Differences in Access to Primary Healthcare – Structures, Equal Opportunity and Prejudice - The Results of an Empirical Study

15 December 2004

In the period September-December 2003, the Hungarian Delphoi Consulting research group conducted a survey commissioned by the Hungarian Ministry of Health, Social and Family Affairs, on Roma access to primary health care entitled "Differences in Access to Primary Healthcare – Structures, Equal Opportunity and Prejudice – The Results of an Empirical Study". In this issue of Roma Rights the ERRC reproduces the authors' summary of the survey's findings translated into English¹.

One of the primary purposes of our research is to establish whether or not various groups in society, including Roma and others that suffer multiple social and economic disadvantages, have full and equal access to primary healthcare services. If there is unequal access to basic services, what are the causes of this inequality, and what are the actual differences in access among the various groups? Because our research focuses partly on the access of Roma, when determining which doctors and health visitors we would question (that is, the actual group that would constitute the subject of our survey) we selected settlements where, on the basis of authoritative estimates², the percentage of Roma inhabitants equalled or exceeded 1%. Consequently, the results are representative only of those general practitioner (GP) practices and health visitor districts that are located in these settlements.

General Practitioners

Structural issues
In the beginning we sought to establish, on the basis of the national statistics, whether the presence or absence of a GP in a settlement is in any way related to the settlement's social standing, the number of its inhabitants, the age distribution of those living there, or to the ratio of unemployed persons or of Roma within its population.

The data suggests that settlements with multiple disadvantages do not offer local practitioner services directly. These settlements, mostly because of an ageing population and the lack of local funds also tend to be lacking in other basic institutional services. If we look at the national picture, we find that the number of pensioners is generally higher in settlements that do not have a local GP. However, the older segment of the population, with its greater health concerns and higher health risks, suffers from the lack of local health services only to a slightly greater extent than does the population as a whole. This slight difference, however, is significant when we look at the actual number of pensioners affected: approximately 128,000 out of over 2 million.

The ratio of the Romani population shows a dramatic difference. Excluding Budapest, 18.6% of the country's total Romani population lives in a settlement without a local GP.

The social and material conditions of Roma and pensioners living in settlements where there is no local GP are significantly worse than average, especially since the social and economic circumstances of these small settlements tend already to be among the worst in the country. The social disadvantages may well compound the problems arising from a lack of direct access to a local GP.

We know that the high rate of health problems among Roma is due directly to poverty, and in this regard, the Romani population of the poorest small settlements – amounting to more than 100,000 individuals – is in an especially grave situation: It simultaneously suffers from poverty, a high incidence of health problems, and the lack of direct and immediate access to the services of a local GP.

To summarise the local inequalities of access to healthcare on the basis of the national statistics, the country is "divided" in terms of the population of smaller settlements, especially small villages. Small settlements with a local GP are well supplied in respect of the patient/doctor ratio, despite the fact that the population of smaller settlements tends to be older, have higher unemployment rates and inadequate funds, and to suffer from poverty. In settlements where there is no GP or where the GP post is unfilled, the ratio of Roma among the general population tends to be significantly higher, and the number of pensioners is also high. The inhabitants of these settlements suffer multiple disadvantages: they are affected by the unfavourable position of the settlement with all its consequences, and by the lack of local and immediately accessible healthcare.

The analysis of national data shows that the significant inequality of access based on location also adds to the doctors' workload. While a little over 80% of doctors work in one settlement and less than 10% work in two settlements, the maximum number of settlements served by one doctor can be as many as eight, according to our data.

The characteristics of a settlement, and the administrative status and size of settlements, fundamentally determine the access of their inhabitants to health services, as well as the workload of their GPs. Just as there are considerable differences in access among patients, so there are significant differences between GPs in terms of their workload, how many patients they serve directly, how long their office hours are and how many hours they are on call.

The distribution by age of doctors is not consistent among practices with considerably differing workloads. The oldest GPs can afford to avoid practice with a higher workload. The youngest ones do not choose practices with higher workloads but are forced to take them in the absence of other options.

An aspect of structural inequalities is the amount of time (attention and work) a GP can spend on a patient. We have observed great differences, which are a result of structural inequalities.

GPs' offices also differ in how well equipped they are, and we have found considerable differences. However, the causes of the presence or absence of equipment are not structural. The practices of the youngest doctors are significantly more well-equipped, middle-aged doctors' practices are more often moderately well-equipped, the offices of older doctors are more often than not below average in equipment. The analysis demonstrated that age is a factor but education is not. Younger doctors have better-equipped offices even when their level of training is lower.

Socially disadvantaged, poor or Roma patients tend to be taken care of by GPs who belong to the younger generation because in settlements where the number of Roma is higher doctors tend to be young. Because young doctors have better-equipped offices, Roma patients are usually served by better-equipped practices. However, the structural advantages or disadvantages seem to be stronger and more significant than, for example, the equipment of a doctor's office.

Equal Opportunity and Social Status

In analysing doctors' attitudes, the issue of whether equal or unequal access is provided to patients of different social status seemed to us more widespread and more complex than simply an issue of prejudice. In our research we considered prejudicial attitudes as a sub-system of mechanisms that promote inequality. We did so because it is obvious from our analyses thus far that one of the most important bases of inequality is structural.

According to our data, indirect discrimination against various social groups, which may not be a result of prejudice, is more frequent than direct discrimination.

Certain GPs offer less expensive medical services to the poor, the unemployed, the Roma or other socially marginalised patients than to others. Their communication with these patients is below average, and conflicts occur with greater frequency than average. The social deprivation of these patients is a causal factor because, among other things, doctors believe that these patients' potential to reduce health risks is low. GPs perceive these patients on the basis of their socio-economic and socio-psychological status, while certain significant dimensions of a GP's practice are defined by these differences in status and not by the patient as a human being.

In addition, GPs determine the level of institutional care³ on the basis of patients' social and socio-psychological status, and therefore the level of institutional care is determined by status and not by a selected protocol.

A certain number of GPs provide therapy at a lower institutional level to patients that are socially marginalised. The social deprivation of patients, as we have seen in relation to the cost of examinations, is a contributing factor. The low assessment of patients' potential to reduce risk to their own health is also an important factor in this regard.

GPs' compassion, or lack thereof, in terms of their taking into consideration the cost of medicine is an independent dimension and has an independent effect on the affordability of the cost of medicine paid by socially disadvantaged patients. A number of GPs can be shown to lack this type of compassion.

A significant number of GPs are not at all or not sufficiently familiar with the considerably higher incidence of disease among Roma and the risks associated with this. Consequently, they do not regard the Romani community as more eligible for increased screening and prevention or intervention which might reduce the incidence of disease among them.

Anti-Romani sentiment or the lack thereof is a measurable factor that impacts the perception of Roma and the level of services provided to them. The causal impact of rejecting anti-Romani sentiments is significant and explains whether a GP has a more or less clear picture of the level of health problems among Roma. It can be proven that the primary cause of the lack of information about the higher incidence of disease among Roma is common and average – not extreme – level of anti-Romani prejudice. On the other hand, it is identified that rejection of anti-Romani feelings is the cause of the clear understanding among doctors of the incidence of Romani health problems.

Anti-Romani sentiments have an impact on medical practice extended to Roma and the attitude towards Roma is to some extent independent of how doctors generally relate to their socially marginalised, poor, and socially disadvantaged patients. This may not be that surprising, since the propensity for anti-Romani feelings appears to have "a life of its own" and is becoming increasingly widespread in society.

Certain versions of anti-Romani feelings do not necessarily result in detrimental situations for Roma with respect to primary healthcare. Even among GPs whose anti-Romani prejudices are strong, there are few who, in comparison with doctors who do not share such prejudices, provide a lower level of services to their Romani patients.

Anti-Romani feelings have a negative, even though not significant, impact on the Roma-doctor relationship. Certain doctors with anti-Romani feelings do not provide the same level of services to their Romani patients as they do to others. However, according to our study, anti-Romani feelings are not a significant factor in primary healthcare services because they can be modified given the right methods.

More important than the damaging effect of negative attitudes towards Roma is the marginalisation of poor, disadvantaged segments, regardless of ethnicity.

We would like to make the following note in closing. It cannot be proven that the apparent inequalities between the level of care received by the social elite and the disadvantaged respectively are caused by direct and open discrimination. In addition, a study conducted among doctors providing the services cannot demonstrate the actual chances for recovery and rehabilitation of socially deprived patients. We can only assume that if the cost and institutional level of care provided to them is lower, if follow-up among them is more infrequent, and the affordability of medications is not always considered, their chances of health maintenance, recovery or rehabilitation will be negatively affected.

Our research has shown, however, that the basic principle that each citizen must receive the same level and the best possible service regardless of social status or ethnicity, suffers.

Recommendations

The writer of this study faces the difficult problem of having to recommend solutions that would ameliorate structural disadvantages and the different degree of disadvantage suffered by healthcare patients in relation to their social status.

The difficulty lies in the fact that structural disadvantages are primarily caused by the structure of settlements in this country, as well as by the resulting economic inequalities, and eliminating them would require considerable long-term inter-ministerial cooperation.

The differences arising from the social status of the patients, namely that certain GPs offer a lower level of services to socially disadvantaged patients, indicate a fundamental deficiency in the solidarity among the various segments of society. Analyses have shown that the number of specialisations or the years of training doctors have, has no bearing on how they relate to socially disadvantaged groups. The level of post-graduate training does not affect the level of anti-Roma feeling either, because it is influenced by deeper causes of socialisation. With that said, we have the following proposals.

• The level of social solidarity demonstrated by GPs should be improved. Each GP, without exception, should regard the members of socially disadvantaged groups as equally valued recipients of services, on a par with the members of the elite who can stand up for their rights. Therefore, courses that focus on the causes and consequences of social stratification must be mandatory (and not elective) in basic and continuing medical training. For this purpose, academic workshops (e.g. ELTE's social work faculty, etc.) and outstanding scholars on poverty in Hungary must be commissioned to prepare targeted course material for basic and continuing medical training. The introduction of suitable course material into medical training must be considered an urgent matter.

• In order to improve the services offered to Roma, new training courses must be prepared and introduced in the framework of continuing education in order to inform GPs of the actual conditions, and the health and social problems of Roma. Concurrently with this effort, a bulletin must be compiled on the basis of available information and research that provides information to GPs about the actual social and health conditions of Roma, including their underlying causes. This bulletin must be distributed among GPs, especially in those settlements where, as far as we are aware, some of the inhabitants are Roma.

• Because it is to be expected that certain GPs will contest the data or claim that all patients receive the same level of services, it is advisable to organise and moderate debates with the participation of appropriate experts (either directly or by creating a specifically targeted Internet portal) which will assist GPs in processing and approving the results.

• Independent of training courses, programs that are effective in creating long-term changes in attitude and in decreasing the existing negative feelings towards the poor and the Roma must be prepared and adapted, after gaining an overview of the relevant international experience.

• Since the most effective way of combating prejudicial attitudes is to penalise the prejudicial behaviour, and the discrimination to which it gives rise, a measuring and monitoring system must be developed for regular application among doctors and patients which is capable of rendering these negative phenomena transparent. Transparency must be followed by indicating that these attitudes are socially unacceptable (socio-psychological punishment).

• A PR programme must be developed which can effectively portray in the media the situation of Roma, as well as the harmful consequences of prejudicial attitudes.

Health Visitors

The designation of health visitors' districts, and the number of health visitors in the various counties and settlements, fail to meet statutory requirements, and in some cases actually contradict them.

Health visitors' tasks are unevenly distributed. While the majority of health visitors work in one settlement on average and perform one basic task at low or moderate levels of intensity, one fifth of health visitors perform several tasks at a high level of intensity in a number of settlements.

Behind the distribution of health visitors' districts within counties and settlements are very serious inequalities in access caused by a structural imbalance. In disadvantaged, poorer areas consisting of small villages, a smaller number of health visitors carry higher workloads and perform extra services, while counties and settlements in more favourable positions employ more health visitors with lower workloads.

More than one fifth of all the health visitors studied carry high workloads and also care for a high number of Roma.

In most cases the high number of Roma is a simple accompanying feature of the settlements' characteristics. The reason why health visitors work with so many patients and in several settlements is not because Roma live there, but the opposite: Roma tend to live in such settlements where health visitors already have a higher workload.

However, the differences between workloads resulting from serious structural imbalances do not mean that health visitors with higher workloads invest less energy in their work or attend fewer training courses. Health visitors in districts with high percentage of Roma did not participate in more hours of training than in other places, and the high number of Roma does not (so far) indicate a greater participation in training.

Therefore, the distribution of health visitors' districts points to serious structural inequalities. In many cases the actual number of patients is three times the optimum number specified in the relevant government regulations (quite apart from the other work commitments). It is a fundamental problem that the local distribution of health visitors' districts and the fluctuating number of patients are both contrary to the letter and the spirit of the law, and do not serve the principle of equal opportunity and equal access.

Health visitors' training and their attitude towards their patients determine the extent to which they take into consideration the needs of their patients. Counselling, the communication of basic information and health-related advice that comprise a health visitor's tasks are interactive processes that greatly depend on the health visitors' attitudes (and not so much on the characteristics of their patients). This observation, however, is more relevant to their attitude to Roma patients than to others.

A fairly large percentage of health visitors are well-trained, care for many persons and are also committed to what they do, which means that they have an excellent grasp of their patients' needs.

A higher percentage of highly trained health visitors who are tolerant towards Roma understand that their Roma patients have numerous healthcare needs.

On the other hand, health visitors with lower levels of training and who are unable to perceive their patients' needs, and health visitors who have some form of anti-Romani attitude have a lesser understanding of their Roma patients' needs. This "blinkered" attitude hinders the true perception of Roma patients' healthcare needs.

The occasional lack of understanding with respect to patients' needs interferes with the provision of equal services because counselling is an interactive activity, which is performed through communication between the counsellor and the patient. If a counsellor creates a communicational space that the patient perceives as inadequate in assessing his/her real needs, the counsellor will be unable to help because an atmosphere of mistrust has been created (towards the potential help).

Health visitors who display some form of anti-Romani attitude have been proven to be less effective in meeting their patients' needs. As a result, on the basis of our knowledge of the communicative dynamics of service-oriented professions, these health visitors are less effective than average in assisting their Roma patients.

A brief summary of our recommendations, aimed at improving, and sometimes creating, equal access to health visitors' services, is as follows:

• Because inequalities in access are fundamentally structural in nature, a new distribution of health visitors' districts must be created which complies more strictly with the stipulations of the relevant decree and is better adapted to the patients' location demographics and socio-economic conditions, as well as to health visitors' work capacity.

• We have two proposals in relation to training which are aimed at improving health visitors' performance with regard to Roma patients: We need to ensure that most health visitors participate in general training courses that encompass all aspects of a health visitor's work, consisting of at least 150 hours of training spread out over a minimum of 5 years. In addition, training courses must be developed and introduced that provide information on the health status and social problems of the Roma population (on the national and local level). These training courses must also increase health visitors' ability to perceive the actual needs of Roma patients (even though they may not be explicitly stated) and to provide appropriate responses for these needs.

• Independent of the training courses, programs must be developed and/or adapted (after gaining an overview of international experiences) that can effectively and permanently modify attitudes and reduce anti-Romani feelings. We emphasise that these programs should be independent of the trainings because the relevant literature, experiences and hypotheses suggest that modifying purely cognitive content and obtaining new information has no bearing on prejudicial attitudes.

• Since the most effective way of combating prejudicial attitudes is to penalise the prejudicial behaviour, and the discrimination to which it gives rise, a measuring and monitoring system must be developed for regular application among doctors and patients which is capable of rendering these negative phenomena transparent. Transparency must be followed by indicating that these attitudes are socially unacceptable (socio-psychological punishment).

• A PR programme must be developed that can effectively portray in the media the situation of Roma as well as the harmful consequences of prejudicial attitudes.

Anti-Romani Attitudes

We examined anti-Romani attitudes among three groups: general practitioners and health visitors who work in settlements where Roma account for more than 1% of the local population, and medical students in Hungarian medical schools.

We treated anti-Romani attitudes as a complex system of attitudes consisting of three basic issues: negative stereotyping of Roma, attitudes to discrimination against Roma, and an emotional distance towards Roma. This concept of measuring anti-Romani sentiment is based on national and international tests that examined prejudicial attitudes against minorities by the majority population.

During the study we identified five markedly different groups. 6.3% of the people studied strongly reject all types of anti-Romani attitudes, 21% do not have anti-Romani attitudes, and 28.3% have no propensity towards accepting discrimination. Consequently, 55.6% cannot be characterised by any form of anti-Romani attitude.

Therefore, only less than half of the people studied have some form of anti-Romani attitude. 14.1% of the people in the study can be characterised as having strongly negative attitudes towards Roma, which means that they engage in negative stereotyping, approve of discrimination, and have a marked emotional distance. Thirty percent have a tendency towards anti-Romani attitudes, which means that they can be characterised by all three components of anti-Romani attitudes but to a lesser degree than those who have strong anti-Romani feelings.

Causal analyses suggest that the tendency towards anti-Romani attitudes is fairly deep-seated in society, and is more widespread among the younger generation than the older. The people we studied belong to the social elite and practice or prepare for service-oriented professions. Therefore the extent, deep roots and pervasiveness among the younger generation of anti-Romani attitudes presents a scary picture.

The intensity of anti-Romani attitudes among GPs and health visitors, in other words those who actively practice a service-oriented profession, is lower than among medical students.

Nevertheless, working with a larger or smaller number of Roma does not have an effect on anti-Romani attitudes. Anti-Romani attitudes are primarily the result of deeply ingrained social values such as intolerance.

Managing and decreasing anti-Romani attitudes is an urgent social problem and is not solely the concern of a particular profession or institution. Because the fundamental cause of anti-Romani attitudes is not lack of information but ingrained, socialised values, decreasing anti-Romani attitudes is not primarily a matter of education. We must create conditions with the help of regulation and education that make anti-Romani attitudes socially unacceptable in both everyday life and in relation to social attitudes. Only then can we expect the prevalence of anti-Romani attitudes to diminish among the next generation.

Endnotes:

1. The translation into English of the survey's findings was commissioned by the ERRC to the Budapest-based Impact Training Ltd. The full text in English is available on the ERRC's website at: http://www.errc.org/uploads/upload_en/file/00/CC/m000000CC.doc. The original document is available at: http://www.delphoi.hu.
2. Based on Gábor Kertesi - Gábor Kézdi: The Roma Population in Hungary (documentation and data, Socio-typo, Bp., 1998), and László Hablicsek: The Demographic Indicators of Roma, Experimental Forecast for 2050. Central Statistical Office (KSH) Institute of Demographics, 1999.
3. This is a question of whether the doctor offers therapy on the spot in his own office, or after the local therapy he refers the patient to a higher institutional and competence level, or refers the patient to a higher institutional level immediately after the diagnosis. We measured this independently from the effects of disease, the patient's age, etc., exclusively in the context of the patients' social and socio-psychological status.