Regional disparity in the provision of rehabilitation services using the open data from the Japanese national claims database: an ecological study

Abstract

Objectives

There are few reports on regional differences in the supply/utilisation balance and provision of rehabilitation services. This study analysed those regional differences in Japan to help policymakers provide more uniform and efficient rehabilitation services and optimally allocate related resources.

Design

An ecological study.

Setting

47 prefectures and 9 regions in Japan in 2017.

Primary and secondary outcome measures

Primary measures were ‘supply/utilisation (S/U) ratio’, calculated by dividing rehabilitation supply converted to service units, by rehabilitation utilisation and ‘utilisation/expected utilisation (U/EU) ratio’, calculated by dividing utilisation by EU. The EU was defined as utilisation expected from the demography in each area. Data required to calculate these indicators were collected from open sources such as the National Database of Health Insurance Claims and Specific Health Checkups of Japan Open Data Japan.

Results

The S/U ratios were higher in Shikoku, Kyusyu, Tohoku and Hokuriku regions, and lower in Kanto and Tokai regions. The number of rehabilitation providers per population was higher mostly in the western part of Japan and lower mostly in the eastern part. The U/EU ratios were also higher mostly in the western part, and lower mostly in the eastern part such as Tohoku and Hokuriku regions. The same trend was seen for cerebrovascular disease and musculoskeletal disorder rehabilitation, which accounted for approximately 84% of rehabilitation services. For disuse syndrome rehabilitation, such a trend did not exist, and the U/EU ratio differed by prefectures.

Conclusions

The large surplus in rehabilitation supply in the western part was attributed to the greater number of providers, while the smaller surplus in Kanto and Tokai regions was due to the smaller amount of supply. The number of rehabilitation services used was lesser in the eastern part such as Tohoku and Hokuriku regions, indicating regional differences in the provision of rehabilitation services.

Correspondence to Professor Katsuhiko Ogasawara; [email protected]

STRENGTHS AND LIMITATIONS OF THIS STUDY

• This study analysed regional differences in the supply/utilisation balance of rehabilitation in Japan.

• This study clarified the regional differences in the provision of rehabilitation services in Japan.

• This study is one of the few that conducted these analyses using data from national claims databases.

• Regional units are limited to prefectures and regions due to a lack of data granularity.

• Rehabilitation services analysed in this study are provided for wide ranges of patients; therefore, further analysis focused on specific diseases or conditions is needed.

Background

As the society is ageing in Japan, medical demand is estimated to increase.¹ Rehabilitation, defined as a set of interventions designed to optimise functioning and reduce disability in individuals with health conditions in interaction with their environment,² is of greater importance in the ageing society for improving activities of daily living and achieving physical independence. Therefore, medical resources related to rehabilitation should be optimally allocated and used uniformly and efficiently.

There are several studies on the supply/demand of rehabilitation. Landry et al predicted the supply/demand of physical therapists (PTs) in the USA and analysed the extent to which attrition rates affected the supply/demand balance.³ Zimbelman et al evaluated the supply/demand balance of PTs in each state in the USA and reported the possibility of future shortages in the southern and the western parts.⁴ Jesus et al compared the supply/demand balances of PTs in four countries and stated that uniform standards for optimal balances did not exist due to differences in factors such as medical systems. Therefore, those balances should be considered for individual countries or regions.⁵

In Japan, the subcommittee on the supply/demand of PTs and occupational therapists (OTs) in the Ministry of Health, Labour and Welfare (MHLW) reported in 2019 that the supply surpassed the demand and would be 1.5 times the demand in 2040, indicating the possibility of oversupply.⁶ Morii et al predicted the supply/demand of PTs in Japan and indicated that there could be oversupply in the future.⁷ However, Morii et al also stated that the relative disparity in PT supply, such as regional disparity and that among rehabilitation services, should also be considered.

Medical services are provided at universal prices in Japan, according to the national fee table.⁸ Health insurance claims are made for each patient at each institution providing medical services, every month. In recent years, those claims are administered by the government in a database called the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB), which researchers can use with approval.⁹ NDB Open Data Japan is open sourced where the results of basic aggregation using NDB are published.¹⁰ The results of basic aggregation, such as rehabilitation utilisation by prefectures and age groups, are available. Therefore, an analysis of regional differences in the provision of rehabilitation services can be conducted using NDB Open Data Japan. There are few reports that analysed regional differences in the supply/demand or supply/utilisation balance of rehabilitation services using national databases regardless of countries.

Therefore, we analysed the regional differences regarding the provision and the supply/utilisation balances of rehabilitation services, to help provide more uniform and efficient rehabilitation services, and optimal allocation of the related resources.

Methods

Subjects and outcomes

Forty-seven prefectures and the nine regions in Japan, in 2017, where rehabilitation services were provided, were included in the study. The geographical locations and brief description on the characteristics of the prefecture and regions are shown in online supplemental appendix 1. By region, Kanto has the highest population (44 million, including Tokyo prefecture and Yokohama in Kanagawa prefecture), followed by Kinki (22.5 million, including Osaka) and Tokai (15 million, including Nagoya in Aichi prefecture). The supply included PTs, OTs and speech therapists (STs), who were the main rehabilitation providers in Japan.

The primary outcomes were ‘supply/utilisation ratio’ (S/U ratio) calculated by dividing rehabilitation supply converted to service units, by rehabilitation services used (utilisation; see the equation), and ‘utilisation/expected utilisation ratio’ (U/EU ratio), calculated by dividing utilisation by EU. The EU was defined as utilisation expected from the demography in each area multiplied by the average amount of utilisation in overall Japan. These indicators’ details are explained in the following sections.

(1)supply/utilisationratio=SupplyconvertedtothenumberofrehabilitationserviceunitsUtilisation(rehabilitationunitsused)

Supply of rehabilitation

Data regarding the number of rehabilitation providers in 2017 were collected from the Survey of Medical Institution by MHLW,¹¹ defined as the amount of ‘supply’ in this study. The supply was converted to the number of rehabilitation service units under the assumption that one rehabilitation provider worked 251 days per year¹² and provided 15 or 18 units of rehabilitation per day (one unit is 20 min, and one rehabilitation provider providing 18 units is a standard in the national fee table).⁸

Moreover, the number of rehabilitation providers per population and the average number of rehabilitation units per provider per day were calculated.

Utilisation of rehabilitation

‘Utilisation’ and ‘expected amount of utilisation’ (EU) were defined and calculated. The former was defined as the number of rehabilitation service units used in 2017. The data on the utilisation were collected from the NDB Open Data Japan.¹⁰

Utilisation of rehabilitation services included the inpatient and outpatient services listed in online supplemental appendix 2. These services were insured for rehabilitation regardless of professionals such as PTs, OTs and STs. For ‘I007 Psychiatry occupational therapy’, 2 hours—six times the length of other rehabilitation services—was considered the standard time, and 25 patients were considered as the standard number treated at a time. Therefore, that service was counted as the 6th of 25 rehabilitation units per utilisation. ‘H008 Group communication therapy’ was counted as one-third rehabilitation unit per utilisation.

The EU indicated the amount of rehabilitation service utilisation expected from the demography in each region or prefecture in 2017. It was calculated for each region and prefecture by multiplying the population by age groups in each area¹³ with the average amount of utilisation per person in Japan, which was calculated by dividing the total utilisation in Japan by age groups¹⁰ by the population in the corresponding age groups in Japan¹³ (equation 2).

(2)$$\begin{gathered} {\displaystyle \begin{array}{l}\mathrm{E}\mathrm{x}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t} \\ \mathrm{o}\mathrm{f}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{s}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}(\mathrm{E}\mathrm{U})\end{array}=\begin{array}{l}\frac{\sum _{i=1}^2\sum _{j=1}^{18}Utilisatio{n}_{ij}}{\sum _{i=1}^2\sum _{j=1}^{18}populatio{n}_{ij}\times \frac{utilisationinJapa{n}_{ij}}{populationinJapa{n}_{ij}}}\end{array}} \end{gathered}$$

where: i=sex, j=age group, utilisation=utilisation of rehabilitation services in a region or prefecture, population=population in a region or prefecture, utilisation in Japan=utilisation of rehabilitation services in overall Japan, population in Japan=population in overall Japan.

The S/U ratio and the U/EU ratio of the overall rehabilitation services were calculated for each region and prefecture. The S/U ratio was calculated by dividing the supply by the utilisation. The U/EU ratio was calculated by dividing utilisation by EU. U/EU ratio was also calculated for disease-specific rehabilitation service. Disease-specific rehabilitation included the five major services numbered H000–H003 in online supplemental appendix 2.

Moreover, to clarify the amount and the time trend of the utilisation, the amount and the increased rates of actual utilisation from the previous year were calculated using data of NDB Open Data Japan.¹⁰ Before 2015, rehabilitation for disuse syndrome was part of cerebrovascular disease rehabilitation (ie, rehabilitation for disuse syndrome was insured for patients with cerebrovascular disease who had disuse syndrome before 2015). Therefore, for the data from 2014 to 2015, cerebrovascular disease rehabilitation was provided for patients with disuse syndrome, and was counted as disuse syndrome rehabilitation service.

Patient and public involvement

No patients were involved.

Results

S/U ratio

The results of the S/U ratio are shown in figure 1A,B and tables 1 and 2. The S/U ratios in overall Japan were 1.19 and 1.42 when one rehabilitation provider provided 15 and 18 units of rehabilitation per day, respectively.

Figure 1. Supply-related indicators by prefectures. (A,B) The supply/utilisation ratio when one rehabilitation provider provided 15 and 18 units of rehabilitation per day, (C) the average number of rehabilitation personnel and (D) the average number of rehabilitation units per personnel by prefectures.

Table 1

The supply/utilisation (S/U)-related indicators by prefectures

Prefecture	S/U ratio by number of rehabilitation units/personnel×day		U/EU ratio	Number of personnel/1000 population	Rehabilitation units/personnel×day	Region
	15 units	18 units
Hokkaido	1.22	1.46	1.13	1.37	12.3	Hokkaido
Aomori	1.31	1.58	0.71	0.97	11.4	Tohoku
Iwate	1.37	1.64	0.64	0.94	10.9	Tohoku
Miyagi	1.25	1.50	0.69	0.78	12.0	Tohoku
Akita	1.52	1.82	0.48	0.87	9.9	Tohoku
Yamagata	1.32	1.58	0.78	1.12	11.4	Tohoku
Fukushima	1.32	1.59	0.72	0.96	11.3	Tohoku
Ibaraki	1.28	1.53	0.77	0.90	11.8	Kanto
Tochigi	1.06	1.27	0.89	0.84	14.2	Kanto
Gumma	1.22	1.47	0.99	1.14	12.3	Kanto
Saitama	1.11	1.33	0.90	0.82	13.5	Kanto
Chiba	1.03	1.23	1.02	0.90	14.6	Kanto
Tokyo	1.10	1.32	0.91	0.79	13.7	Kanto
Kanagawa	1.11	1.33	0.73	0.67	13.5	Kanto
Niigata	1.45	1.73	0.56	0.85	10.4	Hokuriku
Toyama	1.44	1.73	0.71	1.06	10.4	Hokuriku
Ishikawa	1.32	1.59	0.93	1.17	11.4	Hokuriku
Fukui	1.30	1.56	0.97	1.26	11.6	Hokuriku
Yamanashi	1.16	1.39	1.28	1.49	12.9	Kanto
Nagano	1.38	1.65	0.93	1.35	10.9	Hokuriku
Gifu	1.02	1.23	0.82	0.81	14.7	Tokai
Shizuoka	1.06	1.28	0.96	0.97	14.1	Tokai
Aichi	0.92	1.11	1.14	0.85	16.2	Tokai
Mie	1.10	1.32	0.76	0.80	13.7	Tokai
Shiga	1.25	1.50	0.82	0.87	12.0	Kinki
Kyoto	1.25	1.50	1.01	1.16	12.0	Kinki
Osaka	1.16	1.40	1.05	1.06	12.9	Kinki
Hyogo	1.19	1.43	1.04	1.15	12.6	Kinki
Nara	1.21	1.45	1.00	1.16	12.4	Kinki
Wakayama	1.07	1.28	1.16	1.30	14.1	Kinki
Tottori	1.55	1.86	1.06	1.72	9.7	Chugoku
Shimane	1.50	1.80	0.78	1.32	10.0	Chugoku
Okayama	1.37	1.64	1.00	1.35	10.9	Chugoku
Hiroshima	1.16	1.40	1.13	1.23	12.9	Chugoku
Yamaguchi	1.44	1.73	1.02	1.59	10.4	Chugoku
Tokushima	1.39	1.66	1.37	2.02	10.8	Shikoku
Kagawa	1.21	1.45	1.17	1.45	12.4	Shikoku
Ehime	1.17	1.40	1.18	1.46	12.8	Shikoku
Kochi	1.40	1.68	1.96	3.10	10.7	Shikoku
Fukuoka	1.29	1.55	1.52	1.75	11.6	Kyusyu
Saga	1.14	1.36	1.86	2.09	13.2	Kyusyu
Nagasaki	1.23	1.48	1.47	1.89	12.2	Kyusyu
Kumamoto	1.38	1.66	1.37	1.94	10.8	Kyusyu
Oita	1.26	1.52	1.55	2.07	11.9	Kyusyu
Miyazaki	1.40	1.68	0.91	1.33	10.7	Kyusyu
Kagoshima	1.10	1.31	1.99	2.29	13.7	Kyusyu
Okinawa	1.28	1.54	1.62	1.58	11.7	Kyusyu

U/EU, utilisation/expected utilisation.

Table 2

Results regarding the supply/utilisation (S/U)-related indicators by region

Region	S/U ratio per personnel×day (15 units per personnel)	S/U ratio per personnel×day (18 units per personnel)	U/EU ratio	U/EU for H000 Cardiac	U/EU for H001 Cerebrovascular disease	U/EU for H001–2 Disuse syndrome	U/EU for H002 Musculoskeletal disorder	U/EU for H003 Respiratory disorder
Hokkaido	1.22	1.46	1.13	1.30	1.26	1.29	1.12	0.99
Tohoku	1.33	1.59	0.68	0.56	0.78	0.86	0.62	0.65
Kanto	1.11	1.33	0.86	0.88	0.89	1.15	1.05	1.04
Hokuriku	1.38	1.66	0.79	0.85	0.80	0.78	0.80	0.80
Tokai	0.99	1.19	1.00	0.95	0.92	0.61	1.29	1.00
Kinki	1.18	1.42	1.03	1.16	1.00	1.28	1.15	0.94
Chugoku	1.33	1.59	1.03	0.92	1.06	0.54	1.25	1.20
Shikoku	1.29	1.54	1.37	0.92	1.38	1.13	1.69	1.08
Kyusyu	1.26	1.51	1.53	1.53	1.38	1.13	1.99	1.76

U/EU, utilisation/expected utilisation.

When one rehabilitation provider provided 15 units, the S/U ratios were the lowest in Aichi (0.92), Gifu (1.02) and Chiba prefectures (1.03). By region, the ratios were the lowest in Tokai (0.99) and Kanto (1.11). The ratios were the highest in Tottori (1.55), Akita (1.52) and Niigata prefectures (1.45). By region, the S/U ratios were the highest in Hokuriku (1.38), Chugoku and Tohoku (1.33), and Shikoku (1.29).

When one rehabilitation provider provided 18 units, the S/U ratios were the lowest in Aichi prefecture (1.11) and highest in Tottori prefecture (1.86). By region, the ratio was the highest in Hokuriku (1.66), Chugoku and Tohoku (1.59, respectively), and Shikoku (1.54).

U/EU ratio

The results regarding the U/EU ratio are shown in figure 2 and tables 1 and 2. The ratio was the highest in Kyusyu and Shikoku regions (1.37), and lowest in Tohoku (0.68) and Hokuriku regions (0.79).

Figure 2. The utilisation/expected utilisation ratio by prefectures.

The number of rehabilitation providers per population and the number of rehabilitation units per provider

The numbers of rehabilitation providers per population and rehabilitation units per provider are shown in figure 1C,D and table 1.

The number of rehabilitation providers per 1000 population was approximately 1.10. The number was highest in Shikoku (1.87) and Kyusyu regions (1.84), and lowest in Kanto (0.81), Tokai (0.87), Hokuriku (0.87) and Tohoku regions (0.92). The number of rehabilitation providers per 1000 population was 1.37 in Hokkaido, 1.10 in Kinki and 1.38 in Chugoku regions.

The overall number of rehabilitation units per provider per day in Japan was 12.6. The number was highest in Tokai and Kanto regions, and lowest in Tohoku, Shikoku and Chugoku regions (table 1).

Disease-specific rehabilitation

The U/EU ratio of disease-specific rehabilitation services are shown in figure 3A–D and table 2. Specifically, for cerebrovascular disease and musculoskeletal disorder rehabilitation, the U/EU ratio was the highest in Kyusyu and Shikoku regions, and lowest in Tohoku and Hokuriku regions. Kyusyu region had the highest U/EU ratios for all the disease-specific rehabilitation services other than for disuse syndrome.

Figure 3. The utilisation/expected utilisation (U/EU) ratio of disease-specific rehabilitation by prefectures. The results of the U/EU ratio of (A) musculoskeletal disorder rehabilitation, (B) cerebrovascular disease rehabilitation, (C) disuse syndrome rehabilitation and (D) cardiac disorder rehabilitation.

For musculoskeletal disorder rehabilitation, the U/EU ratio was the highest in Kyusyu and Shikoku regions (1.99 and 1.69, respectively), and lowest in Tohoku and Hokuriku (0.62 and 0.80, respectively) regions.

For cerebrovascular disease rehabilitation, the U/EU ratio was the highest in Hokkaido, Kyusyu and Shikoku regions (1.26, 1.99 and 1.69, respectively), and lowest in Tohoku and Hokuriku (0.78 and 0.80, respectively) regions.

For disuse syndrome rehabilitation, the U/EU ratio was highest in Hokkaido and Kinki regions (1.29 and 1.28, respectively), and lowest in Tokai and Chugoku regions (0.61 and 0.54, respectively). By prefectures, the ratio was highest in Okinawa and Kochi (2.56) and Nara (1.62), while it was lowest in Okayama (0.34), Aichi (0.37) and Hiroshima (0.41).

For cardiac disorder rehabilitation, the U/EU ratio was highest in Kyusyu and Hokkaido regions (1.53 and 1.30, respectively), and lowest in the Tohoku region (0.56).

Regarding the amount of utilisation of disease-specific rehabilitation, musculoskeletal disorder rehabilitation accounted for 45.6% of the disease-specific rehabilitation services (H000–H003, H007–H008 in online supplemental appendix 2), followed by cerebrovascular disease rehabilitation (38.6%), disuse syndrome rehabilitation (9.1%), respiratory disorder rehabilitation (2.5%), cardiac disorder rehabilitation (2.2%), cancer rehabilitation (1.3%) and paediatric rehabilitation (0.8%). Cerebrovascular disease and musculoskeletal disorder rehabilitation together accounted for 84% of the disease-specific rehabilitation services.

The results of the time trend are shown in online supplemental figure 1. While the increase rate from the previous year was 1%–3% for cerebrovascular disease, cerebrovascular disease and respiratory disorder rehabilitation, the rate was more than 5% for disuse syndrome, cardiac disorder and cancer rehabilitation.

Discussion

This study analysed the regional differences in the supply/utilisation balance and provision of rehabilitation services by prefectures and regions using NDB Open Data Japan.

In Tokai and Kanto regions, the S/U ratio was the lowest, close to 1. The smaller number of rehabilitation providers per population contributed to little supply surplus. Morii et al¹⁴ compared the regional supply disparity between healthcare occupations in Hokkaido prefecture and reported that there is a possibility of maldistribution of PTs and OTs by smaller number of providers in rural areas within the prefecture. On the contrary, in this study at between-prefectural level, the results indicated that there was less surplus of supply in populated prefectures such as Tokyo and Kanagawa (Kanto region) and Aichi (Tokai region). Since it is desirable to have a certain amount of surplus in supply considering uncertainty in service utilisation, the amount of supply should be increased in these regions, ideally from regions where there is large amount of supply surplus.

Possible ways to increase the supply are adjusting the number of schools or enrolments, decreasing the attrition rates in less-supplied regions (supply-related parameters have been shown in the PT career path model by Morii et al⁷), and an implementation of region-specified enrolments at medical schools in which enrollees are required to work in a prespecified region or prefecture for several years after graduation. Such a system has been implemented for physicians for some schools in Japan.¹⁵

The U/EU ratio was higher mostly in the western part of Japan, such as Kyusyu and Shikoku regions, and lower mostly in the eastern part, such as Tohoku and Hokuriku regions. The result was consistent with the analysis of regional differences in medical expenditure by the MHLW, which indicates that adjusted per-population medical expenditure was greater mostly in the western part of Japan, especially for inpatient medical services.¹⁶ As is shown in online supplemental appendix 1 and table 1, medical resource density, both in general and for rehabilitation, was higher in the western part. Relationships between medical resource density and medical service utilisation have been shown in Japan for outpatient services for chronic disease¹⁷ and inpatient services and outpatient services in general¹⁸ and elsewhere.^{19 20} Izumida et al¹⁸ reported that this relationship can be caused by supplier-induced demand and by patient factors such as consultation rates, and their results indicated a presence of supplier-induced demand. Although the results were not for rehabilitation services, but for inpatient services and outpatient services in general, the trend can be applicable to rehabilitation services since rehabilitation services are provided along with other services, and not in isolation. However, it is still unclear whether the result was attributed to the regional differences in the consultation rates or the number of rehabilitation services provided per patient, or whether the regional differences are explained by supplier-induced demand. It is necessary to clarify whether there is supplier-induced demand on rehabilitation service utilisation, and further analyse it to avoid unnecessary and inefficient treatments.²¹

Economic factors could also influence the utilisation of rehabilitation services. In Japan, Babazono et al reported that average monthly salary had a high positive correlation with outpatient service indicators,²² indicating the possibility of low-income patients being discouraged from seeking medical services. Since the U/EU ratio was the lowest in Tohoku region, where average income is also low, it is possible that economic status could have influenced the utilisation of rehabilitation services.

Although it is difficult to clarify the right amount of rehabilitation required from this study, the result of U/EU ratios indicates that there are regional differences in the provision of rehabilitation services in Japan. Although S/U ratio was higher in Tohoku and Hokuriku regions, which could indicate a seemingly surplus in supply, it may be attributed to the lower amount of utilisation since the amount of supply in these areas was not large (table 2). It is necessary to ensure that sufficient amount of rehabilitation is provided through further research, such as detailed analysis on disease-specific rehabilitation services, in Tohoku and Hokuriku regions since shortages of rehabilitation could lead to worse patient outcomes. It is also necessary to consider whether the amount of rehabilitation provided in the western part of Japan, with larger amount of utilisation and supply, is appropriate from the perspectives of uniform and efficient rehabilitation.

Regarding disease-specific rehabilitation, whereby the results of U/EU showed mostly the same trend as that of the overall rehabilitation, priority should be given to cerebrovascular disease and musculoskeletal disorder rehabilitation since these services accounted for 84% of the disease-specific rehabilitation services. As online supplemental appendix 2 shows, although these services are reimbursed for patients with various types of diseases, data from NDB Open Data Japan do not include what disease each patient using these services has. Therefore, further detailed analysis needs to be conducted for individual diseases, treatments and treatment phases to clarify what exactly contributed to the regional differences, for instance, using NDB database.

Nakanishi et al conducted a survey in 597 hospitals in Japan and reported that even in hospitals providing advanced medical care, cardiac rehabilitation was not sufficiently common.²³ Goto reported that the reasons for not providing cardiac rehabilitation were a shortage of staff, requirements for institutions and proficiency for hospitals.²⁴ The results indicate that the low amount of supply in Tohoku region could lead to the low amount of utilisation of cardiac disorder rehabilitation.

In the Japanese fee schedule, the convalescent rehabilitation ward fee has been insured for patients for whom cardiac disorder rehabilitation is provided since 2022, meaning providing more intensive rehabilitation for those patients will be expected.⁸ Jesus et al reported that not considering unmet needs could affect the amount of required rehabilitation services.⁵ Therefore, it is important to consider unmet needs not only for cardiac rehabilitation but also rehabilitation services overall, and this analysis needs to be refined whenever unmet needs are clarified.

Unlike the trend for the overall rehabilitation services, the U/EU ratio was not different by region but by prefectures for disuse syndrome rehabilitation. This could be because of the differences between prefectures in the root diseases from which disuse syndrome arose, the prevalence and the number of rehabilitation services per patient, and rehabilitation content (other rehabilitation services such as cerebrovascular disease rehabilitation can be reimbursed for patients with disuse syndrome depending on the disease one has). The causes of the regional differences should be clarified by further analysis, such as one for patients with individual conditions who have disuse syndrome, since disuse syndrome rehabilitation can be used by patients with a wide range of root conditions.

This is one of the few studies that analysed regional differences in the supply/utilisation balance and the provision of rehabilitation services using data from a national database in Japan, which is one of the most ageing countries in the world.²⁵ Rehabilitation services will also be of greater importance in other rapidly ageing countries, and its demand will continue to increase; our methodology is applicable to other countries to consider the provision of more uniform and efficient rehabilitation services although design and interpretation should be made as per the context.

However, this study has several limitations. First, subject regional units were prefectures and regions due to a lack of data granularity. Analysis of the finer regional units needs to be conducted as per requirement. Next, since NDB does not include health insurance claims of publicly funded healthcare (eg, healthcare for welfare recipients), the calculated utilisation could have underestimated the actual amount of utilisation by a few per cent. In addition, data from NDB indicate utilisation, not the need for rehabilitation services. Therefore, it is out of scope of this study whether the current rehabilitation supply meets the needs for rehabilitation services or to assess what amount of service utilisation is appropriate. Finally, as shown in online supplemental appendix 2, some rehabilitation services can be provided for patients with a wide range of conditions. Therefore, it is difficult to determine rehabilitation for which specific disease or condition contributed to the results. Further analysis targeting individual conditions should be conducted.

Conclusion

This study analysed the regional differences in the supply/utilisation balance and the provision of rehabilitation services using NDB Open Data Japan to help consider provision of more uniform and efficient rehabilitation services.

The results showed that the S/U ratio was highest in Shikoku, Kyusyu, Tohoku and Hokuriku regions, while it was lowest in Kanto and Tokai regions primarily due to smaller number of rehabilitation providers, which led to little surplus of supply. The U/EU ratio was lowest in Tohoku and Hokuriku regions, indicating smaller amount of utilisation in these regions. Higher S/U ratios in these regions were attributed to the smaller amount of utilisation since the amount of supply was relatively small.

For further analysis on regional differences, priority should be given to cerebrovascular disease rehabilitation and musculoskeletal disorder rehabilitation, which accounted for 84% of the disease-specific rehabilitation services. The U/EU ratio for these rehabilitation services showed the same trend as overall rehabilitation services, higher mostly in the western part of Japan, while lower mostly in the eastern part.

We would like to thank Editage (www.editage.jp) for English language editing.

Data availability statement

Data are available upon reasonable request. The dataset(s) supporting the conclusions of this article is available upon reasonable request.

Ethics statements

Patient consent for publication

Not required.

Ethics approval

Not applicable.

Footnote

Contributors YM, KA, TI, KF, KK and KO made substantial contributions to the conception and design of the work. YM and TI made substantial contributions to the acquisition of data and the analysis. KA made substantial contributions by providing advice from a clinical perspective. TI, KF, KK and KO made substantial contributions by providing advice from human healthcare resourcing. YM drafted the work, and KA, TI, KF, KK and KO substantially revised it. All authors read and approved the final manuscript. KA was responsible for the overall content as guarantor.

Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Map disclaimer The inclusion of any map (including the depiction of any boundaries therein), or of any geographic or locational reference, does not imply the expression of any opinion whatsoever on the part of BMJ concerning the legal status of any country, territory, jurisdiction or area or of its authorities. Any such expression remains solely that of the relevant source and is not endorsed by BMJ. Maps are provided without any warranty of any kind, either express or implied.

Competing interests None declared.

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

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Ministry of Economy. A report of what insurers and firms needs to do for community Healthcare. 2015. Available: https://www.meti.go.jp/committee/kenkyukai/sansei/chiiki_iryo/report_001.html [Accessed 21 Oct 2021 ].

² World Health Organization. Rehabilitation 2030: a call for action. 2017. Available: https://www.who.int/publications/i/item/9789240007208 [Accessed 08 Apr 2023 ].

³ Landry MD, Hack LM, Coulson E, et al. Workforce projections 2010–2020: annual supply and demand forecasting models for physical therapists across the United States. Phys Ther 2016; 96: 71–80. doi:10.2522/ptj.20150010

⁴ Zimbelman JL, Juraschek SP, Zhang X, et al. Physical therapy workforce in the United States: forecasting nationwide shortages. PM R 2010; 2: 1021–9. doi:10.1016/j.pmrj.2010.06.015

⁵ Jesus TS, Koh G, Landry M, et al. Finding the "right-size" physical therapy workforce: international perspective across 4 countries. Phys Ther 2016; 96: 1597–609. doi:10.2522/ptj.20160014

⁶ The Ministry of Health, Labour and Welfare. Plans on the workforce of physical therapists and occupational therapists reflecting the supply/demand projection. 2019. Available: https://www.mhlw.go.jp/stf/shingi/other-isei_348780.html [Accessed 21 Oct 2021 ].

⁷ Morii Y, Ishikawa T, Suzuki T, et al. Projecting future supply and demand for physical therapists in Japan using system dynamics. HPT 2019; 8: 118–27. doi:10.1016/j.hlpt.2019.05.003

⁸ The Ministry of Health, Labour and Welfare. Revision of medical fees. 2022. Available: https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000188411_00037.html [Accessed 09 Apr 2022 ].

⁹ The Ministry of Health, Labour and Welfare. Roles and usage environment of NDB and Kaigo DB. 2018. Available: https://www.mhlw.go.jp/stf/shingi2/0000206551.html [Accessed 21 Oct 2021 ].

¹⁰ The Ministry of Health, Labour and Welfare. NDBDB open data Japan. Available: https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000177182.html [Accessed 21 Oct 2021 ].

¹¹ The Ministry of Health, Labour and Welfare. Survey of medical institutions. 2017. Available: https://www.mhlw.go.jp/toukei/saikin/hw/iryosd/19/ [Accessed 21 Oct 2021 ].

¹² The Ministry of Health, Labour and Welfare. Basic survey on wage structure. 2017. Available: https://www.mhlw.go.jp/toukei/youran/indexyr_d.html [Accessed 21 Oct 2021 ].

¹³ The Ministry of Internal Affairs and Communications. Future population projections. 2017. Available: http://www.stat.go.jp/data/jinsui/ [Accessed 21 Oct 2021 ].

¹⁴ Morii Y, Ishikawa T, Tsuji S, et al. A comparison of distribution equality between health care occupations in Hokkaido. J Jpn Assoc Med Inf 2018; 37: 285–9. doi:10.14948/jami.37.285

¹⁵ Ministry of Health Labour and Weifare. Subcomittee on supply/demand of physicians. summary of region-specific enrollments. 2019. Available: https://www.mhlw.go.jp/stf/shingi/other-isei_318654.html [Accessed 13 May 2023 ].

¹⁶ The Ministry of Health, Labour and Welfare. Analysis on regional differences in medical expenditures Fy2020 (in Japanese). Available: https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/kenkou_iryou/iryouhoken/database/iryomap/index.html [Accessed 13 May 2023 ].

¹⁷ Sekimoto M, Ii M. Supplier-induced demand for chronic disease care in Japan: multilevel analysis of the association between physician density and physician-patient encounter frequency. Value Health Reg Issues 2015; 6: 103–10. doi:10.1016/j.vhri.2015.03.010

¹⁸ Izumida N, Nakanishi S, Urushi H. New entrance regulation of physicians and medical expenditures:expenditure function approach for physician inducement demand hypothesis. Iryoto Shakai 1999; 9: 59–70. doi:10.4091/iken1991.9.1_59

¹⁹ Scott A, Shiell A. Analysing the effect of competition on general practitioners' behaviour using a multilevel modelling framework. Health Econ 1997; 6: 577–88. doi:10.1002/(sici)1099-1050(199711)6:6<577::aid-hec291>3.0.co;2-y

²⁰ Dzampe AK, Takahashi S. Competition and physician-induced demand in a Healthcare market with regulated price: evidence from Ghana. Int J Health Econ Manag 2022; 22: 295–313. doi:10.1007/s10754-021-09320-7

²¹ Rodriguez Santana I, Mason A, Gutacker N, et al. Need, demand, supply in health care: working definitions, and their implications for defining access. Health Econ Policy Law 2023; 18: 1–13. doi:10.1017/S1744133121000293

²² Babazono A, Kuwabara K, Hagihara A, et al. Does income influence demand for medical services despite Japan’s "health care for all" policy. Int J Technol Assess Health Care 2008; 24: 125–30. doi:10.1017/S0266462307080166

²³ Nakanishi M, Nagayama M, Adachi H, et al. Change in implementation of cardiac rehabilitation for acute myocardial infarction in Japan: a nationwide survey. Card Rehabl 2011; 16: 188–92. Available: https://search-jamas-or-jp.ezoris-hokudai.idm.oclc.org/search/do

²⁴ Goto Y. Current status and future perspective of cardiac rehabilitation in Japan. J Jpn Coron Assoc 2015; 21: 58–66. doi:10.7793/jcoron.jcoron.21.006

²⁵ The World Bank. Population ages 65 and above (% of total population). 2022. Available: https://data.worldbank.org/indicator/SP.POP.65UP.TO.ZS?end=2021&start=1960&view=chart [Accessed 13 May 2023 ].