Comparison of the Measurement of Long-Term Care Costs between China and Other Countries: A Systematic Review of the Last Decade

Background: The rapid aging of populations in some countries has led to a growing number of the disabled elderly, creating a huge need for Long-Term Care (LTC) and meeting its costs, which is a heavy economic burden on the families of the disabled elderly and governments. Therefore, the measurement of Long-Term Care (LTC) costs has become an important basis for the government to formulate Long-Term Care (LTC) policies, and academic research on Long-Term Care (LTC) costs is also in the process of continuous development and deepening. Methods: This is a systematic review that aims to examine the evidence published in the last decade (2010–2019) regarding the comparison of the measurement of Long-Term Care (LTC) costs between China and other countries. Results: Eighteen Chinese studies and 17 other countries’ studies were included in this review. Most Chinese scholars estimated long-term care costs based on the degree of disability among the disabled elderly. However, the studies of European and American countries are more and more in-depth and comprehensive, and more detailed regarding the post-care cost of specific diseases, such as Parkinson’s disease, Alzheimer’s disease, and epilepsy. Conclusion: In future academic research, we should fully consider the human value of long-term care providers and further study the differences in the long-term care costs of different chronic diseases. In China’s future policymaking, according to the experience of Germany, Sweden, and other countries, it may be an effective way to develop private long-term care insurance and realize the effective complementarity between private long-term care insurance and public long-term care insurance (LTCI).


Introduction
The increasingly aging population has highlighted the urgency of the crisis in healthcare services for the elderly in China in recent years. According to World Population Prospects (2019), published by the Department of Economic and Social Affairs of United Nations (WPP2019), population aging is expected to increase rapidly (South Korea, 38.1%; Japan, 37.7%; Italy, 36.0%; Germany, 30.0%; and China, 26.1%) by 2050 [1]. According to CSY (China Statistical Yearbook, 2019), the elderly population, aged 65 or above, reached 167 million in 2018, accounting for 11.9 percent of the total population. The ODR (old-age dependency ratio) climbed from 9.9% in 2000 to 16.8% in 2018 [2]. The average life expectancy in China is expected to reach 81.52 years in 2045-2050, which is close to the average of 83.43 years in developed countries (WPP2019). Among them, the number of disabled elderly due to chronic diseases [3], industrial or agricultural environmental pollution [4][5][6][7], accidental injuries,

Literature Search Source
Our review was carried out to show the measurement of Long-Term Care (LTC) costs in related studies. The databases mentioned in our review mainly were: Web of Science, Medline, SCOPUS, EBSCO, PubMed, and CNKI (China). The following combinations of terms were used, with the Boolean phrase "and/or", to maximize the scope and type of material referred in the search: "Long-Term Care cost" OR "LTC cost". The search was carried in Chinese and English. We also added a Chinese research database named VIP (China) to look for any studies that might be missing.

Data Extraction Criteria
Preliminary inclusion criteria: the selection criteria for publications were as follows: (1) studies on the measurement of Long-Term Care (LTC) costs; (2) research conducted in China and other countries, which should be strictly separated to facilitate subsequent analysis; and (3) research published from January 2010 to December 2019. Publications based on opinions or comments, editorials, and summaries of meetings were not included.
Exclusion criteria: these included articles reporting on the results of a qualitative study, quantitative analyses, surveys, feasibility studies, relationship measurements, satisfaction studies, "what-if" analyses, and data collection techniques. Publications based on opinions or comments, editorials, and summaries of meetings were not included. The results were initially extracted by one researcher and then cross-checked by another to ensure that all data had been screened and reviewed. If there was a difference of opinion between the two researchers, a third researcher was invited to express their opinion and finally reach an agreement. All publication information was exported to the Excel database via Endnote, and duplicate sections were removed. We followed some previous special definitions regarding LTC as described previously in this paper. Finally, 18 Chinese studies and 17 other countries' studies were retained for the review.
The information extracted from all the included publications was as follows: time frames, study methodology (design, purpose, participants, and tools), samples, and key conclusions. All the analysis results were analyzed in the following process (as shown in Figure 1). the Excel database via Endnote, and duplicate sections were removed. We followed some previous special definitions regarding LTC as described previously in this paper. Finally, 18 Chinese studies and 17 other countries' studies were retained for the review. The information extracted from all the included publications was as follows: time frames, study methodology (design, purpose, participants, and tools), samples, and key conclusions. All the analysis results were analyzed in the following process (as shown in Figure 1).

Quality Assessment
Study quality was independently assessed by the two researchers using the Risk of Bias Assessment Tool for Non-Randomized Studies (RoBANS) [22]. The criteria included the selection of participants, confounding variables, intervention measurements, the blinding of outcome assessments, the incompleteness of outcome data, and selective outcome reporting. Each criterion was evaluated as "low risk of bias", "high risk of bias", or "unclear". In cases of disagreement, each case was discussed with a third researcher. The quality of the included studies is summarized in Table 1.

Quality Assessment
Study quality was independently assessed by the two researchers using the Risk of Bias Assessment Tool for Non-Randomized Studies (RoBANS) [22]. The criteria included the selection of participants, confounding variables, intervention measurements, the blinding of outcome assessments, the incompleteness of outcome data, and selective outcome reporting. Each criterion was evaluated as "low risk of bias", "high risk of bias", or "unclear". In cases of disagreement, each case was discussed with a third researcher. The quality of the included studies is summarized in Table 1.

A Preliminary Review of the Relevant Literature
It was found that three-quarters of the studies were made up of quantitative research, while qualitative research only accounted for one quarter. Moreover, the qualitative method was used in most research in China, while quantitative methods could be found in nearly every article in other countries, in which the acceptance by and understanding of the readers could be better.
From the publication point of view, the journals that published the most related papers were BMC Health Services Research, Geriatrics & Gerontology International, PLOS One, Public Health Nursing, Nursing Research (Chinese), and Chinese Health Economics (Chinese), respectively. Among them, BMC Health Service Research and PLOS One were tied at 15% each, and Geriatrics & Gerontology International and Public Health Nursing each had 10%. Besides, it was interesting that some of them were nursing and health, biology, physics, and other comprehensive journals, which are not pure medical journals.

Classification by Country Studied
The classification of articles published regarding the measurement of LTC (Long-Term Care) costs in different countries is presented in Figure 2. It can be seen from Figure 2 that the pie chart on the left is divided into eight parts according to the research object for LTC in related articles. It was clear that China has become the most popular research object, accounting for nearly half of all the published articles, and the study of LTC in the UK ranked the second. Meanwhile, the United States came in third place, accounting for 14% of the pie chart. It can be observed that there is a small circle on the right, composed of Spain, Sweden, Germany, and Finland.

Classification by Publication Year
The classification of articles published on LTC (Long-Term Care) costs across the timeline of the study (2010-2019) is presented in Figure 3. It can be seen from Figure 2 that the pie chart on the left is divided into eight parts according to the research object for LTC in related articles. It was clear that China has become the most popular research object, accounting for nearly half of all the published articles, and the study of LTC in the UK ranked the second. Meanwhile, the United States came in third place, accounting for 14% of the pie chart. It can be observed that there is a small circle on the right, composed of Spain, Sweden, Germany, and Finland.

Classification by Publication Year
The classification of articles published on LTC (Long-Term Care) costs across the timeline of the study (2010-2019) is presented in Figure 3. pie chart. It can be observed that there is a small circle on the right, composed of Spain, Sweden, Germany, and Finland.

Classification by Publication Year
The classification of articles published on LTC (Long-Term Care) costs across the timeline of the study (2010-2019) is presented in Figure 3.  It can be seen from Figure 3 that the amount of research on LTC is not large in general and currently in a declining state, with peaks in 2017 and 2018. Nevertheless, it has shown a wavy upward trend. The attention of some scholars had been attracted to LTC since 2010. After a brief decline in 2014, it peaked in 2017 and 2018. Based on the analysis above, it can be inferred that more and more scholars at home and abroad have devoted themselves to the research of long-term care costs since 2014 and gradually achieved excellent results. The result could be attributed to the increasing emphasis on nursing in various countries. At the same time, it was reflected that the development of social structures in various countries is becoming more and more perfect.

Measurement of Long-Term Care Costs in China
In 2010, the cost accounting of home care programs was shown by Jin et al. [23], and the research into long-term care costs is getting more and more diversified (as shown in Table 2).  Most Chinese scholars estimated long-term care costs based on the degree of disability among the disabled elderly. For example, Lu [24], Feng et al. [25], Hu [26], and Yuan [27] calculated the basic LTC costs based on the Activities of Daily Life (ADL) classification of mild, moderate, and severe dysfunction. In addition, Liu and Zhong [28] compared the LTC costs generated by patients with different ADL function obstacles with the living allowances in urban and rural areas, which could be a new reference for the long-term care cost system. Xu et al. used the Bayesian quantile regression method to measure the high, medium, and low levels of long-term care cost prediction for each disability state of the elderly in China from 2020 to 2050 [29].
Other Chinese scholars focus on how to integrate the human value of nursing staff into the measurement system for long-term care costs, so as to achieve the goal of the scientific and reasonable measurement of long-term care costs. The value of nurses was noticed in the studies of Fengyue and Junko [30], Yu et al. [31], and Fan et al. [32], which discussed the LTC costs in Chongqing, Guangdong, and other provinces in China. Qun [33] compared the LTC costs in nine provinces of China with those in Texas, U.S. Li et al. constructed a pricing model for nursing service projects that can measure the value of LTC services and fully reflect the LTC costs [34]. With the development of the nursing industry, a variety of nursing services could be offered by nursing institutions, communities, and even in homes, aside from hospitals. The process of providing home care services in a community of Shanghai was discussed by Du et al. [35]. They calculated the cost of five plates, including manpower, materials, and another three big plates, and it was found that the current charge for the home care services could not cover the necessary expenditures.
The comparison of long-term care costs under different forms of care (such as institutional care, community care, and home care) has become another important topic of academic discussion regarding the development of elderly care institutions in China. Yang et al. [36] and Lu et al. [37] revealed that the total cost of institutional care is far less than the cost of general hospital inspections and other routine expenses. The direct and indirect costs of home care were calculated by Huang et al. [38] and Song et al. [39]. It can be concluded from their research that institutional care is suitable for patients whose daily life is marked by high dependence, while home or community care could be a better option for the less dependent. Then, a survey of the influencing factors for the utilization and cost of formal care in the elderly community of Shanghai was conducted by Fen et al. [40], who found that professional home care was more cost-effective compared with the care provided by family members.
In a word, LTC-related research is a relatively new concept in China. With the pilot practice of the LTCI system in Qingdao, Shanghai, Chongqing and other places, LTC-related research has begun to rise. However, due to the lack of official statistics and uniform standards of LTC costs, these studies have many difficulties, especially in empirical studies.

Measurement of Long-Term Care Costs in Other Countries
Developed countries such as those in Europe and the United States have paid attention to long-term care costs earlier, and the research objects were more diversified and comprehensive (as shown in Table 3). It was found by Ryan Greysen [41] that the LTC costs for the elderly with the highest level of functional disorders were 77% higher than for those without functional disorders. Lagergren et al. [42] predicted the future LTC costs in Japan and Sweden based on age, gender, and degree of ADL dysfunction. Dementia poses a substantial additional burden on the German social security system, and female dementia patients need to be a key target group for health services research in an aging society. 5 Mitsuhiro S et al. [45], 2018

Multiple regression analysis
The Survey of Long-Term Care Benefit Expenditures in Japan The societal cost of dementia in Japan appeared to be considerable. Interventions to mitigate this impact should be considered 6 Hajime T et al. [46], 2019 Cross-section analysis on time series 169 patients with Alzheimer's disease or mild cognitive impairment in Japan As the number of patients with Alzheimer's disease increases, direct social costs will increase. 7 Greg A et al. [47], 2015 Polynomial regression model

veterans hospitalized for ischemic stroke in Veterans Health Administration facilities in 2007
Care trajectories after stroke were associated with stroke severity and functional dependency and then had a dramatic impact on subsequent costs. 8 Ramon L et al. [48], 2018

Multiple regression model
All patients with first-ever incident ACLVI from 2002 to 2012 Hospital care costs were significantly higher than for stroke over the long term and were similar after the inclusion of the costs of institutionalization. 9 Julieta et al. [49], 2016  Costs of LTC decreased as sheltered housing replaced institutional LTC.
As the research on long-term care is more and more in-depth and comprehensive, the research on long-term care costs was more detailed regarding the post-care costs of specific diseases. Some studies on the LTC costs of Parkinson's disease [43][44][45], Alzheimer's disease [46][47][48], colorectal cancer [49], and viral gastroenteritis [50] were conducted. In addition to cancer, which has been a problem for a long time, even research on the LTC costs of alcoholism was carried out by Edward et al. [51]. Overweightness is a public health problem all over the world, and obesity has become a common disease. Thus, Vicky et al. [52] calculated the differences in LTC costs associated with BMI in UK and found that the larger the body mass index, the higher the related costs were. Julie [53] discovered that the cost of care could be decreased after bariatric surgery in Australia. Besides, the LTC cost might be affected by the payment policies of the government, which has been proved by the work of Shota Hamada et al. [54].
Comparisons of long-term care costs under different forms of care, such as institutional care and home care, are also a focus of research in developed countries. Peter et al. [55] compared the LTC costs for high-income people with those for the farmers in institutional care and proved that the LTC cost for farmers was much less. In order to explore whether the increase in the elderly population will make long-term care costs rise, the trajectory of long-term care in 28 European countries was explored by Maria et al. [56], and it was found that the cost reduced as the expenditure of the government on health services was increased. At the same time, the elderly over 65 had turned to home care, which reflects the cost-effectiveness of home care to some extent. Leena et al. [57] studied the cost of all-day use on long-term care from 2002 to 2013 in Finland, and it was discovered that the cost in the shelters is much lower than that in nursing institutions.

Discussion
With the acceleration of the population aging process in most countries, long-term care has gradually been accepted by the global community. After more than 20 years of research and practice, developed countries have generally recognized that the long-term care service system should be effectively separated from the medical system, and a relatively independent long-term care service system should be established. At the same time, developed countries, especially Germany, Japan, and the United States, have initially established long-term care service systems supported by public long-term care insurance system as the main body, and service standards and norms, supplemented by the active participation of family members, social workers, and volunteers. In view of the rapidly aging society in China, it is imperative to actively explore the establishment of an independent long-term care service system. However, China's social pension and nursing function mechanism has not been paid enough attention to and developed well, especially in the sense that medical insurance does not pay long-term nursing expenses. As a result, high nursing expenses are unbearable for ordinary families. Therefore, our review systematically searched the articles on the measurement of long-term care costs from 2010 to 2019, and compared China with other countries, hoping to further expand the academic space for the study of long-term care costs in China in the future and to provide an important policy basis for the government to build a sustainable long-term care service system.
In 2010, cost accounting for home care programs was shown by a study published in the Nursing Journal of the Chinese People's Liberation Army [23], and research into long-term care costs is getting more and more diversified. Until 2017 and 2018, the peak number of articles had been published so far, which means that the study of long-term care costs is gradually becoming more and more acceptable. It is acknowledged that, at the same time, academic circles at home and abroad are also quietly changing. Besides, it can be observed that qualitative research methods are applied in most studies in China, while studies around the world mainly use quantitative research methods, and the number of related articles around the world has exceeded those in China. It is believed that much related research on long-term care in various countries will appear in well-known journals at home and abroad in the future. The research objects are more extensive, and the content is more detailed. The establishment of long-term care systems in various countries is no longer empty talk.
In terms of specific research content or research perspectives, China and other countries show some similarities and differences.
Firstly, most Chinese scholars estimated long-term care costs based on the degree of disability among the disabled elderly [24][25][26][27]. Moreover, Xu et al. used the Bayesian quantile regression method to measure the high, medium, and low levels of long-term care cost prediction for each disability state of the elderly in China from 2020 to 2050 [28], which could be a new reference for the long-term care cost system. On the other hand, there are also a large number of relevant studies based on the degree of ADL dysfunction in European and American countries [41,42]. In fact, studies on long-term care costs in European and American countries were carried out earlier. For example, Martin [58] predicted the future costs for long-term care costs in the United Kingdom, with analysis also based on the degree of disability. It can be seen that both the studies in China and in other countries estimated long-term care costs based on the degrees of disability among the disabled elderly, which implies that the accurate prediction of the scale of disabled and semi-disabled elderly will be important in the measurement of long-term care costs in China in the future.
Secondly, both China's and other countries' scholars are starting to pay attention to how to integrate the human value of nursing staff into the measurement system for long-term care costs, so as to achieve the goal of the scientific and reasonable measurement of long-term care costs [30][31][32]. However, the studies of European and American countries are more and more in-depth and comprehensive, and more detailed regarding the post-care cost of specific diseases, such as Parkinson's disease [43][44][45], Alzheimer's disease [46][47][48], colorectal cancer [49], and viral gastroenteritis [50].
Thirdly, the long-term care costs under different forms of care (such as institutional care, community care, and home care) have become another important topic of academic discussion regarding the development of elderly care institutions in China [36][37][38][39][40] and other countries [55][56][57]. By comparison, the relevant research in China is still a simple analysis of the comparison of long-term care costs under different forms of care, but the relevant research in other countries has begun to delve into the expectations and comparison of long-term care costs of different forms of care among patient groups with different income or education levels [55].
In conclusion, the scientific measurement of long-term care costs is an important basis for the government to formulate a long-term care policy in response to the aging population. Through the comparison between China and other countries in this review, we found that the measurement of LTC costs is more and more refined, which is mainly reflected in two aspects: the comparison of LTC costs after discharge for different chronic diseases, and the comparison of LTC costs for different nursing methods. This also provides us with the countermeasures to solve this problem in the future. Firstly, the accurate prediction of the scale of disabled and semi-disabled elderly is important for the measurement of long-term care costs in China in the future. Secondly, the human value of long-term care providers should be taken into account, and the differences of long-term care costs due for different chronic diseases should be further studied. Moreover, the comparison of long-term care costs based on different care methods (institutional care, home care, or community care) is also an important basis for policymaking. Facing increasing long-term care costs, fund-raising has become a key aspect in the construction of the long-term care service system. According to the experience of the USA, Sweden, and other countries, in the long run, public long-term care insurance (LTCI) separated from medical insurance may not be enough to fully cope with the surging long-term care costs [59]. Therefore, introducing social capital, developing private long-term care insurance (LTCI) in the private market, and realizing the effective complementarity between private long-term care insurance (LTCI) and public long-term care insurance (LTCI) may be an effective pathway [60].
At present, to cope with the increasing cost of LTC, some countries have implemented and gradually improved LTCI systems, which can be divided into two types. The first is the Nordic "welfare state" model of comprehensive public welfare, which is obviously not suitable for China's specific national conditions. The second is the "corporatist-welfare" model, which emphasizes the equivalence of LTCI rights and obligations and is more in line with China's current national conditions [12]. Firstly, publicly providing LTC to all of its citizens without regard to the individual need for public assistance is the basis of the LTC system in Sweden. However, it is an extremely expensive system that is fraught with the potential for abuse [19]. Secondly, home care may be cheaper than institutional care. For instance, Sweden and some OECD countries in Europe focus on providing home care versus institutional care. However, LTC public funding was almost exclusively provided in institutions due to the vast majority of LTC being provided by family and friends in USA. This is similar to in China, where most of the LTC now consists of home care and community care. In addition, how to allocate the value of LTC services, such as the manpower value of service providers, is also an important consideration in the future development of the LTC system in China.

Conclusions
This review includes 20 Chinese articles and 22 articles from other countries. Through comparison, this review draws some valuable conclusions for future academic research and policymaking. In future academic research, we should fully consider the human value of long-term care providers and further study the differences in long-term care costs due to different chronic diseases. This review was limited by the amount of current research on this topic, the search strategy utilized, the number of databases searched, and researchers' and publication bias that may have affected the value and accessibility of the research recognized. We will continue to improve in future research.