631872 research-article2016
SJP0010.1177/1403494816631872VilhjalmssonShort Title
Scandinavian Journal of Public Health, 1–9
Original Article
Public views on the role of government in funding and delivering health services
Runar Vilhjalmsson University of Iceland, Reykjavik, Iceland
Abstract Aims: Public surveys in socialized health systems indicate strong support for the role of government in health care, although different views can be detected. The study considers the public’s views on public versus private funding and delivery of health services. Methods: The study is based on a representative national sample of 1532 Icelandic adults, aged 18 and older, who participated in a national public issues survey. Respondents were asked about government spending on health care and whether the government or private parties should deliver health services. Results: The great majority of respondents thought that the government should spend more on health care, and should be the primary provider of care. Lower age, female gender, countryside residence, and expected high use of health care were related to greater support for governmental funding. Furthermore, countryside residence, less education, lower income, not being a governmental health worker, expected high health care use, and left-wing political ideology were all related to greater support for governmental delivery of health care. Conclusions: Despite sociodemographic variations, the study finds strong overall support for the role of government in funding and delivering health care. Previous perspectives and hypotheses of welfare state endorsement received mixed support, suggesting that further theoretical and empirical work is needed to better account for public views on the role of government in health care. Key Words: Health care, public opinion, welfare state, public funding, public delivery
Introduction Health care is a cornerstone of the welfare state. A review of health care systems in OECD countries shows that they fall broadly into three categories—private, socialized, and national health insurance (NHI) systems [1]. The role of the state differs between these systems with regard to the regulation, ownership, funding, and delivery of services. Socialized health care systems (e.g. the Nordic countries and Great Britain) and NHI systems (e.g. France and Germany) purport to guarantee citizens equal access to health services at time of need [1]. Funding for health services in socialized systems comes mostly from national budgets, and the state or lower levels of government have a substantial direct role in the delivery of services. In NHI systems, sickness funds with employer and employee contributions (mandatory health insurance)
pay most service costs, the role of the state or lower levels of government is more limited, and services are mostly delivered by privately owned and operated service units. Public views on the welfare state In previous studies of public views on the welfare state, respondents have been asked whether they feel that the state or lower levels of government should spend more or less on various public services and benefits. Respondents have also been asked whether they would be willing to pay higher taxes if spent on certain services or benefits. Finally, they have been asked what the role of the state should be as provider of public services [2–6]. The studies have identified
Correspondence: Runar Vilhjalmsson, University of Iceland, Eirbergi, Eiríksgotu 34, Reykjavik, IS-101, Iceland. E-mail: [email protected] (Accepted 5 August 2015) © 2016 the Nordic Societies of Public Health DOI: 10.1177/1403494816631872
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
2 Vilhjalmsson differences of opinion (attitudinal cleavages) as regards the role of government in welfare services and programs. Perspectives and hypotheses of welfare state attitudes Several theoretical perspectives have been proposed to account for variations in welfare state attitudes. According to Power Resources Theory, the development of the welfare state is understood in terms of a struggle between labor and capital. Support for the welfare state and its programs is guided by class interests and corresponding political ideologies. Therefore, lower class or socioeconomic status, and left-wing political ideology, are thought to be key factors underlying greater support for welfare state services [7]. A more recent New Policy Perspective argues that developed postindustrial societies have created new interest groups, including beneficiaries (e.g. the elderly and patients) and employees of the welfare state (predominantly women), that oppose retrenchment of the welfare state, resulting in new welfare state attitudinal cleavages by age, gender, occupational sector (public vs. private), and client status (patients) [8]. Lastly, the Self-Interest Perspective argues that attitudes toward the welfare state are increasingly determined by personal interests rather than the collective values of universalism and solidarity. The degree to which individuals derive direct benefits through receipt of welfare state services, and incur direct costs (e.g. pay taxes), will determine their attitude toward such services, with low-tax-paying beneficiaries having the most positive attitude [2]. Based on these perspectives, several hypotheses have been proposed to explain variations in support for public spending, services, and benefits. In line with Power Control Theory, the political ideology hypothesis posits that individuals’ location on the left– right continuum (which in turn is linked to their socioeconomic status) generally explains their views on the role of government. Those to the left are more likely to stress the importance of solidarity and equality, and thereby support public services and assistance programs. Those to the right, on the other hand, are more likely to stress individual freedom and responsibility, and are less interested in state interference in human affairs, whether in terms of funding, services, or transfers [9,10]. Following the New Policy and Self-Interest Perspectives, the demand hypothesis proposes that endorsement of public services and support is greater among those having used such services in the past, or expecting or hoping to use such services in the future [10]. This could for
example apply to the elderly, expecting parents, individuals with certain diseases, and individuals with relatives that need assistance. Furthermore, the employment sector hypothesis states that public employees (predominantly women) are generally more supportive of public programs than those working in the private sector [3]. It is either argued that this is because public employees are more supportive of public sector colleagues or the clients of public institutions [11], or that public employees have a vested interest in the maintenance of public services [4]. One variation of this hypothesis would be that employees of publicly funded and delivered health services are more ardent supporters of such services than employees working elsewhere. As laid out by the Self-Interest Perspective, the tax burden hypothesis argues that individuals may not oppose (or even support) public programs as long as their contribution to the government is modest or does not cut too heavily into their finances. The hypothesis predicts that support for public programs diminishes as tax payments increase, either in absolute terms or in relation to income [2]. The above perspectives and hypotheses may help explain sociodemographic differences in support for governmental funding and delivery of welfare state programs and services. Studies find that support for governmental funding and delivery of services is generally stronger among women than men [3,5,12]. Women may express greater support than men because they use welfare state services more or expect greater use of such services in the future, are more often employed in the public sector, pay lower taxes, or are ideologically farther to the left. Older individuals and retirees tend to support public financing and service delivery more strongly than their younger counterparts [5,12]. The elderly may be more supportive of public funding and delivery of care because they use welfare state services more, or because they pay lower taxes than their younger and higher earning counterparts. Finally, working class individuals, and individuals with low education or income, are repeatedly shown to be more supportive of public programs than their higher status counterparts [5,10,12]. Lower status individuals may be more supportive because they use welfare state services more, expect greater service use in the future, pay lower taxes, or are ideologically farther to the left [2,3,5]. Government funding and government delivery of health care Public opinion polls repeatedly show that health care is considered one of the top political priorities across different health systems [13,14]. Studies also indicate
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
The role of government in funding and delivering health services 3 that support for public financing and delivery of health services has been quite stable through the years, and is stronger in countries where the government is already heavily involved in the funding and delivery of care [6,9,12,15]. Research in the Nordic countries indicates that the public generally supports the socialized model that has traditionally characterized the health care system in these countries. A large majority believes that the government should be responsible for health care, and support for the role of government is even stronger in the case of larger service units such as hospitals and community health centers where governmental responsibility has been long standing and comprehensive [5,6,10,16]. Nevertheless, there is much to be learned about public views on the proper role of government in health care. For example, it is important to distinguish between public views on government funding of services on the one hand, and government delivery of services on the other. Most previous studies have not made this distinction [but see 5,12]. Furthermore, most existing studies of public views on the role of government have focused on the welfare state or welfare programs in general, or on programs other than health care. These studies often assume that public attitudes relating to the welfare state are largely harmonious across different public programs. It is not clear whether, or to what extent, reported results from these studies apply to the funding and delivery of health services. This should be clarified, as there are indications of greater uniformity in public views on the role of government in health care, as opposed to other services or programs [2]. Finally, although different perspectives and hypotheses have been proposed to explain variations in support for public services and spending, few studies have assessed these hypotheses simultaneously. An example of a socialized health system The current study examines public views toward the funding and delivery of health services in Iceland—a country operating a socialized health system along with other Nordic countries. General hospitals, and most nursing homes and community health centers, are governmentally owned and operated, but doctors’ offices and certain specialty hospitals and rehabilitation centers are privately owned and operated. General hospitals and community health centers are primarily funded directly through the central state budget, but nursing homes, private (non-profit) specialty hospitals, and outpatient specialist medical care and prescription drugs, are mostly funded by the Icelandic Health Insurance Institution (Sjúkratryggingar Íslands), which
in turn receives its funding from the central state budget. Icelanders pay deductibles and coinsurance for outpatient physician visits and prescription drugs according to complex and ever-changing rules (lower rates apply for children, the disabled, and the elderly). Patients are allowed to go directly to specialists without referrals, and the relatively high charges for specialist visits are the same whether or not the patient makes the appointment by herself (which is common in Iceland), or is referred by her family doctor [17]. The purpose of the study The study considers the public’s views on government versus private funding and delivery of health services, and whether the views differ by age, gender, residence, education, and income. Subsequently, the study tests four hypotheses, proposed to explain variations in views on the role of government in health care. These hypotheses refer to differences in political ideology, demand for health services, employment sector, and tax burden. Methods The study uses data from a cross-sectional public issues internet survey of Icelandic adults, aged 18 and older, conducted by the Social Science Institute at the University of Iceland in March/April of 2013 [18]. The random sample was stratified by gender and age using the National Register. The survey complied with regulations and requirements concerning human subjects research, as laid out by the Data Protection Authority in Iceland (Persónuvernd). Respondents were informed that participation was voluntary. They received a link via e-mail to an internet page containing the electronic survey. To enhance participation, a total of three follow-up reminders were sent out to those who had not filled out and returned their questionnaire. The response rate was 74% (N=1532). Demographic comparison between the respondent group and population showed very similar age, gender, and residential composition, although there was a slight underrepresentation of the oldest (60+) age group (a difference of only 1.4%). Variables Sociodemographic variables included age (in years), gender, education (basic, high school, university level), monthly personal income (in thousand Icelandic Kronur, IKr), and residence (Reykjavik area vs. countryside). Employment sector was assessed with two dichotomous variables distinguishing between public vs. private sector employees, and
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
4 Vilhjalmsson Table I. Overall support for governmental funding, and governmental versus private delivery of health services. The government should spend…
Health care should be delivered…
More on health care
Same on health care
Less on health care
Primarily by the government
Equally by the government and private parties
Primarily by private parties
% (n)
% (n)
% (n)
% (n)
% (n)
% (n)
94.0 (1158/1231)
4.8 (59/1231)
1.1 (14/1231)
81.1 (1010/1245)
18.4 (229/1245)
0.5 (6/1245)
governmental health care workers vs. others. Political ideology was measured by asking respondents to place themselves on an 11-point scale ranging from 0 (farthest to the left) to 10 (farthest to the right), with 5 being a neutral (center) position. In cross-tabular analysis, a 3-point variable was created, separating left, center, and right. Tax burden was assessed by asking respondents to estimate how much they paid in income tax relative to their income (in %), using a 10-point scale, ranging from 0 (no income tax) to 9 (50% or more). Three variables assessed realized and potential demand for health services. These included self-assessed overall health (poor or fair vs. good or excellent), number of visits made to a physician in the past 12 months, and expected amount of personal health services use in the next 12 months (little or none vs. substantial or a lot). Two ordinal outcome variables were used in the analysis. The first asked respondents whether they thought the government should spend more, less, or the same amount on health care as it already does. In cross-tabular analysis, a distinction was made between those who believe the government should spend more vs. not more. The second variable asked respondents whether they thought health services should be delivered by the government or private parties (response options: “Only private parties”; “Mostly private parties”; “Both the government and private parties”, “Mostly the government”; “Only the government”). The first response option was extremely rare, and thus the first two were combined to form a 4-point ordinal variable measuring the degree of support for public delivery of services. In cross-tabular analysis a distinction was made between those who thought health services should be primarily delivered by the government (i.e. mostly or only the government), and those who thought otherwise. Statistical analysis The paper first describes the overall distribution of attitudes toward governmental funding and delivery of health care, followed by cross-tabulations of these attitudes by individual predictor variables. Finally, net relationships between variables are assessed by
multivariate ordinal logistic regression. The regression follows a cumulative logistic model for ordinal response data given by
P(Y⩽ i) 1 - P(Y⩽ i)
logit(Y⩽ i) = ln
= α i + β1X1 + . . . + β m X m
i = 1,.....k
with k model equations and one logistic coefficient bj for each covariate [19]. The regression analysis proceeds in two steps: first by including all predictor variables in a full model, and then, through stepwise elimination, only including significant predictors in a trimmed model. Results Table I shows the overall proportion of respondents favoring governmental delivery and governmental spending on health care. The vast majority of respondents (94%) wanted the government to spend more on health care, and wanted health care to be primarily delivered by the government (81%) rather than private parties. Table II shows the cross-classification of views on government spending and delivery of health care by predictor variables. More women than men thought the government should spend more on health care (men and women, however, expressed similar views on governmental delivery of care). Support for governmental delivery was somewhat greater among the less educated, but support for governmental funding did not differ by education. Reykjavik-area residents were somewhat less interested in increased governmental funding, but residential differences regarding governmental delivery were not significant. No age or income differences were observed regarding the role of government in health care. Table II furthermore shows that public sector employees expressed views similar to private sector employees. However, those who worked within governmentally operated health care were less likely to endorse governmental delivery of care. Left-wingers were more ardent supporters of governmental delivery
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
The role of government in funding and delivering health services 5 Table II. Support for governmental funding and delivery of health services by predictor variables. The government should spend more on health care
% (n)
Age 18–39 95.5 (463/485) 93.4 (423/453) 40–59 92.5 (272/294) 60+ Gender Male 90.5 (544/601) Female 97.3 (613/630) Education 95.7 (334/349) Basic High school/vocational 93.3 (434/465) 92.9 (355/382) University level Personal monthly income < 200.000 94.0 (234/249) 94.2 (664/705) 200.000–499.000 500.000 or more 93.5 (260/278) Residence Reykjavik area 92.9 (730/786) 96.6 (424/439) Countryside Public sector employee 95.9 (301/314) Yes 93.5 (828/886) No Governmental health worker 97.0 (65/67) Yes 93.4 (998/1068) No Political ideology 95.4 (355/372) Left 92.4 (218/236) Center 92.8 (448/483) Right Tax burden (% of income) 88.7 (197/222)
SJP0010.1177/1403494816631872VilhjalmssonShort Title
Scandinavian Journal of Public Health, 1–9
Original Article
Public views on the role of government in funding and delivering health services
Runar Vilhjalmsson University of Iceland, Reykjavik, Iceland
Abstract Aims: Public surveys in socialized health systems indicate strong support for the role of government in health care, although different views can be detected. The study considers the public’s views on public versus private funding and delivery of health services. Methods: The study is based on a representative national sample of 1532 Icelandic adults, aged 18 and older, who participated in a national public issues survey. Respondents were asked about government spending on health care and whether the government or private parties should deliver health services. Results: The great majority of respondents thought that the government should spend more on health care, and should be the primary provider of care. Lower age, female gender, countryside residence, and expected high use of health care were related to greater support for governmental funding. Furthermore, countryside residence, less education, lower income, not being a governmental health worker, expected high health care use, and left-wing political ideology were all related to greater support for governmental delivery of health care. Conclusions: Despite sociodemographic variations, the study finds strong overall support for the role of government in funding and delivering health care. Previous perspectives and hypotheses of welfare state endorsement received mixed support, suggesting that further theoretical and empirical work is needed to better account for public views on the role of government in health care. Key Words: Health care, public opinion, welfare state, public funding, public delivery
Introduction Health care is a cornerstone of the welfare state. A review of health care systems in OECD countries shows that they fall broadly into three categories—private, socialized, and national health insurance (NHI) systems [1]. The role of the state differs between these systems with regard to the regulation, ownership, funding, and delivery of services. Socialized health care systems (e.g. the Nordic countries and Great Britain) and NHI systems (e.g. France and Germany) purport to guarantee citizens equal access to health services at time of need [1]. Funding for health services in socialized systems comes mostly from national budgets, and the state or lower levels of government have a substantial direct role in the delivery of services. In NHI systems, sickness funds with employer and employee contributions (mandatory health insurance)
pay most service costs, the role of the state or lower levels of government is more limited, and services are mostly delivered by privately owned and operated service units. Public views on the welfare state In previous studies of public views on the welfare state, respondents have been asked whether they feel that the state or lower levels of government should spend more or less on various public services and benefits. Respondents have also been asked whether they would be willing to pay higher taxes if spent on certain services or benefits. Finally, they have been asked what the role of the state should be as provider of public services [2–6]. The studies have identified
Correspondence: Runar Vilhjalmsson, University of Iceland, Eirbergi, Eiríksgotu 34, Reykjavik, IS-101, Iceland. E-mail: [email protected] (Accepted 5 August 2015) © 2016 the Nordic Societies of Public Health DOI: 10.1177/1403494816631872
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
2 Vilhjalmsson differences of opinion (attitudinal cleavages) as regards the role of government in welfare services and programs. Perspectives and hypotheses of welfare state attitudes Several theoretical perspectives have been proposed to account for variations in welfare state attitudes. According to Power Resources Theory, the development of the welfare state is understood in terms of a struggle between labor and capital. Support for the welfare state and its programs is guided by class interests and corresponding political ideologies. Therefore, lower class or socioeconomic status, and left-wing political ideology, are thought to be key factors underlying greater support for welfare state services [7]. A more recent New Policy Perspective argues that developed postindustrial societies have created new interest groups, including beneficiaries (e.g. the elderly and patients) and employees of the welfare state (predominantly women), that oppose retrenchment of the welfare state, resulting in new welfare state attitudinal cleavages by age, gender, occupational sector (public vs. private), and client status (patients) [8]. Lastly, the Self-Interest Perspective argues that attitudes toward the welfare state are increasingly determined by personal interests rather than the collective values of universalism and solidarity. The degree to which individuals derive direct benefits through receipt of welfare state services, and incur direct costs (e.g. pay taxes), will determine their attitude toward such services, with low-tax-paying beneficiaries having the most positive attitude [2]. Based on these perspectives, several hypotheses have been proposed to explain variations in support for public spending, services, and benefits. In line with Power Control Theory, the political ideology hypothesis posits that individuals’ location on the left– right continuum (which in turn is linked to their socioeconomic status) generally explains their views on the role of government. Those to the left are more likely to stress the importance of solidarity and equality, and thereby support public services and assistance programs. Those to the right, on the other hand, are more likely to stress individual freedom and responsibility, and are less interested in state interference in human affairs, whether in terms of funding, services, or transfers [9,10]. Following the New Policy and Self-Interest Perspectives, the demand hypothesis proposes that endorsement of public services and support is greater among those having used such services in the past, or expecting or hoping to use such services in the future [10]. This could for
example apply to the elderly, expecting parents, individuals with certain diseases, and individuals with relatives that need assistance. Furthermore, the employment sector hypothesis states that public employees (predominantly women) are generally more supportive of public programs than those working in the private sector [3]. It is either argued that this is because public employees are more supportive of public sector colleagues or the clients of public institutions [11], or that public employees have a vested interest in the maintenance of public services [4]. One variation of this hypothesis would be that employees of publicly funded and delivered health services are more ardent supporters of such services than employees working elsewhere. As laid out by the Self-Interest Perspective, the tax burden hypothesis argues that individuals may not oppose (or even support) public programs as long as their contribution to the government is modest or does not cut too heavily into their finances. The hypothesis predicts that support for public programs diminishes as tax payments increase, either in absolute terms or in relation to income [2]. The above perspectives and hypotheses may help explain sociodemographic differences in support for governmental funding and delivery of welfare state programs and services. Studies find that support for governmental funding and delivery of services is generally stronger among women than men [3,5,12]. Women may express greater support than men because they use welfare state services more or expect greater use of such services in the future, are more often employed in the public sector, pay lower taxes, or are ideologically farther to the left. Older individuals and retirees tend to support public financing and service delivery more strongly than their younger counterparts [5,12]. The elderly may be more supportive of public funding and delivery of care because they use welfare state services more, or because they pay lower taxes than their younger and higher earning counterparts. Finally, working class individuals, and individuals with low education or income, are repeatedly shown to be more supportive of public programs than their higher status counterparts [5,10,12]. Lower status individuals may be more supportive because they use welfare state services more, expect greater service use in the future, pay lower taxes, or are ideologically farther to the left [2,3,5]. Government funding and government delivery of health care Public opinion polls repeatedly show that health care is considered one of the top political priorities across different health systems [13,14]. Studies also indicate
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
The role of government in funding and delivering health services 3 that support for public financing and delivery of health services has been quite stable through the years, and is stronger in countries where the government is already heavily involved in the funding and delivery of care [6,9,12,15]. Research in the Nordic countries indicates that the public generally supports the socialized model that has traditionally characterized the health care system in these countries. A large majority believes that the government should be responsible for health care, and support for the role of government is even stronger in the case of larger service units such as hospitals and community health centers where governmental responsibility has been long standing and comprehensive [5,6,10,16]. Nevertheless, there is much to be learned about public views on the proper role of government in health care. For example, it is important to distinguish between public views on government funding of services on the one hand, and government delivery of services on the other. Most previous studies have not made this distinction [but see 5,12]. Furthermore, most existing studies of public views on the role of government have focused on the welfare state or welfare programs in general, or on programs other than health care. These studies often assume that public attitudes relating to the welfare state are largely harmonious across different public programs. It is not clear whether, or to what extent, reported results from these studies apply to the funding and delivery of health services. This should be clarified, as there are indications of greater uniformity in public views on the role of government in health care, as opposed to other services or programs [2]. Finally, although different perspectives and hypotheses have been proposed to explain variations in support for public services and spending, few studies have assessed these hypotheses simultaneously. An example of a socialized health system The current study examines public views toward the funding and delivery of health services in Iceland—a country operating a socialized health system along with other Nordic countries. General hospitals, and most nursing homes and community health centers, are governmentally owned and operated, but doctors’ offices and certain specialty hospitals and rehabilitation centers are privately owned and operated. General hospitals and community health centers are primarily funded directly through the central state budget, but nursing homes, private (non-profit) specialty hospitals, and outpatient specialist medical care and prescription drugs, are mostly funded by the Icelandic Health Insurance Institution (Sjúkratryggingar Íslands), which
in turn receives its funding from the central state budget. Icelanders pay deductibles and coinsurance for outpatient physician visits and prescription drugs according to complex and ever-changing rules (lower rates apply for children, the disabled, and the elderly). Patients are allowed to go directly to specialists without referrals, and the relatively high charges for specialist visits are the same whether or not the patient makes the appointment by herself (which is common in Iceland), or is referred by her family doctor [17]. The purpose of the study The study considers the public’s views on government versus private funding and delivery of health services, and whether the views differ by age, gender, residence, education, and income. Subsequently, the study tests four hypotheses, proposed to explain variations in views on the role of government in health care. These hypotheses refer to differences in political ideology, demand for health services, employment sector, and tax burden. Methods The study uses data from a cross-sectional public issues internet survey of Icelandic adults, aged 18 and older, conducted by the Social Science Institute at the University of Iceland in March/April of 2013 [18]. The random sample was stratified by gender and age using the National Register. The survey complied with regulations and requirements concerning human subjects research, as laid out by the Data Protection Authority in Iceland (Persónuvernd). Respondents were informed that participation was voluntary. They received a link via e-mail to an internet page containing the electronic survey. To enhance participation, a total of three follow-up reminders were sent out to those who had not filled out and returned their questionnaire. The response rate was 74% (N=1532). Demographic comparison between the respondent group and population showed very similar age, gender, and residential composition, although there was a slight underrepresentation of the oldest (60+) age group (a difference of only 1.4%). Variables Sociodemographic variables included age (in years), gender, education (basic, high school, university level), monthly personal income (in thousand Icelandic Kronur, IKr), and residence (Reykjavik area vs. countryside). Employment sector was assessed with two dichotomous variables distinguishing between public vs. private sector employees, and
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
4 Vilhjalmsson Table I. Overall support for governmental funding, and governmental versus private delivery of health services. The government should spend…
Health care should be delivered…
More on health care
Same on health care
Less on health care
Primarily by the government
Equally by the government and private parties
Primarily by private parties
% (n)
% (n)
% (n)
% (n)
% (n)
% (n)
94.0 (1158/1231)
4.8 (59/1231)
1.1 (14/1231)
81.1 (1010/1245)
18.4 (229/1245)
0.5 (6/1245)
governmental health care workers vs. others. Political ideology was measured by asking respondents to place themselves on an 11-point scale ranging from 0 (farthest to the left) to 10 (farthest to the right), with 5 being a neutral (center) position. In cross-tabular analysis, a 3-point variable was created, separating left, center, and right. Tax burden was assessed by asking respondents to estimate how much they paid in income tax relative to their income (in %), using a 10-point scale, ranging from 0 (no income tax) to 9 (50% or more). Three variables assessed realized and potential demand for health services. These included self-assessed overall health (poor or fair vs. good or excellent), number of visits made to a physician in the past 12 months, and expected amount of personal health services use in the next 12 months (little or none vs. substantial or a lot). Two ordinal outcome variables were used in the analysis. The first asked respondents whether they thought the government should spend more, less, or the same amount on health care as it already does. In cross-tabular analysis, a distinction was made between those who believe the government should spend more vs. not more. The second variable asked respondents whether they thought health services should be delivered by the government or private parties (response options: “Only private parties”; “Mostly private parties”; “Both the government and private parties”, “Mostly the government”; “Only the government”). The first response option was extremely rare, and thus the first two were combined to form a 4-point ordinal variable measuring the degree of support for public delivery of services. In cross-tabular analysis a distinction was made between those who thought health services should be primarily delivered by the government (i.e. mostly or only the government), and those who thought otherwise. Statistical analysis The paper first describes the overall distribution of attitudes toward governmental funding and delivery of health care, followed by cross-tabulations of these attitudes by individual predictor variables. Finally, net relationships between variables are assessed by
multivariate ordinal logistic regression. The regression follows a cumulative logistic model for ordinal response data given by
P(Y⩽ i) 1 - P(Y⩽ i)
logit(Y⩽ i) = ln
= α i + β1X1 + . . . + β m X m
i = 1,.....k
with k model equations and one logistic coefficient bj for each covariate [19]. The regression analysis proceeds in two steps: first by including all predictor variables in a full model, and then, through stepwise elimination, only including significant predictors in a trimmed model. Results Table I shows the overall proportion of respondents favoring governmental delivery and governmental spending on health care. The vast majority of respondents (94%) wanted the government to spend more on health care, and wanted health care to be primarily delivered by the government (81%) rather than private parties. Table II shows the cross-classification of views on government spending and delivery of health care by predictor variables. More women than men thought the government should spend more on health care (men and women, however, expressed similar views on governmental delivery of care). Support for governmental delivery was somewhat greater among the less educated, but support for governmental funding did not differ by education. Reykjavik-area residents were somewhat less interested in increased governmental funding, but residential differences regarding governmental delivery were not significant. No age or income differences were observed regarding the role of government in health care. Table II furthermore shows that public sector employees expressed views similar to private sector employees. However, those who worked within governmentally operated health care were less likely to endorse governmental delivery of care. Left-wingers were more ardent supporters of governmental delivery
Downloaded from sjp.sagepub.com at City University Library on April 26, 2016
The role of government in funding and delivering health services 5 Table II. Support for governmental funding and delivery of health services by predictor variables. The government should spend more on health care
% (n)
Age 18–39 95.5 (463/485) 93.4 (423/453) 40–59 92.5 (272/294) 60+ Gender Male 90.5 (544/601) Female 97.3 (613/630) Education 95.7 (334/349) Basic High school/vocational 93.3 (434/465) 92.9 (355/382) University level Personal monthly income < 200.000 94.0 (234/249) 94.2 (664/705) 200.000–499.000 500.000 or more 93.5 (260/278) Residence Reykjavik area 92.9 (730/786) 96.6 (424/439) Countryside Public sector employee 95.9 (301/314) Yes 93.5 (828/886) No Governmental health worker 97.0 (65/67) Yes 93.4 (998/1068) No Political ideology 95.4 (355/372) Left 92.4 (218/236) Center 92.8 (448/483) Right Tax burden (% of income) 88.7 (197/222)
