Research Findings
Close window to return

Issue 4, May 2000

The contribution of social welfare policies in addressing inequalities in health
Margaret Whitehead, Finn Diderichsen and Bo Burström

  • There is increasing political commitment to tackling inequalities in health, coupled with recognition that many of the major causes lie in the wider policy environment beyond the health sector.
  • Although broader policies and interventions are therefore needed to address health inequalities, the effectiveness of many of them has not been evaluated and, indeed, the tools and methodologies to do so are under-developed.
  • The purpose of this Anglo-Swedish collaborative study is to develop ways of assessing the impact of social welfare policies on the health-related circumstances of lone parents and people at risk of or experiencing unemployment.

In relation to lone mothers, our results show that:

  • compared with couple mothers, the health of lone mothers is poor in Sweden as well as in Britain. In relative terms, the size of the health gap between lone and couple mothers is similar in Sweden and in Britain. However, in absolute terms, the trends in the two countries are very different.
  • The pathways leading to the health disadvantage of lone mothers also differ. In Britain, around half of the health gap is accounted for by the mediating factors of poverty and joblessness, whereas in Sweden these factors only account for 13% at most.
  • The findings overall support the thrust of policy recommendations advocated in Britain to tackle the poor socio-economic circumstances of lone mothers. The Swedish findings, however, raise new issues.
  • Swedish social welfare policies have clearly had social and economic benefits for both lone and couple mothers. Our findings indicate that even if such policies are a necessary pre-requisite, they may not be sufficient in themselves to reduce the health disadvantage of lone mothers, or there may be strong influences working in the opposite direction.

In relation to groups at high risk of unemployment our results show that:

  • while Britain has adopted policies since the early 1980s to deregulate the labour market, Sweden, in contrast, has developed strong employment security policies.
  • The impact of these contrasting policies is revealed when analysing how well people with chronic illness fare on the labour markets of the two countries. Contrary to the hypothesis that groups with less skill and with limiting illness should be more easily employed in a deregulated labour market, these groups fared worse in Britain than in Sweden.
  • Furthermore, the inequalities between different socio-economic groups in the social consequences of chronic illness were much smaller in Sweden than in Britain.
  • Active rehabilitation programmes in Sweden - to increase employment among people with chronic illness - improved the participants' quality of life, but failed to have a noticeable effect on employment rates during a period of major economic crisis in the country.

Background
Evidence about the existence and causes of health inequalities is both universal and compelling. In contrast, knowledge about the most cost-effective ways of reducing them or ameliorating their effects is much scarcer. This Anglo-Swedish project, is one attempt to respond to this challenge. The project has three specific aims:

  1. to further develop the methodology for exploiting a "natural policy experiment" in relation to health;
  2. to use the methodology to investigate the contribution of social welfare polices in addressing social
    inequalities in health; and,
  3. to draw out policy implications/policy guidance from the findings.

The study used the framework developed by Finn Diderichsen to map the impact of policy on the social pathways to inequalities in health, refining and adapting the framework for exploiting 'natural policy experiments' (shown in Figure 1). At a theoretical level, the framework provides a way of conceptualising the mechanisms through which the social positions of individuals and the social context of societies influence health. At an empirical level, it gives indications of how some of the various pathways and mechanisms can be measured quantitatively. Finally for policy purposes, it provides a framework for understanding how policies may influence the mechanisms responsible for inequalities in health and for making health impact assessments of policies on population health, including health inequalities.

Figure 1: Framework for researching policy impact on health inequalities

Figure 1


The impact of social position on health through differential exposure (I), differential vulnerability (II), and differential consequences of disease (III). Consequences of diseaese might feed back into a causal pathway (IV). The modifying effect of social context and policy on social stratification (A), differential exposure (B), differential vulnerability (C), and differential social consequences of disease (D).
Source: Diderichsen and Hallqvist, 1999.

To aid the assessment of policy impact, a cross-country comparison was designed. This made use of 'natural policy experiments', in which contrasting policies were in operation in Britain and Sweden and there were also contrasts over time and with changing macro-economic conditions. Two population groups were selected for whom the functioning of the social welfare system was considered particularly important: lone parents and people experiencing unemployment.

Data and methods
The study employed both quantitative and qualitative methods, carried out by British and Swedish researchers in a collaborative, jointly-funded initiative.

Following the refinement of the conceptual framework, this was used to carry out a careful comparative analysis, drawing on official policy documents, on routinely collected statistics, on empirical analysis of household survey data, and on reviews of the research literature in both countries relating to each element in the framework.

For empirical analysis of household survey data, the British General Household Survey (GHS) and the Swedish Survey of Living Conditions (ULF) were employed. Both datasets have 17 years of data, from 1979 to 1995, from a nationally representative sample. At the start of the project, great care was taken to find and create comparable variables for analysis.

Studies relating to lone mothers
Results
The first set of studies concerned social welfare policies in relation to the health disadvantage of lone versus couple parents. First, our results show that the health of lone mothers is poor in Sweden as well as in Britain and, most notably, that the magnitude of the differential between lone and couple mothers is of a similar order in Sweden as in Britain. This is despite the more favourable social policies in Sweden, which our results indicate have protected lone mothers from poverty and insecurity in the labour market to a much greater degree than the policies that have been in place in Britain over the 1980s and 1990s.

Second, our results show that the pathways leading to the observed health disadvantage of lone mothers appear to be very different in the two countries in relation to the identified policy entry points. Overall in Britain, around 50% of the health disadvantage of lone mothers is accounted for by the mediating factors of poverty and joblessness, whereas in Sweden these factors only account for between 3% and 13% of the health gap. This serves to re-emphasise the differences in mechanisms between the two countries.

Policy implications
These findings support the thrust of policy recommendations advocated in recent years in Britain to tackle the poor socio-economic circumstances of lone mothers, including removing barriers to work for those who wish to do so.

The Swedish results, however, add new facets to the policy debate not only within the country itself, but also in Britain and elsewhere. What may be striking, particularly to British observers, is that many of the policies being advocated currently in Britain to improve the life chances and health of lone mothers - including recommendations from the 1998 Independent Inquiry into Inequalities in Health - have been in place in Sweden for many years.

Our findings should not be interpreted as casting doubt on the value of these Swedish policies - it is clear that they have had social and economic benefits for lone and couple mothers in Sweden. Rather, we interpret our study findings as an indication that even if such policies are a necessary pre-requisite, they may not be sufficient in themselves to have a beneficial health impact on the health indicators we use, or there may be strong influences working in the opposite direction. We are now investigating other hypotheses about the underlying causes of the health disadvantage of Swedish lone mothers.

Studies related to people at risk of and experiencing unemployment
Results
In our conceptual framework, a distinction is made between two dimensions in the generation of inequalities in health: the influence of adverse socio-economic factors on ill-health and conversely, the influence of ill-health on socio-economic circumstances. This latter recognises that chronic illness may have social and economic consequences such as social exclusion and poverty. If these consequences vary for people in different social positions, then this may lead to social inequalities. These in turn may widen the health divide still further. One series of studies in this project focused on the social consequences of chronic illness and how these are, or could be, altered by policy intervention.

How well people with chronic illness fare on the labour market depends on several factors, including macroeconomic developments, but also on labour and social policy measures which may vary between countries. Sweden has one of the most regulated labour markets in Europe, Britain one of the least regulated.

In addition, Sweden has launched active retraining and rehabilitation programmes to help unemployed people with chronic illnesses get back to work, as part of its commitment to state support and welfare provision. Retraining and rehabilitation is seen as one potential way in which the health sector can help tackle the continuing rise in the proportion of the population classed as permanently sick.

In recent years there has been a two-way traffic between Britain and Sweden in ideas and political debate about possible policy solutions to pressing employment problems. British policy-makers have been attracted to Sweden's regulated labour market and associated social and health policies, seeing it as providing greater security of employment for older workers and for people with chronic illness. In turn, some Swedish commentators have been attracted to Britain's more flexible, deregulated labour market, seeing it as offering better employment opportunities for unskilled workers, and also for those with chronic illness. Two contrasting hypotheses may be formulated in this context:

  1. that the more flexible, deregulated labour market in Britain would result in higher employment rates than in Sweden for those with and without limiting long-standing illness;
  2. that, because of active labour market measures and associated policies, people with limiting long-standing illness would have a stronger attachment to the labour market in Sweden than in Britain, even during periods of reduced demand for labour.

To explore these hypotheses, we compared the social consequences in terms of attachment to the labour market (employment, unemployment and economical inactivity rates) among people with and without limiting long-standing illness in different socio-economic groups in Britain and Sweden during the period 1979-1995.

Findings from our study lend no support to the first hypothesis: Swedish men with chronic illness fared better than their British counterparts in terms of employment throughout the study period. Furthermore, the inequalities between different socio-economic groups in the social consequences of chronic illness were much smaller in Sweden than in Britain.

Policy implications
There would appear to be no benefit for Sweden in copying British de-regulation policies in terms of opportunities for people with chronic illness to get and to keep jobs.

In relation to the second hypothesis, we then asked what had the Swedish policy experiments on rehabilitation to offer in terms of lessons for Britain and other countries? Evaluations of the Swedish efforts to increase employment among those with limiting long-term illness during the 1990s indicate that these experiments failed when the effects were measured in employment rates.

However, these programmes were conceived in a situation (1990) when labour was in short supply and the high rates of absenteeism were more a result of high employment rates among the ill, rather than ineffective rehabilitation. By the time the programmes were implemented, demand for labour had collapsed and competition for the jobs had sharpened. Hardly surprising then, the powerful macro-economic changes swamped any effects of improved medical and vocational rehabilitation.

Hence, the Swedish experiments are not likely to account for the differences between Britain and Sweden observed in our empirical study, as regards rates of employment, unemployment and economic inactivity among men with chronic illness. The underlying employment protection legislation and traditional political commitment to full employment in Sweden are more likely candidates to explain the higher rates of employment among these men. However, there is some evidence that the Swedish experiments had other beneficial effects - for example improving the quality of life of people with chronic illness following rehabilitation services.

Details of the project
This project was jointly funded: by the ESRC Health Variations Programme from January 1997 to December 1998, by the Swedish National Institute of Public Health, and by Stockholm County Council. Data from the GHS were made available through the Office for National Statistics and the ESRC Data Archive. It was based on a collaboration between researchers at the King's Fund in London (original British team: Margaret Whitehead, Michaela Benzeval, Ken Judge and Suzanna Shouls) and the Karolinska Institute in Stockholm (Swedish team: Finn Diderichsen, Bo Burström, Monica Aberg and Christina Lindholm).

For further information, please contact:

Professor Margaret Whitehead
W.H. Duncan Chair of Public Health
Department of Public Health
The University of Liverpool
Whelan Building,
The Quadrangle
Liverpool
L69 3GB, UK
mmw@liverpool.ac.uk
Telephone 0151-7945576

Selected papers drawn on for these Findings

Whitehead M, Burström B. and Diderichsen F. (2000) 'Social policies and the pathways to inequalities in health: a comparative analysis of lone mothers in Britain and Sweden.' Social Science and Medicine, 50(2): 255-270.

Burström, B., Diderichsen, F., Shouls, S. and Whitehead M. (1999) 'Lone mothers in Sweden: trends in health and socioeconomic circumstances 1979-1995' Journal of Epidemiology and Community Health 53: 750-756.

Shouls, S., Burström, B., Diderichsen, F. and Whitehead, M. (1999)'Lone mothers in Britain - trends in health and social circumstances' Population Trends 95:41-46.

Burström, B., Whitehead, M., Lindholm, C. and Diderichsen, F. (in press, 2000) 'Inequality in the social consequences of illness: how well do people with long-term illness fare on the labour markets of Britain and Sweden?' International Journal of Health Services.

Whitehead M., Diderichsen, F. and Burström, B. (2000) 'Researching the impact of public policy on inequalities in health' in H. Graham (ed) Understanding Health Inequalities, Buckingham : Open University Press.

The findings draw on research funded by the Economic and Social Research Council under the Health Variations Programme. Views expressed are those of the authors and not necessarily those of the ESRC.