Data mainly from one-off surveys clearly show that the health of Roma, the largest ethnic minority of Europe, is much worse than that of the general population. However, results from comprehensive exploratory studies are missing. The aim of our study was to create a complex database for comparative and association studies to better understand the background of the very unfavourable health of Roma, especially the high burden of cardiometabolic diseases. A three-pillar (questionnaire-based, physical and laboratory examinations) health survey was carried out on randomly selected samples of the Hungarian general (HG, n = 417) and Roma (HR, n = 415) populations, and a database consisting of more than half a million datapoints was created. Using selected data, the prevalence rates of metabolic syndrome (MetS) and of its components were determined, and to estimate the risk of insulin resistance (IR), surrogate measures (the homeostasis model assessment of insulin resistance index, quantitative insulin sensitivity check index, McAuley and TyG indices and the TG/HDL-C ratio) were calculated. Receiver operating characteristic curve analysis and Youden’s method were used to define the optimal cut-off values of each IR index. The prevalence of MetS was very high in both study populations (HG: 39.8%, HR: 44.0%) with no statistically significant difference between the two groups in females or males. The prevalence of MetS showed a very marked increase in the HR 35–49 years age group. Among surrogate measures, the TyG index showed the greatest power for predicting IR/MetS at a cut-off value of 4.69 (77% sensitivity, 84% specificity) and indicated a 42.3% (HG) and 40.5% (HR) prevalence of IR. The prevalence of MetS and IR is almost equally very unfavourable in both groups; thus, the factors underlying the high premature mortality burden of Roma should be further clarified by investigating the full spectrum of risk factors available in the database, with a special focus on the access of Roma people to preventive and curative health services.

Background A Model Programme of primary care group practices was implemented in Hungary between 2013 and 2017 – where virtually all GPs had worked in single practices – aiming to increase preventive service uptake and reduce inequalities based on a bilateral agreement between the Swiss and Hungarian governments. Group practices employed a wide variety of health professionals as well as support workers called health mediators. Employment of the latter was based on two decades of European experience of health mediators who specifically facilitate access to and use of health services in Roma minority groups. Health mediators had been recruited from local communities, received training on the job, and were tasked to increase uptake of new preventive services provided by the group practices by personal contacts in the local minority populace. The paper describes the contribution of the work of health mediators to the uptake of two new services provided by group practices. Methods Quantitative analysis of depersonalized administrative data mandatorily reported to the Management of the Programme during 43 months of operation was carried out on the employment of health mediators and their contribution to the uptake of two new preventive services (health status assessment and community health promoting programmes). Results 80% of all clients registered with the GPs participated at health status assessment by invitation that was 1.3–1.7 times higher than participation at the most successful national screening programmes in the past 15 years. Both the number of mediator work minutes per client and participation rate at health status assessment, as well as total work time of mediators and participants at community health events showed high correlation. Twice as many Roma minority patients were motivated for service use by health mediators compared to all patients. The very high participation rate reflects the wide impact of health mediators who probably reached not only Roma minority, but vulnerable population groups in general. Conclusion The future of general practices lays in multidisciplinary teams in which health mediators recruited from the serviced communities can be valuable members, especially in deprived areas.

The risk of premature mortality caused by cardiovascular diseases (CVDs) is approximately three times higher in the Central Eastern European region than in high income European countries, which suggests a lack and/or ineffectiveness of preventive interventions against CVDs. The aim of the present study was to provide data on the relationship between premature CVD mortality, statin utilization as a preventive medication and socioeconomic deprivation at the district level in Hungary. As a conceptually new approach, the prescription of statins, the prescription redemption and the ratio between redemption and prescription rates were also investigated. The number of prescriptions for statins and the number of redeemed statin prescriptions were obtained from the National Health Insurance Fund Administration of Hungary for each primary healthcare practice for the entire year of 2012. The data were aggregated at the district level. To define the frequency of prescription and of redemption, the denominator was the number of the 40+-year-old population adjusted by the rates of 60+-year-old population of the district. The standardized mortality rates, frequency of statin prescriptions, redeemed statin prescriptions, and ratios for compliance in relation to the national average were mapped using the “disease mapping” option, and their association with deprivation (tertile of deprivation index as a district-based categorical covariate) was defined using the risk analysis capabilities within the Rapid Inquiry Facility. The risk analysis showed a significant positive association between deprivation and the relative risk of premature cardiovascular mortality, and a reverse J-shaped association between the relative frequency of statin prescriptions and deprivation. Districts with the highest deprivation showed a low relative frequency of statin prescriptions; however, significantly higher primary compliance (redemption) was observed in districts with the highest deprivation. Our data clearly indicate that insufficient statin utilization is strongly linked to the so-called physician-factor, i.e., a statin prescription. Consequently, statin treatment is poor and represents a significant barrier to reducing mortality, particularly among people living in highly deprived areas of the country.

Reinforcement learning (RL) is a powerful concept underlying forms of associative learning governed by the use of a scalar reward signal, with learning taking place if expectations are violated. RL may be assessed using model-based and model-free approaches. Model-based reinforcement learning involves the amygdala, the hippocampus, and the orbitofrontal cortex (OFC). The model-free system involves the pedunculopontine-tegmental nucleus (PPTgN), the ventral tegmental area (VTA) and the ventral striatum (VS). Based on the functional connectivity of VS, model-free and model based RL systems center on the VS that by integrating model-free signals (received as reward prediction error) and model-based reward related input computes value. Using the concept of reinforcement learning agent we propose that the VS serves as the value function component of the RL agent. Regarding the model utilized for model-based computations we turned to the proactive brain concept, which offers an ubiquitous function for the default network based on its great functional overlap with contextual associative areas. Hence, by means of the default network the brain continuously organizes its environment into context frames enabling the formulation of analogy-based association that are turned into predictions of what to expect. The OFC integrates reward-related information into context frames upon computing reward expectation by compiling stimulus-reward and context-reward information offered by the amygdala and hippocampus, respectively. Furthermore we suggest that the integration of model-based expectations regarding reward into the value signal is further supported by the efferent of the OFC that reach structures canonical for model-free learning (e.g., the PPTgN, VTA, and VS).