Medical cost analysis of family psychoeducation for schizophrenia
Yoshio Mino, MD, Mental Health Section, School of Human Science, Osaka Prefecture University, 1-1 Gakuen-cho, Sakai, Osaka 599–8531, Japan. Email: email@example.com
Abstract Family psychoeducation has been shown to prevent the relapse of schizophrenia. However, whether medical costs are reduced by this approach remains uncertain. The subjects were patients with schizophrenia who lived with high-expressed emotion (EE) families and were at high risk of relapse. A total of 30 patients whose families underwent psychoeducation and intensive family sessions or psychoeducation and subsequent support were regarded as the psychoeducation group. A high-EE group without family psychoeducation made up of 24 patients was used as a control group. The mean outpatient medical cost, duration of hospitalization, inpatient medical cost, and total medical cost during the follow-up period were compared between the psychoeducation group and the control group. The mean inpatient medical cost was ¥270 000 in the psychoeducation group and ¥470 000 in the control group. The mean total medical costs were ¥500 000 in the psychoeducation group and ¥710 000 in the control group. The cost in the psychoeducation group was significantly lower than the control group by Mann–Whitney U-test. The proportion of patients with a total medical cost greater than the median value was 23% in the psychoeducation group and 54% in the control group with a significant difference. The medical cost can be reduced in the psychoeducation group compared with the control group due to the prevention of re-hospitalization by family psychoeducation.
In Japan, the total medical cost in 1998 was ¥29 825 100 000 000 and showed an increase of ¥760 000 000 000 (2.6%) compared with the previous year, and its percentage of the national income became 7.86% (7.41% in the previous year).1
The burden of medical cost has become a major social issue. In these circumstances, efficiency of medical care has been pursued, and evidence-based medicine (EBM) has been proposed. Concerning mental disorders, the medical cost for “mental and behavioral disorders” accounts for 8.7% of the total medical cost of age levels below 65 years in 1998, and cost-conscious approaches are needed also in the field of mental health. Also, according to a patient survey in 1996, the numbers of patients with mental disorders were 720 000 for schizophrenia, 430 000 for mood disorders including depression, and 470 000 for neurotic disorders, all having increased compared with a comparable survey in 1984.1 Therefore, the medical cost of mental health may continue to increase in Japan.
Internationally, 3% of the expenditure of the national health service (NHS) is allotted to schizophrenia in the United Kingdom,2 and 1.5–3% of the national health expenditures are spent on schizophrenia in the Netherlands and the USA.3,4 The cost of the treatment for schizophrenia has become difficult to ignore worldwide.
Studies of the effect of expressed emotion (EE) of the family on the course of schizophrenia began in Western countries5 and have also been conducted in Japan.6–9 On the basis of the results obtained, family psychosocial intervention and psychoeducation have been attempted, and their effects have been confirmed.10–12
In these circumstances, the evaluation of psychoeducation of families of patients with schizophrenia from the viewpoint of health economics has become necessary, but few such studies have been conducted in Japan or abroad.13–15 Also, to date, studies have been carried out primarily in Western countries, and few have been conducted in Japan. Since the inpatient medical cost is lower in Japan than in Western countries, the prevention of hospitalization by psychoeducation may not lead to a reduction in the medical cost. This study was carried out to evaluate the effect of psychoeducation for families of patients with schizophrenia from the viewpoint of health economics. At present, Japan’s Health Insurance system does not pay for family psychoeducation for schizophrenia. If a result can be obtained whereby medical costs in the intervention group are smaller than those in the control group, it could be suggested that the cost difference should be paid for family psychoeducation.
The subjects were 30 patients with schizophrenia who lived with high-EE families and were at high risk for recurrence, and who were admitted to the psychiatric department of Kochi Medical School, Kochi, and to the affiliated Tosa Hospital, Kochi, Japan, between October 1994 and November 1997. Inclusion criteria were: (i) 15–65 years of age; (ii) a diagnosis of schizophrenia made at discharge according to the Diagnostic Statistical Manual for Mental Disorders, version IV and International Classification for Diseases version 10; (iii) living within approximately 15 km from the hospitals, which facilitates home visiting; (iv) residing with their family for 3 months or more before admission; and (v) expected to live with their family after discharge. The effects of psychoeducation were evaluated. A total of 30 patients whose families underwent psychoeducation and intensive family sessions or psychoeducation and subsequent support were regarded as the psychoeducation group,12 and a high-EE group of 24 patients in the authors’ past cohort study was used as the control group.6 The patients in the control group were admitted to the same hospitals and satisfied the same inclusion criteria except for (iii). No particular approach was made to the families of the control group.
Initially, the purpose and the design of the study were explained to the patients and family members, and their consent to participate in the study was confirmed, using a signed form (psychoeducation group) and interview (control group).
The medical cost during the 9 months after discharge was compared between the two groups. The outpatient medical cost during the 9 months after discharge was calculated monthly, and, if the patients were re-admitted, the duration of hospitalization in days and the inpatient medical cost during the follow-up period were calculated. Medical costs for physical disorders were excluded. Medical cost was calculated using medical bills which are reported from the hospitals monthly. The intervention group was treated in 1994, and the control group in 1991–1992. There was no difference in the treatment strategy, and in the method of calculation of medical cost. Atypical antipsychotic drugs were not used for the patients of both groups.
The outpatient medical cost, duration of hospitalization during the follow-up period, inpatient medical cost and total medical cost were compared between the psychoeducation group and control group using Mann–Whitney U-test, because of their non-normal distributions. Also, in consideration of the distribution of the total medical cost, the subjects were divided into two groups at the median value of the total medical cost, and the percentage of patients in whom the medical cost was higher than the median value was compared using the χ2 test.
Characteristics of the patients were compared between the groups. There were no significant differences in: (i) age at admission; (ii) gender; (iii) marital status; (iv) duration of the illness; (v) proportions of subtypes of schizophrenia; and (vi) mean Brief Psychiatric Rating Scale scores at admission and at discharge. There was a significant difference between the groups regarding the number of previous admissions: 7.4 ± 6.6 (mean ± SD) in the psychoeducation group and 9.9 ± 7.2, P = 0.018 by Welch’s test.
As reported previously,12 the relapse risk during the 9 months after discharge was 30% (9/30) in the psychoeducation group, and 58.3% (14/24) in the control group. Although the number of days of hospitalization during the 9 months was smaller in the psychoeducation group, the difference was not significant (Table 1).
Table 1. Comparison of hospital days during 9 months
Table 2 compares the medical costs. The outpatient medical cost was not significantly different between the two groups. The mean inpatient medical cost was ¥270 000 in the psychoeducation group and ¥470 000 in the control group. The mean total medical cost was ¥500 000 in the psychoeducation group and ¥710 000 in the control group. The total medical cost in the psychoeducation group was significantly lower than that in the control group by Mann–Whitney U-test.
Table 2. Comparison of medical costs during 9 months (Yen)
| Psychoeducation group||30||228 264.2|| 8 165–926 960||0.07|
| Comparison group||24||239 139.9||121 892–281 290||–|
| Psychoeducation group||30||273 777.8||0–2 123 333||0.06|
| Comparison group||24||472 013.9|| 0–1 785 000||–|
| Psychoeducation group||30||502 042.0|| 11 696–2 334 333||0.01|
| Comparison group||24||711 153.8||281 290–1 906 892||–|
The median value of the total medical cost in the 54 patients with the psychoeducation and control groups combined was calculated. Table 3 compares the proportion of patients in whom the total cost was above the median value between the two groups. This proportion was 23% in the psychoeducation group and 54% in the control group, with a significant difference.
Table 3. Comparison of proportion of median or more total cost
Four health economics studies of family psychoeducation or psychosocial intervention in patients with schizophrenia have been reported: two from the United Kingdom, one from China, and one from Norway. According to a study in the United Kingdom, the monthly direct medical cost was £1171 (about ¥235 000) in the group that received family intervention, and £1603 (about ¥320 000) in the group that did not receive it during the observation period. The medical cost was able to be reduced in the intervention group because re-hospitalization was prevented.13 A previous study in the United Kingdom also reported similar findings.14 In Norway, the direct cost in the 12 patients of the intervention group during the observation period was 5340 000 krones (¥72 000 000) less than that in the 12 patients of the control group.16 According to a Chinese study, the period of re-hospitalization was shortened, and the medical cost per patient could be reduced by $US170 per year by intervention.15
The results of these studies suggest that family psychoeducation reduces the total medical cost of patients with schizophrenia by preventing their re-hospitalization and reducing the cost of hospitalization. In the present study in Japan, the total medical cost was reduced by the prevention of re-hospitalization. Since, in general, the average inpatient medical cost is smaller in Japan than in Western countries, it was thought that the effect of prevention of rehospitalization on medical cost could be small. However, the total medical cost was significantly reduced by family psychoeducation. This finding suggests that family psychoeducation can reduce the medical cost of schizophrenia in countries where inpatient care is not very expensive.
In Japan, the medical actions that are covered by the insurance policy are specified, and the compensation for each medical action is determined by the national government under the present insurance-for-all-people system. Presently, family psychoeducation for schizophrenia is not included in the insured medical actions, so it is given to only a small proportion of patients at most medical institutions incorporated in the insurance system.
Most medical organizations in Japan are privately run, and medical actions not covered by the insurance policy are difficult to perform in private organizations. Family psychoeducation clearly improves the quality of life (QOL) of patients and their families, and it should be made available to more patients and families by insurance coverage. According to the results of this study, the mean medical cost during the 9 months after discharge was ¥500 000 in the psychoeducation group but ¥710 000 in the control group. Therefore, psychoeducation is considered to reduce the total medical cost if its cost is less than ¥210 000. Also, in the newly determined insurance compensation for psychoeducation, it will not increase the total cost of insurance if it is less than ¥210 000. This estimation is based on the data during the 9 months after discharge, but cost analysis over a longer time span may warrant the allotment of an even greater cost for psychoeducation. In order to improve the QOL of patients and their families, medical insurance should cover family psychoeducation for schizophrenia as a paid medical action.
Only the medical cost was evaluated in this study. However, if other costs are also taken into consideration, a greater cost for psychoeducation may eventually lead to a cost reduction from the viewpoint of the whole community, because the repeated recurrence of schizophrenia means increases in the cost of welfare and other services, in addition to medical costs.
Health economics research for psychoeducation has just begun. Estimation of the cost of psychoeducation has been limited to the direct cost or medical cost. A more comprehensive comparison including indirect cost is needed. Cost-effectiveness analysis, cost-utility analysis, and cost-benefit analysis will also be required in the future.
In this study, patients in the intervention group were treated during 1994–1997, and those in the control group during 1991–1992. This discrepancy might cause differences in medical costs. Although the difference in the treatment strategies could cause medical cost change, there was no treatment strategy change between the periods. Atypical antipsychotics were not used for the subjects.
Average treatment costs per day for inpatients with schizophrenia in the two hospitals were ¥8400 during 1991–1992 and ¥10 713 during 1994–1997. The difference was caused by revision of the medical fee schedule by the government. According to the results of the current study, medical cost for the psychoeducation group was lower than the control group, although the psychoeducation was conducted during the period with higher average inpatient cost for schizophrenia. This means the cost difference in the current study was underestimated.
In the current study, the cost analysis was limited to the medical cost. Since the medical cost is a part of the direct cost, the evaluation of indirect costs as well as other direct costs is also needed.
This study was partly supported by the Japan Ministry of Education, Culture, Sports, Science and Technology (grant number 13470090) and the Japan Ministry of Health, Labor and Welfare (Research on Health Technology Assessment 2002). The authors with to thank Professor Yamamoto, Okayama Science University, for his statistical advice, and to Professor Babazono, Kyushu University, for his comments from the viewpoint of health economics.