Introduction

Schizophrenia, a clinical syndrome with variable psychopathology, affects cognition, emotions, perception and other aspects of behavior, they are invariably severe and usually prolonged; which translates into health expenses. The onset of schizophrenia commonly occurs at the ages of 18-25 years in men and 21-30 years in women.¹ The prevalence of schizophrenia has been estimated at 0.48 % (IQR: 0.34%-0.85%).²

According to the Word bank, out-of-pocket (OOP) health expenses affect 44.7% of the Mexican population, well above the average for Organization for Economic Cooperation and Development (OECD) member states which is 13.7%, while the Latin American average and upper-middle-income countries are 30.0% and 28.3% respectively.³

According to data from the Health Information General Direction (DGIS by its Spanish initials), OOP health expenses represented 52.2% of health spending in the year 2000 and for year 2018 it was 42.1%. In addition to the majority of OOP spending is concentrated in the population without social security.⁴

In Mexico, OOP health expenses has represented a high percentage of total health spending. In 2003, one of the strategies to reduce it was to establish the Social Health Protection System (SPSS by its Spanish initials), whose main objective was to reduce catastrophic and impoverished expenses, that is, to give financial protection to the population without social security.⁵ The first SPSS impact simulations (considering that each household will contribute 5% of available spending) was expected to reduce the percentage of households with catastrophic expenses from 3.4 to 1.6%.⁶

The SPSS was executed by the Seguro Popular (SP by its Spanish initials) and included mental disorders in the Universal Catalog of Health Services. In the case of schizophrenia and other psychotic disorders, patients could access two annual hospitalizations of less than 20 days each one, auxiliary studies diagnostic, outpatient consultation and antipsychotic medication.⁵

The inclusion of interventions to care for patients with schizophrenia was important since specifically for schizophrenia; the demand for health care among patients without social security coverage increased by 33% between 2005 and 2013, with the annual cost per case for the Mexican health system estimated at USD 2 216 during that period.⁷ Studies of other vulnerable groups such as the elderly have documented that the implementation of the SPSS program in Mexico outcomes limited.^8,9 However, no reports were found regarding financial protection in patients suffering from schizophrenia during the period that SPSS operated.

Assessing the health-related OOP expenses, catastrophic expenses (CE) and impoverishment of households of schizophrenia patients is crucial given the dearth of studies on the subject in Mexico and other Latin American countries. The consequences of these events for patients with mental disorders -particularly schizophrenia- and their families can be devastating; because the direct payments for health services can push households into or deepen their poverty. The households facing health expenditures are often compelled to interrupt meeting other essential needs and the households might decide to postpone seeking care for their health needs in order to avoid a financial catastrophe caused by paying for health services, thus creating a vicious circle of illness, disability, and poverty.¹⁰

We therefore undertook this study in order to estimate the magnitude of OOP health expenses, CE and impoverishment caused by medical care expenditures incurred by patients with schizophrenia and their families in the Mexican public-health system.

Materials and methods

Results

Our analysis included 96 patients, two-thirds of whom were men. The majority of participants reported being single and unemployed. Nearly all those who were working were engaged in the informal economy. The mean age in our sample was 37 years, and the average schooling was 10 years. Participants had been in psychiatric treatment for an average of 7.5 years, with 52% reporting over 10 years of clinical evolution (table I).

Regarding the use of health services, 67% of patients had been hospitalized, 100% had received outpatient consultations, 52% had undergone an auxiliary diagnostic study. The patients had access to two outpatient consultation in last year, only 9% of their households obtained support for the purchase of medications and 24% revived free medical consultation. In addition, access to auxiliary diagnostic studies was scarce (laboratory or radiology) (table II).

All households had OOP health expenses for hospitalization, medications, medical visits or diagnostic studies. Purchase of medications accounted for 97% of OOP health expenses, followed by hospitalization and auxiliary diagnostic studies, while outpatient consultation expenses represented 1% (table III).

Similarly, the average annual income for each quintile was lower than the average annual income reported by the ENIGH,¹⁶ the same as the poverty line of the sample (644 USD vs. 915 USD).¹⁷ The payment for medicines was highest for households in the middle quintile; those in the poorest and richest quintiles showed similar figures. Hospitalization expenses were higher for those in the middle-income quintile, while the richest quintile spent more on medical consultations, auxiliary diagnostic studies, and transportation (table IV).

Regarding the distribution of households by income, the data showed that almost all the households in our study fell within quintiles II and I (41 and 30 household respectively). Annual per capita income was ten times higher for those in the richest as compared to the poorest quintile.

The 16% of households incurred in CE, so the OOP representing more the 40% TCP. The percentage of households who became impoverished as a result of OOP health expenses was 6.6%. Nearly half of the households fell within the poorest quintile and a third within the second poorest. More of 80% of household of most poor quintile suffer catastrophic expenses contrary in the middle and upper quintile any household suffer CE or impoverishment (table V).

The capacity to pay exceeds the income of the poorest households, except the richest quintiles, that is, the deficit in income is greater in the poor quintiles; in addition, the TCP illustrate the effect of the OOP better than household spending (table V).

Discussion

Our study demonstrates that 16.6% of households of patients with schizophrenia had OOP health expenses as a percentage of total household capacity to pay. This amount is much higher than 2.3% reported for the general population.¹⁵ In addition, 100% of households had OOP health expenses and the main reason was the payment of medicines (89.5%); these data exceed that reported for the general population which is 45.1% for total OPP health expenses and 53.6% for medicines, according to National Health and Nutrition Survey 2018 (Ensanut by its Spanish initials).¹⁸ Several studies have shown that the purchase of medicines representing roughly half of total household health expenditures.^19,20

Furthermore, the patients in our study sample were could access to health interventions (hospitalization, auxiliary diagnosis studies and medicines) by the SP; according to the Ensanut, 37.4% of the total Mexican population in that period was affiliated with the SPSS.¹⁸ The objective of this public health insurance program, in effect during the study period, was precisely to offer financial assistance with the aim of ensuring access to health services for the uninsured population.

Our findings regarding that the highest percentage of households are concentrated in the poorest quintiles, coincide with that reported by several studies, in which TCP is used as a proportion of OOP health expenses to determine catastrophic expenses; for example, a catastrophic expenditure prevalence of 13.6% has been described, with a high proportion in poor quintiles.^21,14 Another example is the analysis of various insurances schemes in China, in which the incidence of catastrophic spending is 15.53% in rural population with the threshold 40% TCP.²²

Thus, the TCP allows us to identify the effect of financial protection scarcity in poor households, contrary to the method shared budget used by the Objectives to Sustainable Development with thresholds of 10 and 25% of the available budget, which identifies catastrophic expenses more frequently among the richest households.^12,14

In addition, not surprisingly, it has been reported that the households of patients with mental disorders suffer more social disadvantages than households without mentally ill patients.²³ Our study supports these findings, since the poverty line of the population analyzed was lower than that reported in Mexico.¹⁷ This situation could be a consequence of the lack of work activity of patients caused by their illness. Other authors have documented that in households of patients with mental disorders, up to a third are affected by CEs, double that of households without members with this type of health problem.²⁴

Another factor that influences to incur in OOP expenses, is the poor access to health services. According to our findings, this situation possibly reduces the access to medicines and outpatient consultation provided by the SP. It has been reported that the benefits of SP in terms of reducing OOP expenses for health are associated with the type of health facilities to which the user has access, particularly in poor or rural households where the possibility to have catastrophic expenses is high.^25,26

The heterogeneous effect of the SPSS should be adjusted in the recent reforms to the health system.²⁷ Accordingly, our work demonstrates that one of the main challenges facing will be guaranteeing effective and efficient access to medications (antipsychotics, antidepressants, etc.) and hospitalization, to mitigate OOP expenses in backgrounds of schizophrenia patients. In this sense, guaranteeing these interventions could reduce catastrophic expenses between 13 and 19%.²⁸

The present empirical study highlights the effect of OOP expenses in a restricted sample, so that there is a limitation that the results cannot be extrapolated to larger populations; annualizing is another possible bias.