In Mexico, mental disorders constitute a major public health concern, representing a significant proportion of the national burden of disease, and are associated with significant levels of impairment (
1–
5). Despite the impact of mental disorders on quality of life, treatment gaps continue to be significant (
6–
8). Contributing to these large gaps, behavioral health services in Mexico are predominantly accessed through specialized care (
6,
9), suggesting inadequate detection and treatment at the primary level of care.
Community-based primary care in Mexico has a nascent role in the delivery of mental health services, with only 30% of primary care centers having protocols in place to detect and treat mental disorders (
9). In addition, less than 15% of primary care staff report receiving any training regarding mental health concerns. It seems clear, therefore, that the implementation of behavioral health services at the primary level of care will be an utmost challenge for the Mexican health system. Thus this study focused on identifying barriers from the perspective of clinic personnel to integrating behavioral health services in community-based primary care clinics in Mexico City.
The Mexican health system is managed by the Secretary of Health and is divided into three types of services (
10). First, Social Security provides access to health care financed through employer and employee contributions, covering 47% of the population. Second, and the primary focus of this study, are persons not covered by Social Security (49%), who receive care predominantly through community-based primary care clinics and often are the poorest residents of Mexico. Finally, a small percentage of the population (4%) has the financial resources to access private networks of providers.
The organization of the Mexican health care system makes it very difficult to have integrated primary care. The system is fragmented into services provided through Social Security and those provided to the uninsured. The primary care system for the uninsured uses a limited menu of services provided through community-based clinics (
11,
12). Each clinic is responsible for a geographical area and is classified, from smallest to largest, as T-I, T-II, or T-III, based on the size and scope of services. The smallest centers (T-I) typically house only one physician, one nurse, and one social worker and generally serve the most disadvantaged communities.
In the World Health Organization’s (WHO) Mental Health Gap Action Program (mhGAP), a series of strategies was delineated to decrease mental health treatment gaps in low-resource countries (
13). This objective is even more pressing among Mexico’s most disadvantaged communities because low socioeconomic status increases the risk of the development of mental disorders (
14,
15). There are various examples of how the integration of mental health services into primary care has helped close gaps in services (
16). The greatest potential of primary care is the role it often plays as the first contact between patients and the health care system. However, to take advantage of this potential, training of health care teams must be adapted to the needs of the communities they serve (
17,
18).
Therefore, in 2011 the Federal District Mental Health Act was established in Mexico City. It proposed that all community-based primary care clinics in the district target behavioral health and provide appropriate referrals among populations at risk. However, this plan has not had the expected reach because of a lack of financial resources and infrastructure (
19), among other issues. In addition, there has been a lack of organization and coordination among the various entities responsible for implementing policy decisions.
In Mexico, research on barriers to health care, particularly behavioral health services, in primary care settings is limited. In regard to barriers to general medical care, providers often report large caseloads and lack of staff, equipment, resources, and basic medications (
20). Although financial barriers to the integration of mental health in primary care settings have generally been a concern, even among developed countries (
21), these problems are further exacerbated in a developing nation. Providers also report other barriers associated with integration of mental health and primary care, including lack of training, role confusion, and burnout (
22,
23). In addition, primary care staff indicate lacking the necessary training to detect and make appropriate referrals for mental disorders (
24). Therefore, in order to effectively assess and implement the scaling up of mental health services within primary care clinics in Mexico, a broader understanding is needed of the potential barriers to the integration of these services (
25). For example, clinics in Mexico may lack the requisite infrastructure and other key resources necessary for the delivery of high-quality mental health care. It is imperative to identify these barriers to ensure that behavioral health services are integrated effectively and efficiently in primary care settings in Mexico.
To analyze these barriers, we turned to a typology established by Arksey and colleagues (
26), which allows for a global analysis of various barriers to access to care. The typology was originally developed from a literature review of barriers to care for family members of persons with an illness or disability. Although the typology does not focus on the Latin American primary care system, it describes systemic characteristics that are consistent with barriers found in other studies (
27,
28) The typology clearly delineates the nature and characteristics of barriers, grouped into the following five categories: professional characteristics, including professional, day-to-day activities, situations, and characteristics that may interfere with the provision of care; service issues, such as poor coverage by public aid programs and the various difficulties associated with treatment access; language or cultural issues that can lead to service dropout, for example, a disconnect between the quotidian language with which patients express their needs and high stress levels and the tendency of care and treatment to focus on diagnoses; care recipient characteristics, defined as the factors that influence whether patients seek and participate in services, including all aspects of care (
29–
31), timing and prioritizing of help seeking (pursuing care only when symptoms become severe), and seeking other, more informal means of care or using self-medication to help alleviate symptomatology; and knowledge and information issues, for example, a lack of knowledge by patients of services available and how to access those services (
32) as well as a lack of information on their disorders and treatment options.
Using this typology as a guide, the study employed the ethnographic technique of the semistructured interview to elucidate some of the barriers to integrating behavioral health services reported by staff of community-based primary care clinics in Mexico City.
Methods
Permission to conduct this study was sought from Mexico City’s Secretary of Health, who also helped in identifying potential participating clinics for this study. Staff were recruited from 19 primary care clinics by using purposeful criterion sampling (
33). Key personnel at each clinic were contacted, informed about the study’s objectives, and given an opportunity to participate. Maximum variation sampling was used to have a wide representation of both health care roles and personnel from each of the clinic categories (T-I, T-II, and T-III) (
33).
The present sample comprised 25 clinical or administrative staff, each of whom participated in a semistructured interview. The sample included five physicians, three of whom were female (T-1 facility, N=3; T-II facility, N=2); two nurses, both of whom were female (T-I, N=1, and T-III, N=1); eight social workers, all of whom were female (T-I, N=4; T-II, N=1; and T-III, N=3); four psychologists, three of whom were female (T-1, N=1; T-II, N=1; and T-III, N=2); a female psychiatrist (T-III); a male dentist (T-II); and four administrative staff, all of whom were female (T-III). The mean±SD age of staff was 40.9±11.6 years old.
Interview Procedure
The project was approved by the institutional review boards of the National Institute of Psychiatry of Mexico and the University of Southern California. Informed consent was gathered from all participants. Research project personnel, including the authors, conducted the interviews. Key informants from each clinic were interviewed because their experience, knowledge, or role within the clinic was considered most salient for the purpose of this study.
The semistructured interview is one of the most widely used methods to understand the lived experience of research participants, given that it favors a continuous discussion of a topic of the investigator’s choosing (
34). The interview questions reflected each of the following thematic areas: work activities, clinic perceptions, perceptions of work conditions, mental health training and work, stigma, views regarding mental health services, and sociodemographic data. Interviews included an open-ended question asking participants to discuss other topics not covered by previous questions.
Data Analysis
All interviews were transcribed and coded by an author or an undergraduate research assistant. Data analysis was completed in two phases. The first phase employed the meaning categorization method (
35), which allowed for the coding of interview content into mutually exclusive categories. Subcategories were then created by using the same method. The second phase resulted in a comparison of the thematic categories with the typology of barriers to health services developed by Arksey and colleagues (
26). Each interview was coded separately by two study personnel. Interrater agreement was greater than .90, and discrepancies were resolved through discussion between coders. Data analysis was conducted in Spanish, and the quotes included herein were translated to retain their original, literal meaning.
Results
Analysis of the semistructured interview data yielded a number of barriers that fit within Arksey and colleagues’ (
26) typology of barriers. However, no evidence was found regarding barriers related to the category of professional characteristics. In addition, a number of subcategories emerged (
Table 1).
Service Issues
A number of staff reported staff shortages, including a lack of personnel to complete day-to-day tasks, particularly administrative tasks. In particular, staff (mostly T-I) reported being expected to be responsible for extra tasks, such as cleaning. Staff also reported not having necessary medications on hand and being unable to pay for repairs and maintenance of clinics.
Budgetary constraints and various inequities with respect to the mental health budget were also reported. For example, staff explained that substance abuse services were often better funded than were mental health services. Clinics classified as T-III are responsible for distributing funds and resources to other clinics within their catchment zones. However, staff in T-I clinics explained that the funds were often distributed to the T-III and T-II clinics first. Therefore, by the time funds were distributed to T-I clinics, there was little, if anything, left.
Service issues also included a lack of employee benefits. Because staff are often paid on a fee-for-service basis, their positions are unstable. Nonsalaried staff do not have vacation or sick time, often did not get paid on time, and reported having inadequate or no health insurance coverage. In addition, staff explained that they were overworked and lacked the time necessary to adequately evaluate and treat physical ailments. Because of these time management issues, they believed that they would not have the time to adequately detect mental health issues.
Appointment restrictions also limited access to services. Patients in primary care clinics must secure appointments early in the morning (as early as 6:30 a.m.) to be seen later that day, making it necessary to turn away patients who seek treatment later in the day. Services were often incongruent with the needs of the community. For example, most primary care clinics were open only during the hours that patients typically worked (7 a.m. to 5 p.m.). Some clinics offered Saturday appointments but that was not the norm.
Poor health insurance coverage also limited access to services. Health insurance coverage was inconsistent, and some necessary services were not covered.
Language or Cultural Issues
Stereotyping of patients was a barrier to providing services. Staff described patients as being very demanding. In particular, T-I staff explained that residents of the areas where they serve are often untrustworthy and aggressive. They also reported high levels of stigma. Patients are often opposed to behavioral health treatment, staff reported, and do not seek help from behavioral health providers because others may think they are “crazy.” In addition, the staff stated that patients believe being treated in a hospital will afford better services than treatment in primary care.
Care Recipient Characteristics
Staff perceived that patients do not want to engage in treatment and viewed men in particular as being less likely to seek behavioral health services. They also reported that patients lack the necessary financial resources to seek, access, and use services. Patients are often unable to pay for psychotropic medications, according to staff, often because they are not covered by public aid. Various staff reported that patients lack knowledge regarding health-related issues, especially mental disorders. Staff stated that psychoeducation services are available but are rarely used and are ineffective when they are used.
Knowledge and Information
Staff reported that patients lack knowledge about medications. Specifically, staff mentioned ongoing issues regarding the use of similares, or medications that have a similar, but not equal, chemical make-up as generic medications. Similares are more affordable than generic medications, and especially more affordable than brand-name medications. Patients are often unaware that similares are not the same as medications prescribed by physicians and that no evidence exists about their efficacy or potential side effects.
Health insurance coverage is another barrier to mental health treatment. Particularly among patients covered by public aid, according to staff, patients often have little knowledge regarding the services covered by their insurance. In addition, most staff are not trained to deal with behavioral health issues. Of the few training opportunities offered, most are available only to physicians. Case management services are often unavailable, and when a behavioral issue is detected, staff are unsure where to refer the patient.
Discussion
The staff of community-based primary care clinics in Mexico City reported significant barriers not only in the provision of mental health services but also in the provision of health services in general. Many of the thematic categories in Arksey and colleagues’ (
26) typology of barriers to care were present in this sample, except professional characteristics, an absence that was due most likely to our focus on care recipients rather than caregivers. Nonetheless, numerous barriers to care were identified, and, overall, staff painted a grim picture of primary care conditions within Mexico City. To implement the WHO’s mhGAP recommendations would be a difficult task (
13).
Lack of resources, primarily due to fiscal and budgetary concerns, was a consistent theme among staff, an illustration of the scope of the overall lack of health care resources found generally in the Mexican health care system (
20,
22,
23). Budget reductions have limited the staff necessary to carry out day-to-day tasks, including cleaning staff, and limited the availability of treatment necessities, such as medications. The inequitable distribution of resources among clinics that vary in size exacerbates these shortages. The fragility of these community-based primary care clinics does not allow for the inclusion of permanent psychiatric services, principally because of the costs of access to psychotropic medications but also because of the costs associated with hiring specialized mental health staff, such as psychologists. Because these limitations have an adverse impact on behavioral health services, resources, and programming, they are extremely important to consider.
However, the information gleaned from staff provided several recommendations for fiscal and budgetary changes. Some of the reported barriers could be overcome by changes in the distribution of funding. For example, ensuring that maintenance services are handled by ancillary staff, not physicians, would provide more time to engage in patient care, possibly even creating the space to administer brief mental health assessments. The added detection of mental health issues, in the long run, may reduce the burden on clinic resources. For example, if mental illness, such as depression, exacerbates a medical condition, for example, diabetes, proper management of the mental health symptoms may reduce the amount of follow-up and maintenance required for care of the general medical condition. Another possibility is to provide a forum by which plans could be formulated to overcome these barriers. For example, one option is to staff community-based programs with lay workers, among other strategies, to reduce the burden of hiring expensive behavioral health specialists (
18).
In addition, it is evident that staff may have been experiencing high levels of stress because of a lack of resources and the significant needs of the communities they serve. High stress levels are often found among primary care staff in various countries (
36), and although no study has explicitly examined staff stress levels in primary care settings in Mexico, it can be assumed that these staff are at risk of high levels of stress. Therefore, potential targets for intervention include self-care and stress management skills among primary care staff. For example, preliminary evidence suggests that mindfulness-based education programs may be effective tools to reduce stress (
37).
This study included only staff’s perceptions of barriers to the implementation of mental health services at the primary level of care in Mexico, and these barriers may not be generalizable to other settings. It should be emphasized that many of the barriers, particularly those focused on patient characteristics, were identified based on the perspectives of clinic staff and may not be based on actual fact. Regardless, the perceptions are still important to consider in regard to both the evaluation and the implementation of behavioral health services in primary care settings. In addition, it is possible that the perceptions of staff did not include all possible barriers to the implementation of mental health services, given that other barriers may exist but were not discussed by staff.
One strength of this study was its inclusion of a wide range of professional roles within the Mexican primary care system. Future studies of other primary care systems should also employ a comprehensive approach to better hone in on systemwide barriers as well as to identify potential facilitators of integration. Finally, future research should include patients and other associated parties with vested interests in primary care, such as citizen health boards. The addition of interviews with interested parties other than health providers will lead to a more comprehensive picture of the feasibility of implementing mental health services in these primary clinics.
Conclusions
The results provide a preliminary framework for evaluating the feasibility of implementing mental health care in primary care clinics in Mexico. The qualitative information provided allows for the development of evaluative tools to better measure the implementation of mental health services in primary care. The creation of such tools will lead to a more nuanced understanding of how to adapt mental health care to these primary care settings and, more important, increase service access and use for a population with significant unmet mental health needs.
Acknowledgments
The authors thank Critical Research Issues in Latino Mental Health, funded, in part, by the National Institute of Mental Health (R13-MH-066308), and in particular Steven R. López, Ph.D., for providing an opportunity to share and receive commentary on a draft of this article. They also thank the Secretary of Health of Mexico City and especially all the personnel that shared their experiences, without whose help this project would not have been possible.