Impact of the mental health reform in Israel on health care and mortality among people with severe mental illness

doi:10.21203/rs.3.rs-2230826/v1

Download PDF

Research Article

Impact of the mental health reform in Israel on health care and mortality among people with severe mental illness

https://doi.org/10.21203/rs.3.rs-2230826/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 13 Oct, 2023

Read the published version in Social Psychiatry and Psychiatric Epidemiology →

You are reading this latest preprint version

Objective

Reports show disparities in the health care of people with severe mental illness (SMI). Yet, the moderating effect of mental health reforms on the health care disparities remain unexplored. The current study aimed to investigate the outcomes of the mental health reform in Israel on the use of health services among people with SMI.

Method

A case-control epidemiological study comparing the use of health services 3.5 years before and after the mental health reform for service users diagnosed with schizophrenia, schizoaffective disorder, and bipolar disorder. Data on health services included: blood cholesterol test (LDL), hemogalobin-A1C test, and visits to GPs and specialists. Mortality was recorded.

Results

Following the reform the number of visits to GPs was decreased among service users of the three SMI groups, as well as visits to specialists among service users with a schizoaffective or bipolar disorder. Following the reform service users of the three SMI groups showed no-change in the performance of LDL test. Complex findings were noted with regard to the performance of Hemoglobin-A1C test. Mortality rates were higher among service users with SMI and did not change following the reform.

Conclusions

Users of the three SMI groups showed no benefits of the mental health reform in terms of use of health services. Improved health care can be attained by a closer collaboration between the primary physicians and community mental health services.

People with severe mental illness (SMI) including diagnoses of schizophrenia, schizoaffective disorder, and bipolar disorder, are at increased risk for physical comorbidity and premature death [1–5]. Disparities of health care for people with SMI may play a key factor in “the scandal of premature mortality” [6]. Health disparities were reported in the US, including in population groups protected by special insurance systems (e.g., veterans) [7–10], as well as in countries with a universal health coverage such as Australia [11], Canada [12], Denmark [13], Israel [14, 15], New Zealand [16], Sweden [17], and Taiwan [18].

Israel has a national health insurance law that mandates universal and comprehensive health care. Since its legislation in 1995 general health services have been delivered through non-profit health maintenance organizations (HMOs), while mental health services were not included in the basket of services provided by the HMOs, and were provided by the state. In line with the recommendation of the World Health Organization (WHO) to integrate mental health care into general health services in the community [19], in 2015 the mental health reform was implemented and shifted the responsibility of mental health services from the ministry of health to the HMOs.

The integration of mental and general health services were expected to have positive impact firstly on the psychiatric care, including the expansion of ambulatory services, the provision of alternatives for psychiatric hospitalizations and the reduction of the number of unnecessary hospitalizations. In addition, the policy makers expected that the reform would lead to a broader impact, such as reducing the stigma associated with mental illness. Furthermore, they expected that general practitioners (GPs), who are in direct contact with the patients and manage their health care, would enable a better continuum between general and mental care, resulting in improved care on both aspects. To the best of our knowledge, no study on the impact of the mental health reform in Israel was yet published. In the same line, reports summarizing forty years of the mental health reform in Italy (1978) indicated its success in terms of public health concerns, such as lower suicidal rates and less compulsory hospitalizations, but these and other related reports did not relate to general health care aspects [20, 21]. The lack of knowledge on the outcomes of mental health reforms on general health care among persons with SMI is surprising considering the focus given to this aspect by the action plan to reduce care gap in mental health care by the WHO [22].

The aim of the current study is to compare general health care measures provided to persons diagnosed with schizophrenia, schizoaffective disorder, and bipolar disorder, 3.5 years prior to and after the mental health reform in Israel. In addition, this study will compare the consequences of the reform on the mortality rates of persons with the above diagnoses.

Data were gathered using the information systems of Meuhedet Healthcare Services (MHS), the third-largest HMO in Israel serving 1.2 million users. The study was based on a prospective historical design, with data collected from January 1st 2012 to December 31st 2018, representing 3.5 years before and after the implementation of the reform (July 2015). Ethical approval was granted by the head management of MHS. Participants’ identity was undisclosed to the authors.

Identification of service users with SMI. MHS constructed computerized registries of users diagnosed with schizophrenia (ICD-10 code F20), schizoaffective disorder (F25), and bipolar affective disorder (F31). The following inclusion criteria were applied: a. born prior to 1992 (age ≥ 21 at 1.1.20201); b. alive by the year 2012; c. an SMI diagnosis prior to 2012; and d. no change of their respective HMO during the study period.

To generate matched comparison groups we defined an algorithm based on the year of birth, sex, national sector, residential district, and socioeconomic status. This enabled to match between each service user with SMI and five service users with no record of one of the three diagnoses (5:1 ratio). Thus, data analysis was performed on 2,458 service users with schizophrenia (comparison n = 12,290), 1004 with schizoaffective disorder (comparison n = 5,020), and 2,038 with a bipolar disorder (comparison n = 10,190) (Table 1).

Table 1

Sociodemographic characteristics of the sample and clinical data of service users diagnosed with schizophrenia, schizoaffective, or bipolar disorder, and their respective matched comparison groups
	Schizophrenia (n = 2,458)	Schizophrenia Comparison group (n = 12,290)	Schizoaffective disorder (n = 1,004)	Schizoaffective disorder Comparison group (n = 5,020)	Bipolar disorder (n = 2,038)	Bipolar disorder Comparison group (n = 10,190)
Sex (Male)	61.7%	61.7%	53.1%	53.1%	45.7%	45.7%
Age at year 2012^a	44.6 (14.3)	44.5 (14.1)	44.6 (14.7)	44.4 (14.5)	50.1 (16.3)	49.9 (16.1)
National/religious sector Jewish-General Jewish-Orthodox Arab-Israeli	66.0% 24.8% 9.2%	64.6% 24.8% 10.7%	58.0% 33.7% 8.4%	56.2% 33.0% 8.0%	67.4% 28.8% 3.8%	67.1% 28.2% 4.7%
District North Center South Jerusalem district	20.3% 24.8% 30.8% 24.1%	20.4% 24.7% 30.7% 24.2%	18.1% 19.3% 26.8% 35.8%	18.1% 19.0% 27.0% 35.9%	11.7% 28.3% 23.7% 36.3%	11.7% 28.3% 23.7% 36.3%
Socioeconomic index ^a	5.2 (1.8)	5.2 (1.8)	5.1 (1.9)	5.1 (1.9)	5.7 (2.0)	5.7 (1.9)
Age at diagnosis ^a	37.9 (13.9)	--	37.9 (14.4)	--	43.4 (16.0)	--
^a Age, socioeconomic index, and age at diagnosis are presented as mean (SD)

General health services

Information was gathered annually in the years 2012 through 2018.

Socio-demographic: Sex, year of birth, national sector, district of residence, year of death, and socioeconomic status according to the geographic location of the clinic (1–10 scale, 10 indicating higher status).
Medical diagnosis: Data on diabetes, cardiovascular disease (CVD) and cancer were extracted from the service users’ files, including the respective dates of diagnosis using ICD-10 codes. Type 2 diabetes mellitus (E11). CVD diagnoses included: ischemic heart disease (IHD) (I20-I25); congestive heart failure (CHF)(I50); cardiomyopathy (I42); idiopathic hypertrophic sub-aortic stenosis (IHSS)(I42.1-I42.5); pulmonary hypertension (I27); and carotid artery disease (I65.2) (World Health Organization, 1992). Cancer diagnoses included breast (C50), colon (C18), prostate (C61), and rectum (C20), defined as a chronic condition if 5-years had elapsed since diagnosis. Service users with one of the above diagnoses were defined as suffering from a non-psychiatric chronic illness.
Laboratory tests: The number of annual tests was recorded (2012–2018) for blood cholesterol fraction (LDL) and Hemoglobin-A1C. With the increase of awareness to preventive medicine in the context of metabolic syndrome and diabetes such tests are part of the National Quality Indicators Program (NQIP) promoted by the Israeli Ministry of Health (NQIP, 2022). According to NQIP people with SMI should take an annual Hemoglobin-A1C test.
Visits: number of annual visits to GPs and specialists (excluding psychiatrists) were recorded per year (2012–2018).

Data analysis

The associations between SMI groups (schizophrenia, schizoaffective, bipolar vs. respective comparison) and laboratory tests, and visits to GPs and specialists were analyzed using analysis of variance (ANOVA). Univariate analysis was conducted first for all models to test the association between group and medical procedures, and adjusted analysis followed with non-psychiatric chronic illness as a confounding variable. Mortality rates during the follow up period, before and after the reform (1.7.2015) were analyzed using Cox proportional-hazards models to estimate hazard ratios (HR) and 95% confidence intervals (CI). Analysis was performed using the SPSS 25.0 software (IBM Inc., 2017).

Service users diagnosed with schizophrenia

LDL test was recorded more frequently among service users diagnosed with schizophrenia than among counterparts both before and after the reform (Table 2). The mean number of LDL tests did not change following the reform among service users with schizophrenia but was slightly increased in the comparison group. This was supported by a significant group by reform interaction (unadjusted F = 13.5, df = 1, 14476, p < 0.02; adjusted F = 4.0, df = 1, 14473, p = 0.04).

As could be seen in Table 2, Hemoglobin-A1C test was done more frequently among service users diagnosed with schizophrenia. However, when the effect of non-psychiatric chronic illness was added to the model it appeared that Hemoglobin-A1C test was done more frequently only among service users with schizophrenia with no other non-psychiatric chronic illness. In contrast, among service users suffering from a non-psychiatric chronic illness more Hemoglobin-A1C tests were done by service users of the comparison group both before (schizophrenia: 1.7 ± 1.2; comparison: 1.8 ± 1.2) and after the reform (schizophrenia: 1.6 ± 1.2; comparison: 1.8 ± 1.2) compared to service users with schizophrenia. This was supported by a significant group by chronic illness interaction (F = 5.3, df = 1, 14474, p < 0.001), as well as a tendency to significance of the group by chronic by reform interaction (F = 3.8, df = 1, 14473, p = 0.053).

More annual visits to GPs were made by service users with schizophrenia both pre and post-reform than comparison (Table 2). A significant group effect was indicated in both the unadjusted and adjusted models. In addition, while service users with schizophrenia reduced the number of GP visits following the reform, the number of visits was increased among counterparts. This was supported by a significant group by reform interaction (adjusted F = 38.1, df = 1, 14474, p < 0.001).

Less annual visits to specialists were made by service users with schizophrenia both before and after the reform than comparison. A significant group effect was indicated in both the unadjusted and adjusted models.

Table 2

Annual rate of LDL and Hemoglobin-A1C tests and visits before and after the reform among service users with schizophrenia and matched controls
	Schizophrenia		Matched controls		F ^a, p
	Before reform	After reform	Before reform	After reform
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Unadjusted	Adjusted ^b
LDL	0.9 (0.9)	0.9 (1.0)	0.7 (0.8)	0.7 (0.8)	110.0, p < .001	26.2, p < .001
Hemoglobin-A1C	0.7 (1.0)	0.9 (1.1)	0.6 (0.9)	0.8 (1.0)	51.6, p < .001	0.1, p = .9
GP visits	9.9 (8.9)	9.3 (9.2)	6.2 (6.3)	6.4 (6.4)	542.6, p < .001	142.2, p < .001
Specialists visits	2.8 (3.9)	2.9 (3.4)	3.1 (3.5)	3.0 (3.6)	27.1, p < .001	34.6, p < .001
^a Indicates the group effect (schizophrenia, comparison)
^b Adjusted model included non-psychiatric chronic illness

Mortality. The median age of death was lower among service users diagnosed with schizophrenia (63 years) than counterparts (72 years) (t = 6.5, df = 649, p < 0.001). Prior to the reform death was recorded in 2.3% of the service users with schizophrenia and in 0.8% among counterparts, indicating a 2.5 fold risk at any fixed point of time (adjusted HR 2.5, 95% CI 2.1–2.9). After the reform death was recorded in 4.4% of the service users with schizophrenia and in 1.6% among counterparts, indicating a 2.5 fold risk at any fixed point of time (adjusted HR 2.5, 95% CI 2.1-3.0).

Service users diagnosed with a schizoaffective disorder

More annual LDL and Hemoglobin-A1C tests were recorded for service users diagnosed with a schizoaffective disorder than among counterparts both before and after the reform (Table 3). A significant group effect was indicated on LDL in both the unadjusted and adjusted models. The mean number of LDL tests slightly decreased following the reform among service users with a schizoaffective disorder but was increased in the comparison group. This was supported by a significant group by reform interaction (unadjusted F = 5.2, df = 1, 6022, p = 0.023; adjusted F = 9.1, df = 1, 6020, p = 0.003). Hemoglobin-A1C test was done more frequently among service users with a schizoaffective disorder both before and after the reform. A significant group effect on Hemoglobin-A1C was indicated in both unadjusted and adjusted models. No significant reform by group effect was noted (adjusted p = 0.14).

Service users with a schizoaffective disorder made more annual visits to GPs both before and after the reform than comparison (Table 3). A significant group effect was indicated in both the unadjusted and adjusted models. In addition, while service users with a schizoaffective disorder reduced the number of GP visits following the reform, the number of visits slightly increased among counterparts. This was supported by a significant group by reform interaction (adjusted F = 26.3, df = 1, 6020, p < 0.001). The number of annual visits to specialists did not differ between service users with a schizoaffective disorder and comparison both before and after the reform. Service users of both groups slightly reduced the number of annual visits after the reform compared to before.

Table 3

Annual rate of LDL and Hemoglobin-A1C tests and visits before and after the reform among service users with a schizoaffective disorder and matched controls
	Schizoaffective disorder		Matched controls		F ^a, p
	Before reform	After reform	Before reform	After reform
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Unadjusted	Adjusted ^b
LDL	1.0 (0.9)	1.0 (0.9)	0.7 (0.8)	0.7 (0.8)	113.5, p < .001	48.0, p < .001
Hemoglobin-A1C	0.9 (1.1)	1.0 (1.1)	0.6 (1.0)	0.7 (1.0)	82.1, p < .001	35.8, p < .001
GP visits	11.7 (9.5)	10.8 (9.6)	6.6 (6.5)	6.6 (6.5)	397.0, p < .001	245.8, p < .001
Specialists visits	3.3 (4.0)	3.0 (3.3)	3.2 (3.6)	3.1 (3.8)	0.02, p = .88	0.93, p = .33
^a Indicates the group effect (schizoaffective disorder, comparison)
^b Adjusted model included non-psychiatric chronic illness

Mortality. The median age of death was lower among service users diagnosed with a schizoaffective disorder (67 years) than counterparts (71 years)(t = 3.4, df = 243, p = 0.001). Prior to the reform death was recorded in 2.2% of the service users with a schizoaffective disorder and in 0.8% among counterparts, indicating a two-fold risk at any fixed point of time (adjusted HR 2.1, 95% CI 1.6–2.7). After the reform death was recorded in 3.9% of the service users with a schizoaffective disorder and in 1.6% among counterparts, indicating a two-fold risk at any fixed point of time (adjusted HR 2.0, 95% CI 1.4–2.7).

Service users diagnosed with a bipolar disorder

Service users diagnosed with a bipolar disorder performed more annual LDL tests than comparison both before and after the reform (Table 4). A significant group effect was noted in both unadjusted and adjusted models. The mean number of LDL tests slightly decreased following the reform among service users with a bipolar disorder but was increased in the comparison group. This was supported by a significant group by reform interaction (unadjusted F = 14.1, df = 1, 12226, p < 0.001; adjusted F = 22.3, df = 1, 12224, p < 0.001). When the effect of non-psychiatric chronic illness was added to the model a different picture emerged. While service users with a bipolar disorder and no other non-psychiatric chronic illness did more LDL tests both before and after the reform, service users with a bipolar disorder suffering from an additional non-psychiatric chronic illness performed the test more frequently before (bipolar: 1.7 ± 1.2; comparison: 1.6 ± 1.1) but not after the reform (bipolar: 1.5 ± 1.2; comparison: 1.5 ± 1.5). This was supported by a significant group by chronic interaction (F = 26.4, df = 1, 12224, p < 0.001), and a group by chronic by reform interaction (F = 14.4, df = 1, 12224, p < 0.001).

Hemoglobin-A1C test was done more frequently among service users with a bipolar disorder both before and after the reform. A significant group effect was indicated in both unadjusted and adjusted models. In addition, Hemoglobin-A1C test was done more frequently among service users with a bipolar disorder with no other non-psychiatric chronic illness. However, among service users with a bipolar disorder suffering from an additional non-psychiatric chronic illness more Hemoglobin-A1C tests were performed before (bipolar: 1.9 ± 1.2; comparison: 1.8 ± 1.2) but not after the reform (bipolar: 1.7 ± 1.2; comparison: 1.8 ± 1.1). This was supported by a significant group by chronic interaction (F = 8.9, df = 1, 12224, p < 0.01), and a group by chronic by reform interaction (F = 14.5, df = 1, 12224, p < 0.001).

More annual visits to GPs were made by service users with a bipolar disorder both pre and post-reform than comparison (Table 4). A significant group effect was indicated in both unadjusted and adjusted models. In addition, while service users with a bipolar disorder reduced the number of GP visits following the reform, the number of visits increased among counterparts. This was supported by a significant group by reform interaction (adjusted F = 15.8, df = 1, 12031, p < 0.001).

Service users with a bipolar disorder made more annual visits to specialists than comparison both before and after the reform. A significant group effect was noted in both the unadjusted and adjusted models. In addition, when the effect of non-psychiatric chronic illness was added to the model it showed that more visits to specialists were recorded among service users with a bipolar disorder with no other non-psychiatric chronic illness both before and after the reform. However, among service users with a bipolar disorder suffering from an additional non-psychiatric chronic illness more visits to specialists than counterparts were recorded before (bipolar: 5.9 ± 5.9; comparison: 3.5 ± 3.7) but not after the reform (bipolar: 4.6 ± 5.0; comparison: 4.6 ± 4.4). This was supported by a significant group by chronic by reform interaction (F = 6.6, df = 1, 12222, p = 0.01).

Table 4

Annual rate of LDL and Hemoglobin-A1C tests and visits before and after the reform among service users with a bipolar disorder and matched controls
	Bipolar disorder		Matched controls		F ^a, p
	Before reform	After reform	Before reform	After reform
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Unadjusted	Adjusted ^b
LDL	1.2 (1.0)	1.12 (1.0)	0.9 (0.9)	0.9 (0.9)	188.3, p < .001	48.0, p < .001
Hemoglobin-A1C	0.9 (1.1)	1.0 (1.1)	0.7 (1.0)	0.9 (1.1)	55.9, p < .001	8.5, p < .01
GP visits	11.7 (9.0)	11.4 (9.3)	7.2 (6.9)	7.4 (7.0)	601.9, p < .001	325.9, p < .001
Specialists visits	4.4 (4.8)	3.9 (4.3)	3.9 (4.1)	3.6 (3.8)	22.9, p < .001	8.5, p = .004
^a Indicates the group effect (bipolar disorder, comparison)
^b Adjusted model included non-psychiatric chronic illness

Mortality. The median age of death was lower among service users diagnosed with a bipolar disorder (74 years) than counterparts (81 years)(t = 4.0, df = 680, p < 0.001). Prior to the reform death was recorded in 2.5% of the service users with a bipolar disorder and in 1.2% among counterparts, indicating a 1.7 fold risk at any fixed point of time (adjusted HR 1.7, 95% CI 1.5-2.0). After the reform death was recorded in 4.4% of the service users with a bipolar disorder and in 2.1% among counterparts, indicating a 1.7 fold risk at any fixed point of time (adjusted HR 1.7, 95% CI 1.4–2.1).

This study compared general health care utilization and mortality of persons with schizophrenia, schizoaffective disorder, and bipolar disorder 3.5 years before and after the mental health reform in Israel. Following the reform no-change in the performance of LDL test was noted among service users of the three SMI groups. While an increase in the performance of Hemoglobin-A1C test was noted among service users of the three SMI groups, it did not surmount that of counterparts. Among service users with schizophrenia or bipolar disorder diagnosed with an additional chronic physical illness, following the reform a decrease of LDL (bipolar) and Hemoglobin-A1C (schizophrenia and bipolar) tests were observed. In addition, the number of GP visits was decreased following the reform among service users of the three SMI groups, as well as visits to specialists among service users with a schizoaffective or bipolar disorder. Service users with a bipolar disorder diagnosed with an additional chronic physical illness decreased the number of visits to specialists following the reform. Last but noteworthy, with regard to most of the measures the use of services was higher among service users with SMI than their counterparts.

According to our findings, following the reform the number of visits to GPs decreased among service users of the three SMI groups. A differential pattern was noted in visits to specialists between the three SMI groups: lesser visits were seen in those with schizophrenia both before and after the reform, no differences in people with a schizoaffective disorder, and more visits among people with a bipolar disorder. The pattern of a higher tendency to visit GPs together with a lower tendency to visit specialists, which was seen in the current study among service users with schizophrenia, was reported in Israel [24], as well as in other OECD countries [25]. Former studies suggested that this pattern represents socioeconomic inequalities of health services use [24, 26]. This assumption could be relevant to the current findings and may be associated with a lower socioeconomic status of service users with schizophrenia. However, since in the current study the matched samples controlled for such differences (by using the socioeconomic index) other explanations may be raised. Suskolne suggested that service users visit the clinic not only to obtain curative or preventive services but also to satisfy other functions, such as gain an accepting listener [26]. In our case, service users with schizophrenia may have felt more comfortable to communicate with the GP than with a specialist, inasmuch as the former may be more understanding of their personal needs and family context. As suggested in the introduction, following the reform, the GPs, being the case managers of the service users, were expected to play a key role, providing a better continuum between general and mental care. This would imply that for service users with schizophrenia, the number of visits to the GP would be expected if not to increase, at least not to reduce. However, service users with schizophrenia, as well as the two other SMI groups reduced the number of GP visits pre- to post-reform. This may indicate that the mental health reform did not attain the anticipated goal with regard to GPs' function.

In MHS, like other Israeli HMOs, while service users can get an appointment with a GP very easily using online or secretary services, getting an appointment with a specialist sometime needs a referral from the GP, which may be a bureaucratic barrier in reaching the service. In addition, the waiting list for specialists may be long and increase the challenge to perform the visit. These challenges, which were not lessened following the reform, seem to be most difficult to service users with schizophrenia. Visits to specialists among service users with a schizoaffective or a bipolar disorder decreased following the reform, giving another indication that the reform failed to enhance the use of physical health services.

Our findings indicated that service users with SMI performed more LDL and Hemoglobin-A1C tests than their counterparts. Similar findings were previously reported in another Israeli sample of service users diagnosed with a bipolar disorder [15]. According to the NQIP, the treatment of persons with SMI should include the performance of an annual Hemoglobin-A1C test [23]. MHS has implemented NQIP’s recommendations since 2001, and the positive outcome with regard to blood-glucose tests is supported by our study. However, service users with schizophrenia and a comorbid chronic illness performed less Hemoglobin-A1C tests both before and after the reform than counterparts. In the same line, previous studies in Israel reported that while the performance of laboratory tests among service users with schizophrenia slightly deviate from controls [15], more pronounced differences of performance where seen among those with a comorbid CVD [14]. This may be related to a greater burden of health services which are required from those with a comorbid physical condition. These findings suggest that more attention should be given by the GPs in order to manage service users with schizophrenia and an additional physical comorbidity.

While the performance of an annual LDL test is not part of NQIP's recommendations for service users with SMI, it was done more frequently among them. As LDL test is related to risk factors for CVD, this may be explained by other risk factors, such as smoking, overweight, and a high blood pressure, which are known to be more prevalent among people with SMI, and lead GPs to enhance referral to perform the [27–29]. Unfortunately, in our data we had many missing values of these risk factors, thus preventing their inclusion in the analysis.

Along with the above effects, following the reform service users with a bipolar disorder and an additional chronic physical illness performed less LDL and Hemoglobin-A1C tests than their counterparts. This finding is surprising since the current study, as well as a previous report, showed that service users with a bipolar disorder tend to consume relatively more health services [15]. It is worth noting that our definition of chronic physical illness included different physical conditions that require continuous follow up, but may not be relevant for the specific laboratory tests presented here (e.g., cancer). Thus, it is important to further study the specific tests related to each condition.

Higher mortality was seen both before and after the reform among service users of the three SMI groups than counterparts. Notably, while in this study we have found no major disparities in the health care of persons with SMI, it is possible that disparities of care to persons with SMI and comorbid disease did occur in other measures, which could contribute to the excess mortality. For example, another Israeli based study performed in reported on disparities in surgical interventions to service users with schizophrenia with comorbid CVD [14]. Importantly, the increased mortality risk among service users with SMI remained at the exact level (as seen in the HR statistic) after the reform as it was before. However, 3.5 years of follow-up after the reform may be insufficient in order to draw conclusions on the impact of the reform on mortality, and a longer follow-up period may be needed.

This study has several limitations. First, with regard to visits to specialists we cannot conclude whether the findings reflect less referral made by the GPs or less adherence to the GP recommendation. Similarly, we cannot determine whether the high number of GP visits results from a proactive follow-up and higher reaching out by the GPs or self-initiated by people with SMI. Second, no information regarding the causes of death was available, making it difficult to conclude on health care disparities. Third, we did not measure psychiatric visits, so we cannot conclude whether there was a shift of the case managers from the primary physicians to the psychiatrists. We think, however, that these limitations were balanced by the large sample sizes, the well-defined matched comparison groups, the different nature of the measures used, the careful recording of the information collected on the users, and the consistency of the results.

Unfortunately it seems that the mental health reform in Israel did not attain the awaited change to the general health care for people with SMI, as indicated by the performance of laboratory tests and visits. This may implicate that a shift of healthcare management of service users with SMI occurred from the GPs to the psychiatrists, maybe because of the resources allocated to the community mental health services as part of the reform. The reduction in the number of GP visits following the reform, together with the reduction in the performance of the follow-up measures may support this suggestion. Along with the recent call for evidence based policy making, dissemination of the findings of the current study could help the HMOs to improve services to people diagnosed with SMI [30].

To conclude, regarding WHO's call There is no health without mental health, we believe that the opposite it also true. Namely, people with SMI deserve better health programs. Thus, it is imperative that psychiatrists pay more attention to the general medical status and follow-up regiments. This can be done with more profound education and improved collaboration between the primary physicians and the mental health services in the community.

Funding

The study was supported by The Israel National Institute for Health Policy and Health Services Research granted to Prof Gal and Prof Kodesh (grant number r/59/2018).

Correll CU, Solmi M, Croatto G et al (2022) Mortality in people with schizophrenia: a systematic review and meta-analysis of relative risk and aggravating or attenuating factors. World Psychiatry 21:248–71.
Haklai Z, Godberger N, Stein N et al (2011) The mortality risk among persons with psychiatric hospitalizations. Isr J Psychiatry Relat Sci 48:230–9.
Jayatilleke N, Hayes RD, Dutta R et al (2017) Contributions of specific causes of death to lost life expectancy in severe mental illness. Eur Psychiatry 43:109–15.
Lawrence D, Kisely S (2010) Inequalities in healthcare provision for people with severe mental illness. J Psychopharmacol 24:61–8.
Saha S, Chant D, McGrath J (2007) A systematic review of mortality in schizophrenia: is the differential mortality gap worsening over time? Arch Gen Psychiatry 64:1123–31.
Thornicroft G (2011) Physical health disparities and mental illness: the scandal of premature mortality. Br J Psychiatry 199:441–2.
Capasso RM, Lineberry TW, Bostwick JM et al (2008) Mortality in schizophrenia and schizoaffective disorder: an Olmsted County, Minnesota cohort: 1950–2005. Schizophr Res 98:287–94.
Cradock-O'Leary J, Young AS, Yano EM et al (2002) Use of general medical services by VA patients with psychiatric disorders. Psychiatric Services 537:874–8.
Druss BG, Bradford DW, Rosenheck RA et al (2000) Mental disorders and use of cardiovascular procedures after myocardial infarction. JAMA 283:506–11.
Druss BG, Bradford WD, Rosenheck RA et al (2001) Quality of medical care and excess mortality in older patients with mental disorders. Arch Gen Psychiatry 58:565–72.
Lawrence DM, Holman CD, Jablensky AV et al (2003) Death rate from ischaemic heart disease in Western Australian psychiatric patients 1980–1998. Br J Psychiatry 182:31–6.
Goueslard K, Petit JM, Cottenet J et al (2018) Increased Risk of Rehospitalization for Acute Diabetes Complications and Suicide Attempts in Patients With Type 1 Diabetes and Comorbid Schizophrenia. Diabetes Care 41:2316–21.
Scheuer SH, Fleetwood KJ, Licence KAM et al (2022) Severe mental illness and quality of care for type 2 diabetes: A retrospective population-based cohort study. Diabetes Res Clin Pract 190:110026.
Gal G, Munitz H, Levav I (2016) Health care disparities among persons with comorbid schizophrenia and cardiovascular disease: a case-control epidemiological study. Epidemiol Psychiatr Sci 25:541–7.
Gal G, Munitz H, Levav I (2017) Health Care and Mortality among Persons with Severe Mental Illness. Canadian Journal of Psychiatry 62:259–67.
Woodhead C, Cunningham R, Ashworth M et al (2016) Cervical and breast cancer screening uptake among women with serious mental illness: a data linkage study. BMC Cancer 16:819.
Bjorkenstam E, Ljung R, Burstrom B et al (2012) Quality of medical care and excess mortality in psychiatric patients–a nationwide register-based study in Sweden. BMJ Open 2:e000778.
Wu SI, Chen SC, Juang JJ et al (2013) Diagnostic procedures, revascularization, and inpatient mortality after acute myocardial infarction in patients with schizophrenia and bipolar disorder. Psychosom Med 75:52–9.
WHO: The world health report 2001. Mental health: New understanding, new hope. Edited by Organization WH. Geneva, Switzerland: WHO, 2001. http://www.who.int/whr/2001/en/whr01_en.pd Accessed 8 September 2022
Barbui C, Papola D, Saraceno B (2018) The Italian mental health-care reform: public health lessons. Bull World Health Organ 96:731-A.
Barbui C, Papola D, Saraceno B (2018) Forty years without mental hospitals in Italy. International Journal of Mental Health Systems 12.
Saxena S, Funk M, Chisholm D (2013) World Health Assembly adopts Comprehensive Mental Health Action Plan 2013–2020. Lancet 381:1970–1.
The Israel National Program for Quality Indicators in Community. https://en.israelhealthindicators.org/ Accessed 20 October 2022
Yom Din G, Zugman Z, Khashper A (2014) The impact of preventive health behaviour and social factors on visits to the doctor. Isr J Health Policy Res 3:41.
van Doorslaer E, Masseria C, Koolman X et al (2006) Inequalities in access to medical care by income in developed countries. CMAJ 174:177–83,
Soskolne V (2015) Preventive health behaviors and physician visits: relevance to health inequality. Isr J Health Policy Res 4:9.
Casagrande SS, Anderson CA, Dalcin A et al (2011) Dietary intake of adults with serious mental illness. Psychiatr Rehabil J 35:137–40.
Cook BL, Wayne GF, Kafali EN et al (2014) Trends in smoking among adults with mental illness and association between mental health treatment and smoking cessation. JAMA 311:172–82.
Zabeen S, Lawn S, Venning A et al (2021) Why Do People with Severe Mental Illness Have Poor Cardiovascular Health?-The Need for Implementing a Recovery-Based Self-Management Approach. Int J Environ Res Public Health 18
Goldman ML, Swartz MS, Norquist GS et al (2022) Building Bridges Between Evidence and Policy in Mental Health Services Research: Introducing the Policy Review Article Type. Psychiatr Serv 73(10):1165–1168.

No competing interests reported.

Download PDF

Journal Publication

published 13 Oct, 2023

Read the published version in Social Psychiatry and Psychiatric Epidemiology →

Editorial decision: Major revision
30 Mar, 2023
Reviews received at journal
25 Feb, 2023
Reviewers agreed at journal
04 Feb, 2023
Reviews received at journal
09 Dec, 2022
Reviewers agreed at journal
04 Dec, 2022
Reviewers invited by journal
01 Dec, 2022
Editor assigned by journal
16 Nov, 2022
Submission checks completed at journal
03 Nov, 2022
First submitted to journal
02 Nov, 2022

You are reading this latest preprint version

Impact of the mental health reform in Israel on health care and mortality among people with severe mental illness

Status:

Journal Publication

Version 1

Abstract

Objective

Method

Results

Conclusions

Introduction

Methods

General health services

Data analysis

Results

Service users diagnosed with schizophrenia

Service users diagnosed with a schizoaffective disorder

Service users diagnosed with a bipolar disorder

Discussion

Declarations

Funding

References

Additional Declarations

Status:

Journal Publication

Version 1