Barriers of access to care in a managed competition model: lessons from Colombia

Colombia radically reformed its healthcare system with Law 100 of 1993 [3], which created the General Social Security System in Health, with two insurance schemes: the Contributory Regime for formal sector employees and people able to pay, financed by mandatory contributions [3]; and the Subsidized Regime for people without the ability to pay, funded by resources from the Contributory Scheme and other sources of financing, such as taxes (Figure 1). Healthcare Insurers (Empresas Promotoras de Salud - EPS) were introduced for managing the Contributory Regime, as well as Subsidized Regimen (Empresas Promotoras de Salud Subsidiadas - EPS'S). They were to compete for the enrolment of population and received a capitation payment to cover different benefit packages in each regime (Plan Obligatorio de Salud - POS and Plan Obligatorio de Salud Subsidiado - POS-S) [4]. Currently, the contributory market is characterized by the predominance of private insurers - 86.1% of the affiliation - and the concentration in 5 private insurers that hold 50% of markets share [5]. The largest public insurer has been transformed into a mixed company with private capital and 5.8% of membership [5]. Competition for contracts with the insurers was also introduced among public and private healthcare providers (Instituciones Prestadoras de Salud - IPS). Healthcare for the uninsured (vinculados) and services excluded from the POS-S are provided by public hospitals funded by local and regional authorities [6], that represent 31.3% of total healthcare providers [7]. The uninsured have to pay for services and the insured make a co-payment according to their income [8].

The reform of the Colombian healthcare system has been, and still is, presented as a successful experiment in improving access to care [9, 10]. However, it has been a long, complicated process, and the results are controversial [11, 12]. In spite of the significant increase in public health expenditure from 3% to 6.6% of GDP, over the 1993 to 2007 period [13], around 15.3% to 19.3% of the population remains uninsured [14, 15]; and 38.7% are insured under the subsidized regime [15] that covers a range of services (POS-S) greatly inferior to that provided by the contributory one [16, 17]. Approximately 17% of health expenditure is devoted to administrative costs [18], of which more than 50% is spent on supporting daily operations (financial, personnel, and information management) and enrollment processes [19].

Furthermore, several studies seem to indicate a decrease in realized access to services [20, 21], and point to significant barriers related to characteristics of population, such as insurance enrolment [22‐28], income [22, 25, 26, 28], education [22‐27, 29] and, characteristics of services, such as geographic accessibility and quality of care [26, 30]. In 2005, the maternal mortality rate, an indicator that is sensitive to the overall healthcare system, was 130/100.000 in Colombia, compared to 30/100.000 in Costa Rica, while per capita 2004 health expenditure were similar (USD 549 and USD 598, respectively) but a GNP per capita lower in the former (USD 6130 and USD 9220) [31].

In addition, available evidence points to failures in the condition sine qua non for the successful implementation of managed competition, according to its supporters [1]: the existence of an effective regulatory system. These studies [32‐35] reveal deficiencies in regulation authorities in their ability to control a great number of institutions related to insufficient financial resources, lack of control mechanisms and excessive, and sometimes contradictory, regulation norms.

Most studies of the determinants of use of care in Colombia concentrate on personal variables and initial contact with services, and ignore contextual variables - health policy and characteristics of healthcare services. Insurance coverage, measured only by enrolment rate, is often viewed as an independent variable, although in managed competition models, insurers directly influence the provider networks and conditions of access to healthcare [36]. In addition, little research has evaluated access from the point of view of the social actors [26, 37‐39], despite the limited capacity of quantitative models in explaining determinants of use of care, due to methodological difficulties in including contextual variables [40, 41].

The objective of this article is to contribute to the improvement of our understanding of the factors influencing access to the continuum of healthcare services in the Colombian managed competition model, from the perspective of social actors.

Individual semi-structured interviews with a topic guide were conducted. The guides - with a common section and a specific section for each informant group - included opinions on access to the continuum of services across the network, elements that influence it, and strategies for improvement. In the user guide, users' experience with services was also explored. Interviews with users were conducted at their homes, the remainder of informants mostly in the workplace. The interviews lasted between one and two hours and were audio-recorded and fully transcribed.

A content analysis [49] was conducted with the support of the software Atlas-ti 5.0. The process of generating categories was mainly inductive. Data were segmented by case study, informant's group, and themes. Themes were identified, coded, re-coded and classified identifying common patterns by looking at regularities, convergences and divergences in data, through a process of constant comparisons, going back and forth between data and conceptual framework. In order to ensure the quality of results, the information was triangulated between groups of informants and the results were contrasted with the informants and the bibliography. In addition, four researchers with different backgrounds and in-depth knowledge of the context participated in the analysis.

Participants were informed of the objective of the study and that they were free to participate and to leave at any point. Participants gave oral consent for their participation and the interviews to be recorded. The recordings and transcripts were codified in such a way that the individual origin of each one could not be identified, before being appropriately stored. These results are part of a broader research aimed at analyzing the impact of integrated healthcare networks on health care access and efficiency, in Colombia and Brazil. Ethical approval was sought and obtained from the University of Rosario's Ethical Committee in 2005.

The informants of the subsidized regime, in both areas, reported difficulties of access to healthcare related to the insurance design: exclusion from benefits of necessary specialized services, classification of services by levels of care and co-payment (Table 3). The informants, especially secondary care specialists, considered that blocked access to tests classified as lower level in the more complex healthcare levels, as well as the inclusion of only very basic diagnostic tests in the first level, force patients to go back and forth between several health facilities and delay the resolution of the health problem.

Finally, out-of-the-pocket payments - or co-payment for the insured - emerge as one of the biggest barriers. Enrollees of the subsidized regime point to co-payment of services not covered - specialized diagnostic tests and medicines for chronic and high-cost diseases - while the uninsured also mention out-of-the-pocket payment for basic specialized services as a barrier. When confronted with the economic barrier, the informants often desisted in seeking medical care, "... they ordered me those for the thyroid, and there were some exams I need to do that cost about three million [EUR 1.000] (...) the doctor said, 'Well, sell your house or whatever you have', and I said, 'What house? I got the TSH because that is covered, but not the others, because they're too expensive" (user, subsidized regime - rural area). In the discourse of the contributory regime's informants, there strongly emerge norms limiting access to medical care - minimum contributory's period for high cost services or suspension of insurance coverage for lack of payment. As a consequence, they cannot access health services or end up having to make a co-payment (Table 3).

The presence of the insurance companies in the healthcare system emerges, in the interviews with managers and professionals of public facilities, as a barrier to access for those who are insured under the subsidized regime (Table 4). On the one hand, insurers try to maximize their benefits by introducing mechanisms to reduce the use of services. On the other, its presence requires re-routing funds from medical care to intermediation: "all these intermediaries we have earn money based on how much they can avoid directing it to hospitals, so they end up keeping the biggest piece of the pie" (health professional, public primary care provider - urban subsidized network).

The informants from both insurance regimes identify the managed care mechanisms employed by insurers as a barrier to use of services (Table 4). The mechanisms they identified differ according to the healthcare network type. In networks where there is a separation between the insurers and the healthcare providers, mechanisms directly acting upon demand (such a as authorization of clinical services) emerge, whereas if the insurer shares the ownership with the healthcare providers, prevailing mechanisms act upon supply -control of the clinical practice and capitation payment. According to the interviewees, difficulties vary according to the mechanism, and they often refer to access to specialized care. On one side, authorizations delay care, especially in emergency transport and diagnostic testing: "They lose time confirming if the patient is theirs or not [emergency transport], it can last from one minute to four hours just to get an answer. When they answer (...) it could last...from immediately, to six hours, to three days, according to the type of service requested" (administrative, public primary care provider - urban subsidized network). The users from the rural area add costs in time and transportation, to do the paperwork and to obtain authorizations. The limits to clinical practice (medicines, tests and referrals to specialists) delay diagnosis and treatment. Capitation payment to healthcare providers has a direct negative impact on access by creating incentives to decrease the quantity and quality of services provided, and increasing waiting time and travel costs to capitated hospitals that become the first choice of insurers.

Two difficulties are added to subsidized enrollees of the urban area. First, the conflict between insurers and the local health authority in interpreting which services are included in the POS-S, as both want to avoid bearing the costs of the service. This results in more paperwork for patients and further delays in care (Table 4). The second is the way of services purchasing: the insurance companies split the process of care into various sub-processes to be purchased from different hospitals, requiring patients to go back and forth between various centers in order to resolve their health problems (Table 4). In addition, contracted healthcare providers frequently change, generating additional costs for patients due to errors in referral and the need to adapt them to the system (opening hours, documents, etc.).

The majority of public service professionals consider that barriers created by intermediation have made access to the continuum of healthcare services of subsidized enrollees worse than that of the uninsured who use the networks organized by the local health authority (Table 4).

The barriers that predominantly emerge are those regarding the behavior of public healthcare providers in the subsidized regime, and the availability and geographic accessibility of the networks organized by insurers of both regimes. According to the informants, competition has triggered significant change in public hospitals, as they prioritize economic profit over patients' health needs (Table 5). It is considered a common practice for patients with profitable illnesses to be selected, and for patients who are uninsured, or lack the proper documentation to invoice the services, to be denied treatment. This is an especially critical issue in the urban area: "Some diseases are not profitable; you see, for example, all of the chronic, internal medicine ones, because of the length of hospital stay and the cost of medicines. So people tend to provide care for easily resolvable diseases with short hospital stays and appropriate fees (...) and this goes against the population pyramid" (manager, secondary care provider - rural subsidized network).

In both regimes, the shortage in the supply of basic specialties and high technology medical services emerge, as well as limited geographic accessibility, both in the rural area and in the contributory urban network, which is aggravated by the low-income level of the population (Table 5). In the latter, problems in accessibility go all the way down to the first level of care. The causes of the shortage vary according to the regime: in the contributory regime, it is related to insufficient private investment and the closing of public hospitals. In the subsidized regime, informants point to insufficient public funding and to elements of managed competition: added administrative costs and incentives to close non-profitable services (internal medicine) or to insurers not contracting high technology services (Table 5).

Long waiting lists for specialized services, especially in emergency care, are the greatest organizational difficulty in both regimes and areas (Table 5). The informants relate it to insufficient supply and to the authorization requirements of insurance companies. In the subsidized regime, in-person and restricted appointment requirements, and the poor quality of care, are additional barriers to the access of specialized care (Table 5).

Although the main objective of introducing managed competition in Colombia was to improve access to care [3], the results show several barriers of access to the continuum of care. These obstacles are not only due to factors usually taken into account in conceptual frameworks for access - services and population characteristics [47, 50‐52] - but also, and more importantly, to characteristics of the insurance design and to the presence of insurance companies. All are permeated with the values that introduced the managed competition model, that is the search for economic benefit or sustainability, which guides insurers and also healthcare providers' behavior. These values emerge in the discourse of the informants as a direct barrier to access or as a cause of other barriers (Figure 2).

These obstacles produce synergetic effects and increase the negative impact on the use of care, an aspect which is often omitted in quantitative research due to its difficult measurement [53]. For example, the insurers' authorization or in-person appointment requirements increase travel and time costs for patients, which are already high in networks whose services are hard to access geographically. Limits to first level of care tests and authorization requirements increase waiting times for specialized care, that are already long due to insufficient availability of services.

Many difficulties in access emerge in both insurance regimes and areas, but with remarkable differences. In the subsidized regime, financial obstacles predominate and are linked to a smaller specialized care services portfolio. In the contributory regime, structural deficiencies in the network prevail among the obstacles.

In the rural area, structural difficulties emerge more intensively due to insufficient public funds, a common problem in low-income countries with integrated public healthcare systems [54]. In the urban area there are strong barriers related to market functioning - changes of healthcare providers in the networks, conflict between paying entities, fragmentation of service provision across multiple providers, and the rejection of patients - coinciding with other studies [55].

In the rural area, managers and administrative personnel of the EPS'S refused to participate and it was not possible to find a replacement case as it was the only insurance company operating in the area. The missing of these actors' opinions might mean that the information is not complete [71]. In the urban area, the only EPS of the contributory regime that agreed to participate was a non-profit insurance company, whose health services network was near to the area of study. These characteristics should be taken into account in the interpretation and transferral of results to other contexts [72]. With commercial EPS'S, it is reasonable to assume that obstacles would be greater.