] exists on the care needs and service use of older people in the two LTC settings. Based on our assessment of the current situation, as described above, we hypothesized that the care needs of people in LTCHs and LTCFs would be alike, but that service utilization would be affected by the type of institution, even in similar case-mix groups. The purpose of this study was to examine the profile of people in LTCFs under LTCI and LTCHs under NHI in Korea; in particular, we compared the key functional status, case-mix, and service utilization of older people in the two settings using psychometrically sound common measures.

H. Kim et al. / Health Policy xxx (2015) xxx-xxx

2. Materials and methods

2.1. Study design and sample

This is a cross-sectional survey study using a nationally representative sample of LTCHs and LTCFs. We did a two-stage stratified random sampling. LTCHs and LTCFs were randomly selected using region and facility size as strata; and in each sample site, a random sample of about 20% of people aged 65 or older was drawn using the patient/resident roster. Excluded were LTCHs and LTCFs that were in operation one year or less, as their operations were likely to be unstable and in transition. We also excluded small LTCFs whose bed size was below 10, because resident characteristics and service provision in such small-sized nursing homes were likely to differ from those of their counterparts in larger nursing homes. As for older people, we excluded those admitted within 1 month, because their conditions might have been unstable, and the staff might not have fully understood or been familiar with the conditions of such new patients/residents. The characteristics of short-stay patients could also be quite different from those of long-stay patients. The final sample included 1351 older patients in 52 LTCHs and 1462 older residents in 91 LTCFs. The sample institutions represented approximately 6.0% and 4.4%, respectively, of the target institutions. There were no significant differences between our sample and the national population of institutions in establishment year, ownership, nursing staffing level, or physician staffing level (LTCHs only; not shown). This study was approved by the institutional review board for human subject research at the institution the authors are affiliated with.

2.2. Instruments and measures

In order to assess the functional status of older people in the two different settings, LTCHs and LTCFs, using a common tool, we adopted the Korean version of interRAI LTCF, a psychometrically sound, comprehensive geriatric assessment tool [19]. The interRAI LTCF was developed by interRAI (www.interrai.org), a non-profit organization of multidisciplinary researchers from more than 30 countries, and is a widely used needs-assessment and care-quality monitoring tool for institutionalized people with complex health and social care needs, such as the elderly and people with disabilities and/or chronic health conditions. A recent OECD health policy report described the application of the interRAI assessment system, a set of 19 comprehensive and integrated health- and care-information systems, in quality monitoring of LTC in many North American, European, and Pacific Rim countries, including Canada, New Zealand, and Iceland; countries in which the interRAI LTCF or MDS 2.0, an earlier version of interRAI LTCF, is amandatory common instrument for service planning and quality monitoring of LTC.

We examined key functional levels - activities of daily living (ADL), cognition, depression, pain, and condition instability - of older people in LTCHs and LTCFs using the core scales in the interRAI LTCF. ADL was measured with the 4-item Activities of Daily Living Hierarchy scale [19,20],

which measures the extent of physical function in performing acts of personal hygiene, locomotion, toilet use, eating, etc. Cognition was measured by the 4-item Cognitive Performance Scale [21], which is based on items related to decision-making, short-term memory, being understood by others, and eating. Depression was measured by the 7-item Depression Rating Scale [19,22], whose range is from 0 to 14; a higher score means more severe depressive symptoms. Pain was measured by the 2-item Pain Scale rating the frequency and intensity of pain [23]; the scale ranged from 0 to 4, and a higher score means more severe pain. Condition instability was measured by the Changes in Health, End-Stage Disease, Signs, and Symptoms (CHESS) scale [24], which evaluates the clinical stability of the conditions of older people using 10 items in the interRAI LTCF, such as vomiting, dehydration, changes in decision-making, and dyspnea; people with a higher CHESS score are more likely to have ER visits, hospitalization, or death [24].

We compared the case-mix of older people in LTCHs and LTCFs using the current case-mix algorithm for LTCHs in Korea. The current case-mix system for LTCHs in Korea was developed in 2008 by the Health Institute Review & Assessment Service (HIRA) [25], based on the RUG-III case-mix classification system of the MDS 2.0, which used to be the mandatory care needs and quality monitoring tool of skilled nursing facilities in the U.S. [26] and is currently a common quality-monitoring tool in both complex care hospitals and nursing homes in Ontario, Canada [26,27]. The Korean LTCH case-mix system classifies patients into seven categories by medical needs, functional abilities, etc. The Ultra High Medical Care group is the highest resource-use group, the one that needs the most intense medical treatment and observations, followed by the High Medical Care, Medium Medical Care, Behavioral Problem, Impaired Cognition, Mild Clinical Care, and Reduced Physical Function groups. Last, data regarding service utilizations (acute-care hospital admissions requiring a stay of more than one night in a hospital during the last 90 days and physician-ordered rehabilitation service use of more than 15 min during last 7 days) based on chart reviews as well as socio-demographic and clinical conditions were also collected.

2.3. Data collection and analyses

Assessments of older people at the two types of LTC institutions were conducted by staff; most assessors were nurses, though some were social workers who provided care to the elderly day-to-day, as they knew their residents'/patients' conditions well. Because the assessors had the basic knowledge and skills necessary for geriatric assessments, the training focused on assessment using the interRAI LTCF and additional items to calculate the case-mix. Training was delivered at each site by trained research nurses. A standardized packet of training materials prepared by the research team was used, and site-specific questions and concerns about conducting the assessment were discussed at the training sessions.

A logic check and quality control of the data collected were done by the research team. The general characteristics of LTCFs and LTCHs were summarized using descriptive

H. Kim et al. / Health Policy xxx (2015) xxx-xxx

statistics. The level of key functional status and the distributions of case-mix groups were compared between LTCF residents and LTCH patients using t-tests or chi-square tests. Case-mix groups were generated by applying the case-mix algorithm from the HIRA. Rehabilitation service use and acute-care hospital admission were compared between the two groups according to case-mix group using a chi-square test.

3. Results

The key organizational and resident/patient characteristics of LTCFs and LTCHs in Korea are summarized in Table 1. The majority of LTCFs were small or medium size, with 99 or fewer beds (Table 1). Almost all (93.4%) of the LTCFs were private, and about half of them were established in or after 2009, when the public LTCI was introduced. As for LTCHs, the size varied, though the majority were medium size (100-199 beds). Similar to the LTCFs, almost all the LTCHs were private (92.4%), and half of them were established in or after 2009. The nursing staffing levels, including both registered nurses and nurse aides, was about 4.6 per 100 beds for LTCFs and 15.8 per 100 beds for LTCHs. There is no medical staff required in LTCFs, as they are legally welfare institutions for older adults; in contrast, the average physician staffing level in LTCHs was 2.5 physicians per 100 beds.

As for the patient/resident characteristics, about 42% of older residents in LTCFs in Korea were the oldest old (aged 85+). Three fourths of the sample were female, and one fourth of them were low-income people with Medical-Aid. Dementia and stroke were the two most prevalent chronic conditions. LTCF residents in Korea were admitted from home (68.1%), followed by LTCHs (11.3%), while LTCH patients were mostly admitted from home (49.8%), acute-care hospitals (28.1%), and other LTCHs (15.5%).

Table 2 shows the key functional status of older people in LTCFs and LTCHs in Korea. In both settings, more than half of the older people required extensive physical assistance and had moderate/severe or higher cognitive impairment; in addition, 4 out of 10 people experienced depressive symptoms. The proportion of people with unstable health conditions was relatively low in both LTCFs (12.6%) and LTCHs (13.5%), and was not significantly different by setting. On the other hand, the proportion of older people experiencing pain in LTCHs was about 17.4%, which was significantly higher than in LTCFs (8.4%).

The case-mixes of older people in the two long-term care settings were also not highly distinct (Table 3). The largest group in both LTCFs (35.9%) and LTCHs (34.8%) was the Impaired Cognition group. About 45% of older people at LTCHs had a moderate or higher level of clinical care needs (Ultra High, High, or Medium), and about 31% of those at LTCFs had similarly serious clinical care needs. More than half (55.3%) of LTCH patients were categorized in the four case-mix groups with relatively lower clinical and medical needs (the Behavior Problem, Impaired Cognition, Mild Clinical Care, and Reduced Physical Function groups).

Table 4 presents acute-care hospital admission and rehabilitation service use of older people at LTCFs and LTCHs by case-mix group. About 6.5% of older residents

in LTCFs had been admitted to acute-care hospitals at least once in the past 90 days, which was about 2.5 times higher than the prevalence of acute-care hospital admissions (2.6%) in LTCHs. Hospitalizations of LTCF residents were especially high among the Ultra High (25.0%) and High (15.8%) Medical Care groups. As for in-house rehabilitation services under physician orders, overall about half of the older people in both LTCFs (49.6%) and LTCHs (49.1%) had received the services during the last 7 days. The service provision pattern was similar for the respective case-mix groups between the two settings.

4. Discussion

This study is the first survey to profile institutionalized older adults receiving LTC under either NHI or LTCI in Korea with a nationally representative sample using common functional measures and a case-mix classification system. The findings show that a majority of older people at LTCHs and LTCFs in Korea are frail, poor, older people with complex chronic diseases such as dementia or stroke. The current delivery of LTC under the two insurances is not coordinated well. The roles of LTCHs funded by NHI and LTCFs funded by LTCI overlap somewhat, and the two LTC institutions are likely to be in competition with each other for older people with similar care needs. Policy reforms creating a clearer distinction between the roles of the two are necessary to enhance coordination and integration of care for older adults.

A case-mix system classifies people into homogeneous groups according to severity of condition and expected service use. The findings demonstrated the case-mix of older people at LTCHs and LTCFs in Korea is mixed. The proportion of people with a moderate or higher level of clinical care need was somewhat higher at LTCHs than LTCFs, but more than half of LTCH patients had non-medical care needs, such as behavior problems or cognitive impairment, as their primary conditions. A policy policy was adopted in 2009 to cut reimbursement to LTCHs for patients hospitalized for limitations in physical function [7], so the Reduced Physical Function group was relatively low (3.9%) but still existed. Only 13.5% of LTCH patients had unstable health conditions.

Several policy, provider, and consumer factors may contribute to LTCH as a preferred alternative to LTCF, including the following: the unclear admission and discharge criteria of LTCHs; the NHI's financial protection of people with longer hospitalizations by limiting their total copayment amount per year; and a still-extant public perception that sending their older loved ones to LTCHs, an expensive option with more staff and service, instead of LTCFs, regardless of current needs, is an expression of filial piety [18,25]. From a societal perspective, LTCHs could be too expensive for NHI to pay for the care of older people with relatively stable, lower clinical needs; also, hospitalization itself may have a negative impact on the quality of life of frail older people. Similar to LTCHs in Western countries with longer experience in the provision of LTC, LTCH services financed by NHI in Korea should increasingly target people needing post-acute convalescence care and/or intensive rehabilitation service, aiming to return to the

H. Kim et al. / Health Policy xxx (2015) xxx-xxx 5

Table 1

General characteristics.

LTCFs LTCHs p-Value

Facility/hospital Sample (n = 91) n (%) Sample (n = 52) n (%)

Bed sizea

Small 35 (38.5) 17(32.7) 0.5375

Medium 43 (47.3) 24 (46.2)

Large 13(l4.3) 11 (21.2)

Ownership

Public 6 (6.6) 5 (9.6) 0.5142

Private 85 (93.4) 47 (90.4)

Establishment year

Before 2009 47 (51.7) 26 (50.0) 0.8496

2009 and after 44 (48.4) 26 (50.0)

Nursing staff-to-bed ratio 0.046 (0.021) 0.158(0.050)' <0.0001

Physician-to-bed ratio n/a 0.025(0.007) n/a

Patient/resident Sample (n =1462) n (%) Sample (n =1351) n (%)

Age (mean, SD) 82.73 (7.49) 81.20(7.34) <0.0001

65-74 223 (15.3) 257(19.0) <0.0001

75-84 621 (42.5) 630 (46.6)

85 and over 618(42.3) 464 (34.3)

Male 329 (22.5) 340 (25.2) 0.0974

Female 1133(77.5) 1011 (74.8)

Married

Yes 276(18.9) 374 (27.7) <0.0001

No 1,186(81.1) 977 (72.3)

Medical-Aid

Yes 381 (26.1) 293(21.7) 0.0066

No 1,081 (73.9) 1,058(78.3)

Chronic conditions

Dementia 879 (60.1) 852(63.1) 0.1092

CHF 46 (3.1) 59 (4.4) 0.0879

COPD 40 (2.7) 47 (3.5) 0.2555

Diabetes mellitus 240(16.4) 326(24.1) <0.0001

Cancer 21 (1.4) 53 (3.9) <0.0001

Stroke 411 (28.1) 446 (33.0) 0.0048

Admitted from

Home 995 (68.1) 673 (49.8) <0.0001

Acute-care hospital 65 (4.4) 380(28.1)

LTCH 165(11.3) 210(15.5)

LTCF 134 (9.2) 40 (3.0)

Other 103 (7.0) 48 (3.6)

Note: Percentages may not add to 100 due to rounding. a Bed size at LTCFs: small (10-29), medium (30-99), large (100 or more); bed size at LTCHs: small (99 or fewer), medium (100-199), large (200 or more). b Nursing staffing data were missing in 4 LTCHs.

Table 2

Functional status of older people in LTCFs and LTCHs in Korea.3

LTCFs (n =1462) n (%) LTCHs (n =1351) n (%) p-Value

Requires extensive physical assistance (ADLHS 4+) 866(59.2) 800(59.2) 0.9920

Moderate/severe or higher cognitive impairment (CPS 4+) 774(52.9) 697(51.6) 0.4739

Depressive symptoms (DRS 3+) 618 (42.3) 621 (46.0) 0.0486

Mild/moderate or severe pain (Pain Scale 2+)b 122(8.4) 234(17.4) <0.0001

Unstable health (CHESS 2+) 184(12.6) 182(13.5) 0.4853

a The functional statuses in the tables are measured by the following scales or measures based on the interRAI LTCF [19]: ADL, ADL Hierarchy Scale (ADLHS); Cognition, Cognitive Performance Scale (CPS); Depression, Depression Rating Scale (DRS); Pain, Pain Scale; and Condition instability, Changes in Health, End-Stage Disease and Symptoms and Signs (CHESS) scale. b Missing values in items to calculate the pain scale in 4 LTCF residents and 3 LTCH patients.

H. Kim et al. / Health Policy xxx (2015) xxx-xxx

Table 3

Case-mix of older people at LTCFs and LTCHs in Korea.

Case-mix hierarchy LTCFs (n = 1358)a n (Ж) LTCHs (n = 1309)a n (Ж) p-Value

Ultra high medical care 24(1.8) 46(3.5) 0.0086

High medical care 190(14.0) 267 (20.4) 0.0003

Medium medical care 201(14.8) 273(20.9) 0.0009

Behavioral problem 324(23.9) 195(14.9) <0.0001

Impaired cognition 488(35.9) 455(34.8) 0.2825

Mild clinical care 20(1.5) 22(1.7) 0.7576

Reduced physical function 111(8.2) 51 (3.9) <0.0001

a A case-mix group cannot be calculated for some cases due to missing values for key variables. Table 4

Service utilization of older people in LTCFs and LTCHs in Korea.

Acute-care hospital admission3

Rehabilitation service provided'

LTCFs LTCHs p-Value LTCFs LTCHs p-Value

Case-mix Older Acute-care Older Acute-care Older Rehabilitation Older Rehabilitation

category residents hospital patients hospital residents service patients service

admission admission provided provided

(yes = 1, (yes = 1, (yes = 1, row (yes = 1, row

row %) row %) %) %)

Ultra high 24 25.0 43 2.3 0.0036 24 20.8 46 28.3 0.4997

medical

High 190 15.8 262 1.9 <0.0001 190 42.6 267 44.9 0.6236

medical

Medium 200 5.5 266 2.6 0.1117 200 51.0 268 54.1 0.5057

medical

Behavioral 324 4.0 195 1.5 0.1143 316 47.8 194 43.8 0.3826

problem

Impaired 488 4.7 451 3.3 0.2812 488 51.2 454 51.5 0.9236

cognition

Mild 20 5.0 22 4.6 0.9449 20 85.0 22 86.4 0.8996

clinical

Reduced 111 3.6 51 2.0 0.5744 110 57.3 51 45.1 0.1497

physical

function

Total 1357 6.5 1290 2.6 <0.0001 1348 49.6 1302 49.1 0.8752

a Acute-care hospital admission visit during the last 90 days. b Rehabilitation service provided during the last 7 days.

community [8,18,28]. To succeed at such reform, several policies need to be implemented in a well-coordinated way to address the contributing factors discussed above.

For LTCFs under LTCI, strengthening healthcare provision will be a key reform agenda, considering a large portion (30.6%) of current older LTCF residents had serious health conditions. Such reform of LTCI also would be a prerequisite for the success of the suggested reforms described above for LTCHs under NHI. A more skilled LTC workforce for health service is needed; currently, almost three fourths of the workforce at LTCFs are personal care assistants [14]. LTCI law defines RNs and NAs as interchangeable, and some nursing homes hire no RNs [3,18], unlike LTCHs under NHI [7]. It may not be feasible or efficient to increase the staffing level at all current LTCFs, so introduction of a new type of LTCF with strengthened health care, including a minimum RN staffing requirement, might be a policy option. Germany provides nursing care-based LTC at LTCFs, and the United

States has skilled nursing facilities (SNFs); unlike intermediate nursing facilities (INFs), SNFs require 24-h nursing care, with a wide range of short-stay and long-stay health and even rehabilitation services provided by multidisci-plinary teams, including nurses under the supervision of physician [26,27,29].

In addition to the care needs of older people at LTCHs and LTCFs, this study also examined their service utilizations in two settings funded by different insurances. With regard to the use of rehabilitation services under physician orders, this study found there was likely no difference in access to the services by case-mix group across settings, yet the quality and intensity of the services may have differed, which should be examined further in future studies. In addition, a comparison of the effectiveness of rehabilitation in hospital and facility settings for older people with LTC needs is recommended. This could be an interesting question from a policy perspective; the evidence is still inconclusive.

H. Kim et al. / Health Policy xxx (2015) xxx-xxx

We also found that older people with high clinical care needs at LTCFs, where health care service is limited, are at risk for acute-care hospital admission; this could be a safety issue. One fifth of older people in the Ultra High Medical Care group at LTCFs were hospitalized at least once during the last 90 days, about 10 times higher than the incidence among the same case-mix group at LTCHs. Similarly, this incidence rate for the High Medical Care group was about seven times higher among LTCF residents than LTCH patients. This may have been partially related to ineffective regulation preventing these patients from being admitted to LTCFs from the start and/or requiring them to transfer to LTCHs or acute-care hospitals when their conditions became unstable or they had acute problems. In addition, families may prefer to keep their older family members at LTCFs. This may be due to financial issues such as higher out-of-pocket LTCH costs, and/or no preference for life-sustaining medical therapies for their older family members at hospitals. Guidelines and legal enforcement are needed for the protection of the safety of older LTCF residents with high clinical care needs. High-risk groups should be better screened when they enter the LTCI system, and residents with health issues living at LTCFs should be properly monitored and transferred to LTCHs if their conditions worsen.

Another finding to note is a significant portion of institutionalized older people (17% at LTCHs vs. 8% at LTCFs) in Korea experienced pain. Proactive pain management and better awareness of pain as the fifth vital sign and an indicator of quality of life are needed, beyond using pain management as a quality indicator or an item for accreditation review [30,31,32]. The higher prevalence of pain in LTCHs than LTCFs may need to be interpreted with caution: the findings may be the result of higher detection of patients' pain due to better staffing and more proactive pain management in LTCHs, as well as higher proportions of people with more complex medical conditions.

The study findings are not very surprising, in a sense; but with empirical evidence, we confirmed our hypothesis of suboptimal delivery of institutional LTC under NHI and LTCI in Korea, which should be redesigned as a person-centered, integrated system. Currently, LTCFs focus on functional limitations, with clear gate-keeping based on need assessment, while qualification for LTCH services is broadly defined in terms of both medical and functional status, without a gate-keeping (e.g., primary care physician) system. The eligibility for and services of the two institutional LTC settings should be harmonized. The insurer or the provider should offer better guidance supporting older people and their families in navigating the complex LTC system. Setting recommendations based on common assessment across settings, as in Canada [27], could be useful. For the public, education, empowerment, and incentives for rational use of LTC are necessary.

Coordinating and integrating care for older people is an important but challenging policy agenda. Various reforms in post-acute and long-term care systems are in progress in many countries. In order to develop seamless LTC across hospitals and facilities, Ontario, Canada, which has tax-based LTC provision, adopted in the early 1990s a common case-mix system based on interRAI assessment tools [27].

People with high health care needs are advised to use complex continuing care (CCC) hospitals, and those with lower needs, such as cognition impairment and behavioral problems, to use LTCFs providing skilled nursing care, unlike in Korea. This policy has dramatically decreased social hos-pitalizations [27,33]. In the U.S., Medicare and Medicaid, two social insurances, are the major financial sources for post-acute and long-term care (PALC); and common assessment tools across different PALC settings, such as home care, LTCHs, SNFs, and ICFs, have been developed and evaluated [1,34]. The recent establishment of the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014 provides a legal foundation for quality (value)-based Medicare prospective payment reform through developing standardized post-acute assessment instruments and data across Medicare post-acute care (PAC) settings [35].Japan's approach is aimed more toward the community: comprehensive care service centers have been established, aiming to provide patient-centered, integrated health, medical, and social services [1,36]. The evidence related to such policy experiment is still limited.

Several implications and lessons can be drawn from this study. First, regardless of funding sources, LTC needs to be a unified, person-centered system. A cornerstone of this goal is a standardized and integrated assessment system for health and social care needs across services and settings. Such coordination of information is indispensable for coordination of care. Second, while the target population for each funding source should be made as distinct as possible, the service in each setting should be comprehensive, addressing both health/medical and social care needs. Third, there is no clear answer as to the "right" setting for patients with behavioral and cognitive issues, but the literature suggests LTC can be delivered in non-institutional settings for people with stabilized, long-term psycho-behavioral conditions [37,38]. Strengthening the infrastructure for home- and community-based LTC is a prerequisite for the successful implementation of policies to prevent social hospitalizations in LTCHs. Further comparative effectiveness research is necessary, considering the context of each country. More research is also needed on health systems and policies that promote seamless transition across LTC settings.

Conflict of interest statement

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgements

This research was supported by a research grant of the Korean National Health Insurance Services; a grant from the Basic Science Research Program through the National Research Foundation (NRF) of Korea, funded by the Ministry of Science, ICT, & Future Planning (grant number: 2010-0002802); and also a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant

8 H. Kim et al. / Health Policy xxx (2015) xxx-xxx

number:H113C2250). The funding sources had no role in the study design, data collection and analysis, interpretation of the data, writing the manuscript or the decision to submit the manuscript for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding sources.

References

[1] Organization for Economic Co-operation Development (OECD). Help wanted? Providing and paying for long-term care. OECD Publishing; 2011. Available at: (http://www.oecd.org/els/health-systems/help-wanted.htm [accessed 13.08.15].

[2] OECD/European Commission. A good life in old age? Monitoring and improving quality in long-term care. 1n: OECD health policy studies. OECD Publishing; 2013. Available at: (http://dx.doi.org/10.1787/9789264194564-en) [accessed 13.08.15].

[3] Roh YK, Sunwoo D, YoonJL, Won CW, Lee DW, Ki PS, et al. Defining the role of long-term care facilities and geriatric hospitals. Seoul, Korea: Ministry of Health and Welfare, Korean Geriatrics Society; 2010.

[4] Kwon S. Future of long-term care financing for the elderly in Korea. Journal of Aging & Social Policy 2008;20:119-36, http://dx.doi.org/10.1300/J031v20n0K07.

[5] Organization for Economic Co-operation Development (OECD). Health at a glance 2013: OECD indicators. OECD Publishing; 2013. Available at: (http://www.oecd.org/els/health-systems/Health-at-a-Glance-2013.pdf) [accessed 13.08.15].

[6] Organization for Economic Co-operation Development (OECD). OECD health data 2014. In: OECD health statistics (database). Organization for Economic Co-operation Development (OECD); 2015. Available at: (http://stats.oecd.org/index.aspx) [accessed 13.08.15].

[7] Ga H, Won CW. Perspective on long term care hospitals in Korea. Journal of the American Medical Directors Association 2013;14(10):770-2, http://dx.doi.org/10.1016/jjamda.2013.06.007.

[8] Song HJ, Chae JM. Redefining role of long-term care hospital for efficient elderly healthcare system. Seoul, Korea: Health 1nsurance Review & Assessment Service; 2012.

[9] Korean Ministry of Health and Welfare. The white paper on health and welfare, 2006. Seoul, Korea: Korean Ministry of Health and Welfare; 2007.

[10] Korean Ministry of Health and Welfare. The medical service act (article 3). (http://www.law.go.kr/main.html).

[11] Korean National Health Insurance Service. National health insurance statistics yearbook 2013. Seoul, Korea: NH1S; 2013. Available at: (http://www.nhis.or.kr/menu/boardRetriveMenuSet.xx?menu1d= F3321) [accessed 13.08.15].

[12] Seok JE. Public long-term care insurance for the elderly in Korea: design, characteristics, and tasks. Social Work in Public Health 2010;25(2):185-209, http://dx.doi.org/10.1080/ 31937191090354703.

[13] Korean National Health Insurance Services. Long-term care insurance: an overview. NH1S; 2015. Available at: http://www. longtermcare.or.kr/portal/longtermcare/main.jsp 609 [accessed 13.08.15].

[14] Korean National Health 1nsurance Services. Long-term care statistics 2013. Seoul, Korea: NH1S; 2015. Available at: (http://www.nhis.or.kr/menu/boardRetriveMenuSet.xx?menu1d= F332a) [accessed 13.08.15].

[15] Korean National Health 1nsurance Service. Long-term care statistics 2008. Seoul, Korea: NH1S; 2010. Available at: (http://www.nhis.or.kr/menu/boardRetriveMenuSet.xx?menu1d= F332a) [accessed 13.08.15].

[16] Korean National Health 1nsurance Service. A survey on the public's satisfaction with and awareness of the long-term care insurance for the elderly. 1n: Findings report. Korean National Health 1nsurance Service; 2013.

[17] KimJH, Lee HJ, Lee SW, Park MS. Recommendations forthe improvement of the long-term car reimbursement system: short-term approaches. Seoul: Health 1nsurance Review & Assessment Services; 2014.

[18] Kim JS, Sunoo D, Lee KJ, Choi 1D, Lee HY, Kim KA. The roles oflong-term care hospitals and long-term care facilities: a study focus on

focusing coordination strategies. Korea Institute for Health and Social Affairs; 2013.

[19] Kim H, Jung YI, Sung M, Lee JY, YoonJY, YoonJL. Reliability of the interRAI long term care facilities (LTCF) and interRAI home care (HC). Geriatrics & Gerontology International 2015;15(2):220-8, doi: 10.1111.ggi.12330.

[20] Morris JN, Fries BE, Morris SA. Scaling ADLs within the MDS. Journals of Gerontology Series A, Biological Sciences and Medical Sciences 1999;54(11):M546-53.

[21] Morris JN, Fries BE, Mehr DR, Hawes C, Phillips C, Mor V, et al. MDS cognitive performance scale. Journal of Gerontology 1994;49(4):M174-82.

[22] Burrows AB, Morris JN, Simon SE, Hirdes JP, Phillips C. Development ofa minimum data set-based depression rating scale for use in nursing homes. Age and Ageing 2000;29(2):165-72.

[23] Fries BE, Simon SE, Morris JN, Flodstrom C, Bookstein FL. Pain in U.S. nursing homes: validating a pain scale for the minimum data set. Gerontologist 2001;41:173-9.

[24] Hirdes JP, FrijtersDH,TeareGF. The MDS-CHESS scale: a new measure to predict mortality in institutionalized older people. Journal of the American Geriatrics Society 2003;51(1):96-100.

[25] Lee JY, Yoon JY, Kim JH, Song SH, Joo JS, Kim EK. Development of patient classification system in long-term care hospitals. Journal of Korean Academy of Nursing Administration 2008;14(3): 229-40.

[26] Mor V, Leone T, Maresso A. Regulating long-term care quality. New York, NY: Cambridge University Press; 2014.

[27] Fries BE, Fahey CJ, editors. Implementing the resident assessment instrument: case studies of policy making for long-term care in eight countries. New York, NY: Milbank Memorial Fund; 2003. Available at: (http://www.milbank.org/uploads/documents/interRAI/ ResidentAssessment Mech2.pdf) [accessed 13.08.15].

[28] Gage B, Pilkauskas K, Dalton K, Constantine R, Leung M, Hoover S, et al. Long-term care hospital (LTCH) payment system monitoring and evaluation. Baltimore: Centers forMedicare & Medicaid Services;

2007. Available at: (http://cms.hhs.gov/Medicare/Medicare-Fee-for-Service-Payment/LongTermCareHospitalPPS/RTIReports.html) [accessed 13.08.15].

[29] Centers for Medicare & Medicaid Services (CMS). Medicare coverage of skilled nursing facility care. In: CMS product no. 10153. Baltimore: CMS; 2014. Available at: (http://www.medicare.gov/ Pubs/pdf/10153.pdf) [accessed 13.08.15].

[30] Wells N, Pasero C, McCaffery M. Improving the quality of care through pain assessment and management. In: Hughes RG, editor. Patient safety and quality: an evidence-based handbook for nurses. Rockville, MD: Agency for Healthcare Research and Quality (US);

[31] Health Insurance Review & Assessment Service. The results of long-term care hospital assessment in Korea in 2013. Seoul: Health Insurance Review & Assessment Service; 2014.

[32] Korea Institute for Healthcare Accreditation, Ministry of Health & Welfare. Standards for the accreditation review of long-term care hospitals. Seoul: Kore Institute for Healthcare Accreditation; 2013.

[33] Hirdes JP, Mitchell L, Maxwell CJ, White N. Beyond the 'iron lungs of gerontology': using evidence to shape the future of nursing homes in Canada. Canadian Journal on Aging 2011;30(3):371-90, http://dx.doi.org/10.1017/S0714980811000304.

[34] National Quality Forum. Coordination strategy for post-acute care and long-term care performance measurement. National Quality Forum; 2012. Available at: (http://www.qualityforum. org/Publications/2012/02/PAC-LTC_Final_Report.aspx) [accessed 13.08.15].

[35] United Senate Committee on Finance. The improving medicare post-acute care transformation act of 2014: IMPACT Act of 2014. United Senate Committee on Finance; 2015. Available at: (http://www.finance.senate.gov/imo/media/doc/IMPACT %20Summary.pdf) [accessed 13.08.15].

[36] Campbell JC, Ikegami N, Gibson MJ. Lessons from public long-term care insurance in Germany and Japan. Health Affairs (Millwood) 2010;29(1):87-95, http://dx.doi.org/10.1377/hlthaff.2009.0548.

[37] McGill N. As senior population grows, aging in place gains popularity: communities conducting outreach. The Nation's Health;43(8):1-16.

[38] Mestheneos E. Ageing in place in the European Union. IFA Global Aging;7(2);17-24.