The number of persons who are not employed has increased in Sweden since the early 1990s. Unemployment has been found to influence health, especially when unemployment rates are low. The extent to which unemployment affects health when unemployment is high is less clear, and this needs to be further studied. To improve health in the population, the health care system should offer equal access to health care according to need. It is important to study whether the employment status hinders the fulfilment of this goal. This thesis is based on four papers: Paper I and II aimed at analysing self-rated health versus mortality risk in relation to employment status, during one period of low unemployment and one period of high unemployment. Paper III and IV assessed the use of medical health care services and unmet care needs among persons who were unemployed or otherwise not employed. The goal was to analyse what health problems lead people to either seek or abstain from seeking care, and what factors encumber or facilitate this process. The overall results indicate that being unemployed or outside the labour force was associated with an excess risk of poor self-rated health, symptoms of depression, mental and physical exhaustion and mortality. The differences in self-rated health between the unemployed and employed were larger when unemployment levels were high, than when they were low. More groups of the unemployed were also afflicted with poor health when unemployment was high. Thus, poor health among the unemployed seems to be a public health problem during high levels of unemployment. Lack of employment was related to abstaining from seeking care, despite perceiving a need for care, and this was related to psychological symptoms. To deal with the needs of the unemployed and others who are outside the labour force it would be useful to develop and implement interventions within the health care system. These should focus on psychological and psychosocial problems. Future research should analyse how to facilitate health-promoting interventions among persons who are not anchored in the labour market.

Individuals with overweight/obesity have been found to exhibit more negative attitudes toward health care and disproportionate rates of health care delay and avoidance, compared to their healthy weight peers. The present study sought to examine potential mechanisms through which weight status influences health care utilization and attitudes. Six hundred and thirty-three students completed a questionnaire measuring weight status, perceived weight bias, patient-provider relationship, and health care utilization and attitudes. Although the majority of the paths in the proposed theoretical mediation model were supported by the present findings, there was no support for the anticipated link between perceived weight bias and the patient-provider relationship or weight-related embarrassment. Overall, these results corroborated previous findings in a novel sample, but did not provide evidence that perceived weight bias mediates the relationship between weight status and health care outcomes. Possible explanations for these findings are deliberated.

A model for health care utilization was applied to a sample of low income women. Demographic Predisposing, Psychosocial Predisposing, Illness Level, and Enabling indicators were examined separately for African American (n = 266), Anglo American (n = 200), and Mexican American (n = 210) women. Structural Equation Modeling revealed that for African American and Anglo American women, Illness Level, the only significant path to Utilization, had a mediating effect on Psychosocial Predisposing indicators. The model for Mexican Americans was the most complex with Enabling indicators affecting Illness Level and Utilization. Psychosocial Predisposing indicators were mediated by Illness Level and Enabling indicators which both directly affected Utilization. Implications of the results for future research are addressed.

A equidade na utilização de cuidados de saúde deve ser considerada como questão central em qualquer política de saúde que pretenda contribuir para uma sociedade mais justa. Desse modo, o objetivo desta tese é analisar o desempenho da entrega de cuidados no Brasil em termos de equidade por meio de violações do princípio de equidade horizontal na utilização dos serviços de cuidados de saúde e da decomposição dos determinantes da desigualdade na utilização do cuidado relacionada à renda. A desigualdade na distribuição de cuidado médico pela renda é capturada por índices de iniquidade para a utilização de serviços de consultas médicas e internações hospitalares. Esses índices mostram se existem diferenças no uso de serviços de cuidados de saúde entre indivíduos com similares necessidades de saúde. Para explicar as causas da desigualdade, Wagstaff, van Doorslaer e Watanabe (2003) propõem que a medida do grau de desigualdade seja decomposta nas contribuições dos fatores explicativos do uso. A análise também considerou a perspectiva da desigualdade, o que permitiu observar não apenas desigualdades sociais mas também variações regionais na entrega de cuidados de saúde. Os resultados mostraram iniquidade horizontal pró-rico no uso de consultas médicas e pouca evidência de iniquidade no uso de internações. O padrão de iniquidade horizontal no uso se repetiu para todas as regiões, mas regiões menos desenvolvidas como, o Norte e o Nordeste, apresentaram maior grau de iniquidade. A decomposição da desigualdade mostra que contribuições de fatores de necessidades de saúde são principalmente pró-pobre, uma vez que pessoas mais pobres tendem a possuir maiores necessidades de cuidado. Por outro lado, as contribuições dos determinantes sociais foram bastante diversificadas. Renda e escolaridade contribuem para aumentar a distribuição pró-rico no uso de consultas e reduzir a contribuição pró-pobre no uso de internações hospitalares. A contribuição da condição de atividade foi, em geral, pró-pobre, podendo ser explicada pelo maior custo de oportunidade das pessoas ocupadas em procurar cuidados com a saúde. As contribuições dos plano de saúde e das desigualdades regionais são examinadas com maior atenção por serem alvo direto de políticas de saúde. Assim, contribuições pró-rico do plano de saúde e das desigualdades regionais poderiam ser reduzidas, por exemplo, por estratégias com foco em grupos de renda mais baixa e pela ampliação de recursos físicos e humanos das áreas menos desenvolvidas. / Equity in healthcare utilization should be considered as a basic issue at any health policy that you want to contribute to a fairer society. Thus, the aim of this thesis is to analyze the performance of medical care delivery in Brazil in terms of equity through violations of the horizontal equity principle of healthcare use and of the decomposition of the determinants of income-related inequality in the use of medical care. The inequality in the distribution of medical care by income is captured by inequity indices. This indices show if there are differences in the use of healthcare services among individuals with similar health needs. To explain the causes of the income-related inequality, Wagstaff, van Doorslaer, Watanabe (2003) proposed that the measure of the degree of inequality is decomposed into the determinants of use. The analysis also considered the regional perspective of inequality, which allowed us to observe not only social inequalities but also regional variations in the delivery of healthcare. The results find evidence horizontal inequity pro-rich in the utilization of doctor visits and little evidence de inequity in inpatient care use. The pattern of horizontal inequity in use is repeated for all regions, but less developed regions like the North and Northeast had a higher degree of iniquity. The decomposition of inequality shows that contributions of factors of health needs are mainly pro-poor, because poorer people tend to have greater care needs. On the other hand, the contributions of the social determinants were enough diverse. Income and education contributed to increase the pro-rich distribution of use of doctor visits and reduce the distribution pro-poor inpatient care. The contribution of status of activity was mostly pro-poor because probably economically active people have a higher cost of opportunity in seeking healthcare. The contributions from the health insurance and regional inequalities are examined more closely by being a direct target of health policy. Accordingly, contributions pro-rich of health insurance and of regional inequalities could be reduced, for example, by strategies focused on lower income groups and by the expansion of physical and human resources of less developed areas.

The aim of this study was to explore the Health Service Utilization (HSU) of homeless people in Dublin. In particular, it sought to identify a critical realist explanatory model for why the HSU of homeless people differs from that of the general population. Critical realist (CR) ethnography was used as the research methodology and was supplemented with forty-seven semi-structured interviews and two focus groups. The HSU of homeless participants in Dublin is described. When compared to the domiciled population, homeless people were found to have a tendency to present late on in their illness, to have higher utilization of primary care services and lower utilization of secondary care services and to avoid psychiatric services. The factors that influenced participants HSU tendency are identified as external or internal influences on HSU. External factors are described as physical, administrative or attitudinal barriers or deterrents; or external promoters of health service usage. Internalised inhibitors and promoters are illustrated as either cognitions or emotions that are developed in reaction to external circumstances and which either negatively or positively impact on health service usage. Interactions between health professionals and participants that resulted in exclusion (by the health professional or self-exclusion) are described as Conversations of Exclusion. A critical realist model was outlined that offers an explanation for why homeless people’s HSU differs from that of the general population in Dublin. This model included a description of the generative mechanisms identified as producing the HSU tendencies in the study population. The implications of this new model are discussed in the light of the literature and previous models that seek to explain the HSU of homeless people.

<p>Only 12% of the patients died at home. When the period between diagnosis and death was less than one month, every patient died in an institution. Younger patients, married</p><p>patients, and those living within the 40 km radius of the hospital utilized more hospital days. The "length of terminal hospitalisation" and the "proportion of days at home/ total inclusion days" seemed to be feasible outcome varibles when evaluating a palliative support service. The hospital-based palliative support service in this study defrayed its own costs due to a median saving of 10 hospital days/patient, compared with matched historical controls.</p><p>A 19-item quality of life questionnaire (AQEL) was developed which evidenced good signs of reliability and validity. The item most closely correlated to global quality of life was the sense of meaningfulness. This was true for both patients and their spouses. Patients´ levels of pain and anxiety did not increase at the end of life. In this study we could not find convincing evidence for an additive analgesic effect of paracetamol in morphine therapy of pain in cancer patients.</p>

<p>The number of persons who are not employed has increased in Sweden since the early 1990s. Unemployment has been found to influence health, especially when unemployment rates are low. The extent to which unemployment affects health when unemployment is high is less clear, and this needs to be further studied. To improve health in the population, the health care system should offer equal access to health care according to need. It is important to study whether the employment status hinders the fulfilment of this goal. </p><p>This thesis is based on four papers: Paper I and II aimed at analysing self-rated health versus mortality risk in relation to employment status, during one period of low unemployment and one period of high unemployment. Paper III and IV assessed the use of medical health care services and unmet care needs among persons who were unemployed or otherwise not employed. The goal was to analyse what health problems lead people to either seek or abstain from seeking care, and what factors encumber or facilitate this process. </p><p>The overall results indicate that being unemployed or outside the labour force was associated with an excess risk of poor self-rated health, symptoms of depression, mental and physical exhaustion and mortality. The differences in self-rated health between the unemployed and employed were larger when unemployment levels were high, than when they were low. More groups of the unemployed were also afflicted with poor health when unemployment was high. Thus, poor health among the unemployed seems to be a public health problem during high levels of unemployment. Lack of employment was related to abstaining from seeking care, despite perceiving a need for care, and this was related to psychological symptoms. To deal with the needs of the unemployed and others who are outside the labour force it would be useful to develop and implement interventions within the health care system. These should focus on psychological and psychosocial problems. Future research should analyse how to facilitate health-promoting interventions among persons who are not anchored in the labour market. </p>

<p>One of the most vulnerable groups in Swedish society today is single parent families, a group that has increased over the last thirty years in proportion to married and cohabiting parents. The aims of this thesis are to study inequality and inequity in health and health care utilization with regard to whether parents are single or couple (married/cohabiting), to investigate whether the concept of social capital may provide us with further understanding when analyzing inequality and inequity in health and to investigate how the mental health of single parent children may differ from couple parent children and to what extent this difference may be due to parental socio-economic and socio-demographic characteristics, including social capital. The results from the papers on which this thesis is based indicate that both single mothers and single fathers have poorer health than couple mothers and fathers. Single mothers also refrain from seeking medical care to a greater extent. The financial advantage of single fathers, in comparison with single mothers, might have an impact on their medical care utilization, since they seemingly seek and consume health care to an extent that matches their poorer health. Social capital has as robust an association with self-rated health as any traditional social determinant of health. Four parental characteristics were found to be independently associated with children’s mental health; being a single parent, ‘poor parental health’, limited social support and low levels of social capital. The uneven distribution of all investigated determinants of health, including social capital, gives us reason to conclude that our findings indeed raise concerns about equity. Action taken by society to enable single parents to increase their social capital might improve their and their children’s health. It may also be clearly stated that financial status has a major impact on both health and health care utilization. This particular characteristic is also rather accessible to alteration, for example through financial transfers between groups in society. </p>

Objective - To assess the relationship between schizophrenia and cardiovascular disease by evaluating metabolic risk associated with treatment for schizophrenia, prevalence of cardiovascular risk factors (CV-RF) and disease (CV-D), and health care utilization in people with schizophrenia compared to the non-schizophrenic population. Methods Four studies were completed to evaluate the dissertation objectives. A systematic review was completed to quantify the change in metabolic parameters associated with use of atypical antipsychotic agents. The second study utilized a period prevalence design to compare prevalence of CV-RF (diabetes, hypertension, dyslipidemia) and CV-D in people with and without schizophrenia using the administrative databases of Alberta Health and Wellness. General and cardiac specialist health care utilization was evaluated in people with schizophrenia using data from Alberta Health and Wellness. Lastly, results from the Canadian Community Health Survey were used to evaluate prevalence of CV-RF and CV-D while controlling for important lifestyle and demographic variables unavailable in the databases of Alberta Health and Wellness. Results Use of atypical agents, particularly clozapine, resulted in statistically significant weight gain and increases in total cholesterol and blood glucose compared to typical agents. Having schizophrenia was associated with a significantly higher prevalence of diabetes, obesity, smoking, and CV-D compared to people without schizophrenia. Individuals with schizophrenia visited a general practitioner and the emergency department more often, and were more likely to be hospitalized than those without schizophrenia. Despite having a higher prevalence of coronary artery disease, individuals with schizophrenia were significantly less likely to visit a cardiologist or undergo revascularization compared to people with coronary artery disease who did not have schizophrenia. Conclusion Individuals with schizophrenia have a considerable burden of cardiovascular disease compared to people without schizophrenia. This is likely a result of a number of factors, including medications used to treat schizophrenia, the increased prevalence of smoking and other unhealthy lifestyle factors, and the increased prevalence of cardiovascular risk factors in people with schizophrenia. Individuals with schizophrenia utilize the general health care system more frequently than their non-schizophrenic counterparts, therefore the opportunity exists for monitoring for and management of modifiable cardiovascular risk factors in this vulnerable population.

Only 12% of the patients died at home. When the period between diagnosis and death was less than one month, every patient died in an institution. Younger patients, married patients, and those living within the 40 km radius of the hospital utilized more hospital days. The "length of terminal hospitalisation" and the "proportion of days at home/ total inclusion days" seemed to be feasible outcome varibles when evaluating a palliative support service. The hospital-based palliative support service in this study defrayed its own costs due to a median saving of 10 hospital days/patient, compared with matched historical controls. A 19-item quality of life questionnaire (AQEL) was developed which evidenced good signs of reliability and validity. The item most closely correlated to global quality of life was the sense of meaningfulness. This was true for both patients and their spouses. Patients´ levels of pain and anxiety did not increase at the end of life. In this study we could not find convincing evidence for an additive analgesic effect of paracetamol in morphine therapy of pain in cancer patients.