HEALTH INDICATORS FOR SWEDISH CHILDREN

by Lennart Köhler

A CONTRIBUTION TO A MUNICIPAL INDEX

Save the Children fights for children’s rights. We influence public opinion and support children at risk in Sweden and in the world.

Our vision is a world in which all children's rights are fulfilled • a world which respects and values each child • a world where all children participate and have influence • a world where all children have hope and opportunity

Save the Children publishes books and reports in order to spread knowledge about the conditions under which children live, to give guidance and to inspire new thought. and discussion. Our publications can be ordered through direct contact with Save the Children or via Internet at www.rb.se/bokhandel

© 2006 Save the Children and the author

ISBN 10: 91-7321-214-8 ISBN 13: 978-91-7321-214-4 Code no 3332

Author: Lennart Köhler Translation: Janet Vesterlund Technical language edition: Keith Barnard Production Manager, layout: Ulla Ståhl Cover: Annelie Rehnström Printed in Sweden by: Elanders Infologistics Väst AB

Save the Children Sweden SE-107 88 Stockholm Visiting address: Landsvägen 39, Sundbyberg Telephone +46 8 698 90 00 Fax +46 8 698 90 10 [email protected] www.rb.se

Contents

Foreword 5

Background 6 1. Measuring and evaluating the health of a population 6 2. Special conditions in measuring the health of children and adolescents 11 3. Some features of the development of children’s health and wellbeing in Sweden 14 4. Swedish municipalities and their role in children’s health 23

Indicators of children’s health 27 5. Basis for constructing health indicators for children 27 6. Principles for municipal inficators of children’s health 31 7. A set of indicators for children’s health 33 8. From indicators to a municipal index 47 9. What is not available now but should be and perhaps will be soon 54

Results, conclusions and further work 58 10. Situation in the municipalities 58 11. Conclusions 64

References 67

Municipality table 1 77 Individual indicators and children’s health index for the’ 2000’ period (6 indicators). Municipalities in alphabetical order.

Municipality table 2 84 Individual indicators and children’s health index for the ‘1990’ period (5 indicators). Municipalities in alphabetical order.

Appendices 91 Appendix 1. Technical description of indicators used 91 Appendix 2. Technical description of indicators not used 99 Appendix 3. Final indicators 119

Foreword

While I was given the task of constructing a health index for children on the municipal level by Save the Children, this does not mean that I carried out this work entirely on my own. I discussed the design and progress of the work with a reference group with a broad range of competence (Sven Winberg and Eva Svedling, Save the Children Sweden, Anders Hjern, Swedish National Board of Health and Welfare, Marie Berlin, Statistics Sweden, Johanna Alfredsson, Gävleborg County Council) and was encouraged at all times by the senior management of Save the Children, particularly by Project Manager Kalle Elofsson.

In compiling municipal data for the different indicators, I received technical assistance from the Swedish National Board of Health and Welfare (Claes Hedberg, Milla Pakkonen and Anders Åberg), from the Swedish Institute of Infectious Disease Control (Viktoria Romanus) and from Professor Tapio Salonen at Lund University (poverty index).

I had long discussions about constructing indices with Professor Bo Eriksson at the Nordic School of Public Health in Göteborg, who also helped me in processing the raw data and producing an index based on the individual indicators.

I am very grateful for all the help and support I received during different stages of the work.

Göteborg, January 2004 Lennart Köhler

Foreword to the English translation

The original work was written in Swedish and published in 2004. In the beginning of 2006, Janet Vesterlund made the translation into English and, in addition, Keith Barnard made a technical language edition. Allmänna Barnhuset, Stockholm, financed this work. In the interval of two years between the Swedish and the English version, important books were published, relevant reports written and the author may have become wiser. Nevertheless, the decision was made to stick to the original contents, without updating. New references and ideas will be saved for a follow-up report, which will appear in a few years’ time.

Göteborg, May 2006 Lennart Köhler

5 BACKGROUND

1. Measuring and evaluating the health of a population

Health as a component of welfare Being able to measure and evaluate states of health is a central problem in all health care: it is important in each and every contact with individual patients and is a necessary part of epidemiological studies of groups, regardless of whether as a basis for allocating care resources or assessing effects of measures taken, treatment or prevention.

But a population’s state of health is not merely the result of care efforts on different levels; it is also a part of welfare and, as such, is one of the targets of sociopolitical interventions. Already the Romans were aware of the significance of health as a political instrument. Cicero spoke of salus populi lex suprema, the health of the people is the supreme law. The first useful international register of the components of a standard of living was issued by the United Nations in 1961 (International Definition and Measurements of Living). Since then this work has been further developed with ever greater sophistication and impact by other international bodies such as the Organisation for Economic Co-operation and Development (OECD), the Council of Europe and the European Union.

Together with education, employment, income and social relations, health holds a central position in all the registers of the different components of welfare, and thus also in public policy goals. When judging progress in countries’ development, one or more components are regularly included to reflect the health of the population, typically life expectancy and causes of mortality. Even for the World Bank, an organisation focused primarily on economic development, improvements in the health of the population are important goals for aid and development programmes. As such, they must be measured and evaluated (World Bank, Investment in Health, 1993).

Growing internationalisation has increased the need of systems for measuring health that make it possible to compare developments between countries and regions. A project in this vein is the European Commission’s Health Monitoring Programme (HMP), which was started in 1997 to develop the EU’s new public health competence. The programme’s objective is to:

a) measure the state of health, its determinants and trends within the European Union

b) facilitate planning, monitoring and evaluation of EU programmes and actions

c) supply the Member States with appropriate information on health issues so comparisons can be made and national health policies supported

A secondary project in this programme has been European Community Health Indicators, which has resulted in the development of a set of health indicators for use

6 in each member state (ECHI 2001). These indicators constitute a matrix that covers the most important areas of population health. The proposal is now under consideration in the EU’s Commission and Parliament and if adopted will in time be promulgated as a Directive binding all Member States.

Individual countries have developed their own systems for monitoring the health of their populations as a part of their living conditions, and the results are regularly published in in the form of statistics, investigations and scientific reports. They are a rich source of information for politicians, planners and researchers.

Sweden was one of the first countries to begin mapping its population’s living conditions. The first initiative was taken in 1965, the so called Low Income Survey (Johansson 1973), the terms of which were: “to make a measure-related survey of the living conditions of the Swedish people, with the specific purpose of identifying persons with low incomes and studying how they live”. The theoretical projections made then came to have major significance for the development of later studies, in Sweden and other countries. A so called resource perspective was applied, whereby it was taken as given that it was the responsibility of society to ensure that resources are equitably distributed, and largely up to individuals to determine how those resources should be used. State of health was given a prominent role as one of these resources, “with the help of which the indivdiual can control and consciously steer the conditions under which he or she lives”, to paraphrase the Low Income Survey (Johansson 1973) and the most recent Swedish Welfare Balance Sheet (Välfärdsbokslut SOU 2001).

Data on the living conditions of Sweden’s adult population have been published regularly since 1976. Welfare and its distribution are reported in the following areas: education, employment and working hours, work environment, health, financial situation, housing, transport and communication, leisure time, individual and shared activities, political resources, and security and safety. In October of 2003 Statistics Sweden presented a new analysis of developments in Swedish welfare, Welfare and Hardship in the 1990s (SCB 2003). This is however still based only on the adult population.

In 1996 the Swedish Association of Local Authorities, the Federation of Swedish County Councils and the Swedish National Institute of Public Health initiated a joint project, Local Welfare Balance Sheets (Välfärdsbokslut Folkhälsoinstitutet 1999). Its purpose was to develop and test methods for guiding and monitoring municipal activites in a local and realistic perspective, where the focus is the health and welfare of the population. In the model that was developed and tested in some ten municipalities, welfare was described in terms of 13 components that were in turn expressed as some 60 measurable key numbers. Good health, support for healthy lifestyles, social relationships, safety and security, and access to medical care were some of these components.

The annual reports of district medical officers started in the mid-nineteenth century and became a valuable source of information about the population’s state of health. They were used by the National Board of Medicine (Medicinalstyrelsen), forerunner to today’s National Board of Health and Welfare, to compile national reports. This system of reporting continued until the beginning of the twentieth century when the main interest moved towards individual-based care instead of population health. It was only

7 much later, reflecting a growing international commitment to initiatives with a public health perspective, and stimulated in part by the World Health Organisation’s (WHO)‘health for all’ movement, that the Swedish Government gave the National Board of Health and Welfare the task of preparing national public health reports. The first came out in 1987 and has since been followed by a further five reports. It was decided at the outset that the reports would include certain overall indicators to reflect the commitment to “greater equity (as) an overall objective in determining priorities in matters related to public health” (Public Health Bill 1990). Since then the Government’s instructions to the National Board of Health and Welfare have been expanded to cover the whole of the social sector, in order to identify and analyse the ways in which social welfare and social problems develop.

In later years, social reports were published simultaneously with the national public health reports. The latest Public Health Bill (Bill 2002/2003:35) states that there has been no overall national monitoring or evaluation of public health measures and their effects on factors that affect health. The Government therefore places great importance on monitoring and evaluating the health of the population, and proposes that the efforts made in target areas should be carefully analysed and assessed. The intention is to build up an extensive system for public health reporting to explain and understand how ill health arises and to clarify the different aspects of health development in a population-wide perspective.

Further, the National Board of Health and Welfare’s National Public Health Reports, which describe the underlying basis for its information, are to be supplemented by a public health policy report prepared by the the Swedish National Institute of Public Health. This will describe the development of health indicators, particularly those that determine population health, and propose and analyse the measures that should be taken.

What do we mean by ‘health’? The better the knowledge we have about the lifestyles and living conditions of the population, the greater are our opportunities to plan public policy interventions, and to evaluate the effect or impact of the actions we take. Sustainable, consistent and systematic monitoring of the population’s living conditions allows us to make reliable analyses of developments and hence to avoid opportunistic and hasty actions. It is important that the methods and measurements used are tried and tested, and relevant, valid and reliable. In terms of health components, however, this is far from straightforward. Health has many dimensions, and opinions differ as to what is important and what should be measured.

Overall definitions of the concept of health are utopian rather than operational and, more importantly, unmeasurable This applies very obviously to the most often cited definition given in the preamble to the 1946 WHO Constitution: “a state of complete physical, mental and social wellbeing and not merely the absence of disease or infirmity”. Of course such definitions are not intended to be operational. The WHO definition is there in the Constitution as a statement of its vision and a reference point for the Organization’s Objective, (Article 1), which is ‘the highest attainable level of health for all peoples’. Even so, to be meaningful this objective requires measurement of its progress, more specifically the outcomes of all the actions taken in pursuing the objective. Later, moving towards a more operational definition and recognising health as a resource for

8 the individual and society, WHO proposed that the target should be that all peoples enjoyed a level of health enabling them to “live a socially and economically productive life” i.e. giving them the capacity to function in society. (WHA 30.43, the ‘health for all resolution’ 1977).

Later still in the same vein, a WHO Conference on health promotion concluded that “health is a resource that enables the individual to identify and to realise aspirations, to satisfy needs, and to change or cope with the environment” (Ottawa Conference, 1986).

While more specific in their understanding of the practical significance of good health as a resource in everyday life, these definitions still present a problem of measurement, and should therefore be seen as a point of departure, rather than a handbook for everyday life. Crucially, and in common with the definitions in the action plans and programmes of other organizations, such as the Council of Europe and the EU, they make clear that ‘functional’ health implies much more than the absence of disease. Any operational definition of health must also include a positive element, encompass mental and social as well as physical aspects, and have both objective and subjective components.

This approach also accords with the general public’s view of what health is. Social and personal resources are as important as physical capacity. Health is a positive resource and a part of our total life experience, and not necessarily something associated with freedom from sickness (Fugelli & Ingstad 2001).

According to WHO reports, many though not all European countries have drawn up health and medical care plans, or at least worked on strategies, that apply such thinking. In Sweden, the opening paragraphs of the law on health and medical care, and more recently the Government’s Public Health Bill (2002), set out overall goals of this kind. Of course we have a long way to go before the goals are attained, but this in no way detracts from the importance of the work of defining objectives and strategies for achieving them.

The problems Sweden faces in the welfare state of today are different from those of a century ago and different also from those that poorer countries face today. In the new panorama we see more elderly people, greater demands for medical care, static or shrinking resources and overworked staff. These are the defining problems that characterise modern health care. The problems of an uneven distribution of resources and the use of them remain unresolved, ethical considerations and dilemmas are becoming increasingly complex, and demands by patients for a real say are becoming more strident.

The population’s health problems are not only medical , nor are the solutions to them. They require a societal response. Professionals and others working in sectors outside medicine will often have a significant role to play, and an inter-disciplinary knowledge base is essential for the development of effective responses to problems.

This knowledge base for addressing health problems is called public health science, defined as the inter-disciplinary study of the importance of the structure of society, working life, the environment, and the health care system and its effectiveness, or ”the science and art of preventing disease, prolonging life and promoting health through organised efforts of society” (Acheson

9 1988). It also includes studies of the effects of different policy measures and the effect of public health operations on society and on different population groups. The work to improve general standards of public health and to eliminate differences between different groups in society is called public health practice (Modeste & Tamayose 1996). These measures are systematic and goal-oriented, with an view to achieve good and equitable standards of health throughout the population. The Government’s latest Public Health Bill (2002) states that the overall objective is “to create social conditions for good health on equal terms for the whole population”.

At the same time that health, in the spirit of WHO’s 1977 definition, is a resource for the individual to live a satisfying and productive life, public health is identified as one of society’s welfare goals. Sweden’s public health policy is therefore directed toward those social factors, common resources and environmental exposures that are difficult or, indeed, impossible for individuals themselves to influence.

An important means to this end is health promotion, which starts with individuals’ resources and attempts to reinforce protective factors rather than simply identifying risk factors.. Theoretically, this approach is based in part on Aaron Antonovsky’s work on “sense of coherence” (SOC), which tries to assess whether existence is experienced as manageable, meaningful and comprehensive from childhood onwards (Antonovsky 1987).

Medical care is but one of many factors in improving the health of a population, and it has been claimed that historically its role has been marginal in comparison with the role played by the general improvement in welfare, thanks primarily to better nutrition, better housing and less physical hardship (McKeown 1976). “Peace, potatoes and vaccine,” were what Esaias Tegnér, an early 19th century Swedish poet, academician, bishop and politician, claimed to be most important to the health of the population.

Today, when improved welfare in the privileged part of the world has created the conditions for a long life, the provision of appropriate effective medical care is nontheless essential for good population health, in treating and where possible curing infirmities, in enhancing our ability to function, and in providing relief when cure is not possible

The concept of health is thus enormously complex, and our understanding of what it is we must measure is still imperfect. It should come as no surprise that we are still uncertain about the measurement methods to use and how to make sense of the measurements we make. This leads to the choice of measurement being determined to a great extent by what data are available. The end result is that on both national and international levels, we most often measure and compare ill health, that is, disease.

And then we are immediately on firmer ground: we know a great deal about disease and death. We have well defined professional groups there, whose task it is to treat diseases and abnormalities, who, under the cloak of science have gained strong positions in society and who have thereby established precedence over others to interpret findings. The WHO’s International Classification of Disease (ICD) listing some 8 000 diagnoses is regularly updated. It is now used all over the world, to the great benefit of standardisation and comparisons of diagnoses. However, these diagnoses only provide a picture of diseases and abnormalities, and only from a professional perspective. They

10 leave no room for the positive aspects of health and well-being, and very little for people’s subjective experience of ill health.

Still, these non-medical dimensions of health are important to both individuals and society. They cannot be ignored if our aim is to establish complete and relevant measurements for comparisons between groups and areas and over time. The public health perspective provides a natural starting point for the construction of a system of health indicators for populations and groups within populations.

2. Special considerations in measuring the health of children and adolescents

Health, welfare and children Even though measuring welfare has thus long been an important task for politicians and researchers, it is nonetheless obvious that the welfare of the population has been considered to start earliest at the age of 16 – children’s welfare has not attracted the same interest. It is only possible to find data on children as a sort of appendage to adults. An international project on childhood as a social phenomenon (Qvortrup 1994) summarises the results gathered in 18 countries as regards collecting information about children in the following way:

¾ children are seldom units of observation ¾ the information available about children is fragmented ¾ the age categories reported are inconsistent ¾ information about children is not published on an ongoing basis

Sweden’s several hundred year old, unbroken series of population statistics and the more recent introduction of personal identification numbers offer unique opportunities to gather comprehensive, reliable data on the population and its living conditions. But these early welfare surveys have made it easy for themselves by excluding children (Köhler & Jakobsson 1991), in spite of the fact that this group should be given high priority. In the Welfare Balance Sheet from the 1990s, for example, children are named as a key group for welfare policy (SOU 2001a).

There are several reasons why children are especially important as a target group for public health efforts, in both in their own right and in their future as adults.

• Children constitute a large portion of the population. In Sweden, children 0 to 17 years make up over a fifth of the population. In many developing countries children make up more than half of the population. • Children constitute a vulnerable group, exposed to considerable health risks and dependent upon others for their protection and care. • Children’s health reflects the care that countries offer their citizens.

11 • The advances achieved in this area represent a measure of countries’ general social standard and are frequently used in international comparisons. • Children have no political power and are not represented in formal or informal pressure groups, as is the case, for example, with elderly or disabled persons. • The knowledge, attitudes and behaviour of adults in health issues, and often their basic state of health as well, are founded and established during childhood and adolescence. • With the ratification of the UN Convention on the Rights of the Child in 1989, Sweden obligates itself to give all children within its borders the rights of the child “to the enjoyment of the highest attainable standard of health” (article 24), “to a standard of living adequate for the child's physical, mental, spiritual, moral and social development” (article 27) and “to education” (article 28)

Thus arguments are not lacking for the urgency of the task of carefully monitoring and following up children’s health and welfare in a society that wishes to call itself child- friendly.

There has been a comparatively large number of studies and reports in Sweden about welfare, which in modern time have endeavoured also to include children. Some have even concentrated on children and young people as the sole target group. A series of reports were prepared at the end of the 1960s and beginning of the 1970s that surveyed children’s environment and needs in different contexts, e.g. reports on daycare, on the social care of children and young people, on children’s leisure time, on the school, on child health care and school health care.

The broadest and most comprehensive of these was Bror Rexed’s report on children’s environment, whose purpose was to give a systematic report of children’s living conditions. Nine volumes totalling over 1 300 pages reported on children’s development, upbringing and health status, as well as on the financial status of families with children, their physical, cultural and social environment and their working conditions. This was thus in essence an historical study of children’s welfare up to the 1970s (Children Report of the Children’s Environment Investigation (Barnmiljö- utredningen) 1975).

Statistics Sweden (SCB) later published several reports on children’s living conditions, and other reports have also contributed to satisfying relatively well the need for knowledge about children’s living conditions (see e.g. Welfare Balance Sheets for the 1990s (SOU 2001a), Bremberg, Level of Living Surveys/Children (SOU 2001b), Statistics Sweden’s Children’s Living Conditions, National Board of Health and Welfare’s Children’s Health in Sweden, the Reports of the Children’s Ombudsman, Follow-up of the UN Convention of the Rights of the Child.).

The first Public Health Report commissioned by the Parliament in 1988 gave no information at all on children. In the second report in 1991, a section was devoted to children, based on a specially developed investigation of basic information (Köhler &

12 Jakobsson 1991), and individual chapters on children’s health were included in later public health reports.

The Children’s Ombudsman and Statistics Sweden (SCB) publish “Up to 18” every third year, a popular statistical reference book on children, and the SCB annually reports register data on children from a welfare perspective.

In the Local Welfare Balance Sheet (Folkhälsoinstitutet 1999), a joint project of the Swedish Association of Local Authorities, the Federation of Swedish County Councils and the Swedish National Institute of Public Health, some of the participating municipalities focused on children’s welfare, for example Göteborg and Haninge.

Developing this work further, the Swedish National Institute of Public Health constructed a database of municipal facts as a basis for public health work, and this treats a number of children’s issues (Folkhälsoinstitutet 2003, www.fhi.se). Proposals have been made to facilitate monitoring of children’s health by using a standard form for central registration of e.g. vaccinations, injuries and health examinations in pre- school and schoolchildren (Socialstyrelsen Memorandum, March 2002).

Preparations are ongoing at Statistics Sweden for an expansion of routine statistics on children’s health that will include about 40 group indicators. The goal is to regularly present simple health measures in cooperation with other authorities to give an overview of health developments among children, particularly on the municipal and regional levels.

A three-year project has been carried out at the National Board of Health and Welfare to survey the mental health of children and young people with the intention of offering a complete, concrete and tested system for regular surveys of Swedish children’s mental health (Hagquist 2004).

The Government’s new public health policy statement highlighted secure and favourable conditions for children and young people as one of the target areas, which will thus be an area of careful study and will include the development of indicators and analyses of measures.

Although all these ambitions are certainly praiseworthy, we have not yet come so far that these reports, either by themselves or together, give a complete picture of children’s state of health. In preparing these statistical reports, it is often pointed out that a very large part of the information about health in different registers is not based on personal identification numbers and that they lack information about social background. The conclusion we drew after our earlier survey of children’s health in Sweden (Köhler & Jakobsson 1991) still applies today: “One looks in vain for a systematic, continuous and complete reporting of children’s health, in a child perspective and related to a social context”.

A sub-project, carried out in the EU Health Monitoring Programme mentioned above, and its set of health indicators (ECHI) treated health indicators for children on a national level (Child Health Indicators of Life and Development, CHILD). The project gathered specialists from all EU member states, plus Norway and Iceland, and the final report was submitted to the EU Commission in September of 2002 (Rigby & Köhler

13 2002). The proposal will now be considered by the Commission and its technical agencies and be incorporated, at least in part, if and when an overall health surveillance programme is initiated.

Child Public Health We can give thought to how the broad concepts of health, public health science and children’s special needs of protection for their health and welfare can be combined into a concept that could be called Child Public Health (Köhler 1998). Action areas in Child Public Health include educational programmes, research and practical activities. One of its most important practical tasks should be to create centres of knowledge about children, to conduct programmes in concert with politicians and other decisionmakers in children’s health, and to work with action groups, voluntary organisations and different occupational groups in other sectors of society.

Setting up a centre of knowledge about children’s and adolescents’ health and health behaviour in municipalities, county councils or on a national level is a long term and relevant goal that certainly requires much work but is not unrealistic. A knowledge centre must be the natural entity to which politicians, the press and the general public can turn for up-to-date, objective information on issues that have to do with children’s and adolescents’ health and welfare. Components of these centres already exist in child health care, school health care, paediatric care, the social services, schools, children’s and young people’s clinics, child psychiatry and children’s ombudsmen. What is lacking are common visions, overall goals, an established structure with clear strategies for cooperation and inspired leaders for collective efforts in this vital area of society.

The strength of child public health work is that it is based on the concepts and values of the WHO Health for All Strategy, adopted by the World Health Assembly, it employs methodologies used in public health, and it places children’s and young people’s health in their full social, economic and political context. This means that its activities – in education, research and practice – will be practical and relevant and will include insight and experience from a number of different professions and sciences. The approaches and methods in child public health will form the basis for the review of suitable health indicators for children that is given in this report. The proposal is also coloured by experience from the recently completed EU project on health indicators for children.

3. Some features of the development of children’s health and wellbeing in Sweden

There are several ways to describe developments in children’s health. We can start from the degree of severity or the frequency of health problems in children; we could use diagnoses or organ systems or medical specialties or areas in which there have been great changes; we could use investigative methods or sources of information as our starting point, or survey children’s health and ill health according to age groups or sex. Or we could use a combination of these methods.

As this review will be a part of a report about health indicators and a health index, I have chosen to group health variables according to generally available concepts and

14 measures. Measures of this kind that are most often used in studies of health or ill health in child populations are easily available via registers or other means, usually mortality, morbidity and growth and development. More modern and complete descriptions also include wellbeing, health behaviour and perhaps quality of life.

We are used to seeing Sweden and the other Nordic countries as leaders in terms of welfare states and thus as populated by healthy, flourishing citizens. This may be so now, but has not always been the case. In truth it was only during the twentieth century that major advances were made in population health. This goes for children as well, during the century that the Swedish author Ellen Key called the children’s century. To take a longer historic perspective of children’s health status we have to look at indicators that are easily available in population registers or medical officers’ reports, and these deal with mortality and to some extent morbidity.

Mortality The most frequently used variable for mortality is infant mortality, i.e. the number of deaths under one year of age per 1 000 live births. Sweden has records dating back to the middle of the eighteenth century that show that the numbers have steadily decreased. Perinatal mortality is also often used (stillborns + deaths during the first week of life). Neonatal mortality is sometimes also used, that is, deaths during the first month per 1 000 live births.

The under five mortality rate (U5MR) is also used, especially in international comparisons, and is considered by UNICEF (the United Nation’s Children’s Fund) to be the single measure that best illustrates children’s situation in a country. Sweden and the other Nordic countries have the lowest figures in the world, together with Japan.

Mortality statistics, especially infant mortality and U5MR, are generally used as a measure of a country’s social and economic standard, more sensitive and appropriate than the gross national product (GNP), for example. As a further measure of socioeconomic conditions, WHO uses the proportion of live births with low birthweight, that is, less than 2 500 g. These numbers are about 4 percent in Sweden while they are 30-50 percent in many developing countries.

In general statistics it is also simple to relate mortality to specific diagnoses. In diagnosis-related mortality, diseases in newborns, injuries and cancer take the leading positions. In industrialised countries, injuries are now the leading cause of death among children over one year old, and a large portion, often the majority, of these are deaths in traffic.

In spite of the strong growth in road traffic in Sweden, the number of deaths from traffic injuries, among both children and adults, has decreased, as have the number of deaths due to drowning, suffocation and poisoning. There has also been a shift toward less severe injuries. In the 1950s, 400 children died annually as a result of injuries, of which 140 died in traffic. The corresponding figures during the 1990s were between 60 and 30 deaths. The child population was approximately the same, but the number of cars had increased from one to four million (Sylwander 2001).

Another leading cause of death among children is malignant diseases. The number of children with cancer has been relatively stable, but mortality has decreased and length

15 of survival has increased with effective treatment. It is estimated that about 300 children under 15 years of age in Sweden receive a diagnosis of cancer each year, of whom over 70 percent survive (Public Health Report, Socialstyrelsen 2001).

While Sweden has traditionally had a reputation as a country with a high suicide rate, the suicide frequency declined after an increase in the 1960s and 1970s. Sweden is now in a middle group among countries in Europe. There was a clear decrease in suicides in the age group 15 to 24 years between 1987 and 1996, but suicide attempts have shown a tendency toward an increase. An explanation for the rise in suicide attempts could be that poisonings that now occur are often related to intake of paracetamol rather than acetylsalicylic acid. Intake of paracetamol leads more often to hospital care. However, it is difficult to make a clear interpretation of the rise in registered suicide attempts.

Average length of life is in a way also a mortality statistic and indicates how long an individual of a particular age can expect to live. Sweden is at the top here as well, surpassed only by Japan. A newborn boy in Sweden has a life expectancy of 77.6 years and a girl 82.1 years. However the number of years that a 50-year-old man has left to live is only about six more than it was in the nineteenth century, which of course illustrates that the greatest gains in terms of life expectancy have been achieved in younger people.

Morbidity Use of care Morbidity among children can be described in many ways. Health care register data can give a picture of how society’s health care resources are used in terms of hospitalisations, visits to primary care etc., in other words of the use of care among children. This is a very common method in adults and is now also used in children. Surveys of this kind are used foremost as a basis for planning resources and are a questionable method for measuring morbidity.

Important factors that are not directly related to morbidity affect the need for care, demand for care and use of care, e.g. knowledge, attitudes, values, belief in the system and financial resources on the part of the patient, and organisation, treatment practice and availability of the health care system.

Availability of care has been shown to be a particularly powerful means to control utilisation of care. With the general restructuring of inpatient care in Sweden the number of hospital beds decreased from above 2 000 in the 1980s to somewhat over 1 100 in 1998 (Socialstyrelsen 2001a). More and more children are instead treated in ambulatory care. The average length of hospital stays decreased from 4.6 days in 1981 to 3.7 days in 1998. The majority of care days was used for diseases in newborns, psychiatric disorders and injuries. Not counting neonatal care, the mean care time in 1998 was 2.8 days.

Somewhat over 11 percent of county councils’ collective costs for health care in 1999, excluding dental care, went to the care of children between 0 and 19 years, who at the same time represented 22 percent of the population. There has also been a marked change in the disease panorama since the beginning of the twentieth century. Treatment of diseases that previously led to rapid death, such as cancer and neonatal diseases, has improved and survival times have increased. Undernourishment and

16 serious infections have disappeared, replaced by allergies, diabetes and other chronic conmditions.

The most common chronic diseases in children are now allergies and asthma. It is estimated that one third of schoolchildren in Sweden have or have had an allergic disorder. The prevalence of asthma is now 8-10 percent among schoolchildren, which is three times higher than the rate 25 years ago (Formgren 1998, Socialstyrelsen 2001a). Diabetes is very much on the rise in Sweden The incidence has risen by an average of over 2 percent each year, and is even higher in children under ten years of age. More than 500 cases of diabetes in children are diagnosed each year and, despite systematic and intensive treatment, there is a large risk of serious complications later in life. The cause of this increase has not been clarified, but there is a strong corrrelation with the increase in obesity and overweight in children (Dahlquist & Mustonen 2000).

As chronic conditions have become more important for families, they have come to occupy a more prominent place in health and social care. Demands for treatment and care have increased among both professionals and lay persons. Children with disabilities and and long term conditions have received much greater attention in recent decades. Still, information about the occurrence of disabilities in children remains uncertain because the definitions and boundaries are diffuse and no comprehensive surveys have been made. Existing data are based on estimations and prognoses from local studies. According to a compilation by the Swedish Disability Institute using a number of different sources, a total of 225 000 children between 0 and 17 years of age have a long term condition or some mild to serious disability, that is, 13 percent of that segment of the population (Hjälpmedelsinstitutet 2002). The material does not allow any deeper analyses or breakdown on the municipal level.

Dental diseases in children are particularly interesting from three perspectives. First, they are easy to diagnose, at least in their more serious forms. We can tell the general dental health status of a child by simply looking in its mouth. Secondly, dental diseases, primarily caries, are closely related to socioeconomic status. While there is generally a clear relationship between the family’s social class and children’s health, one of the clearest relationships is seen for caries. Thirdly, preventive measures are most often the best way to counteract dental disease. Good regional studies of children’s dental health have been carried out in Sweden since the 1960s, and the National Board of Health and Welfare has followed developments throughout the country since the 1970s, via the National Dental Service and later also via private dentists. Annual reports are published. Local and regional prevention programmes have been conducted throughout the country at child health centres, in pre-schools and in schools, and national care reforms have made dental care easily available and free of charge for children. There have been significant improvements in dental status in all groups of children and young people, although considerable regional and social differences still exist. Reports to the National Board of Health and Welfare are not based on personal identification numbers and connections are not routinely made to the child’s socioeconomic background and home municipality.

Children are large users of ambulatory care, especially in the ages one to four years, and the most problems in pre-school ages are ear infections and upper respiratory infections. It appears that infectious diseases have increased in younger age groups and decreased in older groups, possibly because children come together in groups at pre-

17 school and not when they start school as was previously the case. Among older children, the more common problems are injuries and psychosomatic symptoms. In a study of children and families with children in the Nordic countries, parents reported that about half of children aged two to 18 years had visited a doctor as an outpatient during the most recent three months, while during the same time approximately 60 percent of the children had been absent from school or pre-school as a result of illness, usually four to five days.

About 15 percent of mothers and 10 percent of fathers had been absent from work to care for a sick child. 40 percent of the children in the Nordic countries were ‘completely’ healthy, i.e. had made no use of society’s care resources, while 3 percent of children were heavy users, i.e. had used hospital care, doctors and outpatient care (Köhler & Jakobsson 1991).

In the 1996/97 investigation of living conditions based on a representative sample of Swedes, 23 percent of children, 0-15 years of age, had visited a doctor as an outpatient during the most recent three months. Over a third of children aged 0-2 years old had visited a doctor, a frequency exceeded only among retired people (Socialstyrelsen 2001a).

Health examinations Another method for identifying sickness and health is health examinations. Sweden has long had a well organised system of child health surveillance that is used by practically all children. One might thus think that there would be large quantities of data on children’s health (or ill health) derived from check-ups, screening investigations and individual assessments.

Unfortunately, however, the data are seldom systematised or compiled. The results presented in official reports have mostly to do with structural features (e.g. the organisation, number of doctors and nurses, number of children per doctor) and performance measures (e.g.number of visits, number of vaccinations), not with outcome and effectiveness measures, i.e. how health is influenced by these activities (Köhler & Sundelin 1985; Medical Research Council (Medicinska forskningsrådet) 1999).

Data compiled from routine child health and school health services are thus relatively sparse, and the data that exist are not reliable (Hagelin 2001). It is necessary instead to consult special health examinations with specific questions and a built-in evaluation to form a picture of children’s everyday health. A growing number of screening methods have gradually been introduced in health examinations, i.e. simple, fast and cheap methods for identifying those children who should undergo more extensive diagnostic investigations.

Screening is used primarily in mass investigations to find hidden health problems that have a significant effect on the individual and whose course can be helped by early detection and treatment. Among children, systematic screening has been used most for vision and hearing and for certain laboratory investigations of blood and urine. The methods have been critically examined and evaluated and there is now good knowledge about the value of these investigations. Studies that have followed groups of children over several years have shown that certain complaints disappear spontaneously and

18 others make their debut. For certain individuals it can of course be important that abnormalities are detected, but the effect of early detection and treatment on the health of the group seems to be marginal, except possibly as regards impaired vision and early hearing deficits (Hall 1996).

The value of formal screening of mental, behavioural, linguistic, social and moral development has not yet been satisfactorily documented. The complex biological and social dynamics that lie behind a child’s development make it difficult to find simple and unambiguous methods for early detection of disorders. Furthermore, their treatment has not been particularly successful (Sundelin & Sonnander 2000). National Board of Health and Welfare has proposed a computerised standard record that can be used to create a central database with good quality (Socialstyrelsen Memorandum, March 2002).

All modern surveys of population health include self-assessed health as an important feature. This captures aspects of health that are not otherwise detected. Subjective evaluations of health have also shown good agreement with objectively observed disease and survival. Interviews, dialogues, questionnaires and diaries can help to give an idea of how people experience their health or their diseases.

Studies of this kind can be designed and analysed using quantitative or qualitative methods, methods that have commonly been used for adult populations but are now also beginning to surface more and more often in investigations of older children.

An example is the Health Behaviour of School Children (HBSC), an international study initiated by WHO of schoolchildren’s health habits. It has been conducted every fourth year for some years, and now encompasses more than 30 countries (WHO 2000c). The proposed system for continuous surveillance of children’s mental health is also based on information from children themselves (Hagquist 2004).

Self-assessments are sometimes combined with health examinations or studies of health care utilisation, making it possible to relate subjective and objective measures of health. This is often done in child and school health care but, in that case, it is generally parents that give subjective information. The younger the children, the more difficult it is to use them as sources of information.

In some projects children and parents have been asked to keep their own registers of episodes of illness or health activities. Studies of this kind can show that small children have many episodes of illness but that the episodes are largely managed by the parents, most commonly the mother, without professional help (Uldall 1986).

Mental health It is not difficult to argue that children’s health status has improved dramatically in the last century and that this group is now the most healthy in our population. There are solid data to confirm this and the knowledge of it is well rooted in our thinking. An exception is children’s and young people’s mental health and the psychosocial environment in which they grow up. Many express their anxiety that children are not feeling well mentally and are not able to achieve optimal development, and that society’s increasingly crass priorities, the demands made in working life, and the dissolution of families have led to more worry,

19 alienation, burn-out, substance abuse, criminality and marginalisation from working life or society. The picture in the mass media is clear. The public is convinced, and many health, social welfare and school professionals are deeply concerned about the developments they see in their everyday work – and politicians want to do something.

However, it is not a simple task to confirm these ideas in scientific studies or with official statistics. It is difficult to make comparisons with how things were earlier, in part because many of today’s studies have never been carried out before,.and in part because methods, definitions, populations and questions change. It has been shown for instance that the very strong increase in reports to the police of battered children is not based on a true increase in abuse but on a greater tendency to report cases. Furthermore, physical punishment as a method for bringing up children has decreased drastically since the 1980s, and the expressed attitudes of the general public to physical punishment of children show clearly that it has become less accepted (Janson 2001b).

At the same time, there are a number of reports that give a completely different picture of Swedish children and young people today, one that seems to be positive, especially in an international context. Mental health, thus, is a very complex issue. It is however clear that children’s mental health has become an important area for professional and political initiatives, and it is an area that should continue to be carefully followed.

A Government investigation has been carried out, and a number of scientific studies are being done.The National Institute of Public Health has become involved in this work, and the National Board of Health and Welfare has proposed regular measurements of children’s and young people’s mental health (see chapter 9).

Growth There is a close connection between children’s growth, development and maturity on the one hand and their ill health (in terms of mortality, morbidity and disability) on the other. A sufficient and balanced diet is important for growth and resistance to disease and “normal” growth and maturity profiles are thus often used as indicators of good health status.

By analysing the detailed recorded data on children’s height, weight and sexual maturity, we can show a general tendency toward larger physical size in all ages and a lower age for onset of puberty, the so called secular changes. The growth pattern shows great variation, however, not only between different times but also between different geographical areas and social classes (Brundtland & Walløe 1973; Cernerud 1991; Elmén 1995).

The greatest threat to children’s health today is not insufficient nutrition but the rise in overweight. In 18-year-olds registering for military service in Sweden, overweight (Body Mass Index, BMI, in kilograms/m2 over 25) increased from 7 percent to 18 percent between 1971 and 1998, and the proportion of obesity (a BMI over 30) increased from 1 percent to almost 4 percent (Rasmussen et al. 1999). A dramatic increase in overweight has also been noted in schoolchildren and younger children. This is considered to be caused primarily by changes in lifestyle, notably diet and physical activity. However, WHO’s investigations in schoolchildren have not shown any great changes over time in the physical activity of Swedish children. And an investigation in the Nordic countries did not show a decrease in schoolchildren’s activity in sports and

20 physical activity between 1984 and 1996; to the contrary, it had increased somewhat and was highest in Sweden and Finland (Nordhagen & Nystad 2000).

Obesity in both children and adults is considerably more common among those with poorer socioeconomic conditions (Socialstyrelsen 2001b). Even if the effects of both preventive and treatment efforts have thus far been limited, obesity and overweight have such a strong negative impact on physical and psychological health that even small gains must be seen to offer hope in efforts to limit the explosive obesity epidemic (SBU 2002).

Wellbeing and quality of life Wellbeing is subjective satisfaction with different areas of life, such as external conditions, relations with others and internal, individual characteristics. The concept of ‘quality of life’ is often used synonymously. Two studies of Nordic children’s health (Lindström 1994) have used both objective and subjective estimations of quality of life that include”:

a) socioeconomic conditions, b) structure and function of the child’s and family’s network and c) the child’s wellbeing

In the first study, in 1984, children showed a generally high level of quality of life. Despite the decline in economic conditions during the 1990s, which had a great effect on families with children, it was found in the 1996 study that objective quality of life, such as income and time for children, had increased in all the Nordic countries. However, subjective quality of life in the personal sphere, such as psychosomatic complaints, poor self-confidence and bullying, had become worse in most countries, including Sweden. We could say that children and their families now feel worse than their socioeconomic status would indicate (Berntsson, Köhler & Vuille). This kind of findings open up for discussions of the importance of issues such as social capital, reference anxiety and status syndrome.

Other studies of children’s and young people’s wellbeing have shown similar results: satisfaction with the family, school, friends and life in general is good on an overall level, but there is a clear tendency to greater unhappiness about certain things and in certain groups. This is also reflected in the growing proportion of children that say they have mental and psychosomatic problems and that seek professional help for their complaints (Olsson 1998; Berntsson 2000; Danielsson & Marklund 2000).

Problems with substance abuse among young people in Sweden is still limited in an international perspective, and no great changes in consumption habits have been noted during the past decade.

Alcohol consumption is however considerable among many young people, particularly drinking for the purpose of getting drunk. Politicians and researchers have expressed a great deal of worry that the liberal tax and customs policies in the EU will lead to a deterioration in population health, both in children and adults.

21 Use of tobacco among young people in Sweden is the lowest in Europe. Use of narcotics, which was earlier on the rise, has now stabilised and is relatively moderate seen in an international perspective (CAN 2002 a, b).

Inequity It is an old truth that there is a close and consistent relationship between living conditions and health, and it is still seen today, especially in developing countries and countries with large differences between social classes. It is generally less well accepted that this is relevant in rich welfare states as well. It has not been a central theme in the political debate on health until recent decades. It has been even more difficult to accept that there is a considerable social difference in the health of Swedish children. However, in the past ten years, convincing evidence has been gathered to show that this is the case. There are clear social differences in most health problems in the childhood years, and in certain cases, very large differences (Bremberg 2002). Children’s social and economic conditions have an effect on their health and wellbeing, and on the occurrence and course of their illnesses, both in childhood and later on in life. There is clear ebvidence that social, economic and health inequity has increased in Sweden (Socialstyrelsen 2001b, SCB 2003). But social differences can be reduced and parents’ situation strengthened, and both the environment and health of children improved by a variety of political, economic and organisational efforts.

There are thus no conclusive reasons for not always including children’s socioeconomic background in studies of their health, neither in scientific reports nor in official statistics. This would offer the possibility to follow social developments in society in an area that is of the utmost significance for the future.

Conclusions The conclusions that can be drawn about children’s and young people’s health and wellbeing in Sweden after this review can be summarised as follows:

Generally positive features • Children are generally healthy and feel well. Or, to cite the Children’s Ombudsman in her report to UN on Sweden’s compliance with the UN Convention of the Rights of the Child: “In an international perspective, the situation among children and young people in Sweden is generally good. The material standard is high and children’s physical health is good. Parents care about their children and respect their opinions, and children have rich leisure time” (Barnombudsmannen 2001).

• The welfare society has largely functioned well in its ambitions to provide the population including children and families.with the conditions for a good life.

Potential threats • All children do not feel well. Long term illnesses, especially allergies, overweight and diabetes, and psychosomatic complaints have increased. The children who fare worst are those in families that are for different reasons at risk, parents with low incomes, poor education, or unemployed, single parents and families seeking asylum.

22 • Economic, social and health inequity has increased in Swedish society.

• Families with children are among those who are affected first and worst by financial crises.

If a growing segregation of society is permitted, there is an imminent risk of an increase in health problems, primarily among those who are already the weakest. Children in these situations will be at particularly high risk.

In short, if the development that began in the twentieth century continues and children are given greater importance and greater rights in our democratic society, not only rhetorically, in formal speeches and manifestos, but also in realistic policy measures, the 21st century can truly be a Children’s Century. Sweden has a heavy and binding responsibility for leading developments in the right direction, and our starting point is in fact exceptionally favourable.

4. Swedish municipalities and their role in children’s health

Swedish municipalities have historically been very independent of the Government. This gives them a great responsibility for supplying services and utilities in important areas of their inhabitants’ everyday lives, chiefly in environmental and health protection, health care and other types of care, community planning, education and child care, culture and leisure time. Much of these activities are regulated by law, e.g. the Social Services Act, community planning and building laws, school laws, and environmental acts. Other activities can be voluntary, often in cooperation with other authorities such as the police, the National Road Authority, county councils or other organisations. The different boards in municipalities often have descriptions of welfare goals that are important to health, in e.g. environmental and building plans, programmes for substance abuse, violence and bullying. The activities of the municipalities thus span over large areas, each of which are important in their own way to the lives of their inhabitants. They can be said to be the core areas of welfare policy and thus to have great significance for population health. Municipalities can help to create the conditions for a good living environment and possibilities for their inhabitants to live good and healthy lives, for example by offering good housing, opportunities for physical activity and recreation, communication, access to health care and other types of care, and good education. Municipalities have recently come to be important actors in health promotion and have particularly good prerequisites for taking responsibility for general, primary preventive and inhabitant-oriented efforts. Municipalities can be said to have a special position in terms of creating conditions for a good life (SOU 2001a).

Municipalities are not alone in their responsibility for welfare policy and provision of services, however, and can of course not influence all the factors that help to promote health and quality of life. The Government has the overall responsibility for welfare issues by forming legislation, by the economic possibilities it offers, by social policies, labour market policies etc. County councils and regions have the operative responsibility for the greater part of health care and for transport questions and regional planning. The municipalities’ capacity to carry out their tasks has recently been questioned. Most importantly, many small municipalities are experiencing difficulties as

23 a result of their decreasing populations, which must finance services for the growing number of elderly persons through municipal taxes.

A Government committee was established in 2003 to analyse the organisation of Swedish society, demographically, socioeconomically and technologically. One of the factors being studied is the municipalities’ future role in relation to central Government and the county councils. “We must ask ourselves whether many of our municipalities are too small to be able to survive in the long run” (Mats Svegfors, Chairof the Government committee, in Dagens Nyheter, the Stockholm newspaper, 25 January 2003).

The municipalities’ most important efforts for the health of its citizens are currently in social services.They provide individuals, chiefly elderly persons, with care and service, usually to compensate for their inability to manage their daily lives in some area. whether running the home , nutrition, personal hygiene or self-care.

It is difficult to draw a clear line between social services and health care. Attempts are being made by WHO and the EU to create an overall international system to determine the boundaries between efforts to maintain health, to care for the sick and to provide social care in order to allow comparisons between countries and over time.

By analysing the activities of different personnel categories in the municipalities, the Swedish National Board of Health and Welfare calculated that about 15 percent of personnel time in care financed by the municipalities is spent on tasks within the areas of health maintenance and care for the sick. This means that the cost of municipalities’ efforts in health care during 2000 was SEK 14.6 billion, or 0.7 percent of GNP.

At the same time, the Government’s and the county councils’ expenditure on health care was 7.7 percent of GNP. It can be assumed, although there is no complete proof of this, that the municipalities’ health care efforts grew during the 1990s because of medical, technical and structural changes and shifts in assessments during difficult economic times (Socialstyrelsen 2001c).

The municipalities have a special responsibility for children that is expressed most clearly in the Social Services Act and the Education Act. These are also the areas that cost most; the annual cost for one child in pre-school in 2002 was over SEK 90 000, for one young person in upper secondary school SEK 80 000 and for one child in compulsory school SEK 60 000 (Socialstyrelsen 2003). (10 000 SEK is about 1 250 USD).

Together with music school, after-school activities and support to non-profit associations, child care and schools are the areas in which the municipalities can generally report their costs for children as separate categories. Other municipal activities, such as libraries and care for individuals and families, combine different age groups, and thus costs cannot be reported separately with respect to children’s utilisation of these resources.

In a study of all welfare activity for children in the county of Stockholm, each child was estimated to cost society SEK 3.400 000 in 1996. The family’s cost in terms of time and money was estimated at SEK 2.100 000. The municipality was the largest public actor with SEK 1.200 000, primarily for daycare and schooling. The county council and

24 voluntary organisations came in third and fourth, with SEK 120.000 and 90.000, respectively.

Differences in the scope of activity were relatively large between municipalities and could be explained only to a small degree by social conditions (Dalman & Bremberg 1999). During the economic recession in the beginning of the 1990s, municipalities were forced to make serious cut-backs in their activities, over 7 percent during 1993 and 1994. Activities for children and adolescents were hit hardest: daycare by almost 15 percent and music school by 12 percent (Socialstyrelsen 1994). The cut-backs that most affected children were in personnel reductions that resulted in children being cared for and taught in larger groups .

The municipality has a great responsibility in child care and schools for making children’s environment as supportive and stimulating as possible. As stated in the guidelines for the municipalities’ social welfare board: “The Social Welfare Board shall work to give children and adolescents secure and beneficial conditions.” This is of course a highly important determinant for children’s health and wellbeing and should as such be included in a system of indicators. However, as the Child Municipality Index of Save the Children will include a third part relating specifically to education, which will be presented in a later stage, education indicators will not be treated in the health context here.

It is possible however to include in a health index an indicator that reflects deviations in children and young people’s social behaviour. If a child or adolescent is at risk of unfavourable development, the social welfare board can take over care and place the child outside the home, in a family home or a group home. These placements can be made on the basis of the Social Services Act (voluntary placements) or of LVU, the law regulating the care of children and adolescents (where the placement is in conflict with the desires of the children or of the guardian). The municipalities report their interventions to the National Board of Health and Welfare, which compiles the results on an annual basis and publishes a social report every third year that also includes information on the structure of the municipal activities. This social report is thus a counterpart to the public health report and is intended to give information about and analyse how social welfare and social problems develop.

As mentioned in Chapter 2, there is a joint project – Local Welfare Balance Sheets (Folkhälsoinstitutet 1999) – involving the Swedish Association of Local Authorities, the Federation of Swedish County Councils and the Swedish National Institute of Public Health. Some of the participating municipalities, such as Göteborg and Haninge, focus on children’s welfare. There is now a database with basic information on the municipal level that gives a foundation for work in public health, and attention is also given to some child areas (www.fhi.se). It includes a number of concepts that are interesting from a child perspective.

One is “safe communities”, which means that the municipality fulfils six criteria drawn up by WHO. The municipality works in an intersectoral, structured way to prevent injuries in all risk groups.

The other is the “allergy-adapted municipality”, which means that the municipality has taken a political decision to work for the prevention of allergy, that there is an allergy

25 committee with broad representation and an action plan for the work. In October of 2002, 38 municipalities out of 289 complied with these requirements. The goal is for all municipalities in Sweden to be allergy-adapted. Both these indicators were included in the first version of this proposal for indicators for children (a “wish list”). Another indicator was also included that reflects the municipalities’ interventions among children and adolescents. The indicator was selected that shows the total number of children with at least one intervention during the year in question: “Children and adolescents who have been the subject of one or more interventions at some time during the year.”

In total, in all of Sweden’s municipalities, 16 000 children and young people were the object of some measure in 1998, and the information has already been broken down on the municipal level and calculated per 1 000 children in the population according to age groups 0-12 years, 13-17 years and 18-20 years (Socialstyrelsen 1999).

This indicator describes the municipalities’ efforts and not actually children’s health status. It is a process indicator and, as such, is affected by the resources, ambitions and prevailing ideologies of the local social welfare board, and of course by the composition and needs of each municipality’s population. The purpose of placing a child outside the home however is to improve the child’s growth and development and to prevent harmful development. The action can thus be expected to be significant for the child’s health and wellbeing. For this reason it can be justifiable to attempt to document these measures and use them as an indicator of the municipality’s support to children and adolescents and their families. The indicator is presently difficult to interpret but, in relation to other indicators that give information about the municipality’s social eonomic and ethnic structure, it should be valuable in the future. The quality of the data is not yet satisfactory and work is being done to improve it .

26 Indicators of children’s health

5. A basis for constructing health indicators for children

The previous chapters analysed the foundations upon which a municipal child index should be based. A picture was presented of the historical development of population health measurement both in Sweden and internationally, together with the special considerations that come into play when measuring the health of children and adolescents.

As the point of departure the municipalities, the rights of their inhabitants and their corresponding responsibilities with respect to health were described. A short review was also given of the health status of children in Sweden at the present time. There has been reference to the various factors pointing to the need for and the problems to be confronted in creating a health index for children on the municipal level.

The relevance of health measurement as an aspect of welfare measurement is increasingly recognised. Several international organisations, UN, WHO, UNICEF, OECD and EU,. regularly publish reports in the area, Nevertheless, the availability of reliable, comparable health data is still very limited, and the reports contain only the simplest and crudest measurement points. However, systematic efforts are now being made by WHO, OECD and EU to establish unified and comparable measurement methods. The most important development in Europe at this time is EU’s Health Monitoring Project (HMP), which works on a number of different fronts to set up a common system of health indicators for the Member States. A number of countries have introduced regular health measurements and tried to create health indicators for children, e.g. Australia (Waters et al 2002), Canada (The Health of Canada’s children), Italy (Tamburlini 2001), New Zealand (Our Children’s Health), UK (Health of Young People) and the US (America’s Children).

In all these cases the chosen indicators are used to describe the general situation in the country. The results of studies of representative samples of the national child population are extrapolated to reflect conditions in the country as a whole. But sub- nationally, the smaller the population one wishes to study, the more detailed and complete the information must be. Most national data are not detailed enough to be broken down to the municipal level. Another feature of these national reports is that no one has tried to combine several indicators into one index. To the contrary, it is advised that this should not be done.

Sweden has a long tradition of official demographic statistics, and in recent decades welfare measurements have been developed into a relatively complete system involving different national authorities. However, most of the data collected relate to the health and welfare of the adult population. While there has been some improvement in recent years, information about children’s welfare remains limited. There is currently no system that can survey and follow children’s health and wellbeing over time and compare developments between different parts of the country or between one part and the national population. However, there are some welfare and health projects that include children or even focus on children (Statistics Sweden, National Board of Health

27 and Welfare, Swedish National Institute of Public Health). This is further discussed in chapter 9.

Save the Children’s initiative, to create a child index that can be used to track or monitor and evaluate the situation in Sweden, comes at a very opportune time. It is important for several reasons that the Convention on the Rights of the Child has been used as a basis.

First, it follows an international set of regulations in whose formulation Sweden has invested resources and prestige. Secondly, it takes the Convention ‘s broad view of children’s health and wellbeing, i.e. it has not only to do with the right to life but also the right to survive and develop. “This means physical, mental, spiritual, moral, psychological and social development that must be provided for so that the child is prepared for an independent life in a free society” (Children’s Ombudsman, Barnombudsmannen 2001). It is important that central responsibility is placed on the municipalities. They have a major influence on the environments in which children live; they have the greater part of the responsibility for addressing the factors that promote the wellbeing of children and families.

But in terms of identifying potential indicators to reflect children’s health status, there is a problem in going down to the municipal level. Most of the so called health indicators in common use have to do with disease and treatment measures, and this is normally the county councils’, not the municipalities’ area of responsibility. However, indicators should not focus on purely medical features. A broad multidimensional and multisectoral concept of health must be emphasised. It is important to identify determinants of health (both risk and protective factors) as well as manifestations of disease. It is also important that the indicators focus on children’s conditions, that the perspective is that of children and not of adults. There is a strength in using local indicators because they can so easily be connected with local activities. They can form the foundation for future work, spurring local people’s involvement and debate.

Health problems, especiaslly those causing impairments, disabilities and handicaps, are very important for the child’s capacity to function in society .They carry a risk for secondary psychological disorders – and for a long period of time, as we have a long life expectancy. But they also impact on the child’s family. The life of the whole family can change. Everyday care often requires much time and great effort. The child’s needs require the attention and concern of the other family members, and the family’s social activities may need to be altered. Siblings also run a higher risk of psychosocial problems.

Society and its resources are also burdened as the problem tests the functioning of the medical and social care system, in small matters and large. The effects of children’s disorders can reach far into the world of adults and can even affect the next generation. However, it is important to emphasise that children are individuals and citizens with their own rights. Childhood is not simply a road to adulthood; it has a value of its own and should be as healthy, constructive and happy as possible. Monitoring and protecting children are thus central tasks for society, not least because children are not able themselves to claim their share of welfare or express their preferences.

28 It is evident that the most vulnerable children, those who do not have the advantages of good health or of living in families with good financial, social and cultural resources, need to be most carefully monitored. These “risk children” require particular interventions and it is important that they are identified and given support as early as possible. On the other hand, new threats to health can arise which change the current picture of problems and services requireed. Risks can change, new “risk groups” can arise and old ones disappear. Furthermore, even if there is an over-representation of “cases” within risk groups, the large volume of “non-risk children” makes most of the “cases” still appear outside the risk groups. It is thus necessary to follow the entire population.

Children’s health is consequently a very large and complex area, and there are many different aspects to take into account. It can be difficult to find the exact number of children for a special indicator, for example, both as numerators and denominators.

The fundamental requirement for any indicator is that it must relate to children who are covered by the Convention on the Rights of the Child, i.e. between the ages of 0 and 17. The Convention ceases to apply at the age of 18, and also at that time children in Sweden come of age. This can cause problems when using data from existing routine registers. Most registers are kept according to demographic principles, i.e. grouped according to age bands 1-4, 5-9, 10-14 and 14-19. There is no cut-off at 18 years.

Other registers, such as the diabetes register, include only patients receiving care at children’s clinics, which used to be children under 16 years, meaning that we must either abandon the limit of 18 years, or request special treatment of data where this is possible. This must be determined case by case, but it is in children’s interests that the data and the knowledge derived from them, are adapted to the child’s needs and not the reverse. Hopefully, in the long run, all data , whether registered on the central, regional or local level, will be treated and presented in ways consistent with the Convention on the Rights of the Child. It would mean that children are made the unit of focus in the statistics in question.

Even though interest is said to be in children, it is often the adult perspective that can be glimpsed behind. For instance, when reporting the occurrence of passive smoking with reference to foetuses or infants, it is important to use the number of children who are exposed to tobacco as the numerator and not the number of parents or households that smoke.

There are further aspects of health indicators in general, and indicators for children in particular, that must be considered before a realistic proposal can be presented. A health indicator shall be a defined and most often quantitatively measurable dimension of an important part of health, the health care system or related factors. An indicator can be said to quantify and simplify phenomena and facilitate the understanding of complex realities. It must be informative and sensitive to changes over time and between areas. It must fit logically within a framework and there must be a general understanding as to what is being measured and why (ISO/TS 2002). Thus the chosen indicators must satisfy strict criteria and properly reflect the parts of the system one wishes to measure.

29 As Wolfson writes: “Data and facts are not like pebbles on a beach, waiting to be picked up and collected. They can only be perceived and measured through an underlying theoretical and conceptual framework, which defines relevant facts and distinguishes them from background noise” (Wolfson 1994).

In formulating an indicator, both quantitative and qualitative criteria must be applied.

1. The indicator must reflect conditions which are

a) significant to the population’s health, in the short or long term, i.e. it shall be widespread and/or serious, b) an economic burden to the individual, the family and/or society and c) treatable.

In Sweden, the most important health problems in childhood, estimated as causes of disability-adjusted lost years of life, are diseases in infancy, congenital malformations, mental ill health, injuries, infections, allergies and cancer. The primary determinants are less favourable social conditions, use of tobacco, deficiencies in the immune system, poor breastfeeding and several other factors having to do with mental ill health and injuries (Bremberg 1999).

2. The indicator must be flexible and it must be possible to change it according to changes in conditions, for example in the composition of the population or scientific advances.

3. There must be access to the indicator. To be able to suggest indicators that are meaningful, it must of course be possible to measure and collect them. In point of fact, many indicators are constructed on the basis of available data sources, i.e. that which is not present in existing data sources is not considered. And things that are at a particular time impossible to measure or too expensive to measure are not included. We look for the coin we’ve lost under the light of the street lamp and not where we actually lost it. For this reason it is important not only to propose indicators but also to identify conceivable sources of data. There must also be a vision and preparedness for the future. But if we keep simply to what exists, we will be unable to create anything new, however necessary it becomes to do so.“If we always do what we’ve always done, we’ll always get what we always have gotten” (Kevin Dunbar).

4. Scientific requirements and qualitative criteria must be fulfilled.

a) There must be high validity, both as regards the ability to measure that which it is intended it shall measure (face validity), the ability to cover important aspects of that which it shall measure (content validity) and the ability to co-variate with other indicators that measure the same thing (construct validity).

b) Reliability must be high, i.e. repeated measurements in different areas and over time must show as small differences as possible. This is achieved by strict definitions and by precision in collecting and reporting data.

30

c) Sensitivity and specificity must be high, i.e. the risk must be as small as possible of a measurement missing those who according to the definitions belong to the group, and of including individuals who do not belong in the group in question.

Thus, there must be solid, knowledge-based insights behind each and every indicator. Together, they should build a system for surveying children’s health and wellbeing on the municipal level, a system that will allow practical measures that improve their health. This is“measuring for doing, not measuring for measuring.” Although these indicators are not being constructed for use in scientific studies; scientific surveys and analyses should be carried out. A number of such studies are recommended at the end of this report that can contribute to expanding the scope and improving the quality of the next generation of indicators. The criteria laid down here for the construction of indicators must be followed up and evaluated on a continuing basis.

6. Principles for municipal indicators of children’s health

On the basis of the goals set by Save the Children and the analyses and considerations introduced above, we can formulate a number of principles for the construction of health indicators to be included in a municipal child index.

1. The index shall measure and evaluate children’s health status and the factors that determine their health status on a municipal level and follow them over time.

2. The index shall facilitate planning, monitoring and evaluation of municipal efforts to protect and promote children’s health.

3. The index shall cover all important aspects of children’s health.

4. The index shall aim for an ideal set of indicators.

The ideal outcome would be a set of indicators that reflect the broad concepts of health, wellbeing and development, from birth to 18 years. From this ideal (and perhaps utopic) level, a group of realistic, available and at the same time functional indicators have been chosen for the final report. A proposal for future work to develop the indicators and their availability will also be made, so that the ideal outcome may in time be reached.

The important reason for this approach is that care of the health and wellbeing of our citizens is an overall goal for both individual countries and international organisations such as the UN, WHO and EU. The most vulnerable group among all citizens is children, who are most at risk and who are politically the weakest. Supervising developments in children’s health is therefore an important task for all countries and should be followed as energetically and effectively as possible. For several practical and economic reasons, however, it is not possible to immediately achieve the goal of a complete, systematic and detailed surveillance, and so we must lower our ambitions. Instead of basing work on the indicators that are already available in general registers,

31 this analysis is based on what we want to achieve, seen in terms of what is best for children. The final proposal for indicators will then be a compromise between goals and possibilities, but there must not be any doubt as to what the final goal of health indicators for children is: a comprehensive, relevant and high quality surveillance of children’s health, development and wellbeing, placed in a social and political context.

5. The index shall be based on a public health perspective. In order to follow developments in children’s health and protect and promote it, it is not sufficient to use factors that reflect medical circumstances or weaknesses in individuals. Broad, multidisciplinary and intersectoral aspects must be taken into account, not only those that reflect morbidity and mortality. With this perspective, the factors that determine health will also be meaningful. This is in agreement with the modern interpretation of public health as it is referred to in national and international policy documents, and elaborated in chapter 1.

6. The index shall be based on a general population perspective. The indicators must cover the entire child population and not be oriented towards risk groups only. In a population perspective, the gains to be made with risk group thinking are small. A large proportion of high risk group children will usually remain healthy, and a not insignficant proportion of those who become sick will come from low risk groups. Furthermore, the definitions of risk groups are often imprecise and can vary over time and area, which creates considerable problems in agreeing common and sustainable definitions.

7. The index shall be characterized by a child perspective. Children are an important investment for the future and a healthy childhood is the best foundation for a healthy adult life. Nonetheless, children must be seen as an important group in society in themselves, a valued group with its own needs and rights, here and now, not merely future adults. The often used ‘family perspective’ is based on the needs of adults, which do not necessarily coincide with the needs of children. Children need and deserve their own perspective. “A system for surveying children’s health. should be at the centre of Child Health, not at the periphery of monitoring” (Rigby & Köhler 2002).

8. The index shall, if possible, be based on outcome variables, and at times on process variables, while the use of structure variables is questionable. An indicator based on an outcome variable is more reliable and generally has a higher validity than other types of variables. They are better suited to measurements of health status and of the effects of health and sociopolitical interventions. However, when measuring the effects of plans and strategies, especially in the short term, process variables can be valuable. They can give an idea about the factors that affect outcomes, and they are usually easy to change in the plans. Structure variables often lie far from outcomes, and it is not always easy to find direct causal connections between structure and outcomes. Confounders, disturbing factors, often appear and make it difficult to interpret the situation.

9. The index shall include socioeconomic, cultural and demographic background factors. All health indicators must be seen in their social context in order to analyse what factors determine health, which is usually unevenly distributed in the population, and to

32 set priorities for interventions. As far as possible, indicators must be reported according to age group, sex, socioeconomic status and ethnicity.

10. The index shall permit systematic monitoring, quality assurance and cost estimations.

11. The index shall be open and allow future additions and changes.

These principles are intended to guide the construction of an ideal set of indicators for children’s health and wellbeing. It is obvious that the economic, staffing and organisational resources needed for constructing an ideal set do not exist today and perhaps not tomorrow either. But that is not the crucial point. In a progressive society that values and protects children, there must be a comprehensive system for monitoring and evaluating children’s health and wellbeing that is set in a social and political context and. adopts the children’s own perspective. Its construction will require perseverence and commitment from everyone involved, from people at the grass roots to the highest political leadership, and it may take a long time.

7. A set of indicators for children’s health

Children’s health and its determinants can be structured according to particular domains. The following division into four domains is based on analyses adopted by the EU project Child Health Indicators for Life and Development (CHILD). They are also well suited for a Swedish system of municipal level indicators.

A) Demographics and socioeconomics B) Health status and wellbeing C) Determinants (risk and protective factors) D) Service, support and health policy

The following set of indicators within these four domains accords with the principles and criteria presented in the previous chapters. In full use it would present a comrehensive picture of children’s health and welfare.However, as of today, this set remains in the nature of a “wish list” As will become clear below , it is not yet possible to employ all of them. Some will have to wait.

A. Demographics and socioeconomics The indicators in this domain have two purposes. They give an overall picture of the social and population structure, which varies markedly between municipalities. Most of them will be used as background variables for other indicators on the list. As explained under principle 9 in Chapter 6 above, the reason is that health and ill health and the factors that determine them do not occur randomly in the population.

In all societies – even in rich welfare states – they are unevenly distributed in the population, so that children from families in a good financial position and a high education level have the best health, and children from families in a poorer financial position and a lower education level have higher mortality and morbidity and poorer

33 wellbeing. Extreme social, economic and cultural factors make immigrant and refugee children particularly vulnerable.

Ideally, to understand the problems and to determine appropriate priority interventions, the indicators should be reported according to age group, sex, socioeconomic class and ethnicity.

In this context it has been decided to use parents’ social group, education and ethnicity as well as family structure. All these variables have been well tested in a welfare context and have been shown to be strongly connected to health problems of many different kinds (Hjern et al 1998, 2001 Socialstyrelsen 2001b, Bremberg 2002, SCB/National Institute for Working Life 2002).

B. Health status and wellbeing Mortality is one of the most commonly used indicators of ill health in an international perspective as it has such a great impact on the individual and on society. It is also definitive, easy to determine, and allows fairly exact comparisons over time and betrween areas.

However childhood mortality is very low in Sweden; a little over 600 children under 20 years of age die each year. If these deaths are distributed over different ages and socioeconomic groups, the numbers become too small to be usable on the municipal level, even if several years are pooled. Thus mortality cannot be used in the municipal index, and even less, cause specific mortality. Instead of mortality, injuries treated in hospitals has been chosen.. They can be prevented by different types of interventions, not least on the municipal level. Diagnoses of injuries are available in the National Board of Health and Welfare’s inpatient register and in sufficiently large numbers to be disaggregated to the municipal level.

Morbidity is a considerably more complicated measure of ill health than mortality. The concept of ill health is not clear and may be defined differently by the general public and professionals. The measurement methods are also more uncertain, and results reflect more than only morbidity. Geographic, social, economic, cultural and organisational factors affect the availability of care and are also decisive for how questions about morbidity are asked, and the answers registered and used. Centrally available data on local use of care resources are limited to outpatient activities in hospitals. Data from primary care units are still lacking.

The physical diseases included as indicators are either those whose incidence is increasing most in the Western world and have a great effect on the individual, family and society (asthma, diabetes, overweight/obesity) or those whose prevention is simple (caries).

One of the most common complaints in childhood is infections, which can also cause significant disruption in the child’s and the family’s daily lives. Most infections are now shortlived and mild and are treated by parents without the help of professional care (Uldall 1986). To capture serious infections, one must have data on children treated in hospital. But fewer and fewer children receive hospital care for their infections, an effect of successful ambulatory treatment and the provision of fewer hospital beds.

34 In comparisons over time, it is difficult to know what causes changes in the use of care: is it the panorama of disease that has changed or is it the care resources, care organisation or treatment routines? To capture common and mild infections and their consequences for the family and society, we must have information from parents, pre- schools and schools. Studies of this kind have been done in limited areas, but information is not available for the majority of municipalities and not systematically.

Extensive and dynamic vaccination programmes prevent the most serious infections . The number of children vaccinated for the most important infectious diseases has been suggested as an indicator under the “Determinants” group below. The most important risk factor for upper respiratory infections – passive smoking – is also included as a further indicator among the determinants.

Mental health is much more difficult than physical health to capture with simple and easily available indicators, and yet this is a part of children’s health that may be most affected by the social environment in the municipality (Bremberg 1999, Hwang 2002). The methods available (questionnaires, interviews) have been used only in particular areas or in more or less representative samples of the national population (Children/Swedish Level of Living Survey, Health Behavior of Swedish School- children). They cannot be used to describe the situation in individual municipalities. This will only be possible when the proposal for regular measurements of children’s and young people’s mental health becomes a reality (Hagquist 2004). Positive aspects of mental health will then also be given greater attention in formulating appropriate indicators.

The present set of indicators has included for mental health attempted suicide among children and children in the care of the municipality, both of which have a negative connotation.

C. Determinants (Risk and protective factors) This domain contains a number of indicators that all have a clear and documented effect on health status in childhood and later. It is also possible to influence them through health promotion, not least on the municipal level. Two of them are positive (children who are breastfed and children who are vaccinated), and five are negative (low birthweight, passive smoking, active smoking, alcohol habits and teenage abortions).

The educational system should be able to add more indicators to this group, such as the number of children who attend pre-school and children who complete the compulsory school years. The national, regular measurement of children’s mental health will be able to give further suggestions in this area (Hagquist 2004).

Certain indicators could also appear in this section as risk factors, as they are both signs of ill health in themselves and determinants for other health problems, e.g. refugee children and children in the care of the municipality.

D. Service, support and health policy Three municipality-related indicators have been identified, ‘safe communities’, ‘allergy-adapted communities and ‘community action programme against bullying in schools’. The first two are part of the National Institute of Public Health’s municipal database. The educational system

35 and the new measurements of children’s mental health should be able to generate further policy indicators in this domain.

A wish list A list follows below of indicators that would reflect children’s health and wellbeing and allow comparisons over time and between municipalities. It is not strictly an idealistic wish list ; certain adaptations that acknowledge practical realities have been made, and are explained in the text.

Domain A. Demographics and socioeconomics

Indicators Definition

Children’s Proportion of children that live in households where parents social class (primarily the father and mother or single-parent families): are not professional workers, lower white-collar workers, middle- level workers and above, entrepreneurs, farmers or students. Distributed according to boys and girls and total as well as age groups 0-4, 5-9, 10-14, 15-17.

Parents’ Proportion of children that live in families with an educational education level (primarily the father and mother or single-parent families): low education level (compulsory school), middle- level education (upper secondary school or further education less than 3 years) or high education level (3 years of school post secondary school or longer or post-graduate education). Distributed among boys and girls and age groups 0-4, 5-9, 10-14, 15-17.

Children’s Proportion of children who live in families with only one parent structure or guardian, male, female and total, and percent of children who live in families with co-habitating parents in age groups 0-4, 5-9, 10-14, 15-17.

Children Proportion of children born in: the Nordic countries, from other European countries outside the Nordic area, and others. countries Distributed among boys and girls and total as well as age groups 0-4, 5-9, 10-14, 15-17.

Refugee Proportion of children seeking asylum, alone or as part of a children family, per 100 children in the municipality, distributed according to boys and girls and total as well as age groups 0-4, 5-9, 10-14, 15-17.

36 B. Health status and wellbeing

Indicators Definition

Children with Proportion of children 0-17 years who have received external hospital care for external injuries, distributed according to injuries boys and girls and total.

Children with Proportion of children with a diagnosis of diabetes (type 1, diabetes insulin-dependent), distributed according to boys and girls and total, and age groups 1-4, 5-9, 10-14, and total.

Children Proportion of children who at 6 years of age are caries-free without caries in the primary dentition, distributed according to boys and girls and total.

Suicide attempts Incidence of suicide attempts, defined as release from among children hospitals with a diagnosis of attempted suicide, distributed according to boys and girls and total, and age group 10-17.

Children with Prevalence of asthma among boys and girls and total, and in asthma age groups 0-4, 5-9, 10-14, 15-19, and by social group.

Overweight and Proportion of children with an age and sex standardised Body obese children Mass Index (BMI) of at least 25 and 30, measured at the time they begin school and at 18-19 years (boys).

Children placed Proportion of children and young people who have been in municipal the object of some municipal intervention one or more times care during the year. Distributed according to boys and girls and total, and to age groups 0-12 and 13-17 years, per 1.000 children in corresponding population groups.

C. Determinants (Risk and protective factors)

Indicators Definition

Children with Proportion of infants with birthweight <2 500 g, birthweight as a percentage of the number of live births, according to socioeconomic class and family structure.

Children that Proportion of children that are breastfed solely or partially at are breastfed 4 months of age as a percentage of live births and distribution among social groups.

Foetal exposure Proportion of children exposed to tobacco in the womb to tobacco by mother’s smoking, per social group.

37 Teenage Proportion of abortions per 1 000 women 10-17 years, abortions per social group.

Tobacco Proportion of children that report that they smoke every smoking among week. Distributed according to boys and girls and total, and children and according to ages 11, 13 and 15, and social group. young people

Alcohol Proportion of children in the 9th grade that report intensive consumption consumption of alcohol some time during a month or more among young frequently. Distributed according to boys and girls and total, people and according to social group.

Vaccination Proportion of children who have complete vaccination protection for measles, mumps and rubella at the age of 2, as percent of all children in the age group.

D. Service, support and health policy

Indicator Definition

Safe Municipality has been named a ‘safe community’ for the current community year by the National Institute of Public Health/Rescue Services Agnecy.

Allergy- Municipality has been named an ‘allergy-adapted community’ adapted or the current year by the National Institute of Public Health. community

Action pro- Proportion of children attending a school with a gramme written, implemented and monitored action programme to against counteract bullying, as percent of all schoolchildren. bullying

Excluded indicators While keeping in mind the ideal presented in chapters 5 and 6, indicators must meet other requirements. To have a chance of being adopted in municipal planning, indicators must be understandable and reliable as well as readily and economically available, and all who handle them should realise their value. They must be meaningful to those who provide and collect the data and manageable to those who then process the data and interpret the results. It is not possible to use all indicators that are desirable, at least not at the present. But the indicators that are possible are not necessarily desirable if, for example, they do not fulfil the quality requirement.

Thus the list must be examined and shortened so that the indicators that are finally selected are relevant, valid, reliable and available. The following indicators listed according to domain will now be excluded from the original list.

38

Domain A. Demographics and socioeconomics The indicators here are intended to serve two purposes: to give a general description of the municipalities and their structure and to function as background and explanatory variables to the other outcome indicators. The same or very similar indicators were already used in the first report in Save the Children’s series on Municipal Child Index, Child Poverty in Sweden (Salonen 2002). They are thererfore all excluded here.

Instead, the poverty index in the report will be used, for both the purposes given above. Salonen’s definition is based on a combination of two indicators ”children who live neither in economically poor households nor in households with means tested social assistance”.

Domain B. Health status and wellbeing

Indicator Reason for exclusion

Children with Diabetes is in many ways an excellent indicator: it reflects a diabetes serious disease with a unified diagnosis and treatment that is on the rise in Sweden, as in the rest of the Western world. However, the number is still too low for a meaningful report per municipality, even if age groups are pooled. Nonetheless it is a disease that must be followed carefully on the national level, and this is being done in national medical registers, as well as in several research projects.

Children Caries is in many ways an important disease: it can be painful without inextreme cases, it is related to other diseases, it is easy to caries diagnose, it clearly reflects socioeconomic differences, it is easily available for preventive efforts and it is used internationally as an indicator. In other words it is an almost ideal health indicator for children and should as such be included in a complete health surveillance system. Unfortunately, there is still not information on all children on the municipal level and it has thus been excluded here. These data may however become available within a few years.

Suicide Suicide reflects both individual and social conditions attempts It is fortunately rare during childhood, and even suicide among attempts, an important predictor for later suicide, are not children sufficient in number to be broken down on the municipal level in any meaningful way. They are however part of indicator B 1 children injured by external causes.

Children with Asthma and allergies are now the most common long term asthma conditions in children and young people and there has a been strong increase in practically every Western country. There has been great interest in tracing, treating and preventing them. However, there are no clear and standardised criteria for diagnosis and the large geographical

39 variations in prevalence are partly due to variations in diagnostic criteria. Neither is there good agreement between the occurrence of symptoms and use of care. Inpatient care of children 5-18 years for asthma has e.g. in Göteborg decreased by 90 percent in the past 10 years, which might indicate that there is effective outpatient care (Wennergren & Strannegård 2002), but outpatient data are not generally available. The indicator has been excluded because the diagnosis is uncertain and varies between different clinics and because it is unclear what the indicator reflects.

Children with Overweight has become one of the fastest growing crucial overweight public health problems in the West. There are internationally and obesity accepted definitions but complete data are not yet available, neither in pre-school ages nor school ages. All 18-19-year-old boys are weighed and their height measured when they enrol for military service, presently about 90 percent of this subpopulation. Girls do not enrol and hence are not measured. The indicator has been excluded because data collection is incomplete. Enrolled boys represent only half of the actual population and moreover often fall outside the limit of 18 year.

Children This indicator describes community efforts rather than children’s health placed in status. It is a process indicator influenced by the resources, ambitions municipal and ideologies of the local social welfare board, and by the composition care and needs of the population in question. The quality of the data has been criticised for being poor because the indications for interventions vary and because all municipalities do not report their interventions to the National Board of Health and Welfare. Work is ongoing to improve quality. The indicator has been excluded because the quality of the data is questionable and their interpretation unclear.

Domain C. Determinants (Risk and protective factors)

Indicator Reason for exclusion

Tobacco The indicator is well documented and very often used in studies of smoking young people’s health habits. It is clearly related to a number of among serious health problems, both in adolescence and in adulthood. children and All studies so far have been conducted on more or less representative adolescents samples and are not available for all children on the municipal level. The indicator has been excluded because data are not available on the municipal level.

Alcohol Alcohol consumption is clearly related important health problems , and comsumption is very often included in surveys of young people’s health. This is done among in random testing among representative samples and not in complete adolescents population groups. The indicator has been excluded because data are not available on the municipal level.

40 Domain D. Service, support and healthy policy

Indicator Reason for exclusion

Safe muni- WHO and the National Institute of Public Health/Rescue cipality Services Agency have revised the criteria for this indicator and now focus on 6 instead of the previous 12 criteria. However, it is still not entirely clear how the criteria shall be interpreted and there is no established plan for systematic monitoring of how municipalities apply the criteria each year. The indicator has been excluded because the concept of safe municipality is vaguely defined and unclear in terms of its continuity and because the evidence on any direct relation to children’s health is weak.

Allergy-adapted The indicator reflects the municipality’s efforts to work in community such a way as to prevent allergy. This includes actions that are diffuse and difficult to capture. What actions such as a political decision to appoint a committee to establish an action plan actually imply for allergic diseases is difficult to say. They may reflect the municipalities’ ambitions, but have largely unproven effects on health. Furthermore, even the best intentions can lead to actions that are later found to be completely meaningless or even harmful, which has been the case in many well meaning medical recommendations in the area of allergy. The indicator has been excluded because the criteria are not clear and the consequences for children’s health are doubtful.

Action This is again a process indicator that does not directly reflect programme children’s health status but instead describes how seriously against bullying the municipality views a grave and widespread health problem in schools and its attempts to prevent it. This indicator is included in the proposal for a common European indicator system for children’s health. It has been excluded here because it may come to be treated within the framework of Save the Children’s third part of the project concerning the educational system.

Remaining indicators Now that the above indicators have been removed, there remains one indicator that reflects significant health problems in children and young people, namely hospitalised external injuries including suicide attempts. As determinants, there are five indicators that are significant for current or future health, of which two are positive (breastfeeding and vaccination) and three negative (low birthweight, early exposure to tobacco and teenage abortions).

41 Domain A. Demographics and socioeconomics No indicators remain

Domain B. Health status and wellbeing Indicator Rationale

Children External injuries treated in hospital reflect a serious injured by complaint that could to some extent be prevented by social external interventions, not least by municipalities. Data have long been causes available.

Domain C. Determinants (Risk and protective factors) Indicator Rationale

Children with There is a connection between low birthweight and difficult low birth- socio- economic family conditions, but also low birthweight weight in itself increases risks for later medical and developmental complications. It can be influenced at least to some extent by local efforts. Used world wide as a health and welfare indi- cator.

Children that Although not intimately connected with survival, as are breastfed it is indeveloping countries, breastfeeding is a positive factor for health and wellbeing in rich industrial countries as well. Efforts to stimulate breastfeeding can be initiated on the local level. It is used internationally as a health and welfare indicator.

Foetal More and more knowledge is being gathered on the harmful effects exposure to of tobacco, especially on immature and growing organisms. Great tobacco efforts must be made to protect newborns and small children from this health threat, and much can be done within the municipalities.

Teenage Teenage abortions are a sign of the effectiveness or failure of abortions preventive efforts, primarily those in schools and young people’s clinics. The increasing numbers in recent years and the uneven geographical distribution indicate that local actions are significant.

Vaccination Vaccination is one of the most used and most cost-effective methods of preventing disease. The percentage of children that are vaccinated describes how good the protection against infection is in the child population in question and how well health information given at children’s health centres and other local actors has reached parents and gained their trust. Local vaccination data are collected both regionally and centrally and can be broken down on the municipal level via the central register at the Swedish Institute for Infectious Disease Control.

42 Domain D. Service, support and health policy No remaining indicators

Final selection of indicators1 (1 See appendix 1 for an exact definition and technical description)

1. Children injured by external causes 2. Children exposed to tobacco in the womb 3. Children with low birthweight 4. Children that are breastfed 5. Teenage abortions 6. Children vaccinated against MPR

Sources of data The indicators that have met the requirements for quality and availability on the municipal level are all based on data produced within the local health care system and then collected and processed by central authorities. The results are presented as reports and tables to various target groups, such as Parliament, the Government, county councils, municipalities, and to the general public. These reports are available in printed versions and electronically on the authorities’ websites.

The results are most often disaggregated to county council areas. But because most information is connected to personal identification numbers, on special request it is very easy technically to match it to smaller units, such as municipalities.

The reliability of published reports depends primarily on how carefully and uniformly the data have been collected on the local level. The primary collection of data is by individual doctors nurses, secretaries and others working in the care system.. The task requires precision and interest on the part of individuals following detailed instructions, functioning routines and an effective control system. These prereqisites do indeed exist, but the amount and quality of the data still vary between different areas of the country, e.g. data on drop-out rates. However, quality is generally good and the reliability of the data high, and there is a continuous supervision of quality.

Information on the data used in this report, the sources, background, history, drop-out and general reliability, now follows.

The information on “injuries treated in hospital”, “low birthweight”, “breastfeeding”, “smoking during pregnancy” and “teenage abortions” was taken from the National Board of Health and Welfare and on “children vaccinated” from the Swedish Institute for Infectious Disease Control.

National Board of Health and Welfare The National Board of Health and Welfare is the central authority for public health surveillance. For this purpose it has established the Centre for Epidemiology (EpC), whose task is to monitor, analyse and report on health and social conditions in Sweden. The EpC is responsible for several registers, including the cancer register, medical birth register and patient register, as well as for classification and national reports on population health and social conditions.

43 Patient Register The Patient Register includes information on all patients that have been treated in hospitals and is thus the basis for the indicator children injured by external causes. Central compilations of citizens’ illnesses and treatments have been made in Sweden for over 100 years, and the present register has covered the entire country since 1987. The most recent years include information on 1.5 – 1.7 million treatment episodes each year. There is information on diagnoses, surgery, external causes of injury, sex, age, residence, hospital, specialty and means of admission and discharge.

The Patient Register is a so called health data register and is regulated by the law on health data registers (1998:543), the National Board of Health and Welfare regulation on patient registers (2001:707) and the National Board of Health and Welfare’s instructions for compulsory information to the patient register (2001:1). The legislation makes it possible for the National Board of Health and Welfare to keep a register based on personal identification numbers.

Regular quality controls of the data submitted by the health care authorities are made by the EpC and show generally high quality for the variables of interest in this context. The drop-out in primary diagnoses in 2001 was 0.9 percent and in personal identification numbers 0.4 percent. Validation studies have also shown good reliability (www.sos.se/epc).

The Medical Birth Register (Medicinska födelseregistret, MFR) was introduced throughout the country in 1973 and uses individual records from antenatal care, obstetrical care and neonatal care. Copies of the records are submitted on a running basis from every clinic/ward to the National Board of Health and Welfare, where they are compiled and analysed. The purpose of the register is to provide a basis for analyses of risks in women and children during pregnancy, childbirth and the neonatal period and to study mortality, morbidity, injuries and care routines. The register has the permission of the Swedish Data Inspection Board to base the information on personal identification numbers and can thus be connected to other registers with data relating to children and parents. It contains information on the age of the mother, previous pregnancies, smoking during pregnancy, period of care, pain relief, interventions during delivery, weight and length at birth, malformations and the parents’ social conditions.

Errors can occur in reporting when records are filled in, when infants are transferred to children’s clinics, in the event of stillbirths and in registering children born outside the country.

Repeated analyses have been made of the quality of the data and the drop-out rate is estimated to be between 0.5 and 1.5 percent per year. In respect of records of the Medical Birth Register (MFR) and the population register kept by Statistics Sweden (SCB) a more detailed analysis of the quality of the MFR has been published in English (Cnattingius et al. 1990).

Statistics on breastfeeding Information on breastfeeding has been collected for many years at child health centres (BVC), but there has only been complete information since the end of the 1960s. Because of the importance of data on breastfeeding, the National Board of Health and Welfare and the Federation of County Councils now recommend that county councils

44 compile regional breastfeeding statistics that are then forwarded to the National Board of Health and Welfare, which compiles the information on a national basis every year. The information is collected on a running basis during the mother’s regular visits to the child health centre and is recorded in the child’s record that is kept there. This record follows the child when they move to a different area of the country. Since computerised information on the municipal level was not available until the end of the 1990s, breastfeeding at the age of 4 months can not be used as an indicator for the beginning of the 1990s.

Contrary to data from (e.g.) patient statistics, breastfeeding statistics do not count the total number of children as a denominator but use instead “the number of children registered at the child health centre”. This means a certain drop-out since some children do not use the child health centres. This portion of children is however very small, about 0.5 percent, and does not affect reliability. Information in the records is still entered by hand, which means that a further small number of records are illegible.

Statistics on mothers’ smoking habits This register is also based on child health centre records. Information is collected and treated in the same way. Smoking habits are reported by the mothers at the time of their registration at the antenatal clinic. The same calculation bases are used and the limitations are the same. Note that it is children who breastfeed and are exposed to tobacco and not the mothers who breastfeed and smoke that are used as the basis for calculation.

Teenage abortions The new abortion law of 1975 required new routines for data collection to be able to provide statistics that would allow a rapid follow-up of its effects. A special form was drawn up for reporting to the National Board of Health and Welfare that is submitted from all clinics or units at which abortions are carried out. The form has been revised a couple of times, and information on the woman’s home district was added in 1985. The information on the form is anonymised, i.e. it can not be traced to an individual, but it can be broken down to the municipal level because it contains information on the home district. The register is compiled on a continuous basis, and preliminary reports are published every six months. There is a very small drop-out that has to do with errors in filling out the form and transferring data. These errors are assumed to be relatively constant and there is nothing to indicate that the drop-out affects the purpose of the register, i.e. to measure changes over time.

Information on vaccination The current reporting on vaccination from children’s health centres to central authorities began in 1981. It is done using special forms. Each county’s Head of Child Health Services collects information from all the centres in their districts and then transfers it electronically to Children’s Health Services in Stockholm which then passes it to the Swedish Institute for Infectious Disease Control to be compiled and analysed. Reporting is based on vaccination status in children who had their second birthday during the previous calendar year and is calculated as the proportion of children of the same age who at the start of the year were registered at a children’s health centre.

The indicator shows the combined vaccination for measles, mumps and rubella (MMR) (Swedish ‘morbilli, parotitis and rubella’- MPR).

45 During the 1990s a measles vaccine was also given separately, but only to a small extent, 0.2 – 0.3 percent. These cases have been counted in the MMR numbers.

The reporting frequency has increased over the years, but is still not complete for the whole country. Two of 21 county councils (Örebro and Uppsala) are missing for the year 2000. These counties have chosen to keep their own registers, based on personal identification numbers, and the statistics are thus not fully comparable with those of other counties and do not exist on the municipal level. Thus, in the calculation of an index and in statistical treatments, the municipalities missing in these counties have been given values for this indicator that correspond to the mean value of the respective municipal group (see Table 2, chapter 8).

Reporting from municipalities in other parts of the country also varies . The explanation is in part that the boundaries of the children’s health centres catchment areas and the municipalities are not always the same. It is calculated that in total in the counties reporting there are reports for over 99 percent of the children for each calendar year, or 93 percent of the total number of children in the entire country (Smittskyddsinstitutet 2001).

Ordering data Extensive discussions have been held with the responsible persons at the various authorities regarding availability, suitability of format and data processing methods. Two groups of data were finally ordered, consisting of the latest available and complete annual compilations, and a group of corresponding data ten years back in time. There is complete information on most indicators most recently for 2001, for one indicator up to 2002. The most recent year when all indicators were available was 2000.

To achieve a uniform starting point, it was decided that 2000 would be used as a base year; and, to avoid the risk of values from one single year deviating from the usual pattern, the values were pooled to obtain a mean value for the most recent years, i.e. usually 1998, 1999 and 2000. This group is called the 2000 group or the 2000 period. The same calculation was made for information ten years back in time, i.e. the mean value for three years, usually 1988, 1989 and 1990, was used to form the 1990 group or 1990 period.

The indicator breastfeeding at 4 months of age was not available so many years back and thus comparisons between the two periods are made only for the other five indicators.

46 Table 1. Years from which data were used for statistical treatment

2000 period 1990 period

B1. Children receiving inpatient 1998, 1999, 2000 1988, 1989, 1990 care for external injuries

C1. children exposed to tobacco 1998, 1999. 2000 1988, 1989, 1990 in the womb

C2. Children with low birthweight 1998, 1999, 2000 1988, 1989, 1990

C3. Children who breastfedd 1998, 1999, 2000 1988, 1989, 1990

C4. Children vaccinated for MRP¹ 1998, 1999., 2000 1988, 1989, 1990

C5. Teenage abortions 1998, 1999, 2000 1988, 1989, 1990

¹ The years for the vaccination indicator are years during which data were collected and cover children who turned two years old during the previous calendar year.

The cost for extracting these data from the registers for each of the 289 municipalities amounted to a few hundred dollars.

Professor Tapio Salonen provided data from his poverty study, which used the years 2000 and 1991.

8. From indicators to municipal index

The review thus far should have made it plain that health measurements are a complicated business, in part because the concept of health is so multifaceted and in part because the methods we use are uncertain. Nonetheless it is a very important task to measure and evaluate the health of the population and to monitor it over time.

Health is a central component of welfare and as such is used as a way to characterize a country or a region. It is also important to measure health in parts of the population and in parts o a country because living conditions can be so different that it is meaningless to pool results; we get simply a porridge of numbers. This is one of the reasons why observing children’s health in municipalities is justified in it self.

Earlier chapters have made it clear that children deserve their own health surveillance system (chapters 2 and 3) and that municipalities are suitable units in which to work to promote health and prevent ill health (chapter 4). The first problem is then to identify children’ s health indicators that are valid, reliable and relevant on the municipal level. This is discussed in chapter 5.

47

Few indicators fulfil all these requirements today. But intense development efforts are being made in several areas of the country, often involving close cooperation between different government authorities (see chapter 9). It is reasonable to believe that within a few years there will be possibilities for a qualitatively strong and comprehensive surveillance system for children’s health, even on the municipal level.

This system will also include the difficult task of mapping children’s mental health in detail. It will be easier, although still complicated, to create a children’s health index that expresses the municipalities’ goal achievement on a scale from 0 to 100. Until the planned indicators work in reality, we can only do the best with the opportunities we have.

The municipalities’ ultimate health goal must be to ensure that their inhabitants can enjoy their highest possible or attainable level of health. At the least it must be to provide the conditions for good health.

This does not mean eternal life or a life without injury or sickness. There must be realistic goals. And room must be left for human variations, even among children. In terms of children’s economic conditions, it may be justifiable to use the goal that “no child shall live in an impoverished household” (Salonen 2002), which is an operationalisation of article 27 of the Convention on the Rights of the Child.

Article 24 of the Convention states that children have the right to “enjoyment of the highest attainable standard of health and facilities for treatment of illness and rehabilitation of health”

Right to the highest attainable standard or level of health means in this context that consideration is given both to the individual’s physical and socioeconomic circumstances and to the municipality’s resources and possibilities. The child and its parents themselves have a responsibility for any choice to engage in potentially dangerous activities. And of course, they have their own set of genetic conditions for health and disease.

Society has an obligation to facilitate, promote and provide for the right to health of its population. Society has the responsibility to provide health services of different kinds (disease prevention as well as care and treatment) to as great an extent as possible (UN Declaration on Human Rights 1948).

The latest public health bill from the Swedish Government states that work in public health, on the national, regional and local levels, shall focus on the factors that determine health (Mål för folkhälsan, Bill 2002/03:35). This of course has to do with how available rresources are used, as there are no absolute answers to what is sufficient. This leads naturally to judgements about priorities and decisions on actions to be taken, and these are not always objective and not always agreed upon by all.

Thus, it is no simple matter for a municipalitiy to operationalise its tasks. Many tasks are linked with responsibility for health care which is borne primarily by the county councils and the Government. The proposed indicators have been formulated such that

48 they reflect areas in which the municipalities have at least a certain responsibility and influence.

The municipalities should set challenging goals. It is manifestly desirable, for example, that no child should need hospital care for injuries, or undergo an abortion ,or be exposed to tobacco smoke. And although such ideal outcomes do not stand up as credible operational targets, it is reasonable to make use of a “zero vision” as a reference point, and then to calculate how far each municipality has come in comparison with it .

For example, the Swedish Road Administration states as a “zero vision” that “no child shall die or suffer a long term loss of health or experience insecurity in moving to or from school as a result of deficiencies in the road transport system” (Vägverket, 1999, 2002).

The WHO European Region has used this approach in its HEALTH 21 policy document, which it commended to Member States as a model for their consideration. The targets set are more loosely formulated, and there is only an attempt at quantification, a matter that is left to countries to determine for thermselves. In Goal 4, Young People’s Health, it is stated that “By the year 2020 young people in the Region should be healthier and better able to fulfill their roles in society”. According to the strategy, this will happen for instance by “reducing mortality and disability from violence and accidents by at least 50 percent and by reducing teenage pregnancies by at least one third” (WHO 1999b).

A collective measure of population health development has been produced by Statistics Sweden and is used in the National Board of Health and Welfare’s Public Health Reports as a Health Index. Here, the population is classified according to four morbidity groups based on a combination of the responses in Statistics Sweden’s investigation of living conditions (ULF) to questions on self-assessed health, long term illness and impaired capacity for work Its use is limited to the ages 16-84 years, no report is made of diagnoses or determining factors and it is based on an investigation of a sample of the population. It is thus not suitable for several reasons as a child index on the municipal level.

A more advanced way to present a collective measure of health development in a country is to use the concept of Disability Adjusted Life Years (DALY). One DALY corresponds to one year of health lost as a result of illness. It was developed by WHO and the World Bank to measure the burden of illness in a population and make comparisons between different parts of the world (World Bank 1993). It has also been applied in Sweden to provide a basis for decisionmaking in health policies, as a kind of “ill health GNP” (Diderichsen et al. 1998).

Although DALY has been seen as a breakthrough for making possible internationally comparable health measurements among populations, it is not unproblematic. The focus is on professional diagnoses of diseases according to defined codes, and there is a lack of a holistic perspective on health (see chapter 1). Self-experienced ill health is not included, and neither usually are risk or protective factors. Furthermore, calculations are most often made for adults, primarily because there is less knowledge about the incidence, prevalence and duration of illnesses among children and there are greater difficulties in extrapolating the weights of the functions. Other consequences of illness among children that have not been considered are that hospital stays can mean losses in

49 school time and play time. These in turn can lead to losses in education, poorer socialisation and perhaps lower social competence (Rigby et al. 2002).

Another difficulty with the DALY is that both the intricate weighting and the subjective evaluations – the ethical values and social preferences – must be carried out having in mind the significance of different health problems for individuals and groups of individuals. The DALY, in spite of its enticing apparent objectivity, is not free of valuation The values are simply expressed in numbers and carried out in a consistent manner (Diderichsen et al. 1998, SOU 2000).

More recently, WHO has used another international measurement method, DALE (Disability Adjusted Life Expectancy), which is an even more crude calculation that does not involve specific weights but is directly comparable with the measure of life expectancy (WHO 2000b).

DALY and similar measurement methods are now being developed all over the world, also in Sweden. To make them useable in a municipal index for children, however, there needs to be much more detailed analyses of the data, including extensive evaluations by panels of experts.

It would also be possible to use all the indicators that are included to construct a health profile for each municipality, with the values of the indicators put into a figure with lines or columns. By placing the profiles on top of or next to one another, the municipalities can compare their profiles with others ,and identify their particular strengths and weaknesses. They can focus on what they consider to be important for their municipality and its population without combining them into a single integer and without needing to relate the results to a hypothetical or unrealistic goal.

More refined comparisons can be made between all or selected municipalities by expressing the values as risk quotas and by e.g. standardising for socioeconomic class. It is not correct to assume that all municipalities are like one another and can be compared with each other in a simple way. The municipalities are far too different in size and structure. To make more valid comparisons, we can use the Swedish Association of Local Authorities’ division of municipalities into nine groups, which takes into consideration population size, degree of urbanisation and economic structure. This division is based on the 1990 population and housing census (FoB 90) but is updated according to need.

Table 2. Swedish Association of Local Authorities’ division of the 289 municipalities

Urban area Municipality with a population over 200 000 (3 municipalities).

Suburban municipality More than 50 percent of the registered population commutes to work in another municipality. The most common commuting destination being an urban area (36 municipalities).

50 Larger city Municipalities with 50 000 to 200 .000 inhabitants and less than 40 percent of the registered population employed in the industrial sector (26 municipalities).

Middle size city Municipality with 20 000 to 50 000 inhabitants; over 70 percent urban area and less than 40 percent of the population employedin the industrial sector (40 munici- palities).

Industrial municipality Municipality with more than 40 percent of the registered population employed in the industrial sector (53 munici- palities).

Semi-rural municipality Municipality with more than 6.4 percent of the registered population employed in the agricultural and forestry sectors, urban area of less than 70 percent and not a rural municipality (30 municipalities).

Rural municipality Municipality with less than 5 inhabitants per square kilometer and less than 20 000 inhabitants (29 munici- palities).

Other larger Other municipalities with 15 000 to 50 000 (31 municipalites inhabitants (31 municipalities).

Other smaller Other municipalities with less than 15 000 municipalities inhibitants (41 municipalities).

This classification is used in the comparisons between the municipalities reported in the following chapters. Over the years, certain changes have been made in the municipalities. Since 1992, five new municipalities have been formed that are areas that earlier belonged to other municipalities (Gnesta (0461), Trosa (0488), Bollebygd (1535), Lekeberg (1814) and Nykvarn (0140). This has been taken into account in the calculations in chapter 10 and in the tables.

By using health profiles of this kind we avoid the difficulties of weighting indicators. But of course, having several dimensions of welfare combined into one single measure to describe children’s health also creates difficulties, something which welfare researchers have often warned against (see e.g. SOU 2001a; SOU2001b, WHO 2000a and Rigby et al. 2002).

A municipal index If nonetheless, in spite of all reasonable objections, we want to create an index (and it was one of the tasks given by Save the Children), should it then be an index oriented toward the ideal? Or should it perhaps be based on comparisons between actual conditions in the municipalities, and the average or even the best values (benchmarking) then be used as the target? Both approaches are entirely possible and they can moreover be combined, as will be seen below.

51 In the first option it is decided that the long term goal toward which the municipality should work shall be optimal health, which means that diseases are eradicated, injurious determinants do not exist and the beneficial determinants are present for all inhabitants. The maximum value for each indicator is 100, which is awarded when the goal has been reached. The values of all the indicators are added together and divided by the number of indicators to give a number between 0 and 100.

What is simple about this system is of course that it is not a complicated business to add and remove indicators as new knowledge is gained or the availability of data changes. A disadvantage is the difficulty in evaluating and weighting the different indicators and domains, and then relating them to generally accepted and realistic goals (see above under DALY). In truth, it is not possible, even with the help of weights, to compare such different indicators as e.g. breastfeeding up to 4 months, teenage abortions and serious injuries. In a changing world, any attempt to weight these indicators would lead to ethical and practical difficulties, both immediately and in the long term. Giving all indicators the same value is certainly not correct, but it is uniform and less complicated.

OECD has reasoned in a similar way in constructing the so called Gini coefficient (OECD 2000). This indicates the position on a comparative scale that a country holds in terms of its income distribution. If income in the country is distributed such that all people have an equal income, the index is 0; if one person has all the income, the index is 1. This is sometimes multiplied by 100 to obtain manageable numbers, i.e. the index varies from 0 to 100. A scale without weights is also used here, and both endpoints are equally unrealistic.

With all these considerations in mind, the report presents the outcome of the six indicators in all municipalities in the country in the form of a common health index. Thus, to create a municipal health index for children, the results for each indicator are added and the sum is divided by six, i.e. the number of indicators.

The number of children in the municipality that have not received inpatient care for external injuries is calculated from Domain B, which consists of this single indicator. If all children from 0-17 years in the municipality (100 percent) fulfil this criterion, that is, no child received inpatient care for external injuries, the municipality receives the value 100. If no child fulfils the criterion, i.e. all children have received inpatient care for external injuries, the value is 0.

The proportion of children in the municipality in each age group is calculated in the same way for the five indicators in Domain C, breastfed up to 4 months, did not weigh less than 2500 g at birth, did not have a teenage abortion, was not exposed to tobacco in the womb and has complete vaccination against MPR. The goal is again that all the children in the municipality in the respective age groups fulfil these requirements. If this is achieved, the municipality’s value is 100.

Pooling the indicators and dividing by six gives a common municipal index between 0 and 100, where 100 is the desired goal. This index thus is an aggregated number calculated from the municipality’s achievement in each indicator. No attempt has been made to weight the different components – each indicator has been given the same weight, i.e. the outcome and determinant indicators are of equal worth in the final

52 judgment of how far the municipality has succeeded in creating healthy children and good conditions for healthy children.

Pooling of data In order to avoid distortion of the results by variations from year to year, a mean integer has been calculated for the results of pooled years, one period around 1990, normally 1988, 1989 and 1990, and one period around 2000, normally 1998, 1999 and 2000 (Table 1, page 47). However, individual indicators and indices in small municipalities must be compared with some caution. All estimations, and thus also comparisons between estimations, are affected by random variations. The size and direction of the variations depend on the size of the municipality and the frequency of the variable in question. Small municipalities and rare events have a larger random distribution. The municipalities’ indices can therefore vary between different measurement periods, especially among the small municipalities.

Breastfeeding at 4 months is not available at all as an indicator in the 1990 period. This has been taken into account in comparisons between periods; thus, the comparisons made between the periods cover only the five indicators that exist for both periods.

Presentation The results are presented in chapter 10 in tables that show the present situation and the situation ten years previously as well as the change over time for each individual municipality, for groups of municipalities, for those that rank highest and those that rank lowest.

The results are presented both for each of the six indicators and for the combined child health index. In this way we can form an understanding of each municipality’s current situation and compare that with the situation one decade earlier. It is also possible to compare one municipality with other municipalities, now and ten years earlier, and to make statistical calculations of different relationships, e.g. between different indicators and between the health index and the poverty index.

This ought to be interesting for both those who want to see the overall situation and those who are interested in detailed statistics. Small municipalities and infrequent events have a larger random distribution. The municipalities’ indices and ranks can thus show a great variation between different measurement periods, especially in small municipalities.

A number of questions remain unanswered, such as whether there is a concentration of particularly vulnerable children and families in a municipality. It is known that it is often the same children that are exposed to tobacco, have low birthweight and are not breastfed. May it perhaps be that these are often the same adolescents that are injured and who have abortions?

A deeper analysis and a detailed examination of the indicators in each municipality are beyond the scope of this report.. It is up to each municipality to examine its own situation and the reasons why it may have moved up or down in rank, what factors

53 have been decisive for its rank and, not least, what actions can be taken to make things better.

9. What is not available but should be and perhaps will be soon

The indicators presented in the final version are a compromise between what is desirable for the surveillance and monitoring of children’s health and wellbeing in the municipalities and what is possible on the basis of available data sources and resources. With further research and development, other indicators could be established that would give municipalities even better means for this task.

The greatest lack of valid, reliable and available data is in children’s mental health – despite the fact that this area is central to an individual’s wellbeing and society’s costs, and despite the fact that the problems in this area show the greatest increase and cause great concern among professionals, politicians and the general public. The reasons of course are that mental health, particularly among children, is a difficult phenomenon to capture, and we lack measurement methods to grasp all aspects of it at the same time. Many studies have been made of children’s mental health and behavioural problems and many methods have been tested. But this has not led to any systematic information on the mental health of all Swedish children or to data that can be broken down to the municipal level. There is broad agreement however that prevention is the most effective way to deal with the problem and that it should preferably be in children’s closest surroundings. It is primarily efforts made in the family, pre-school and school that can be expected to have the greatest influence on mental health (Bremberg 1998, Lagerberg 2002, Hwang 2002). Of particular interest is when co-ordinated efforts are made within a municipality (Berg Kelly et al. 1993).

Another issue in this complex of problems is social capital and its significance for children’s health and development. It has been argued in many publications during the past decade that beyond a certain base level, it is not absolute welfare in material terms but rather its distribution, relative welfare, which creates health and wellbeing in populations (Wilkinson 1996). The mechanisms have not been entirely clarified, but certain phenomena, the degree of social solidarity, the prevailing social norms and networks between citizens that constitute social capital, are considered to play an important role (Coleman 1988). In the investigation of Nordic children’s health between 1984 and 1996, referred to earlier, it was found that changes in health during this period co-varied positively with changes in social capital but negatively with changes in economic capital (Berntsson, Köhler & Vuille). There is growing interest in how social capital can influence, and indeed improve, human relationships, which argues for making children’s and families’ social networks a priority area in future analyses of children’s wellbeing (Vimpani 2000; Bing 2003).

As mentioned in an earlier chapter, there is now intensive work in several areas of the country to improve child health surveillance. Three initiatives in particular are expected to improve the situation considerably within a few years.

One initiative is the collaboration between the Swedish National Institute of Public Health, the Swedish Association of Local Authorities and the Federation of the

54 Swedish County Councils to create local welfare balance sheets, that focus on public health. A ‘municipal database’ has been developed with attention given to certain areas related to children (Folkhälsoinstitutet 2003, www.fhi.se). A growing number of municipalities have begun to work according to its guidelines. In February 2002, 29 municipalities had taken the first step, a formal decision to draw up a welfare balance sheet. In 49 municipalities the process was under way, and ten municipalities had actually drawn up a balance sheet.. When the system has been finalised and quality assessed, there will be several more indicators that can be used for children. The second initiative is a project, commissioned by the Government to the National Board of Health and Welfare’s Centre for Epidemiology. This project will develop and test a model for continuous surveillance of children’s and adolescents’ mental health (Hagquist 2004). It will be based on regular cross-sectional investigations every third year in schoolchildren 11-16 years of age (grades 6 and 9). The intention is to collect data in a complete investigation of all schoolchildren, which allows the possibility to disaggregate the results to specific municipalities. The questionnaires will be answered anonymously in classrooms and will contain items on wellbeing and mental ill health and on children’s conditions in the family, among friends and at school. A final proposal was submitted to the Government in 2004. It now awaits the Government’s decision to allocate funding.

The third is a proposal for a standard form for electronic child health centre records that has been developed to simplify the collection of data at health examinations, growth, screening investigations, counseling and vaccination at child health centres and in school health care (Socialstyrelsen, Memorandum, March 2002). This will allow systematised information to be gathered on individual children, will improve the quality of this information, and will make possible epidemiological surveillance of children’s health on both the local and national levels. One further source will give important information about children’s health. This deals with children’s education, which will be Save the Children’s third component in the creation of a municipal child index. The educational system – pre-school and compulsory school – are the most important public arenas as children grow up and have a great significance for their current and future health. For this reason it is obvious that indicators in these areas must be considered in a complete set of indicators, chiefly as determinants. The proposal for health indicators for children that was submitted to the EU Commission (Rigby & Köhler 2002) suggests: • proportion of children that leave school before the statutory school leaving age, • proportion of children enrolled in pre-primary education or kindergarten programme, • proportion of children attending schools with a written anti-bullying policy in operation.

Göteborg’s welfare balance sheet also uses proportion of children registered in pre- school and proportion of children with a passing grade in core subjects in the 9th grade (Välfärdsbokslut 2002).

For over ten years, the Council of Europe, EU and theWHO Regional Office for Europe have run the European Network for Health Promoting Schools. The goal is to develop all everyday aspects of the school to support and promote the physical and

55 psychosocial environment for health and learning and to strengthen and develop health education. Over 40 countries are cooperating in the project. and the Swedish National Institute of Public Health was initially the responsible authority in Sweden (Folkhälsoinstitutet 1997a). The Swedish National Agency for Education took over this responsiblity in 2002.

What is lacking and will be lacking in the immediate future is an investigation or any large scale project in Sweden on pre-school children’s mental health, behaviour, well-being and quality of life. These questions are extremely important, but methological problems and costs have so far been obstacles to country-wide surveys.

A project (KIDSCREEN) is being carried out in ten European countries to develop a screening instrument to survey the mental health and quality of life of schoolchildren ten years and older (Ravens-Sieberer et al. 2001). The instrument was translated into Swedish and used in an investigation among a random sample of children for validation and standardisation.

It is said that appetite comes with eating. When a reasonable set of indicators has been identified and tested in the municipalities and found to be operationally feasible, there will probably be a desire to have more, better and more detailed indicators to survey children’s health and wellbeing even more effectively. Thus we will surely want to identify children’s injuries in more detail so that we can start countermeasures, even for less serious injuries. While many municipalities have already started projects of this kind, we have far to go before they cover all parts of the country.

Similarly, we will probably want to keep closer watch on the physical environment and its impact on children’s health in the areas of air, water and noise pollution. Some municipalities, particularly those in our large cities, make detailed measurements of air quality. Data exist for relatively small geographical areas and their inhabitants, so called ecological methodology, and are often used to shed light on the effect of the environment on the population (Elliot et al. 1992).

Eating habits and nutrition are important components of the health behaviour of the growing numbers of overweight and obese children, in the long run also at increased risk of diabetes and cardiovascular disease. It is important to construct surveillance systems with appropriate reliable, sensitive and available indicators as a complement to the traditional system of weighing and measuring children at child health centres and in schools.

The offering of cultural experiences for children and participation in cultural events, opportunities for play, sports and other leisure activities are all areas that have not been possible to bring into this proposal. This is not because they are not important to children’s and young people’s health and wellbeing but because we do not have reliable methods and routines for measuring these activities and their effects.

Particularly vulnerable are those children with disabilities or other long term disorders. These conditions limit children in their normal function and activity and they increase psychosomatic problems. Children in this group are also exposed more to bullying; a Nordic study showed that these children had a twofold risk of being bullied over other

56 children (Nordhagen 2000). It is well documented and generally accepted that families with disabled children bear a greater burden and are more vulnerable than other families. The stresses can be physical, mental, social and economic. For these children and their families, childhood is often a very worrisome period with frequent hospitalisations and visits to doctors, complicated and difficult treatments, limitations in lifestyle, financial difficulties, and uncertainty and anxiety over the child’s future (Köhler & Jakobsson 1991). Even though more recent studies indicate that we have succeeded in improving conditions for these children and their families, they still bear a very large burden and much remains to be improved in terms of psychosocial support. It would thus be desirable to be able to include children with long term illnesses in a systematic health surveillance system and as a part of a health index – actions that are currently not possible. The proposal for measurements of children’s mental health, which will be a complete investigation of two school grades, may be able to contribute to this survey.

57 Results, conclusions and further work

10. Situation in the municipalities

As described in earlier chapters, the values of each indicator are put together to form one single index with a maximum and optimum value of 100. For the sake of simplicity, this means that the values can be expressed as percentages. All six indicators are included in the 2000 period – the mean value for three years around the year 2000 – while the mean value for the three years around 1990 includes only five indicators since breastfeeding data are not available for that time. The two periods are therefore not directly comparable (see below).

Children’s health index for individual municipalities and groups of municipalities in the 2000 and the 1990 periods The values in the 2000 period vary in the municipalities by a maximum of 9.5 percentage points. With a maximum possible value of 100, they go from 93.0 in Ekerö, a Stockholm suburban municipality, to 83.5 in Tanum, a small municipality in the northern part of the Westcoast. The mean value for all municipalities is 88.0 and the median is 88.8. The 20 municipalities that rank at the top and the bottom are shown in Table 3.

Table 3. Municipalities with the lowest and the highest indices, 2000 period (6 indicators)

Municipalities with the Municipalities with the lowest index highest index in the 2000 period in the 2000 period

Municipality Index Municipality Index Ekerö 93 Tanum 83.5 Täby 92.8 Degerfors 84 Vaxholm 92.6 Ljusnarsberg 84.6 Danderyd 92.6 Kumla 84.8 Hammarö 92.3 Älvkarleby 84.8 Lidingö 92.3 Storfors 84.8 Nacka 92.2 Östra Göinge 85 Sollentuna 92.2 Landskrona 85 Härjedalen 92.2 Mellerud 85.2 Solna 91.7 Grums 85.6 Munkfors 91.7 Bjuv 85.7 Berg 91.6 Sorsele 85.8 Järfälla 91.5 Söderhamn 85.8 Lomma 91.5 Bromölla 85.8 Ydre 91.5 Örkelljunga 85.9 Karlstad 91.3 Ödeshög 86 Lerum 91.3 Norberg 86.2 Åre 91.3 Markaryd 86.3 Emmaboda 91.3 Munkedal 86.3 Heby 91.3 Lilla Edet 86.4

58 Ten years earlier, in the 1990 period, the values varied between the municipalities by 7.9 percentage points, from 94.4 in Åsele to 86.5 in Jokkmokk. The mean value for all the municipalities was 90.3 and the median 90.4. The 20 municipalities that rank at the top and the bottom are shown in Table 4.

Table 4. Municipalities with the highest and lowest indices, 1990 period (5 indicators)

Municipalities with the highest Municipalities with the lowest index in the 1990 period index in 1990 period

Municipality Index Municipality Index Åsele 94.4 Jokkmokk 86 Lomma 93.3 Orsa 86.8 Danderyd 93.3 Åstorp 87.2 Essunga 93.2 Osby 87.6 Lycksele 93 Älvkarleby 87.6 Habo 92.9 Burlöv 87.6 Tranås 92.9 Södertälje 87.7 Strömstad 92.7 Svalöv 87.7 Sollentuna 92.6 Landskrona 87.9 Alvesta 92.5 Kävlinge 88 Krokom 92.4 Norberg 88 Uppsala 92.4 Malmö 88 Torsås 92.4 Sala 88.1 Örnsköldsvik 92.3 Bjuv 88.1 Lidingö 92.3 Kramfors 88.2 Sunne 92.2 Söderhamn 88.2 Nässjö 92.2 Kiruna 88.3 Lund 92.2 Sorsele 88.3 Falun 92.2 Grums 88.4 Karlstad 92.2 Norrtälje 88.4

The gap between the highest and lowest municipality increased somewhat, from 7.9 percentage points in the 1990 period to 9.5 in the 2000 period. This means that the addition of the sixth indicator, breastfeeding status, increased the deviation between the municipalities and also decreased the mean value. Tables 3 and 4 are not directly comparable, as they include a different number of indicators.

The municipal tables in Appendices 1 and 2 give the health index for all municipalities for both the 2000 and the 1990 periods. As discussed earlier, the municipalities differ a great deal among themselves in terms of number of inhabitants, business and industry profile etc. For this reason the municipalities are often divided into nine different groups (see chapter 8).

The large cities ranked somewhat worse than other groups of municipalities in the 1990 period (almost 3 percentage points). The suburban municipalities ranked highest, together with the rural municipalities. In the 2000 period, the large cities climbed in rank as a group, while the industrial municipalities are lowest and the suburban

59 municipalities again highest. The differences between the groups of municipalities are not great, however, neither in the 1990 period nor the 2000 period.

Table 5. Mean values of the children’s health index in the 1990 and 2000 periods according to groups of municipalities

2000(6 1990 (5 2000 (5 Difference 1990- indicators) indicators) indicators) 2000 (5 indicator)

Municipality Mean S.D. Mean S.D. Mean S.D. Mean S.D. group Urban (3) 89 1.9 89.2 1.1 93.1 1.1 4.0 0.5 Suburban 90.3 1.6 90.5 1.2 93.9 1 3.4 0.8 (36) Larger 89.5 1.3 91 2.2 93.6 0.9 2.6 0.8 cities (26) Mid size 88.7 1.2 90.1 1.1 93.1 0.9 3.0 1.1 cities (40) Industrial 88.2 1.8 90.2 1.1 93.2 1.1 3.0 1.3 (53) Semi-rural 88.4 1.8 90.3 1.3 93 1.1 2.7 1.3 (30) Rural (29) 89.2 1.7 90.5 1.8 93.3 1.3 2.8 1.8 Other 88.7 1.3 90.2 1.3 93.3 1 3.1 1.2 larger (31) Other 88.2 1.5 90.3 1.2 92.5 1.6 2.2 1.5 smaller (41) All muni- 88.8 1.7 90.3 1.3 93.2 1.2 2.9 1.3 cipalities (289)

When the values are grouped according to geographic region, the children’s health index is found to be highest, 90.4, in the Stockholm region and lowest, 87.0, in the Örebro region. Mean values of the children’s health index in the regions of Götaland, Svealand and Norrland show no differences, however. (Southern, Middle and Northern regions of the country).

Comparison between the periods The numbers must be corrected to be able to compare the two periods and judge changes in the health index between them because data on breastfeeding are lacking for the 1990 period. A correct comparison can thus only be made when breastfeeding data are removed from the 2000 period and the five indicators remaining for both periods are used.

In the comparison, consideration has also been given to the changes in the number of municipalities so that the five municipalities that did not exist in the 1990 period have been given an average value for their municipality groups.

60 It is found then that the index, after having been made comparable, has shown an improvement during these ten years, with an increase in the mean value from 90.3 to 93.2. All municipalities except Ljusnarsberg (-3.1 percentage points) and Åsele (-2.3 percentage points) improved their health index between the two measurement periods.

The greatest improvement, over 5 percentage points, took place in 11 municipalities (see table 3). The difference between the maximum and minimum was 7.9 percentage points in the 1990 period (from 94.4 to 86.5). In the comparable 2000 period (5 indicators), the difference between the municipalities’ maximum and minimum values was 9.2 (from 96.3 to 87.1), i.e. an increase of 1.3 percentage points.

Table 6. Municipalities that improved their children’s health index between the 1990 period and the 2000 period by more than 5 percentage points (5 indicators)

Municipality Improvement in percentage points

Jokkmokk 7.6 Norberg 7.2 Sala 6.3 Osby 5.6 Ragunda 5.6 Åstorp 5.5 Kävlinge 5.5 Munkfors 5.4 Kiruna 5.3 Robertsfors 5.3 Sigtuna 5.1

Seen together, there is a clear consistency between the periods, which means that those who ranked high in the 1990 period did so also in the 2000 period.

In terms of municipality group, the greatest improvement took place in the three large cities of Stockholm, Göteborg and Malmö, 4 percentage points, and the smallest change was seen among the 41 municipalities with less than 15 000 inhabitants.

Individual indicators in the index When the health index is broken down into its components it is seen that it is more difficult to achieve the goal in the case of some indicators than in others (see table 7).

Teenage abortions show the best average value; this is the least frequent negative health indicator. Less than two of 100 girls <18 years underwent an abortion. Immunisation and low birthweight also show good values. In the entire country, the low birthweight indicator is just below 96, i.e. only four of 100 newborns weigh less than 2 500 g. The Breastfeeding indicator can be improved most; an average of only two thirds of infants are breastfed at four months.

61 Table 7. Highest and lowest values in individual indicators, 2000 period and 1990 period

2000-period Indicator Municipality High Municipality Lowest Range Mean value value value Injuries Storfors 96.3 Ödershög 81.9 14.4 90.2 Tobacco Danderyd 97.7 Ljusnarsberg 68.7 29.0 86.1 Birth- Orsa 99.4 Ockelbo 91.4 8.0 95.7 weight Breast- Emmaboda 85.2 Storfors 41.1 44.1 66.8 fed Vaccina- Eda 99.8 Ljusnarsberg 81.6 18.2 95.8 tion Abortions Lomma 99.6 Bjurholm 95.8 3.8 98.3

1990-period Indicator Municipality High Municipality Lowest Range Mean value value value Injuries Ludvika 96.1 Sala 82.0 14.1 89.9 Tobacco Lycksele 88.2 Eda 59.3 28.9 73.4 Birth- Habo 99.5 Vaxholm 91.8 7.7 95.6 weight Breast- ______fed Vaccina- Kungsör 99.7 Gagnef 80.8 18.9 94.5 tion Abortions Olofström 99.6 Sundbyberg 96.3 3.3 98.3

The greatest variation in the municipalities is found in the breastfeeding indicator, where there is a difference of 44.1 percentage points between Storfors, 41.1, and Emmaboda, 85.2. There are also strong differences in individual municipalities in mothers’ smoking during pregnancy, from 97.7 in Danderyd to 68.7 in Ljusnarsberg (29 percentage points). The variation in teenage abortions is very small, however – only 3.8 percentage points – between Lomma, with 99.6, and Bjurholm, with 95.8.

Regardless of the varying weights of the indicators purely in terms of health, their mathematical values are also significant. Even if we consistently use average values to construct indicators and add them to an index and thus give them identical weights in all calculations, the variations between municipalities are important: the greater the variation an indicator shows between municipalities, the more it affects the comparison between the municipalities. Thus teenage abortions and low birthweight, for example, show little variation between the municipalities and therefore have little explanatory power in terms of the differences between the municipalities. The opposite is true for breastfeeding, with its strong variation.

The above reasoning applies in principle for both the 2000 and the 1990 periods. The tobacco indicator is alone in showing a strong change between the two measurement

62 periods. Only five municipalities demonstrate poorer values (Markaryd, Smedjebacken, Hallstahammar, Ljusnarsberg and Upplands Väsby). This health threat has decreased in all the other municipalities, in certain cases by 25-30 percentage points (Munkfors, Jokkmokk, Ragunda, Robertsfors, Hällefors, Norberg and Ydre).

Significance of the children’s health index Are these differences between the municipalities significant and are the results important to children and society? They are significant in the sense that there is a systematic difference. The correlation between the municipalities’ rankings in the 1990 and 2000 periods is strong (p < 0.01). The differences are notable, although not particularly large. The difference in the 2000 period was at most 9.5 percentage points and the deviation from the maximum was at most 16.5 percentage points. This can be compared with the difference between the municipalities in the poverty index presented earlier by Salonen.

There is a greater variation in the municipalities with regard to the “proportion of children who live in households that are neither poor nor draw means tested social assistance” – from 62.3 percent in Malmö to 94.5 percent in Nykvarn, , a range of 32 percentage points. In 1999, 38 municipalities had less than 80 percent while 17 municipalities showed over 90 percent. It must be kept in mind, however, that the index can be affected in small municipalities by temporary variations in individual indicators (see also chapter 8).

We group municipalities according to types, but the results are still given as an average for the municipalities included in each group. This is a considerable twisting of the truth, especially for the large cities, since there are marked differences between different areas within these cities, which often have sub-populations that represent both the most healthy and the most sick, the richest and the most impoverished in the country. However, we are not able at this time to break down the data to units smaller than the municipality. It must be a task for the municipalities themselves to investigate their populations in greater detail or for future special studies of the large cities.

The results are important because this kind of compilation of data on children’s health on the municipal level has not been done before. It gives municipalities a unique basis for further analysis of where they have succeeded and where there is room for improvement and where thus efforts should be made.

Health index and child poverty index The differences between the municipalities are significant because they show that health differences coincide with other differences in society. If we relate the health index to the poverty index, which was calculated for the same populations (Salonen 2002), we find a consistent association. There is a statistically significant correlation over all municipalities between the poverty index and the children’s health index on the 0.01 level in both the 1990 and the 2000 periods.

This is true even when the index is compared within each group of municipalities. However, there is no given relationship between a low children’s health index and child poverty in each individual municipality, just as it would not be true to say that all poor individuals are sick or that all rich individuals are healthy.

63 Individual indicators also show a statistically significant relationship with the child poverty index in the 2000 period. This concerns the indicators of breastfeeding (p < 0.02), tobacco use during pregnancy (p < 0.001) and teenage abortions (p < 0.001). These relationships are so strong that they have an impact on and give statistical significance to the index as a whole, even if the other indicators are not themselves significantly correlated to the child poverty index (injuries, low birthweight and vaccination). Again, the calculations are based on mean values and pooled data.

Conclusions It can thus be confirmed that, measured in this way, children’s health in Swedish municipalities is very good and that the differences in the health index are lower than the differences in the poverty index. However, there is still room for improvement. Using the goal of 100 percent, there is still a considerable number of children who do not achieve this optimal health.

Even if we were to be satisfied with the best municipality as a target for children’s health (benchmarking), there are many children in other muncipalities that do not reach this level. It is not possible to determine exactly how many children this means without more detailed analyses.

These calculations are difficult even when data are put together on the individual level because the indicators that together comprise the index cover different periods of the childhood years and thus different groups of children and in part also different denominators. Furthermore, the statistics on abortions are anonymised and cannot be connected to specific individuals.

11. Conclusions

This report must be seen in the light of the long term work of Save the Children to create an index based on rights that provides information about different aspects of the welfare of children and adolescents in Sweden on the municipal and national levels. Earlier reports have shed light on the economic dimension of children’s welfare. (See e.g. Child Poverty in Sweden – a contribution to a municipal children’s index and subsequent annual follow-up reports, all written by Professor Tapio Salonen for Save the Children).

The purpose of this report is to shed light in a systematic way on children’s health in Sweden and on methods that can be used to measure it. On the basis of the Convention on the Rights of the Child, children’s right to the highest attainable standard of health (article 24) has been placed in a Swedish perspective with a focus on the municipal level. This is the first time an attempt has been made to construct a municipal health index for children in a broad public health perspective. The results of this extensive review can be presented under three headings: gaps in knowledge, ongoing efforts and current results.

Gaps in knowledge The report clearly shows that there is a great deal that we do not know about children’s health. A great deal of the usefulness of the review presented in this report is that it shows the difficulties and shortcomings in measuring children’s health. The

64 municipalities currently have too few and too incomplete indicators. We thus lack information to systematically follow pre-school children’s mental health, behaviour, wellbeing and quality of life.

Information about children’s physical environment and its impact on their health is also very poor. Other areas in which there is a lack of reliable data are eating habits and nutrition and cultural offerings and experiences.

The conclusion that was drawn after an earlier review of children’s health in Sweden (Köhler & Jakobsson 1991) is essentially still true today: “One looks in vain for a systematic, continuous and comprehensive reporting of children’s health, viewed in a child perspective and related to a social context”

Ongoing efforts As stated, there is a gaping absence of available information with which to monitor children’s health. Several promising initiatives have however been taken to systematise the surveillance of the health and wellbeing of the population in Sweden, although few have a child perspective. Three initiatives particularly worthwhile mentioning are:

(i) Local welfare balance sheets. The Swedish Association of Local Authorities and Regions together with the Federation of Swedish County Councils and the Swedish National Institute of Public Health are the organisations that have taken this initiative. Not all municipalities participate and the set of indicators used differs from municipality to municipality. Some of the participating municipalities have focused on children’s welfare, however. Work is now continuing to develop a national database of municipal basic facts as a foundation for public health work. Some of the indicators chosen also cover children.

(ii) The National Board of Health and Welfare’s Centre for Epidemiology has presented a system for continuous surveillance of children’s and adolescents’ mental health. The plan is to perform repeated surveys of a few age-groups of all children, giving the municipalities access to the results. A decision to implement this proposal has not yet been made.

(iii) A third initiative is a proposed nationally standardised form for computerised health records from the Child Health Sevices and the School Health Services, which would facilitate the collection of important information.

Current results As is obvious from the previous discussions, it has not been easy to construct a relevant health index for children on the municipal level. In many cases data have been insufficient to describe the broad and complex area that children’s health represents. Of the 22 indicators originally proposed, six remain (children injured by external causes, foetal exposure to tobacco, children with low birthweight, children that are breastfed, MPR vaccination and teenage abortions. After careful examination only these six were found to fulfil the requirements that were set for quality and availability. In spite of shortcomings, on the basis of the analyses of the six indicators (for all of Sweden’s municipalitie) we can confirm that:

65

(i) Children’s health is good in Sweden. Between the two measurement periods, the 1990 and 2000 periods, the pooled health index increased from 90.3 to 93.2. During the same period, 11 municipalities improved their health index by more than five percentage points (pp) while the index fell in only two municipalities. There is a high national average for several individual indicators. For example, only two of 100 girls under 18 years of age undergo an abortion and slightly less than 4.5 percent of all newborns weigh less than 2 500 grams.

(ii) There are large differences between the municipalities. The difference between the municipality with the highest index value and that with the lowest value in the 2000 period was almost ten percentage points (pp). The differences between the municipalities in individual indicators are also large. The greatest variation is seen in the following indicators: breastfeeding (44 pp), foetal exposure to tobacco (29 pp), vaccination (18.2 pp) and injuries (14.4 pp). Moreover, comparison of the two periods shows that the deviation has increased, with the exception of the vaccination indicator.

(iii) As is almost always the case, there is a strong correlation between children’s health and their economic vulnerability. Obviously, this cannot always be demonstrated for individual municipalities, but there is a statistically significant correlation between the general poverty index and the general health index, both the 1990 period and the 2000 period. The same is true when the indices are compared within the different groups of municipalities.

The report does not offer recommendations for measures that municipalities and/or county councils should take to improve children’s health. It should be the responsibility of the individual municipalities and county councils to analyse in detail the data presented in municipality tables 1 and 2.

66 References

Antonovsky, A. (1987) Unrevelling the mystery of health. How people manage stress and stay well. Josey-Bass, San Francisco

Barnmiljöutredningen (1975) (Children. Report of the Children Environment Investigation) Liber förlag, Stockholm

Barnombudsmannen (2001). Med barnkonventionen som karta och kompass i kommuner och landsting.(Using the Convention of the Rights of the Child as map and compass in local work) Barnombudsmannen, Stockholm.

Barnsäkerhetsdelegationen (2002). Sociala skillnader i skaderisker. En rapport om den socioekonomiska fördelningen av skador bland barn och ungdomar i Sverige. (Social differences in injuries. A report on socio-economic distribution of injuries among children and adolescents) SOU. Fritzes, Stockholm

Ben-Arich, A. & Wintersberger, H. (1997). (Eds.) Monitoring and measuring the state of children - beyond survival. The European Centre for Social Welfare Policy and Research, Vienna.

Berg Kelly, K. (1993). Ungdomars hälsovanor och riskbeteenden. Stark kommunal satsning ger bättre hälsa. (Health behaviour in adolescence). Läkartidningen, 90710-717.

Berntsson, L. (2000). Health and well-being of children in the five Nordic countries in 1984 and 1996. Thesis. Nordic School of Public Health, Göteborg.

Berntsson, L., Köhler, L. & Vuille, J.-C. Health, economy and social capital in Nordic children and their families. A comparison between 1984 and 1996. Child:Care, Health and Development

Bing, V. (2003). Små, få och fattiga. Om barn och folkhälsa. (Small, few and poor. On children and Public Health.) Studentlitteratur, Lund.

Bolin A-K (1997). Children's dental health in Europe. An epidemiological investigation of 5- and 12 year-old children form eight EU countries. Thesis. Swedish Dental Journal, Suppl, 1-88.

Bremberg, S. (1998). Barnrapporten. Kunskapsbaserat folkhälsoarbete för barn och ungdom i Stockholms län. (Public Health work among children and adolescents in Stockholm County). Stockholms läns landsting, Stockholm.

Bremberg, S. (1999). Bättre hälsa för barn och ungdom. (Improved health for children and adolescents). Folkhälsoinstitutet, Stockholm.

Bremberg, S. (2002). Social inequities in health in Swedish children and adolescents – a review). National Institute of Public Health, Stockholm.

67 Brottsförebyggande rådet (2000). Barnmisshandel. En kartläggning av polisanmäld misshandel av små barn. (Child abuse reported to the police). Brottsförebyggande rådet, Stockholm

Brundtland, G.H. & Walløe, L. (1973). Menarchal age in Norway. Nature, 241, 478-479.

CAN (2002 a). Drogutvecklingen i Sverige. (Development of drug abuse in Sweden) Centralförbundet för alkohol- och narkotikauppplysning, Stockholm

CAN (2002 b). Skolelevers drogvanor 2002. (Drug abuse among schoolchildren 2002). Centralförbundet för alkohol- och narkotikaupplysning, Stockholm

Cernerud, L. (1991). Growth and social conditions. Height and weight of Stockholm schoolchildren in a public health context. Thesis. Nordic School of Public Health, Göteborg

Cnattingius, S., Ericson, A., Gunnarskog, J. & Källén, B. (1990). A quality study of a mediccal birth registry. Scandinavian Journal of Social Medicine, 18, 143-148.

Coleman, J. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, 95-120.

Dahlquist, G. & Mustonen, L. (2000). Analysis of 20 years of prospective registration of childhood onset diabetes - time trends and birth cohort effects. Acta Paediatrica, 89, 1231-1237.

Dalman, C. & Bremberg, S. (1999). Hur satsar vi på barnen? Välfärdsinsatser för barn i Stockholms län mätt i kronor.(Social support for children in Stockholm county). Stockholms läns landsting, Stockholm.

Danielsson M, Marklund U.(2000) Svenska skolbarns hälsovanor 1997/98. (Health behaviour of Swedish schoolchildren.) Folkhälsoinstitutet Stockholm de Vylder, S. (1998). Barn och barnfamiljer - 90-talets förlorare. Sverige efter välfärdskrisen - mellan hot och hopp (ed. B. Starrin, & B. Svensson). (Children and families – the losers of the 1990’s. In Sweden after the economic crisis – between threat and hope). Borea, Umeå.

Diderichsen, F., Eriksson, C., Hellman, L., Bosse, P. & Måns, R. (1998). Sjukdomsbördan i Sverige. En svensk DALY analys. (Burden of diseases in Sweden. A Swedish DALY analysis). Karolinska Institutet, Stockholm

ECHI (2001). Design for a set of European Community Health Indicators. European Commission, Brussels

Elliot, P., English, D. & Stern, R. (1992). (Eds.) Geographical and environmental epidemiology. Method for small area studies. Oxford University Press, Oxford.

Elmén, H. (1995). Child health in a Swedish city. Mortality and birthweight as indicators of health and social inequality. Thesis. Nordic School of Public Health, Göteborg

68 EURODIAB ACE Study Group (2000). Variation and trends in incidence of childhood diabetes in Europe. Lancet, 355, 873-876.

European Commission. (2000). Report of the state of young people's health in the European Union. European Commission, DG V, Brussels

European Environmental Agency & WHO regional Office for Europe (2002). Children's environmental health: review of the evidence. WHO, Copenhagen

Eurostat (2000). Key data on health 2000: Data 1985-1995. Eurostat, Luxembourg.

Folkhälsoinstitutet (1997). Hälsofrämjande skola. Ett samlat grepp för visionen om skolan som en stödjande miljö för hälsa och lärande. (Health Promoting School) Folkhälsoinstitutet, Stockholm

Folkhälsoinstitutet (1997). Determinants of the burden of disease in the European Community. National Institute of Public Health, Stockholm

Folkhälsoinstitutet (1999). Institute of Public Health (1999) Välfärdsbokslut, (Balance sheet of welfare). National Institute of Public Health, Stockholm

Folkhälsoinstitutet ( 2003). Kommunala basfakta som underlag för folkhälsoarbete. (Basic facts of the municipalities – foundation for public health work). National Institute of Public Health, Stockholm

Formgren, H. (1998). Allergisjukdomar hos barn och unga. I. Omfattning av allergi och annan överkänslighet. Vetenskaplig sammanställning. (Allergies in children and youth). Folkhälsoinstitutet, Stockholm

Forum on Child and Family Statistics (2002). America's children: Key national Indicators of Well-Being 2002. Federal Interagency Forum on Child and Family Stastistics, Washington, DC

Fugelli P, Ingstad B. (2001) Helse – slik folk ser det. (Health seen by ordinary people). Tidskr Nor Legeforen 121:3600-4.

Hagelin, E. (2001). Child health records as a data source. An analysis of completeness, reliability and accuracy of recording. In: Protection Prevention Promotion. Development and future of Child Health Services (ed. G. Norvenius, L. Köhler, G. Wennergren, & J. Johansson). Nordic School of Public Health, Göteborg.

Hagquist, C. (2004). Mätning av barns och ungdomars hälsa – förslag till nationella och återkommande undersökningar. (Measurement of mental helath among children and adolescents. Proposal for national and recurrent investigations). Socialstyrelsen/EpC, Stockholm.

Hall, D.M.B. (1996). Health for all children. Oxford University Press, Oxford.

Hanratty B, Holt T, Duffell E, Patterson W, Ramsay M, White JM, Jin L & P., L. (2000). UK measles outbreak in non-immune anthroposophic communities: the implications for the elimination of measles from Europe. Epidemiol Infect, 124, (2), 377- 383.

69 Hanson LÅ (2002). Breastfeeding, a complex support system for the offspring. Pediatr Int, 44, (4), 357-352.

Hasselgren M, Arne M, Lindahl A, Janson S & Lundbäck B (2001). Estimated prevalence of respiratory symptoms, asthma and chronic obstructive pulmonary disease realted to detection rate in primary health care. Scandinavian Journal of Primary Health Care, 19, 54-57.

Hjern, A., Allebeck, P. (1997). Health examinations and health services for asylum seekers in Sweden. Scandinavian Journal of Social Medicine, 25, 207-209.

Hjern, A., Angel, B. & Jeppson, O. (1998). Political violence, family stress and mental health of refugee children in exile. Scandinavian Journal of Social Medicine, 26, 18-25.

Hjern, A., Ringbäck-Weitoft, G. & Andersson R (2001). Sociodemographic risk factors for home-type injuries in Swedish infants and toddlers. Acta Paediatrica, 90, 61-68.

Hjälpmedelsinstitutet (2002). Så många barn har någon långvarig sjukdom eller något funktionshinder.(Statistics on children with disabilities) Statistiksammanställning, Stockholm

HMSO (1997). The Health of Young People 1995-97. Health Survey for England. HMSO, London

Hwang, P., Wickberg, B. (2002). Föräldrastöd och spädbarns psykiska hälsa. (Mental health of infants and support to parents). Statens Folkhälsoinstitut, Stockholm.

International Organization for Standardization, T.C. (2002). ISO/TS Health informatics - Health indicators conceptual framework. International Organization for Standardization, Technical Committee 215,

ISAAC (1998). The International Study of Asthma and Allergies in Childhood. Worldwide variations in the prevalence of of asthma symptoms. European Respiratory Journal, 12, 315-335.

Janlert, U. (2000) Folkhälsovetenskapligt lexikon. (Public Health Dictionary) Natur & Kultur, Stockholm

Janson, S. (2001). Barns och ungdomars hälsa. In: Folkhälsorapport 2001 (Health of children and adolescents. In Public Health Report 2001). Socialstyrelsen, Stockholm.

Janson, S. (2001). En rapport om kroppslig bestraffning och annan misshandel i Sverige i slutet av 1900-talet. (Physical punishment and other abuse in Sweden at the end of the 1990’s) Fritzes, Stockholm.

Johansson, S. (1973) The Swedish Level of Living Survey: a presentation. Acta Sociologica, 16, 211-219

Karolinska institutet (2002). Safe Communities, Stockholm (www.phs.ki.se/csp)

70 Köhler, L. (1990). (Ed.) Barn och barnfamiljer i Norden. (Children and families in the Nordic countries). Studentlitteratur, Lund.

Köhler, L., Jakobsson, G. (1991). Children’s health in Sweden. An overview for the 1991 Public Health Report) National Board of Health and Wellfare, Stockholm.

Köhler, L. (1998). Child Public health. A new basis for child health workers. European Journal of Public Health 82, 53-55

Köhler, L. (2000). (Ed.) Det är bra men kan bli ännu bättre. En studie av barns hälsa och välfärd i de fem nordiska länderna från 1984 till 1996. (It is well, but could be better. A study of children’s health and wellbeing in the five Nordic countries from 1984 to 1996). Nordic School of Public Health, Göteborg

Köhler, L., Sundelin, C. (1985). Hälsoundersökningar i förskoleåldern - praktisk verksamhet och vetenskaplig grund. (Health examinations of preschool children – practice and science). Läkartidningen, 82, 874-879.

Lagerberg, D. (2002). (Ed.) Measurement and monitoring of mental health in children and adolescents. Proceedings of a conference at Sätra Bruk. Allmänna Barnhuset, Stockholm.

Lagerberg, D., Mellbin, T., Claes, S. & Vuille, J.-C. (1994). Growing up in Uppsala: The "New morbidity" in the adolescent period. A longitudinal epidemiological study based on school data and some external sources. Acta Paediatrica Scandinavica, 83, Suppl 398.

Lindström, B. (1994). The Essence of Existence. On the Quality of Life of Children in the Nordic Countries. Thesis. Nordic School of Public Health, Göteborg

Lindström, B. (1995). Measuring and improving quality of life for children. In: Social Paediatrics (eds. B. Lindström, & N. Spencer), Oxford University Press, Oxford, New York.

Macfarlane, A. (1995). Outcome and performance measures in child health. In: Social Paediatrics (eds. B. Lindström, & N. Spencer), Oxford University Press, Oxford, New York.

Martin, J.F. (2002). Tobacco smoking as a form of child abuse. European Journal of Public Health, 12, 236-237.

McKeown, T. (1976) The role of medicine. – Dream, Mirage or Nemesis. Nuffield Provincial Trust, London

Medicinska forskningsrådet (1999). Barnhälsovårdens betydelse för barns hälsa - en analys av möjligheter och begränsningar i ett framtidsperspektiv. (The importance of Child Health Services – an analysis of opportunities and limitations). Medical Research Council , Stockholm

Meltzer, H., Gatward, R., Goodman, R. & Ford, T. (2000). Mental health of children and adolescents in Great Britain. Office for National Statistics, London.

71 Ministry of Health (1998). Our Children's Health. Key findings on the health of New Zealand children. Ministry of Health, Wellington

Morgenstern, H. (1982). Uses of ecological analysis in epidemiological research. American Journal of Public Health, 72, 1336-1344.

NASP (2002). Nyhetsbrev. NASP Nationellt centrum för suicidforskning och prevention av psykisk ohälsa (Newsletter. National centre for suicide research), Stockholm

Nationella folkhälsokommittén (2000). Hälsa på lika villkor- nationella mål för folkhälsan. Slutbetänkande. (Health and equity. National goals for the citizen’s health) . Ministry of Health , Stockholm.

Nordhagen, R. (2000). Mobbning och mobbare. (Bullying and being bullied). In: Det är bra men kan bli ännu bättre. En studie av barns hälsa och välfärd i de fem nordiska länderna från 1984 till 1996 (It is well, but could be better. A study of children’s health and weellbeing in the five Nordic countries from 1984 to 1996) (ed. L. Köhler). Nordic School of Public Health, Göteborg.

Nordhagen, R., Nystad, W. (2000). Idrott och motion hos barn. (Physical activities in children) I: Det är bra men kan bli ännu bättre. En studie av barns hälsa och välfärd i de fem nordiska länderna från 1984 till 1996 (It is well, but could be better. A study of children’s health and weellbeing in the five Nordic countries from 1984 to 1996) (ed. L. Köhler). Nordic School of Public Health, Göteborg.

Nygaard Christoffersen, M. (1999). Barn og ungdoms levekår i Norden. Sociale indikatorer på barns levekår. (Conditions of life for children and adolescents in the Nordic countries. Social indicators) Nordic Council of Ministers, Copenhagen.

OECD (2000). A system of health accounts. OECD, Paris

Olin, P. (2001). Svenska barns vaccinationsskydd. (Immunisation of Swedish children) Läkartidningen, 98, 3654-3657.

Olsson, G. (1998). Adolescent depression. Epidemiology, nosology, life stress and social network. Uppsala Universitet, Uppsala

Ottawa Conference on Health Promotion. (1986). Health Promotion – An International Journal vol 1, no 4

Pape, H. & Hammer, T. (1996). How does young people's alcohol consumption change during the transition to early adulthood? A longitudinal study of changes at aggregate and individual level. Addiction, 91, 1345-1357.

Qvortrup, J. (1994). (Ed.) Childhood matters: social theory, practice and policy. Avebury, Aldershot

Rasmussen, F., Johansson, M. & Hansen, H.-O. (1999). Trends in overweight and obesity among 18 years old males i Sweden between 1971 and 1995. Acta Paediatrica, 88, 431-437.

72 Ravens-Sieberer, U., von Rueden, U., Power, M., Rajmil, L., Cloetta, B., Erhardt, M. & Group, t.E.K. (2002). The KIDSCREEN project: results from the pilot test in 7 European countries. Quality of Life Research, Submitted.

Ravens-Sieberer, U., Gosch, A., Abel, T., Auquier, P., Bellach, B.-M., Bruil, J., Dur, W., Power, M., Rajmil, l.L. & Group, t.E.K. (2001). Quality of life in children and adolescents: a European public health perspective. Sozial- und Präventivmedizin, 46, 294- 302.

Rigby, M., Köhler, L. (2002). Child Health Indicators of Life and Development (CHILD). European Commission, Luxembourg

Rigby, M., Köhler, L., Blair, M. & Mechtler, R. (2002). Child Health indicators for Europe. A priority for a caring society. European Journal of Public Health, 13 (suppl 3):38- 42

Ringbäck Weitoft, G., Hjern, A., Haglund, B. & Rosén, M. (2002). Mortality, severe morbidity and injury among children of lone parents in Sweden. Lancet, 361, 289-285.

Salonen, T. (2002). Barns ekonomiska utsatthet under 1990-talet. Bidrag till ett kommunalt barnindex. (Child poverty in Sweden in the 1990’s. Contribution to a municipality index.) Save the Children, Stockholm.

SBU (2002). Att förebygga karies. En systematisk litteraturöversikt. (To prevent caries. A systematic literature review) SBU, Stockholm.

SCB (2002). Statistik om barns hälsa. Förslag,. (Statistics on children’s health. Proposal) Statistics Sweden. Stockholm

SCB/National Institute for Working Life (2002) Integration till svensk välfärd? Om invandrares välfärd på 90-talet. (Integration into Swedish welfare? On the welfare of immigrants in the 1990’s) Statistics Sweden, Stockholm

SCB (2003) Välfärd och ofärd på 90-talet (Welfare and misfortunes in the 90’s). Statistics Sweden, Stockholm

Schmidtke, A. (1996). Attempted suicides in Europe: rates, trends and sociodemographic characteristics of suicide attempters during the period 1989-1992. Results of the WHO/Euro Multicentre Study on Parasuicide. Acta Psychiatrica Scandinavica, 93, (5), 327-338.

Skolverket (1999). Nationella kvalitetsgranskningar 1999. (National quality assessments 1999) Skolverket, Stockholm

Smittskyddsinstitutet (2001). Vaccinationsstatistik från barnavårdscentralerna, (Vaccination statistics from Child Health Centres). Stockholm

Socialstyrelsen (1999). Insatser för barn och unga, (Contribution towards children and adolescents). Stockholm

73 Socialstyrelsen (1999). Socialtjänsten i Sverige 1999. Behov - Insatser - Utveckling. (Social services in Sweden. Needs –Actions – Development). Stockholm

Socialstyrelsen (2001a). (Yearbook of health services 2001). Socialstyrelsen, Stockholm.

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm.

Socialstyrelsen (2001) Den kommunala hälso- och sjukvårdens omfattning. (Health and social services in the municipalities). Stockholm

Socialstyrelsen (2002). Memorandum on the need for an electronic record for Child Health and School Health ServicesStockholm

Socialstyrelsen (2002a). Fakta om mammor, förlossningar och nyfödda barn. Medicinska födelseregistret 1973 till 2000. (Facts about mothers, deliveries and newborn infants. Medical Birth Register 1973-2000). Epidemiologiskt centrum, Stockholm

Socialstyrelsen (2002b). Amning av barn födda 2000. (Breastfeeding of children born in 2000). Stockholm

Socialstyrelsen (2002c). Rökvanor bland gravida och småbarnsföräldrar (Smoking among pregnant women and families with small children). Socialstyrelsen. Epidemiologiskt centrum, Stockholm

Socialstyrelsen (2002d). Aborter i Sverige 2002, januari - juni. (Abortions in Sweden 2002, January-June). Epidemiologiskt centrum, Stockholm

Socialstyrelsen (2003). Folkhälsa och sociala förhållanden 2002. (Public health and social conditions 2002). Socialstyrelsen, Stockholm.

SOU (1998). Det gäller livet. Stöd och vård till barn och ungdomar med psykiska problem. (Support and care for children and adolescents with mental problems). Fritzes, Stockholm.

SOU (1999). Barnombudsmannen - företrädare för barn och ungdom. (Ombudsman for children). Fritzes, Stockholm.

SOU (2000). Hälsa på lika villkor- nationella mål för folkhälsan. Slutbetänkande av Nationella folkhälsokommittén. (Equal opportunities for health. Final report from National Public Halth Committee ). Appendix B. SOU 2000:91. Socialdepartementet, Stockholm

SOU (2001a). Välfärdsbokslut för 1990-talet. Slutbetänkande. (Balance sheet for welfare. Final report) Fritzes, Stockholm.

SOU (2001b). Barns och ungdomars välfärd. Antologi. Kommittén Välfärdsbokslut. (Welfare of children and young people. An antology). The National Committee of Welfare Balance Sheet Fritzes, Stockholm.

74 SOU (2002). Sociala skillnader i skador hos barn och ungdom. Rapport från Barnsäkerhetsdelegationen. (Social differences in childhood injuries. Report from the Child Safety Delegation). Fritzes, Stockholm.

Spencer, N. (1996). Poverty and child health. Radcliffe Press, Oxford.

Sundhedsstyrelsen (1999). Indberetning på börne- og ungdomstandplejeområdet, København (Report on dental health of children and adolescents). Copenhagen

Sylwander, L. (2001). Child Accident Prevention. A Swedish success story. In: Protection Prevention Promotion. Development and future of Child Health Services (ed. G. Norvenius, L. Köhler,G. Wennergren, & J. Johansson). Nordic School of Public Health, Göteborg.

Tamburlini, G., Ronfani, L. & Buzzetti, R. (2001). Development of a child health indicator system in Italy. European Journal of Public Health, 11, 11-17.

Trost, A.-C. (1984). Abort och psykiska besvär. Abortion and mental problems). International Library, Västerås.

Uldall, P. (1986). Spaed- og småbørns almindelige sygelighed, (Morbidity of infants and toddlers) Institut for Socialmedicin. Københavns Universitet, Copenhagen van den Hof S, van den Kerkhof JH, ten Ham PB, van Binnendijk, R., Conyn- vanSpaendonk, M. & JE, v.S. (2001). Measles epidemic in the Netherlands 1999-2000. Ned Tijdschr Geneeskd, 145, (25), 29-33.

Välfärdsbokslut för Göteborgssamhället 2001.(Balance sheet for welfare in Göteborg) (2002). Stadskansliet, Göteborg.

Waters, E., Goldfield, S. & Hopkins, S. (2002). Indicators for child health, development and wellbeing. A systematic review of the literature and reecommendations for population monitoring. Centre for Community Health, Melbourne.

Wendt, L.-K., Hallonsten, A.-L. & Koch, G. (1999). Oral health in pre-school children living in Sweden. Swedish Dental Journal, 23, 17-25.

Wennergren, G. & Strannegård, I.-L. (2002). Asthma hospitalizations continue to decrease in schoolchildren but hospitalization rates for wheezing illnesses remain high in young children. Acta Paediatrica, 91, 1239-1245.

WHO. (1978) Primary health care. Report of the International Conference on Primary Health Care. Geneva: WHO

WHO (1985). Teenage Pregnancies in Developed Coutries 1985. WHO, Geneva.

WHO (1993). Targets for Health for All. The Health Policy for Europe. WHO, Copenhagen.

WHO (1998). Obesity - preventing and managing the global epidemic. Report of a WHO consultation. WHO, Geneva

75 WHO (1999). Health for All indicators for monitoring and evaluation of Health21. WHO Regional Office for Europe, Copenhagen.

WHO (1999). Health 21. Health for All in the 21st century. WHO/EURO, Copenhagen.

WHO (2000a). Health for All indicators for the new health policy in Europe. WHO Regional Office for Europe, Copenhagen.

WHO (2000b). World Health Report 2000. WHO, Geneva.

WHO (2000c). Health and Health Behaviour among Young People, Copenhagen

WHO (2001). Infant and young child nutrition. Resolution of the World Health Assembly WHA 54.2. WHO, Geneva

Wilkinson, R.G. (1996). Unhealthy societies: the afflictions of inequality. Routledge, London.

Vimpani, G. (2000). Child development and the civil society - Does social capital matter? Journal of Health and Social Behavior, 21, 44-47.

Wolfson, M. (1994). Social proprioception: Measurement, data and information from a population health perspective. In: Why are some people healty and others not? (eds. R. Evans, M. Barer, & T. Marmor). Aldine de Gruyter, New York.

World Bank (1993). Investing in health. World Development Report 1993. World Bank/Oxford University Press, Oxford

Vägverket (1999). Förslag till Regeringen om trafiksäkerhetsplan för 1999-2001. (Proposal for a national road safety plan for 199-2001). Vägverket, Borlänge

Vägverket (2002). Barn och närsamhälle. (Children and society) Vägverket, Borlänge

Yngve A, Kylberg E & Sjöström M (2001). Breastfeeding surveillance in the EU and EFTA: Recommendations adopted at the Breastfeeding Surveillance Conference. Breastfeeding Surveillance Conference, Stockholm

76 Municipality table 1

Individual indicators and children’s health index for the 2000 period (6 indicators). 100 is the highest value and 0 the lowest value. Municipalities in alphabetical order.

Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weight feeding nation

Ale 87.8 81.5 96.0 62.2 96.1 98.2 87.0 Alingsås 90.5 85.6 96.6 74.0 94.6 98.5 90.0 Alvesta 91.5 87.5 94.5 67.4 99.1 98.7 89.8 Aneby 86.6 84.7 95.6 59.2 95.3 98.8 86.7 Arboga 94.9 87.1 97.0 70.3 97.5 97.8 90.8 Arjeplog 88.6 86.3 94.4 74.7 81.8 98.2 87.3 Arvidsjaur 89.1 84.3 93.4 68.5 94.2 98.6 88.0 Arvika 93.9 82.8 96.0 61.6 95.8 98.8 88.2 Askersund 89.9 86.1 95.5 56.5 94.2 98.3 86.7 Avesta 91.0 85.5 96.0 70.5 99.1 98.3 90.1 Bengtsfors 89.8 82.4 96.7 61.8 97.5 98.5 87.8 Berg 87.1 92.1 96.4 80.9 94.7 98.2 91.6 Bjurholm 83.2 91.0 93.7 66.7 95.9 95.8 87.7 Bjuv 91.2 78.9 97.1 51.8 97.3 97.6 85.7 Boden 89.1 85.0 94.6 75.8 95.7 98.1 89.7 Bollebygd 90.0 88.2 95.8 73.1 98.9 98.0 90.7 Bollnäs 92.3 81.8 96.9 64.3 90.9 98.6 87.5 Borgholm 89.5 87.2 95.9 75.0 97.3 99.0 90.6 Borlänge 88.3 82.9 95.9 61.6 96.1 97.7 87.1 Borås 87.0 83.7 96.2 67.1 95.0 98.3 87.9 Botkyrka 93.6 85.4 94.3 62.3 95.6 98.0 88.2 Boxholm 85.3 82.5 98.0 65.5 96.1 98.6 87.7 Bromölla 86.0 78.7 95.4 62.4 94.1 97.9 85.8 Bräcke 91.2 87.3 95.5 63.4 97.7 97.4 88.8 Burlöv 91.8 82.0 93.8 57.8 95.7 98.1 86.5 Båstad 92.5 86.1 95.6 75.9 95.7 98.6 90.7 Dals-Ed 85.4 84.5 92.0 65.7 99.5 97.6 87.4 Danderyd 93.4 97.7 96.5 77.2 91.6 99.1 92.6 Degerfors 92.9 79.1 92.5 49.4 92.9 97.4 84.0 Dorotea 91.9 90.8 93.3 73.4 93.1 97.5 90.0 Eda 90.9 73.7 94.0 68.0 99.8 96.4 87.1 Ekerö 94.3 92.4 97.6 81.5 93.3 98.9 93.0 Eksjö 84.9 93.7 95.6 70.2 99.2 98.9 90.4 Emmaboda 87.0 84.9 94.3 85.2 98.3 98.0 91.3 Enköping 92.7 87.1 96.2 67.5 91.9 98.5 89.0

77

Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weight feeding nation

Eskilstuna 86.4 83.0 94.5 64.5 97.0 98.2 87.3 Eslöv 91.2 83.0 93.4 61.8 96.2 98.1 87.3 Essunga 89.4 87.2 96.7 68.1 97.5 98.7 89.9 Fagersta 94.5 82.9 95.0 60.7 98.1 96.2 87.9 Falkenberg 89.4 87.6 95.5 66.8 96.6 98.6 89.1 Falköping 88.5 87.7 95.8 62.2 97.2 98.3 88.3 Falun 88.2 89.5 96.2 71.3 97.0 98.9 90.2 Filipstad 90.8 82.7 97.1 67.4 98.4 97.8 89.0 Finspång 90.5 89.5 96.7 64.3 99.3 99.0 89.9 Flen 90.6 85.0 97.6 66.1 97.9 98.0 89.2 Forshaga 92.0 82.9 95.8 73.0 97.7 97.4 89.8 Färgelanda 91.6 82.3 94.0 58.6 97.4 98.4 87.0 Gagnef 87.3 87.3 96.5 67.9 89.7 98.5 87.9 Gislaved 91.4 87.9 96.3 65.6 95.0 98.9 89.2 Gnesta 89.9 79.9 96.6 66.5 92.0 98.2 87.2 Gnosjö 89.7 89.1 96.7 66.5 99.1 99.1 90.1 Gotland 86.7 85.1 94.7 72.5 95.4 98.2 88.8 Grums 91.3 81.8 97.1 58.0 88.4 97.0 85.6 Grästorp 88.4 89.7 93.4 72.6 90.4 97.7 88.7 Gullspång 90.5 84.5 94.8 59.0 97.3 98.3 87.4 Gällivare 90.1 85.5 95.0 68.0 97.1 98.3 89.0 Gävle 89.6 85.3 94.8 67.7 97.6 97.9 88.8 Göteborg 85.9 88.4 95.1 66.9 95.9 98.0 88.4 Götene 91.1 85.6 95.5 64.5 97.8 98.2 88.8 Habo 93.5 88.9 94.0 66.0 95.9 98.9 89.6 Hagfors 92.2 83.8 97.2 60.9 99.0 98.6 88.6 Hallsberg 89.4 87.2 96.3 59.1 94.3 98.3 87.4 Hallstahammar 95.5 84.4 94.6 63.7 98.3 98.7 89.2 Halmstad 89.2 85.7 94.8 66.2 98.4 99.0 88.9 Hammarö 91.1 94.2 95.1 76.7 98.5 98.2 92.3 Haninge 92.8 82.3 94.8 66.0 96.1 97.1 88.2 Haparanda 91.8 80.3 95.3 57.9 96.0 97.0 86.4 Heby 95.3 81.8 95.8 79.0 97.0 98.6 91.3 Hedemora 89.5 84.3 96.1 63.1 93.6 98.0 87.4 Helsingborg 90.2 85.2 94.9 68.5 95.4 98.0 88.7 Herrljunga 89.3 87.5 95.6 61.8 91.7 98.6 87.4 Hjo 87.4 90.9 95.4 62.9 99.4 98.6 89.1 Hofors 88.8 81.8 94.1 58.6 98.0 98.6 86.7 Huddinge 92.0 87.5 95.1 70.4 95.4 97.4 89.6 Hudiksvall 87.6 86.1 94.4 68.4 96.2 98.0 88.4 Hultsfred 89.6 87.2 95.9 71.9 98.3 99.3 90.3 Hylte 90.0 83.8 97.7 65.9 98.6 97.7 88.9 Håbo 93.6 86.8 95.8 72.4 93.9 98.0 90.1 Hällefors 92.1 88.2 94.1 64.9 93.2 97.0 88.2 Härjedalen 88.5 90.8 97.2 80.9 97.6 98.0 92.2

78

Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weight feeding nation

Härnösand 91.1 86.7 95.6 69.8 96.3 98.7 89.7 Härryda 85.3 92.0 95.3 75.3 97.2 98.4 90.6 Hässleholm 88.7 85.3 97.4 64.5 95.2 98.8 88.3 Höganäs 91.4 86.5 94.1 70.0 93.5 98.4 89.0 Högsby 89.5 85.8 97.5 60.5 98.2 98.0 88.2 Hörby 90.0 83.6 94.3 61.7 96.1 98.8 87.4 Höör 90.6 77.3 94.0 66.6 95.2 98.3 87.0 Jokkmokk 89.9 89.9 95.2 58.4 97.4 98.3 88.2 Järfälla 93.1 88.6 96.6 76.0 96.2 98.5 91.5 Jönköping 91.7 89.6 95.4 68.9 99.0 98.9 90.6 Kalix 90.6 84.5 94.9 61.9 93.1 98.4 87.2 Kalmar 87.6 86.9 95.4 73.4 98.0 98.6 90.0 Karlsborg 87.0 84.7 97.5 61.6 96.7 98.7 87.7 Karlshamn 92.7 87.1 94.3 62.8 98.7 98.9 89.1 Karlskoga 94.2 88.0 94.7 57.6 98.3 98.6 88.6 Karlskrona 95.0 87.7 95.1 68.7 98.2 98.3 90.5 Karlstad 90.8 88.2 96.1 76.7 97.7 98.6 91.3 Katrineholm 92.4 85.9 94.1 65.8 97.8 98.7 89.1 Kil 91.6 84.0 96.2 64.4 97.9 99.0 88.8 Kinda 88.8 86.4 97.0 68.9 97.9 98.9 89.7 Kiruna 91.8 88.0 94.2 60.5 95.4 98.5 88.1 Klippan 90.6 81.0 92.4 59.5 96.7 98.4 86.4 Kramfors 87.4 85.3 95.9 56.3 98.7 98.2 87.0 Kristianstad 87.8 86.4 97.0 62.8 98.1 98.5 88.4 Kristinehamn 95.6 84.5 93.3 67.8 99.3 97.5 89.6 Krokom 87.0 89.5 98.6 76.8 96.8 98.7 91.2 Kumla 89.0 86.4 95.4 42.2 98.0 98.1 84.8 Kungsbacka 89.7 91.1 95.8 70.6 91.7 98.7 89.6 Kungsör 95.5 85.3 97.7 67.4 98.4 97.9 90.4 Kungälv 90.4 90.6 96.7 70.6 93.0 98.4 89.9 Kävlinge 91.8 85.7 95.9 66.0 96.1 98.3 89.0 Köping 94.1 82.8 95.3 70.2 98.0 97.6 89.7 Laholm 89.4 83.8 92.7 59.6 97.1 99.1 86.9 Landskrona 91.4 76.7 94.6 53.4 96.0 98.1 85.0 Laxå 91.6 86.7 95.9 63.8 93.7 99.1 88.5 Lekeberg 89.0 87.1 92.4 66.7 92.4 98.4 87.7 Leksand 89.1 89.9 95.1 70.7 94.9 98.9 89.8 Lerum 89.5 89.9 96.3 78.2 96.0 98.2 91.3 Lessebo 89.9 84.9 96.6 57.8 97.8 97.9 87.5 Lidingö 93.0 95.8 97.0 77.9 91.2 98.7 92.3 Lidköping 92.1 90.6 96.3 65.8 99.1 98.6 90.4

79

Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weigt feeding nation

Lilla Edet 85.3 82.9 95.3 62.6 94.2 98.0 86.4 Lindesberg 90.6 83.3 97.0 54.1 96.1 98.7 86.6 Linköping 88.9 91.3 96.1 71.2 95.2 98.8 90.2 Ljungby 93.8 88.5 92.8 62.8 98.1 98.4 89.1 Ljusdal 90.0 83.6 96.8 71.1 96.8 98.4 89.5 Ljusnarsberg 91.4 68.7 97.3 72.2 81.6 96.6 84.6 Lomma 91.8 95.3 93.7 71.4 97.2 99.6 91.5 Ludvika 91.5 83.0 96.7 66.5 96.6 97.8 88.7 Luleå 91.0 87.3 96.6 73.5 95.0 98.6 90.3 Lund 91.4 92.2 96.0 72.8 92.5 98.5 90.6 Lycksele 88.3 95.2 98.0 66.8 91.6 98.8 89.8 Lysekil 91.8 85.5 96.9 60.2 98.3 98.9 88.6 Malmö 88.5 84.8 95.3 63.3 95.7 97.7 87.5 Malung 89.4 90.1 96.4 65.5 97.4 98.7 89.6 Malå 92.6 92.5 96.5 68.7 94.9 98.5 90.6 Mariestad 88.5 83.9 95.6 62.9 97.8 98.8 87.9 Mark 89.8 85.5 95.0 64.1 96.7 98.3 88.2 Markaryd 91.4 81.4 94.6 54.7 97.3 98.6 86.3 Mellerud 89.2 78.8 95.8 52.9 96.5 97.8 85.2 Mjölby 84.0 85.2 94.3 64.4 98.9 98.3 87.5 Mora 85.3 88.0 96.2 73.2 92.7 98.4 89.0 Motala 84.3 83.3 95.5 71.4 98.0 98.5 88.5 Mullsjö 93.8 86.4 98.6 66.1 98.1 99.0 90.3 Munkedal 89.1 76.2 95.8 63.3 95.5 97.7 86.3 Munkfors 92.6 93.9 95.0 72.0 98.5 97.9 91.7 Mölndal 86.5 91.2 95.3 70.2 96.8 97.9 89.6 Mönsterås 91.0 82.1 92.7 73.0 97.3 98.3 89.1 Mörbylånga 85.7 83.5 94.6 70.3 98.5 98.1 88.4 Nacka 93.4 91.1 95.7 81.2 93.9 98.2 92.2 Nora 91.3 85.2 95.6 66.8 98.4 98.4 89.3 Norberg 93.8 88.7 96.5 41.7 98.6 98.2 86.2 Nordanstig 84.3 79.8 96.2 67.5 96.1 98.4 87.1 Nordmaling 85.9 89.1 98.5 61.1 91.7 97.6 87.3 Norrköping 88.6 86.2 95.6 67.4 94.2 97.8 88.3 Norrtälje 92.4 82.8 96.6 66.7 92.9 97.6 88.1 Norsjö 90.0 92.1 96.1 51.6 92.3 98.3 86.7 Nybro 87.4 86.1 93.7 65.2 98.0 98.3 88.1 Nykvarn 92.1 82.7 93.4 73.2 87.5 97.6 87.7 Nyköping 89.4 85.4 96.1 68.2 98.5 98.2 89.3 Nynäshamn 92.4 82.6 95.8 67.7 94.5 96.8 88.3 Nässjö 89.4 88.1 93.8 65.5 98.1 98.9 89.0

80

Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weight feeding nation

Ockelbo 92.9 88.5 91.4 64.2 99.1 98.5 89.1 Olofström 92.8 85.6 93.9 60.4 98.8 99.1 88.4 Orsa 88.4 77.0 99.4 69.0 90.1 98.5 87.1 Orust 89.2 86.1 97.4 72.3 89.1 98.7 88.8 Osby 86.8 86.0 97.2 65.0 97.2 98.9 88.5 Oskarshamn 91.5 86.3 96.2 70.8 98.1 98.6 90.2 Ovanåker 92.8 87.7 96.9 63.4 95.4 98.5 89.1 Oxelösund 90.2 83.8 95.8 76.0 99.3 96.9 90.3 Pajala 95.4 84.9 93.7 68.0 95.9 98.7 89.4 Partille 85.6 89.9 95.4 70.7 97.7 98.0 89.6 Perstorp 89.5 83.4 95.3 65.1 96.6 97.7 87.9 Piteå 91.6 86.3 96.3 71.0 96.4 98.5 90.0 Ragunda 85.6 91.9 97.6 74.5 97.3 97.8 90.8 Robertsfors 87.6 94.8 93.1 61.3 96.1 98.9 88.6 Ronneby 86.1 95.1 63.7 97.8 98.8 89.4 94.9 Rättvik 87.0 93.8 78.2 92.3 98.6 89.7 88.3 Sala 85.6 94.9 66.1 97.6 98.2 89.7 95.9 Salem 89.1 95.2 73.1 95.7 97.2 90.4 92.1 Sandviken 85.0 93.9 62.2 98.6 98.2 88.3 91.9 Sigtuna 93.6 86.7 95.4 68.1 95.8 98.4 89.7 Simrishamn 91.8 80.6 96.2 63.6 94.2 98.6 87.5 Sjöbo 92.0 78.3 94.9 66.6 93.3 97.8 87.1 Skara 88.3 88.3 94.5 62.4 98.8 98.4 88.5 Skellefteå 89.1 90.4 95.9 64.7 96.1 98.9 89.2 Skinnskatteberg 93.9 78.8 96.0 74.6 94.3 98.0 89.3 Skurup 92.4 79.5 95.1 68.8 97.4 98.0 88.5 Skövde 88.1 88.2 96.2 68.2 98.3 98.2 89.5 Smedjebacken 91.5 84.0 94.0 61.7 96.8 98.1 87.7 Sollefteå 85.5 86.7 96.5 63.0 96.3 98.3 87.7 Sollentuna 93.5 92.7 96.2 76.1 95.7 98.9 92.2 Solna 94.0 91.4 96.3 76.1 95.1 97.7 91.7 Sorsele 87.8 83.3 98.9 57.0 90.3 97.6 85.8 Sotenäs 88.7 82.7 97.0 57.7 93.9 98.6 86.4 Staffanstorp 91.3 89.0 94.6 63.9 96.5 98.8 89.0 Stenungsund 89.2 87.9 96.8 63.3 85.5 98.1 86.8 Stockholm 93.7 90.8 95.7 75.4 93.5 98.1 91.2 Storfors 96.3 82.4 92.0 41.1 99.4 97.4 84.8 Storuman 93.0 94.5 95.5 70.9 95.8 97.7 91.2 Strängnäs 89.0 83.1 94.9 70.3 95.4 98.5 88.5 Strömstad 89.1 85.5 96.3 65.6 95.3 99.0 88.5 Strömsund 88.9 91.1 96.6 65.3 98.2 97.8 89.7 Sundbyberg 93.7 89.0 95.5 70.8 95.3 98.0 90.4 Sundsvall 86.8 87.4 96.4 70.0 98.5 97.9 89.5 Sunne 91.7 86.4 93.4 65.8 95.7 97.7 88.4 Surahammar 94.4 82.0 96.3 56.2 96.9 96.7 87.1

81

Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weight feeding nation

Svalöv 91.1 79.9 94.9 58.0 97.3 98.2 86.6 Svedala 90.0 82.2 96.6 67.3 91.9 98.8 87.8 Svenljunga 89.2 91.8 95.9 63.2 99.8 98.0 88.0 Säffle 89.9 83.2 95.0 72.8 97.5 97.3 89.3 Säter 88.6 83.0 95.5 73.0 96.0 98.5 89.1 Sävsjö 85.7 97.4 65.7 97.5 98.8 88.9 88.6 Söderhamn 94.0 78.5 96.5 58.2 88.4 99.0 85.8 Söderköping 88.3 87.0 96.0 78.9 97.4 98.5 91.0 Södertälje 93.3 86.8 94.6 63.2 85.8 98.4 87.0 Sölvesborg 93.6 85.2 97.5 61.8 97.2 98.9 89.0 Tanum 89.3 79.3 95.7 42.1 95.7 98.6 83.5 Tibro 87.0 88.1 95.6 62.5 98.9 98.3 88.4 Tidaholm 88.6 86.9 85.2 62.7 97.7 98.1 88.2 Tierp 94.2 86.1 95.0 64.4 93.9 98.9 88.7 Timrå 86.1 81.4 96.3 64.4 97.8 97.9 87.3 Tingsryd 90.2 82.7 95.2 61.8 96.5 98.9 97.6 Tjörn 89.9 87.3 95.7 66.3 90.4 98.4 88.0 Tomelilla 91.6 82.3 95.0 64.3 96.6 98.7 88.2 Torsby 92.1 84.8 96.8 62.3 97.6 97.6 88.5 Torsås 90.5 81.7 94.9 68.6 96.7 98.9 88.5 Tranemo 88.4 85.8 97.2 70.4 95.7 99.1 89.4 Tranås 89.0 90.4 95.8 70.7 99.0 98.9 90.6 Trelleborg 91.1 81.7 94.9 61.3 97.7 98.5 87.5 Trollhättan 86.0 84.2 95.0 66.8 98.3 98.2 88.1 Trosa 89.1 83.6 95.6 70.0 92.0 98.6 88.2 Tyresö 93.5 88.7 95.2 77.1 95.8 96.8 91.2 Täby 93.5 95.9 96.2 81.4 90.8 98.9 92.8 Töreboda 88.5 83.9 98.8 65.2 97.7 98.6 88.8 Uddevalla 86.7 86.9 95.5 64.2 95.4 98.0 87.8 Ulricehamn 89.5 87.0 97.3 69.3 98.0 98.9 90.0 Umeå 86.4 94.3 95.5 74.3 95.0 98.6 90.7 Upplands-Bro 94.4 85.5 96.2 65.6 95.3 98.1 89.2 Upplands- 93.4 84.8 95.3 66.9 96.1 98.2 89.1 Väsby Uppsala 95.1 91.9 95.6 74.1 91.1 98.8 91.1 Uppvidinge 89.2 86.2 95.1 61.3 95.6 98.1 87.6 Vadstena 84.7 88.0 96.8 71.6 96.3 97.8 89.2 Vaggeryd 92.4 90.7 94.4 71.1 96.1 99.3 90.6 Valdemarsvik 88.8 88.0 98.0 70.1 95.6 97.9 89.7 Vallentuna 93.4 92.0 96.6 74.1 91.0 97.9 90.8 Vansbro 90.4 87.3 97.1 71.5 97.3 98.7 90.4 Vara 90.5 85.3 95.7 69.1 98.6 99.2 89.7 Varberg 85.3 89.3 95.1 67.8 97.0 98.9 88.9 Vaxholm 93.5 94.5 96.0 82.7 90.6 98.3 92.6 Vellinge 89.3 91.6 96.2 69.7 93.3 99.1 89.9 Vetlanda 88.4 90.6 95.0 70.2 98.4 99.0 90.3

82 Municipality Injuries Tobacco Birth- Breast- Vacci- Abortion Index weight feeding nation

Vilhelmina 88.0 96.2 74.4 92.8 98.7 89.7 88.1 Vimmerby 91.4 88.8 95.1 71.7 98.2 98.7 90.6 Vindeln 85.2 89.6 96.5 79.7 94.0 99.0 90.7 Vingåker 89.3 78.5 94.0 69.8 98.8 98.8 88.2 Vårgårda 89.7 80.5 95.0 64.7 96.8 99.2 87.7 Vänersborg 86.1 82.9 96.1 63.2 98.5 97.9 87.5 Vännäs 85.3 93.7 95.7 63.8 95.3 99.2 88.8 Värmdö 93.6 86.9 96.1 72.0 94.0 97.1 90.0 Värnamo 90.9 89.5 95.7 71.6 97.2 98.7 90.6 Västervik 86.9 85.6 93.9 75.3 98.8 98.2 89.8 Västerås 95.1 87.0 95.1 66.8 97.0 98.4 89.9 Växjö 91.0 92.2 94.9 69.9 97.2 98.0 90.6 Ydre 89.9 95.1 98.2 67.4 99.4 98.9 91.5 Ystad 92.2 87.6 94.5 69.2 96.0 98.7 89.7 Åmål 91.4 83.6 98.3 61.4 96.9 98.7 88.4 Ånge 85.5 87.3 97.0 66.3 98.9 97.2 88.7 Åre 89.1 91.3 96.9 79.2 93.4 98.1 91.3 Årjäng 92.0 84.1 95.3 75.8 98.2 99.1 90.7 Åsele 88.6 88.4 97.9 61.9 89.2 96.3 87.1 Åstorp 90.8 80.7 95.0 56.3 98.4 98.6 86.6 Åtvidaberg 90.9 90.0 95.0 59.3 96.4 98.3 88.3 Älmhult 93.3 91.3 95.4 71.5 96.9 99.0 91.2 Älvdalen 89.0 81.2 98.0 67.2 95.5 98.5 88.2 Älvkarleby 89.4 73.5 95.0 65.0 89.5 96.6 84.8 Älvsbyn 89.0 87.0 96.5 65.9 93.9 98.3 88.4 Ängelholm 91.7 87.2 95.6 71.9 95.1 98.7 90.0 Öckerö 87.6 90.9 97.4 79.5 91.4 97.4 90.7 Ödeshög 81.9 84.5 93.3 59.2 98.7 98.4 86.0 Örebro 87.7 88.1 95.4 60.2 93.4 98.1 87.2 Örkelljunga 91.1 75.6 97.8 55.3 97.2 98.5 85.9 Örnsköldsvik 87.9 91.9 96.5 69.7 98.1 98.8 90.5 Östersund 87.5 91.1 95.1 74.9 95.5 98.4 90.4 Österåker 94.5 87.9 97.6 80.2 88.7 98.3 91.2 Östhammar 95.2 86.1 96.2 68.1 90.8 98.3 89.1 Östra Göinge 84.5 79.3 98.1 54.3 95.4 98.7 85.0 Överkalix 94.5 86.7 97.8 58.0 90.3 98.6 87.6 Övertorneå 93.1 85.5 98.7 71.9 93.7 98.1 90.2

83 Municipality table 2

Individual indicators and children’s health index for the 1990 period (5 indicators). 100 is the highest value and 0 the lowest value. Municipalities in alphabetical order.

Municipality Injuries Tobacco Birt- Vacci- Abortion Index weight nation

Ale 88.5 71.9 96.3 95.7 98.0 90.1 Alingsås 87.4 75.5 96.0 94.6 98.9 90.5 Alvesta 92.2 79.5 96.4 96.0 98.5 92.5 Aneby 86.4 79.3 98.5 93.2 98.0 91.1 Arboga 90.6 75.3 95.0 94.8 98.6 90.9 Arjeplog 92.4 66.2 96.7 90.0 98.4 88.5 Arvidsjaur 91.6 72.0 96.2 86.8 97.9 88.9 Arvika 98.1 74.0 95.4 89.8 98.3 89.9 Askersund 94.6 74.9 95.2 94.5 99.3 91.7 Avesta 93.4 75.3 96.3 92.8 97.3 91.0 Bengtsfors 89.0 69.4 95.5 94.7 97.2 89.2 Berg 89.6 74.3 96.9 97.0 97.9 91.1 Bjurholm 86.3 81.0 97.9 96.7 97.1 91.8 Bjuv 87.5 63.6 95.0 96.5 98.0 88.1 Boden 88.0 71.7 95.5 95.4 97.9 89.7 Bollebygd ------Bollnäs 88.5 70.2 96.3 94.9 98.7 89.7 Borgholm 90.3 78.3 96.7 96.8 98.7 92.2 Borlänge 94.8 72.7 96.3 94.0 97.8 91.1 Borås 93.5 72.8 95.4 94.4 98.1 90.8 Botkyrka 90.2 71.7 94.4 93.0 97.7 89.4 Boxholm 87.4 80.1 95.3 97.3 97.7 91.6 Bromölla 84.8 72.6 95.7 94.5 98.7 89.2 Bräcke 90.6 70.1 96.1 97.0 97.4 90.2 Burlöv 89.8 62.9 95.3 90.8 99.1 87.6 Båstad 89.2 77.7 96.5 95.3 98.9 91.5 Dals-Ed 90.2 80.7 95.1 85.7 97.7 89.9 Danderyd 92.1 85.2 96.1 94.1 98.9 93.3 Degerfors 90.0 66.4 94.3 94.5 98.2 88.7 Dorotea 89.8 74.8 95.9 97.1 98.3 91.2 Eda 93.1 59.3 95.9 95.5 98.4 88.4 Ekerö 91.2 77.6 97.8 90.5 98.2 91.1 Eksjö 86.1 79.7 94.8 95.9 98.9 91.1 Emmaboda 87.8 72.6 93.8 95.7 97.7 89.5 Enköping 89.5 71.5 96.1 94.7 98.8 90.1 Eskilstuna 86.8 69.9 94.4 97.6 97.8 89.3 Eslöv 90.8 66.6 94.3 92.2 98.6 88.5 Essunga 89.8 82.7 96.7 98.9 98.1 93.2 Fagersta 89.5 68.8 94.7 97.1 97.8 89.5

84

Municipality Injuries Tobacco Birth- Vacci- Abortion Index weight nation

Falkenberg 91.4 73.6 95.3 95.3 98.4 90.8 Falköping 85.7 74.3 95.9 96.1 98.8 90.2 Falun 93.4 78.8 95.9 94.8 98.0 92.2 Filipstad 90.6 71.4 95.4 97.9 98.1 90.7 Finspång 88.8 73.8 96.8 97.9 98.4 91.1 Flen 91.3 67.4 94.4 97.0 97.8 89.6 Forshaga 93.7 75.3 96.8 96.4 98.2 92.1 Färgelanda 87.7 74.4 95.5 96.6 96.9 90.2 Gagnef 93.0 78.1 97.8 80.8 98.2 89.6 Gislaved 89.3 73.4 96.2 92.3 98.8 90.0 Gnesta ------Gnosjö 88.6 76.0 95.6 96.0 97.9 90.8 Gotland 87.0 68.2 95.6 95.6 98.0 88.9 Grums 89.5 65.3 95.9 92.5 98.9 88.4 Grästorp 87.1 72.7 94.0 94.6 98.2 89.3 Gullspång 90.4 74.2 98.1 93.3 98.9 91.0 Gällivare 87.4 69.9 96.0 94.8 96.9 89.0 Gävle 88.1 71.2 95.3 98.1 97.6 90.1 Göteborg 85.2 73.3 95.4 94.9 97.6 89.3 Götene 87.4 72.8 95.2 97.9 99.1 90.5 Habo 89.6 76.0 99.5 97.2 99.2 92.9 Hagfors 90.3 66.5 94.7 96.4 98.6 89.3 Hallsberg 94.5 75.8 96.4 94.5 98.3 91.9 Hallstahammar 94.5 68.4 95.9 97.1 98.6 90.9 Halmstad 92.8 73.2 95.3 98.0 98.8 91.6 Hammarö 91.6 77.9 95.6 97.3 97.8 92.0 Haninge 90.8 69.7 95.9 93.6 97.2 89.4 Haparanda 91.8 69.0 94.9 95.5 98.5 90.0 Heby 83.6 71.6 95.5 97.9 99.0 89.5 Hedemora 94.1 70.0 96.9 93.2 98.2 90.5 Helsingborg 87.6 70.2 95.1 92.6 97.9 88.7 Herrljunga 91.5 73.6 94.8 86.6 99.0 89.1 Hjo 88.5 73.9 95.5 91.6 97.4 89.4 Hofors 91.2 74.7 95.9 91.6 98.7 90.4 Huddinge 90.7 71.9 94.7 94.7 97.2 89.8 Hudiksvall 82.1 74.2 96.6 92.9 97.2 88.6 Hultsfred 90.6 73.2 95.6 98.3 99.0 91.3 Hylte 92.6 72.4 93.0 97.0 99.3 90.8 Håbo 91.9 67.8 94.6 92.9 98.4 89.1 Hällefors 93.2 65.9 96.2 94.5 97.7 89.5 Härjedalen 90.2 72.2 94.5 96.0 98.5 90.3 Härnösand 88.0 73.2 94.8 93.3 98.5 89.6 Härryda 90.7 73.9 96.1 96.2 98.3 91.0 Hässleholm 87.0 72.0 95.5 94.6 99.0 89.6 Höganäs 89.8 69.3 95.4 90.7 98.4 88.7 Högsby 90.2 67.7 96.9 94.7 98.0 89.5

85 Municipality Injuries Tobacco Birth- Vacci- Abortion Index weight nation

Hörby 90.3 67.4 96.4 92.4 98.4 89.2 Höör 90.8 74.7 95.2 95.2 98.7 90.9 Jokkmokk 88.4 61.6 93.9 91.0 97.9 86.5 Järfälla 91.8 74.6 96.5 96.1 97.8 91.4 Jönköping 87.4 78.7 95.6 97.4 98.7 91.6 Kalix 87.6 69.5 96.3 94.8 98.3 89.3 Kalmar 88.8 74.3 94.0 97.6 98.7 90.7 Karlsborg 90.6 75.5 95.8 96.2 99.2 91.5 Karlshamn 92.5 72.7 94.9 93.3 99.2 90.5 Karlskoga 89.4 73.6 95.9 94.5 97.9 90.3 Karlskrona 91.5 76.1 95.0 96.7 98.1 91.5 Karlstad 92.2 76.8 95.9 97.9 98.0 92.2 Katrineholm 92.3 73.9 95.6 97.7 98.5 91.6 Kil 91.5 74.5 96.4 96.6 97.8 91.4 Kinda 91.1 77.8 95.5 97.2 98.0 91.9 Kiruna 87.8 67.0 95.2 93.7 97.7 88.3 Klippan 89.3 67.1 94.4 95.1 98.5 88.9 Kramfors 85.2 69.4 96.0 92.5 98.0 88.2 Kristianstad 86.1 75.6 95.3 96.2 98.5 90.1 Kristinehamn 87.6 69.5 95.1 98.8 97.7 89.7 Krokom 92.3 79.9 94.8 96.0 99.1 92.4 Kumla 93.6 73.3 96.4 94.5 98.9 91.3 Kungsbacka 90.2 78.1 95.1 95.5 98.2 91.4 Kungsör 90.4 72.6 95.2 99.7 98.5 91.3 Kungälv 87.9 76.2 96.5 93.1 98.2 90.5 Kävlinge 89.6 67.5 95.6 89.5 97.9 88.0 Köping 88.9 68.8 95.9 98.0 97.9 89.9 Laholm 93.4 73.0 95.6 96.7 99.0 91.5 Landskrona 89.3 62.4 95.0 95.2 97.9 87.9 Laxå 90.3 74.3 96.4 94.5 97.5 90.6 Lekeberg ------Leksand 92.9 82.5 96.3 89.6 99.1 92.1 Lerum 88.6 79.4 95.0 94.1 98.7 91.1 Lessebo 88.9 74.0 96.5 95.4 97.9 90.5 Lidingö 91.4 82.5 94.5 94.4 98.7 92.3 Lidköping 86.9 78.8 95.8 95.9 98.9 91.3 Lilla Edet 86.3 70.4 95.7 94.3 97.4 88.8 Lindesberg 91.7 74.9 95.9 94.5 98.5 91.1 Linköping 89.3 77.7 94.7 95.8 98.5 91.2 Ljungby 90.2 74.9 95.2 97.1 98.8 91.3 Ljusdal 88.9 70.4 95.4 93.2 98.3 89.2 Ljusnarsberg 92.3 70.0 95.6 94.5 98.8 90.3 Lomma 92.9 84.0 96.6 94.6 98.4 93.3 Ludvika 96.1 67.4 96.0 91.5 97.9 89.8 Luleå 90.4 74.1 95.8 94.8 97.4 90.5 Lund 91.2 83.2 95.7 92.5 98.3 92.2

86

Municipality Injuries Tobacco Birth- Vacci- Abortion Index weight nation

Lycksele 86.2 88.2 97.1 94.6 99.1 93.0 Lysekil 90.1 74.1 95.5 97.5 99.0 91.2 Malmö 87.9 69.1 94.0 91.7 97.2 88.0 Malung 92.2 75.1 96.0 95.1 97.6 91.2 Malå 90.5 81.8 94.9 95.2 98.2 92.1 Mariestad 89.4 76.6 95.7 98.2 98.2 91.6 Mark 87.6 75.9 95.2 94.1 98.2 90.2 Markaryd 92.2 84.2 94.4 88.9 98.8 91.7 Mellerud 91.0 78.0 95.4 91.9 98.1 90.9 Mjölby 89.9 74.3 95.4 97.8 98.4 91.2 Mora 90.9 77.1 95.3 88.4 98.3 90.0 Motala 85.9 71.8 94.5 98.2 98.4 89.7 Mullsjö 88.7 73.1 95.4 94.7 99.2 90.2 Munkedal 84.9 69.5 96.1 93.9 98.4 88.6 Munkfors 92.0 65.2 98.8 98.2 96.9 90.2 Mölndal 91.1 76.6 96.0 96.6 98.4 91.7 Mönsterås 90.5 75.1 95.2 97.5 99.4 91.5 Mörbylånga 86.5 70.5 95.9 97.0 98.6 89.7 Nacka 91.5 75.3 95.8 92.9 97.5 90.6 Nora 91.0 74.7 94.2 94.5 98.1 90.5 Norberg 88.1 66.5 95.5 92.3 97.5 88.0 Nordanstig 83.6 73.1 95.0 93.3 97.7 88.5 Nordmaling 87.9 77.4 94.8 94.4 98.3 90.6 Norrköping 92.1 71.9 95.2 94.2 97.9 90.3 Norrtälje 88.5 67.7 95.4 92.6 97.8 88.4 Norsjö 91.0 78.7 98.9 92.0 98.9 91.9 Nybro 88.1 71.2 93.8 95.3 98.8 89.4 Nykvarn ------Nyköping 95.4 73.4 95.4 95.9 98.2 91.6 Nynäshamn 90.8 69.8 95.3 91.3 97.7 89.0 Nässjö 89.9 79.1 96.7 96.8 98.4 92.2 Ockelbo 91.2 71.9 95.1 97.9 96.8 90.6 Olofström 93.4 74.6 95.7 90.3 99.6 90.7 Orsa 88.3 68.1 94.8 85.4 97.6 86.8 Orust 87.7 75.8 96.9 95.5 98.4 90.8 Osby 84.9 69.4 95.4 90.4 98.1 87.6 Oskarshamn 88.1 77.4 93.3 97.9 98.3 91.0 Ovanåker 90.0 77.7 96.4 90.3 98.6 90.6 Oxelösund 92.3 72.7 96.3 98.8 99.4 91.9 Pajala 92.6 73.4 93.5 91.4 97.9 89.8 Partille 85.3 77.1 94.5 96.0 98.1 90.2 Perstorp 87.1 69.6 93.9 97.2 99.3 89.4 Piteå 90.2 72.6 95.4 94.6 99.3 90.4 Ragunda 87.5 65.7 94.6 95.9 98.5 88.4 Robertsfors 84.7 70.6 94.7 95.4 98.9 88.9 Ronneby 91.2 75.7 95.3 94.3 98.0 90.9

87 Municipality Injuries Tobacco Birth- Vacci- Abortion Index weight nation

Rättvik 92.3 76.8 96.4 87.3 98.9 90.3 Sala 82.0 68.1 95.6 96.2 98.7 88.1 Salem 90.8 70.3 95.7 95.3 97.7 90.0 Sandviken 90.2 68.8 95.4 98.1 98.1 90.1 Sigtuna 92.0 71.2 95.4 88.8 97.2 88.9 Simrishamn 91.3 69.4 95.2 92.0 98.3 89.2 Sjöbo 88.2 70.5 94.8 91.5 98.3 88.7 Skara 88.7 71.7 95.0 95.7 98.3 89.9 Skellefteå 86.0 77.9 95.8 93.5 99.0 90.5 Skinnskatteberg 92.4 65.6 95.7 92.8 98.8 89.0 Skurup 87.8 69.6 94.0 95.3 98.6 89.1 Skövde 89.2 74.8 95.9 96.5 98.1 90.9 Smedjebacken 93.2 73.8 95.5 96.8 97.9 91.4 Sollefteå 85.3 72.1 95.9 90.7 98.8 88.6 Sollentuna 92.2 80.0 95.6 97.1 98.3 92.6 Solna 91.7 73.9 95.2 94.7 97.0 90.5 Sorsele 85.4 66.2 95.0 97.0 97.8 88.3 Sotenäs 86.9 66.3 96.8 95.5 97.8 88.6 Staffanstorp 90.7 72.6 96.4 93.6 98.6 90.4 Stenungsund 87.9 73.2 94.8 94.4 97.8 89.6 Stockholm 91.2 76.3 95.5 90.8 97.5 90.2 Storfors 91.2 65.7 96.9 95.8 98.0 89.5 Storuman 88.5 78.7 95.8 96.0 99.0 91.6 Strängnäs 89.8 68.2 94.9 91.5 97.8 88.4 Strömstad 94.2 78.4 95.8 96.0 98.9 92.7 Strömsund 91.5 75.7 96.1 96.5 99.0 91.8 Sundbyberg 93.3 74.4 96.3 93.8 96.3 90.8 Sundsvall 87.3 72.3 95.4 98.2 98.2 90.3 Sunne 91.4 79.2 96.2 95.3 99.1 92.2 Surahammar 93.6 69.7 96.2 95.0 97.5 90.4 Svalöv 88.0 64.4 93.5 94.4 98.0 87.7 Svedala 89.7 67.9 94.6 93.2 98.3 88.8 Svenljunga 89.0 70.9 94.0 94.9 98.4 89.4 Säffle 91.0 75.3 92.3 93.7 98.7 90.2 Säter 92.9 72.9 95.5 94.0 98.9 90.8 Sävsjö 88.1 79.9 95.3 90.9 97.9 90.4 Söderhamn 87.6 66.0 96.0 93.4 97.8 88.2 Söderköping 92.6 75.3 97.3 94.1 98.2 91.5 Södertälje 92.6 69.2 95.8 83.6 97.5 87.7 Sölvesborg 90.7 67.9 97.0 95.5 99.0 90.0 Tanum 91.7 72.3 96.4 91.1 98.0 89.9 Tibro 88.9 76.3 97.1 94.0 99.0 91.1 Tidaholm 89.8 76.1 95.8 95.0 98.4 91.0 Tierp 91.7 74.6 97.0 94.9 98.5 91.3 Timrå 87.0 68.5 94.2 95.0 98.3 88.6 Tingsryd 91.3 67.2 93.2 94.4 98.5 88.9

88

Municipality Injuries TobaccoBirth- Vacci- Abortion Index weight nation

Tjörn 87.8 77.4 95.5 92.1 99.1 90.4 Tomelilla 89.9 69.4 96.3 91.6 98.4 89.1 Torsby 87.1 71.3 95.9 95.8 97.5 89.5 Torsås 89.2 79.9 95.7 98.1 99.0 92.4 Tranemo 94.0 71.6 96.3 95.5 98.1 91.1 Tranås 93.0 79.8 96.5 96.0 99.0 92.9 Trelleborg 90.0 66.3 94.1 95.6 98.6 88.9 Trollhättan 85.9 71.9 96.4 95.9 97.4 89.5 Trosa ------Tyresö 90.3 73.0 95.5 90.5 97.8 89.4 Täby 92.0 81.2 94.2 92.9 98.7 91.8 Töreboda 87.6 75.3 96.3 92.9 97.9 90.0 Uddevalla 86.1 76.1 96.0 96.9 98.6 90.7 Ulricehamn 94.4 76.1 95.3 94.8 98.7 91.9 Umeå 88.0 80.5 95.4 93.1 98.2 91.0 Upplands-Bro 91.0 69.3 95.1 93.8 97.4 89.3 Upplands-Väsby 91.5 71.4 96.3 91.9 97.8 89.8 Uppsala 92.6 80.4 96.2 95.0 98.2 92.4 Uppvidinge 91.2 73.2 96.4 96.9 98.5 91.2 Vadstena 87.2 71.3 92.5 98.0 98.6 89.5 Vaggeryd 88.9 78.7 93.3 95.1 98.9 91.0 Valdemarsvik 91.0 70.7 96.5 97.1 97.7 90.6 Vallentuna 90.5 74.3 95.8 90.0 98.2 89.7 Vansbro 93.7 70.0 96.8 95.2 98.3 90.8 Vara 87.8 73.8 95.6 95.5 99.3 90.4 Varberg 88.8 73.4 96.0 97.8 98.0 90.8 Vaxholm 89.4 79.6 91.8 89.2 98.7 89.7 Vellinge 90.3 75.2 95.5 89.7 98.6 89.9 Vetlanda 89.9 75.2 96.1 94.1 99.1 90.9 Vilhelmina 91.2 73.3 96.4 93.5 98.2 90.5 Vimmerby 90.9 74.4 95.3 97.3 98.9 91.4 Vindeln 88.6 78.0 97.4 92.7 98.3 91.0 Vingåker 93.2 71.4 96.0 96.2 98.9 91.1 Vårgårda 88.3 75.0 94.4 93.9 99.0 90.1 Vänersborg 86.5 73.0 96.5 96.7 981 90.1 Vännäs 89.8 74.1 93.3 96.1 98.9 90.4 Värmdö 90.5 71.6 95.1 94.7 97.9 90.0 Värnamo 87.5 77.1 95.8 92.2 98.8 90.3 Västervik 87.4 75.7 95.6 97.8 98.8 91.1 Västerås 95.3 75.0 95.4 96.7 97.9 92.1 Växjö 92.1 78.5 95.6 95.8 98.4 92.1 Ydre 92.2 74.2 97.5 97.0 98.4 91.9 Ystad 90.7 73.0 95.4 95.0 97.8 90.4 Åmål 90.8 77.2 97.5 93.0 98.3 91.4 Ånge 86.2 72.7 94.0 97.5 98.2 89.7 Åre 90.6 76.6 97.6 96.9 97.6 91.8

89 Municipality Injuries Tobacco Birth- Vacci- Abortion Index weight nation

Årjäng 93.9 71.9 93.8 94.1 97.7 90.3 Åsele 94.4 84.2 97.8 97.6 98.0 94.4 Åstorp 89.1 61.7 94.2 92.9 98.1 87.2 Åtvidaberg 89.1 77.3 95.9 94.4 99.1 91.1 Älmhult 92.5 78.8 95.3 92.8 98.8 91.6 Älvdalen 91.5 76.7 96.9 90.5 96.9 90.5 Älvkarleby 86.8 63.5 95.6 93.6 98.5 87.6 Älvsbyn 87.7 75.7 96.3 90.5 98.5 89.7 Ängelholm 91.2 72.9 93.8 94.3 99.0 90.2 Öckerö 91.5 74.4 96.1 92.4 99.0 90.7 Ödeshög 89.2 75.0 95.0 92.7 99.3 90.2 Örebro 94.1 77.4 95.1 94.5 98.2 91.9 Örkelljunga 90.9 68.6 96.8 95.4 98.6 90.1 Örnsköldsvik 89.4 79.9 95.9 97.9 98.4 92.3 Östersund 90.0 78.1 95.7 97.2 98.4 91.9 Österåker 90.8 76.2 96.3 91.2 98.8 90.6 Östhammar 92.2 72.2 95.9 95.7 97.7 90.7 Östra Göinge 85.4 72.5 95.7 93.2 98.7 89.1 Överkalix 93.2 69.4 92.6 94.4 98.5 89.6 Övertorneå 92.9 79.1 96.0 92.8 98.5 91.9

90 Appendices

Appendix 1. Technical description of the indicators used

Domain B. Health status and wellbeing

B 1. Indicator Children hospitalised for injuries by external causes.

Operational definition Proportion of children 0 – 17 years receiving inpatient care for external injuries, per 1 000 children.

Background Sweden has long been a leading country in efforts to prevent accidents among children. Fatal injuries have decreased dramatically since the 1950s, from 30 per 100 000 children to 5 per 100 000, despite that the number of children is approximately the same and that the number of automobiles has risen from 1 million to 4 million (Sylwander 2001). Nonetheless, external injuries cause the largest proportion, approximately one third, of mortalities in children and adolescents, primarily in traffic. It is also one of the causes of mortality that varies most between socioeconomic groups and between boys and girls (Hjern et al. 2001; SOU 2002; Bremberg 2002).

Injuries can be further reduced by protective and preventive measures, many of them on the local level. Despite its relative importance, the number of deaths caused by injuries in the childhood years is low; a total of 94 children and young people under 18 years of age died in 2000 (according to the register for causes of mortality of the National Board of Health and Welfare). Even when results for several years are pooled, there is a very small number to be broken down to the municipal level – such a small number that it can hardly be seen as meaningful. However, as external injuries are still such an important and preventable cause of ill health, injuries caused by external factors that are treated in hospital can be used instead. These injuries are in principle serious and they are listed in the patient register of the National Board of Health and Welfare. In 2000, approximately 20 000 children between 0 and 17 years received hospital care for accidents and premeditated injuries, i.e. a sufficiently large number of cases to be able to distribute them on the municipal level.

Data from outpatient care will be available for the entire country in a few years if there should be a wish in the future to extend the indicator to include less serious injuries as well. The National Board of Health and Welfare has recently been given the task by the Government to investigate conditions for a national information system to follow developments in personal injuries. The Government has also (October 2001) appointed a delegation to examine and work with questions concerning safety and prevention of injuries in children’s and adolescents’ environment (SOU 2000).

91 Technical criteria • Useability The advantage of including all injuries is of course that the number of cases would then be large enough to make it meaningful to break them down to the municipal level each year without needing to pool cases into periods of several years. For use on the national level within the EU, it is suggested to concentrate on fractures of long bones as an injury indicator as these injuries are considered to always require hospital care and can hardly be sensitive to changing treatment policies between countries (Rigby & Köhler 2002). By including all children treated as inpatients for injury diagnoses, we can not of course exclude the possibility that varying resources and care policies in different parts of the country will affect the number of cases. However, routine annual statistics kept by the National Board of Health and Welfare show that the distribution between county councils is relatively uniform. • Robustness The indicator is well documentated as being sensitive to health outcome, socioeconomic distribution and preventive efforts. • Understandability The indicator is widely used and is easy to understand.

Data sources Patient register of the National Board of Health and Welfare. (includes personal identification number)

Data availability to the municipalities As the register contains information about the patient’s home municipality, data can be reported on the municipal level.

Key references Bremberg, S. (2002). Social inequities in health in Swedish children and adolescents – a review). National Institute of Public Health, Stockholm

Hjern, A., Ringbäck-Weitoft, G. & Andersson R (2001). Sociodemographic risk factors for home-type injuries in Swedish infants and toddlers. Acta Paediatrica, 90, 61-68

Rigby, M., Köhler, L. (2002). Child Health Indicators of Life and Development (CHILD). European Commission, Luxembourg

Socialstyrelsen (2001b). Folkhälsorapport 2001.(Public Health Report 2001). Socialstyrelsen, Stockholm

SOU (2002). Sociala skillnader i skador hos barn och ungdom. Rapport från Barn- säkerhetsdelegationen. (Social differences in childhood injuries. Report from the Child Safety Delegation.) Fritzes, Stockholm

Sylwander, L. (2001). Child Accident Prevention. A Swedish success story. In: Protection Prevention Promotion. Development and future of Child Health Services (ed. G. Norvenius,L. Köhler, G. Wennergren, & J. Johansson). Nordic School of Public Health, Göteborg

92 Domain C. Determinants (Risk and protective factors)

C 1. Indicator Children with low birthweight

Operational definition Proportion of newborns with a birthweight below 2 500 g, per 100 newborns

Background Low birthweight was previously an important cause of high mortality in infancy and still is in developing countries. Improvements in nutritional status and generally high living standards have drastically reduced the occurrence of low birthweight in Sweden and decreased risks among survivors. Somewhat over 4 percent of children born today have a birthweight below 2 500 g and less than 1 percent a birthweight below 1 500 g. With modern neonatal care, even the smallest infants survive but as a group run greater risks for neurological damage, difficulties in school and behavioural disorders. While these risks have also markedly decreased, a certain elevated risk still remains.

The causes of low birthweight are complex, but there is a clear relation to poorer socioeconomic conditions in the parents, maternal smoking and stress during pregnancy (Elmén 1994, Bremberg 2002). These factors can be influenced at least to some extent on the local level.

Technical criteria • Useability The definition is widely used internationally. Information is collected from the entire country and is compiled annually by the National Board of Health and Welfare. Information is available from 1973 and onward. The quality is high and drop-out low. • Robustness Despite the fact that the consequences of low birthweight have decreased, it still indicates an elevated risk for different health problems in childhood and possibly later in life. • Understandability The indicator is easy to understand and use.

Data sources Medical birth register, the National Board of Health and Welfare

Data availability to the municipalities The information is based on personal identification numbers and can be broken down to the municipal level (approximately 4 000 children with a birthweight lower than 2 500 g are born each year). Children with birthweights under 2 500 g are included in the Swedish National Institute of Public Health’s database on municipal facts for public health planning as three-year mean values.

Key references Bremberg, S. (2002). Social inequities in health in Swedish children and adolescents – a review). National Institute of Public Health, Stockholm

93 Elmén, H. (1995). Child health in a Swedish city. Mortality and birthweight as indicators of health and social inequality. Thesis. Nordic School of Public Health, Göteborg

Socialstyrelsen (2002a). Fakta om mammor, förlossningar och nyfödda barn. Medicinska födelseregistret 1973 till 2000. (Facts about mothers, deliveries and newborn infants. Medical Birth Register 1973-2000). Epidemiologiskt centrum, Stockholm

C 2. Indicator Children that are breastfed

Operational definition Proportion of children who are breastfed at 4 months of age, per 100 Children registered at a child health centre

Background Breastfeeding is the most natural and most often simplest way to give infants food. Breastmilk has an ideal nutritional content and protects the child against infectious diseases and possibly also allergies and in the longer run against overweight and type 1 diabetes and guarantees close physical contact between mother and child. Breastfeeding prevalence is used internationally as an important health indicator. It also reflects national, regional and local success in efforts to disperse information and promote health. The most important factors behind a high breastfeeding frequency are considered to be motivation, social stability and possibility for support in the event of difficulties with breastfeeding (National Board of Health and Welfare 2002).

Technical criteria • Useability Current Swedish criteria differ from those of WHO used internationally. For instance “exclusively breastfed” in Sweden allows small amounts of other food (“tasting portions”), while the international definition allows only supplementary vitamins. Breastfeeding is followed internationally up to 12 months; in Sweden it has been followed up until now for 6 months. The international definition is used in the new proposal for EU indicators, which will probably be introduced in Sweden as well. To limit the amount of data, the age of 4 months is used here, which is a reasonable compromise between the necessary (2 months) and the desirable (6 months). • Robustness The indicator is generally known as a base indicator for optimal development of the child. • Understandability It is somewhat difficult to understand why the Swedish definitions should differ from those used internationally.

Data sources Information on breastfeeding is gathered from mothers on a running basis at regular visits to the child health centre. Via the records at the centres, the data are then compiled by the head of the Child Health Sevices of the county and submitted to the National Board of Health and Welfare, which presents annual reports. Results in areas within the counties are available after special data processing at the Centre for Epidemiology.

94 Data availability to the municipalities At this time, annual data are regularly presented by the National Board of Health and Welfare according to health care areas, i.e. county councils, and can be broken down to the municipal level.

Key references Socialstyrelsen (2002b). Amning av barn födda 2000. (Breastfeeding of children born in 2000). Stockholm

Hanson LÅ (2002). Breastfeeding, a complex support system for the offspring. Pediatr Int, 44, (4), 357-352

WHO (2001). Infant and young child nutrition. Resolution of the World Health Assembly WHA 54.2. WHO, Geneva

Yngve A, Kylberg E & Sjöström M (2001). Breastfeeding surveillance in the EU and EFTA: Recommendations adopted at the Breastfeeding Surveillance Conference. Breastfeeding Surveillance Conference, Stockholm

C 3. Indicator Foetal exposure to tobacco

Operational definition Proportion of children exposed to tobacco in the womb because of mothers’ smoking habits, per 100 infants Registered at a child health centre

Background ETS (Environmental Tobacco Smoke) or passive smoking is the single most significant form of air pollution in the indoor environment. It is related to a series of acute and chronic health effects. In the foetus, it increases the risks for mortality and low birthweight. In the infant, it increases the risks for breathing difficulties, infections and allergies (EEA 2002). Information can considerably reduce the occurrence of ETS. In Sweden, smoking among pregnant women has dropped by 50 percent during the past 10-year period and is now at an average of 12 percent. Pregnant teenage girls smoke most, nearly one third of this group. Nine percent of mothers of infants smoke and 14 percent of fathers of infants smoke. When children reach the age of 8 months, 11 percent of the mothers and still 14 percent of the fathers smoke (National Board of Health and Welfare 2002). These levels are low in international comparisons but are still too high for children’s best. Parents in lower social groups smoke more than others.

Technical criteria • Useability The definition used internationally is exposure to non-temporary tobacco smoke i.e. at least each week. Data are collected by specific questionnaires or routine questions at antenatal clinics/child health centres. Many different definitions have been used, which make international comparisons difficult. Smoking habits among pregnant women in Sweden are

95 identified at the first visit to the antenatal clinic and are documented in the records there. • Robustness The indicator has clearly been shown to be related to health problems in small children. • Understandability The indicator is easily understood.

Data sources Medical birth register, National Board of Health and Welfare

Data availability to the municipalities Information about smoking habits in pregnant women in Sweden is collected at the antenatal clinics and is documented in the records there, which then follow the child to the child health centres. The information is collected regularly by the National Board of Health and Welfare and is presented per health care area but can be broken down to the municipal level. The register is constructed such that the proportion of children who are exposed to tobacco smoke is presented, and not the proportion of women who smoke. Pregnant women who smoke are included in the Swedish National Institute of Public Health’s data base for municipal facts.

Key references European Environmental Agency & WHO regional Office for Europe (2002). Children's environmental health: review of the evidence. WHO, Copenhagen

Socialstyrelsen (2002a). Fakta om mammor, förlossningar och nyfödda barn. Medicinska födelseregistret 1973 till 2000. (Facts about mothers, deliveries and newborn infants. Medical Birth Register 1973-2000). Epidemiologiskt centrum, Stockholm

Socialstyrelsen (2002c). Rökvanor bland gravida och småbarnsföräldrar (Smoking among pregnant women and families with small children). Socialstyrelsen. Epidemiologiskt centrum, Stockholm

C 4. Indicator Teenage abortions

Operational definition Proportion of abortions carried out per 1 000 women under 18 years of age

Background An overall goal in society is that children shall be wanted and for society to take responsibility for providing information and the means for people to freely and responsibly plan the births of their children. Both unwanted pregnancies and abortions imply a form of life crisis for many people. Thus a goal in general work to promote health among young people is to prevent unwanted pregnancies (Action Plan for the Prevention of Unwanted Pregnancies, 2001). On the other hand, it is not certain that a pregnancy that is initially experienced as unwanted actually results in an unwanted child. Up to 15 percent of women who early in their pregnancies wanted to abort changed their minds and gave birth (Trost, 1984). In practice, it is difficult to differentiate between wanted and unwanted pregnancies, and to document the effects of preventive

96 measures it is necessary to rely on factors that are related to unwanted pregnancies, e.g. abortions. Swedish abortion policy has as its goal in part to decrease the number of unwanted pregnancies and in part to ensure women’s right to medically safe abortions and good psychosocial care (Action Plan for the Prevention of Unwanted Pregnancies, 2001).

Childbirth among teenagers has decreased considerably during the latest decades and now represents less than 5 percent of all women who give birth. Until 1995, the number of abortions also decreased but has since increased for seven straight years, altogether by 50 percent, of which the increase was 12 percent in the most recent years (Abortions in Sweden, 2002). An especially large increase has been noted among 14 year olds. This indicates that preventive efforts have not had the impact they did in previous years. The earlier successful decrease in teenage abortions has been ascribed to the broad efforts in instruction in sex and cohabitation in schools and at adolescent clinics (WHO 1985). There is an uneven distribution of abortions among young women regionally and socioeconomically (Bremberg, 2002).

Technical criteria • Useability The number of teenage abortions is now over 6 000 per year or approximately 20 per 1 000 teenage women. • Robustness The indicator shows changes in adolescents’ values and in their sexual behaviour, at the same time that it is closely connected to social and economic conditions in society. Preventive and health-promoting efforts, especially on the municipal level, have been judged to be successful. • Understandability The indicator is easy to understand and to manage.

Data sources National Board of Health and Welfare, Centre for Epidemiology

Data availability to the municipalities The data are included in the Swedish National Institute of Public Health’s municipal base facts for public health planning as five-year mean values in the municipalities.

Key references Socialstyrelsen (2002d). Aborter i Sverige 2002, januari - juni. (Abortions in Sweden 2002, January-June). Epidemiologiskt centrum, Stockholm

Trost, A.-C. (1984). Abort och psykiska besvär. (Abortion and mental problems) International Library, Västerås

97 C 5. Indicator

Proportion of vaccinated children

Operational definition Proportion of children who at the age of 2 years have completed the vaccination programme for measles, mumps and rubella (MPR), per 100 children in that age group

Background The significance of focal or mass immunity to prevent and limit epidemics of childhood diseases is well known and has recently been demonstrated by breakouts of measles in areas with children who have not been vaccinated in Great Britain and Holland (Hanratty et al. 2000; van den Hof et al. 2000). Concern has grown in Sweden that the decreasing tendency among parents to vaccinate their children will lead to poorer focal immunity and epidemic breakouts (Norrby 1999; Vaccination of children, Smittsskyddsinstitutet 2001; Olin 2002, Mass immunity is lacking in half our municipalities 2002). This is considered to be able to take place when fewer than 90 percent of the population in question is vaccinated.

Vaccination is one of the most powerful and cost effective forms of primary prevention (Vaccination of children 2001). The responsibility for vaccinations rests with the health care organisations, primarily the child health centres, and information efforts on the municipal level can be successful.

Technical criteria • Useabillity The MPR vaccination is the most important vaccine in early childhood and gives a good picture of the level of protection against serious infectious diseases and at the same time a picture of parents’ trust in society’s care for its citizens. • Robustness A level of vaccination of less than 90 percent of the population in question is considered to pose risks for epidemics or endemics.

Data sources Information on individual vaccinations is registered at local child health centres. Aggregated data that can be broken down to the municipal level exist at the county councils’ child health care units and at the Swedish Institute for Infectious Disease Control.

Data availability to the municipalities Data are normally reported per health care area but are also available for each municipality.

Key references Hanratty B, Holt T, Duffell E, Patterson W, Ramsay M, White JM, Jin L & P., L. (2000). UK measles outbreak in non-immune anthroposophic communities: the implications for the elimination of measles from Europe. Epidemiol Infect, 124, (2), 377- 383

98 Olin, P. (2001). Svenska barns vaccinationsskydd. (Immunisation of Swedish children) Läkartidningen, 98, 3654-3657 van den Hof S, van den Kerkhof JH, ten Ham PB, van Binnendijk, R., Conyn- vanSpaendonk, M. & JE, v.S. (2001). Measles epidemic in the Netherlands 1999-2000. Ned Tijdschr Geneeskd, 145, (25), 29-33

Appendix 2. Technical description of the indicators not used

Domain A. Demographics and socioeconomics

Indicator Children’s social class

Operational definition Percent of children who live in households in which the parents belong to social group (highest among the father and the mother or single parent): 1. Non-professional; 2. Professional; 3. Lower level white-collar; 4. Middle level and higher white-collar; 5. Self-employed; 6. Farmer; 7. Student. Distributed according to sex and age groups: 0-4, 5-9, 10-14, 15-17.

Background Children’s social background is decisive in most areas for their later lives. The family’s social position has been estimated to explain 20-40 percent of children’s ill health in Sweden (Bremberg 2002). Social groups determine families’ place in the social hierarchy as they combine occupation, education and income with prestige, privileges and power.

This indicator shall describe the child population in each municipality and constitutes a background variable for later health variables. It is in this report replaced by the poverty index, used in the first Save the Children report, Child Poverty in Sweden (Salonen 2002).

Technical criteria • Useability The indicator is generally used in this form in national public health reports, and in similar forms in other national and international reports on health. However, “unemployed persons” should be included as a special category. • The occupational categories are intimately associated with the economic and social structure. Shifts in the relative size and significance of occupational groups can thus make comparisons over time difficult. • Robustness The indicator is well documented to be strongly discriminatory for health outcome. • Understandability The indicator is widely used and easy to understand.

Data sources Statistics Sweden

99 Data availability to the municipalities The indicator is a part of standard statistics, even on the municipal level.

Key references Bremberg, S. (1999). Bättre hälsa för barn och ungdom. (Improved health for children and adolescents). Folkhälsoinstitutet, Stockholm

Bremberg, S. (2002). Social inequities in health in Swedish children and adolescents – a review). National Institute of Public Health, Stockholm

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

Indicator Parents’ education

Operational definition Percent of children who live in families with education levels (highest of the father and the mother or single parent): low education (maximum of compulsory school); medium level education (upper secondary school or less than 3 years of education past upper secondary school); high education (3 years or more education past upper secondary school or post-graduate education). Distributed according to sex and age groups: 0-4, 5-9, 10-14, 15-17.

Background Parents’ education, particularly mothers’ education, has been shown to be a reliable predictor of children’s health and is frequently used both nationally and internationally. The education level affects not only children’s health status of a physical, mental and social nature but also the utilisation of care. It is applicable also for parents not participating in the labour market. The indicator is especially robust for marginalised children. It is in this report replaced by the poverty index, used in the first Save the Children report Child Poverty in Sweden (Salonen 2002).

Technical criteria • Useability The indicator is generally used, in this form in national public health reports, and in similar forms in other national and international reports on health. • Robustness The indicator is well documented as a strongly discriminating factor for health outcome. • Understandability The indicator is widely used and easy to understand.

Data sources Statistics Sweden

Data availability to the municipalities The indicator is included in standard statistics, even on the municipal level.

100 Key reference Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

Indicator Children’s family structure

Operational definition Percentage of children who live families with only one parent or guardian, male and female and total, and percentage of children who live in families with co-existing parents, in age groups: 0-4, 5-9, 10-14, 15-17

Background Although there seem to be considerably smaller differences between children in families with one or two parents in Sweden than are reported in other parts of the world, there is still sufficient evidence that children who live in families with only one parent are more vulnerable than other children, both economically, socially and culturally. Many studies show a greater burden among these children, resulting in mental disorders and psychosomatic complaints.

This indicator shall describe the child population in each municipality and constitute a background variable for later health variables. It is in this report replaced by the poverty index, used in the first Save the Children report Child Poverty in Sweden (Salonen 2002).

Technical criteria • Useability The indicator is generally used, in this form in national public health reports, and in similar forms in other national and international reports on health. • Robustness The indicator is well documented to be strongly discriminatory for health outcome. • Understandability The indicator is widely used and easy to understand.

Data sources Statistics Sweden

Data availability to the municipalities The indicator is included in standard statistics, even on the municipal level.

Key references Berntsson, L. (2000). Health and well-being of children in the five Nordic countries in 1984 and 1996. Thesis. Nordic School of Public Health, Göteborg

Meltzer, H., Gatward, R., Goodman, R. & Ford, T. (2000). Mental health of children and adolescents in Great Britain. Office for National Statistics, London

Ringbäck Weitoft, G., Hjern, A., Haglund, B. & Rosén, M. (2002). Mortality, severe morbidity and injury among children of lone parents in Sweden. Lancet, 361, 289-285.

101 Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

Indicator Children of foreign origin

Operational definition Proportion of children born outside the country according to area: 1. the Nordic countries; 2. Europe, outside the Nordic countries; 3. other. Girls and boys and total, per 100 children in the municipality, distributed according to age groups: 0-4, 5-9, 10- 14, 15-17

Background In Sweden at this time approximately 6 percent of children were born outside the country and a further 18 percent have at least one immigrant parent. Thus, together, one fourth of children in the country have a foreign background, although there are large regional differences. A large number of scientific studies and other surveys have shown that these children overall have poorer health, poorer social roots and poorer resources. Children seeking asylum and children who reside illegally in the country are particularly at risk (Hjern et al. 1997, 1998). The indicator is in this report replaced by the poverty index, used in the first Save the Children report Child Poverty in Sweden (Salonen 2002).

Technical criteria • Useability The indicator is generally used, in this form in the 2001 Public Health Report, and in similar forms in other national and international reports and scientific studies of health. • Robustness The indicator is well documented to be a strongly discriminatory factor for general health outcome. • Understandability The indicator is widely used and is easy to understand.

Data sources Statistics Sweden

Data availability to the municipalities The indicator is included in standard statistics, even on the municipal level.

Key references Hjern, A., Allebeck, P. (1997). Health examinations and health services for asylum seekers in Sweden. Scandinavian Journal of Social Medicine, 25, 207-209.

Hjern, A., Angel, B. & Jeppson, O. (1998). Political violence, family stress and mental health of refugee children in exile. Scandinavian Journal of Social Medicine, 26, 18-25.

SOU (2001b). Barns och ungdomars välfärd. Antologi. Kommittén Välfärdsbokslut. (Welfare of children and young people. An antology). The National Committee of Welfare Balance Sheet Fritzes, Stockholm.

102

Indicator Children seeking asylum

Operational definition Proportion of children who have sought asylum, alone or as part of a family, per 100 children in the municipality, boys and girls and total, distributed according to age groups: 0-4, 5-9, 10-14, 15-17

Background One of the most important goals of the UN Convention on the Rights of the Child is to improve treatment and living conditions for children who are at extreme risk for their health and development. Children in exile have been forced to leave their home countries, many also owing to military and political violence, and they run a great risk for physical and mental problems. Up to 40-50 percent of refugee children with non- European backgrounds have been shown to have mental problems (Hjern et al. 1998). The majority of refugee families have left their social network in their home countries. Very often, the adults who should support these children have psychological difficulties themselves. There is great reason for refugee children to be given immediate help when they come to the new country (Hjern et al. 1998). The indicator is in this report replaced by the poverty index, used in the first Save the Children report Child Poverty in Sweden (Salonen 2002).

Refugee children are also a risk population for serious infections, such as tuberculosis, hepatitis and intestinal parasites. Children without permanent residence permits were previously excluded from preventive and curative medical care, with the exception of emergency treatment. This was changed in 2000 and all these children now have full access to the same care as other children.

Over 400 children from about 50 different countries come to Sweden alone, without their families. Of all those who sought asylum in 2001 (23 000 individuals), 32 percent or over 7 000 individuals were under 20 years of age. The largest group among these were 0-4 years old. Added to this is an unknown number of illegal immigrants and refugees.

The Swedish Migration Board is responsible for people seeking asylum, even in the case of children who come to the country alone. However, according to the Social Services Act, the municipalities are responsible for all people living there, and have a particular responsibility for children and young people. It is unclear how these different responsibilities should be interpreted in individual cases. The Government has thus given several different authorities the task of generally improving the reception of children seeking asylum and of clarifying the division of responsibility between them (2002).

103 Technical criteria • Useability Information on the number of persons seeking asylum is not completely reliable, especially in the case of illegal immigrants. While the number of persons seeking asylum thus does not give a completely correct picture of the scope of the problem, the indicator reflects at least one of the most high risk and vulnerable groups in our society. The responsibility of the municipalities for these children is not clear but work commissioned by the Government is ongoing to clarify this. • Robustness The indicator focuses on one of the most vulnerable groups in society, a group that has especially been identified in the Convention on the Rights of the Child. • Understandability Easy to understand but difficult to document without provoking objections.

Data sources National Board of Health and Welfare Statistics Sweden

Data availability to the municipalities Personal identification numbers exist and data can in principle be broken down to the municipal level. The municipalities’ responsibility is not clear.

Key references Hjern, A., Angel, B. & Jeppson, O. (1998). Political violence, family stress and mental health of refugee children in exile. Scandinavian Journal of Social Medicine, 26, 18-25

Hjern, A., Ringbäck-Weitoft, G. & Andersson R (2001). Sociodemographic risk factors for home-type injuries in Swedish infants and toddlers. Acta Paediatrica, 90, 61-68

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

Domain B Health status and wellbeing

Indicator Children with diabetes

Operational definition Proportion of children with a new diagnosis of diabetes (type 1, insulin dependent). Distributed among boys and girls and total, and in age groups: 0-4, 5-9 and 10-14 and total, and according to socioeconomic group

Background Diabetes has increased strongly in Sweden and more than in most other countries (although the numbers are highest in Finland and Sardinia).

The frequency has increased an average of over 2 percent per year and even more in children under 10 years of age. Approximately 600 children are diagnosed in Sweden

104 each year and, despite systematic and intensive treatment, the risks for serious complications in the long term are greater, particularly for kidney, eye, heart and nervous system complications. The cause of the increase has not been clarified, but there is a strong association with the equally large increase in welfare diseases such as obesity and overweight among children (Dahlquist & Mustonen 2000).

The strong increase, the complex causality and the serious consequences of the disease make it very important to keep developments under close surveillance.

The indicator is defined in the same way as in the proposal for health indicators for EU’s member states.

Technical criteria • Useability A uniform definition of the diagnosis is used throughout the country and all children newly diagnosed receive treatment at children’s clinics. Sweden has had a national diabetes register for children up to 15 years of age since 1977, and the register is based on reports from all paediatric clinics in the country (this is the reason for the limit of 15 years of age). Professor Gisela Dahlquist at Umeå University is responsible for the register. The quality of the register has been judged to be good in several evaluations (Dahlquist, personal communication 2002). • Robustness The indicator is well documented as being significant for children’s current and future health status. • Understandability The indicator is widely used and easy to understand.

Data sources Children’s Diabetes Register, Umeå University

Data availability to the municipalities Personal identification numbers exist but the reports from the register are based on anonymised information and are presented according to region. Detailed studies have been made of parts of the material down to the parish level and even down to the individual house level. The entire material is not available on the municipal level, and the number of new diagnoses is small.

The indicator is most suitable for surveillance on the national level owing to the relatively limited number of cases. It is excluded here since data are available only for a limited number of municipalities.

Key references EURODIAB ACE Study Group (2000). Variation and trends in incidence of childhood diabetes in Europe. Lancet, 355, 873-876.

Dahlquist, G. & Mustonen, L. (2000). Analysis of 20 years of prospective registration of childhood onset diabetes - time trends and birth cohort effects. Acta Paediatrica, 89, 1231-1237.

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm.

105 Rigby, M., Köhler, L. (2002). Child Health Indicators of Life and Development (CHILD). European Commission, Luxembourg

Indicator Children without caries

Operational definition Proportion of children who are caries free in temporary dentition at the age of 6, per 100 boys and girls and total, and according to socioeconomic group

Background Children’s dental health has radically improved during the latest decades by systematic preventive efforts. In 1967 only 17 percent of Swedish 4-year-olds were caries free (Wendt et al. 1999). Follow-ups done by the National Board of Health and Welfare show that 83 percent of 3-year-olds were caries free in 1985 and that 94 percent were caries free in 1999. During the same time, the proportion of caries free 6-year-olds increased from 45 to 72 percent and the proportion of caries free 12-year-olds from 22 percent to 61 percent (Public Health Report 2001).

Viewed internationally, there are large differences in preventive efforts and diet patterns (Bolin 1997).

Dental diseases in children are a health problem in themselves, but they are also related to other health problems, in childhood and later in adulthood.

Dental health status varies between different areas and shows an easily accessible and clear relationship with socioeconomic conditions.

The good results achieved in children’s dental health are the result of great efforts of a primarily preventive character. These results are not automatically permanent, however, but must be maintained and improved with continued preventive measures, particularly among groups most at risk.

Technical criteria • Useability The indicator of freedom from caries, often combined in national and international studies and reports with DFS-a, is widely used and relatively easy to apply, even on a base level. (DFS-a means “decayed and filled surface” in the approximal surface (contact surface in the dental arch) of permanent teeth.) To limit the amount of data and simplify calculations, information has been concentrated to caries free 6-year-olds who have had temporary dentition for a number of years. Information is collected on a national level by the National Board of Health and Welfare via the National Dental Service and private dentists, although not together with personal identification numbers. Approximately 65 percent of all 6-year-olds are currently reported. • Robustness The indicator is well documented to be discriminatory for dental health and is strongly associated with socioeconomic status. • Understandability The indicator is widely used and easy to understand.

106 Data sources The National Board of Health and Welfare

Data availability to the municipalities The National Dental Service and certain private dentists have data that in certain cases are available on the municipal level. In the register of the National Board of Health and Welfare, information can be presented for groups of municipalities. However, in a few years’ time information will be available for individual municipalities (Hans Sundberg and Agenta Ekman, National Board of Health and Welfare, personal communication). It is excluded here since data are not now available for all municipalities.

Key references Bolin A-K (1997). Children's dental health in Europe. An epidemiological investigation of 5- and 12 year-old children form eight EU countries. Thesis. Swedish Dental Journal, Suppl, 1-88

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

Wendt, L.-K., Hallonsten, A.-L. & Koch, G. (1999). Oral health in pre-school children living in Sweden. Swedish Dental Journal, 23, 17-25

Indicator Suicide attempts in children

Operational definition Incidence of suicide attempts, defined as discharge from hospital with the diagnosis of suicide attempt, per 100 boys and girls and total, and according to age group 10-17 years and to social group

Background Sweden has traditionally had a reputation as a country with a high frequency of suicides, but, after an increase during the 1960s and 1970s, the number of suicides decreased. Sweden now belongs to a middle group among European countries. However, suicide is still the most common cause of death among 15-44-year-olds and the second most common among 15-24-year-olds.

The incidence has been almost unchanged during the past 20 years (NASP 2002). During the period 1980-1998, 40-50 adolescents in ages up to 19 years took their lives each year. In the youngest age group, boys and girls under 15 years of age, there was an average of six suicides per year between 1980 and 1999. Even if the numbers are pooled in five-year periods, they are too small to be able to be broken down to the municipal level. Registered suicide attempts are however an important predictor for later suicide and can thus be used as an indicator. The numbers continue to increase among young people and are estimated now to be 3 percent among boys and 8 percent among girls 16-17 years old.

107 Several investigations show that 5 percent of teenagers carry out one or more serious attempts to take their lives during their adolescence (NASP 2002).

Technical criteria • Useability Defined by ICD 10 codes X60 – X84, Y87.0. The diagnosis suffers from some uncertainty; roughly 20 percent are usually counted as “uncertain suicide”.Very few cases are treated in hospital for suicide attempts. For the year 2000, for example, 714 children/adolescents were treated for this diagnosis, and only 192 municipalities had one or more cases; the other municipalities had no cases. Suicide attempts, or more correctly “intentionally self-destructive actions”, have therefore been included in the injury indicator, B1. • Robustness The indicator is considered to reflect the general psychic instability present during the teenage years and is an important predictor of later successful suicide. • Understandability The indicator is easy to understand.

Data sources National Board of Health and Welfare’s Patient Register NASP, The National Centre for Suicide Research and Prevention of Mental Ill Health

Data availability to the municipalities Personal identification numbers exist but the number of cases is too small to be used alone on the municipal level. Suicide attempts are however included in the indicator “External injuries”.

Key references NASP (2002). Nyhetsbrev. NASP Nationellt centrum för suicidforskning och prevention av psykisk ohälsa, (Newsletter. National centre for suicide research). Stockholm Schmidtke, A. (1996). Attempted suicides in Europe: rates, trends and sociodemographic characteristics of suicide attempters during the period 1989-1992. Results of the WHO/Euro Multicentre Study on Parasuicide. Acta Psychiatrica Scandinavica, 93, (5), 327-338

Indicator Children with asthma

Operational definition Prevalence of asthma per 100 boys and girls and total, and in age groups: 0-4, 5-9, 10- 14, 15-17, and according to socioeconomic group

Background Asthma and allergies are now the most common chronic illnesses among children and adolescents. There has been a strong increase in nearly all countries in the West (Formgren 1998). In Sweden, about 40 percent of schoolchildren have some form of allergy, and children with asthma account for half of all the patients of physicians in outpatient care (Public Health Report 2001). The causes of the illness and its increase are not completely known, but both hereditary and environmental factors are responsible, probably in a complex interplay.

108 Conditions during childhood and the local environment probably play an important role for the occurrence of the disease. Modern treatment can most often keep the illness in check, but asthma is still a serious illness with a certain mortality and not infrequently functional impairments. Occurrence decreases in adulthood but the illness has begun in many cases in childhood.

Technical criteria • Useability There is no standard definition of the disease and comparisons between areas and over time are difficult. International attempts at standardisation have been made but are still not useable (ISAAC 1998). The National Board of Health and Welfare’s inpatient register may perhaps be able to be used, but the reliability of the diagnosis is uncertain. It is thought that there is considerable underdiagnosis in the register, both in inpatient and outpatient care (Hasselgren et al. 2001). Several studies are currently being carried out on the regional level. • Robustness Owing to problems having to do with the definition, it is not certain what the indicator reflects. • Understandability There are inclarities in definitions.

Data sources The National Board of Health and Welfare’s inpatient register

Data availability to the municipalities The inpatient register of the National Board of Health and Welfare is based on personal identification numbers and can theoretically be broken down to the municipal level.The indicator is excluded here because the diagnosis is uncertain and varies between different clinics and because it is unclear what the indicator reflects.

Key references Formgren, H. (1998). Allergisjukdomar hos barn och unga. I. Omfattning av allergi och annan överkänslighet. Vetenskaplig sammanställning. (Allergies in children and youth) Folkhälsoinstitutet, Stockholm

Hasselgren M, Arne M, Lindahl A, Janson S & Lundbäck B (2001). Estimated prevalence of respiratory symptoms, asthma and chronic obstructive pulmonary disease realted to detection rate in primary health care. Scandinavian Journal of Primary Health Care, 19, 54-57.

ISAAC (1998). The International Study of Asthma and Allergies in Childhood. Worldwide variations in the prevalence of of asthma symptoms. European Respiratory Journal, 12, 315-335.

Socialstyrelsen (2001b). Folkhälsorapport 2001.(Public Health Report 2001) Socialstyrelsen, Stockholm.

109 Indicator Obese and overweight children

Operational definition Proportion of children who have an age and sex standardised Body Mass Index (BMI) of at least 25 and 30, respectively, measured at the start of school and at 18-19 years of age (among boys), per 100 children and according to sex and socioeconomic group

Background Overweight and obesity are associated with a number of health problems of a physical and mental nature, both in childhood and later in adulthood, e.g. diabetes, cardiovascular diseases, joint diseases and poorer quality of life (SBU 2002). The occurrence is rising rapidly all over the world and has become an important public health problem that requires preventive and treatment efforts. In Sweden, the proportion of overweight (BMI kg/m2 > 25) boys registering for military duty increased from 7 to 18 percent between 1971 and 1998 and the proportion of obese boys (BMI > 30) from 1 to 4 percent (National Public Health Report 2001). The increase is greatest in families with low socioeconomic status.

Technical criteria • Useability Internally used definition with fixed limits. The method shows some weaknesses among children, however, as it underestimates the degree of overweight in short children and overestimates the degree of overweight in tall children (SBU 2002). • Robustness The indicator is well documented to be discriminatory for health outcome. • Understandability The indicator is widely used, nationally and internationally, and is easy to understand.

Data sources National Service Administration Children’s Health Care School Health Care

Data availability to the municipalities Growth data collected regularly from the child health centres and school health services are not yet available throughout the country. Military registration data exist only for 18- 19-year-old boys, who furthermore per definition are no longer children. Several studies on smaller populations are ongoing. The indicator is excluded here because data collection is incomplete.

Key references Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

Rasmussen, F., Johansson, M. & Hansen, H.-O. (1999). Trends in overweight and obesity among 18 years old males i Sweden between 1971 and 1995. Acta Paediatrica, 88, 431-437

110 WHO (1998). Obesity - preventing and managing the global epidemic. Report of a WHO consultation. WHO, Geneva

Indicator Children in the care of the municipality

Operational definition Proportion of children and adolescents for whom interventions have been made by the municipality one or more times during the year, boys, girls and total, in the age groups 0-12 and 13-17 years, per 1 000 children in corresponding population groups

Background The social welfare board of the municipality shall act such that children and adolescents are able to grow up in secure and favourable conditions. If children and adolescents risk developing in an unfavourable way, the municipality can provide care and upbringing outside the home, in family homes or homes for care or residence. These placements can be made with the support of the Social Services Act (voluntary placement) or with the support of LVU (the law that gives the municipality the right to place children and adolescents in care outside the home against the child’s or guardian’s wishes). The municipalities report their interventions to the National Board of Health and Welfare, which compiles the results on an annual basis. In all of Sweden’s municipalities, a total of 16 000 children and adolescents were the object of interventions in 1998, and the information is already broken down to the municipal level and calculated per 1 000 children in the population, distributed among the ages 0- 12, 13-17 and 18-20 (Insatser för barn och unga (Interventions among children and adolescents) 1999).

Technical criteria • Useability This indicator describes the municipalities’ interventions and actually not the health status of the children. It is a process indicator and, as such, is affected by resources, ambitions and ideologies in the local social welfare board, and of course by the composition and needs of the population in question. Placement outside the home is however intended to improve the children’s upbringing and to prevent harmful development. It can thus be expected to be significant to the child’s health and wellbeing, and it can therefore be justifiable to include these efforts as an indicator of the municipality’s support measures for children and adolescents and their families. Social group is not included in the reporting, but the great majority of measures have to do with children from families with social and economic problems. • Robustness Alone, the indicator is probably difficult to interpret but, when put in relation to other indicators that give information on the municipality’s social, economic and ethnic structure, it can be valuable. The quality of the data is not yet completely satisfactory and work is going on to improve it (Insatser för barn och unga 1999). • Understandability It is easy to understand but not always easy to interpret.

111 Data sources National Board of Health and Welfare, Social statistics

Data availability to the municipalities All municipalities report to the National Board of Health and Welfare, which compiles the results by municipality and age group. The indicator has been excluded because the quality of the data is questionable and their interpretation unclear.

Key references Socialstyrelsen (1999). Insatser för barn och unga. (Contribution towards children and adolescents). Stockholm

Socialstyrelsen (1999). Socialtjänsten i Sverige 1999. Behov - Insatser - Utveckling. (Social services in Sweden. Needs –Actions – Development). Stockholm

Domain C. Determinants (Risk and wellness factors)

Indicator Children’s and adolescents’ tobacco smoking

Operational definition Proportion of children who report that they smoke each week, per 100 boys, girls and total, at ages 11, 13 and 15 and according to socioeconomic group

Background Tobacco smoking is a well documented risk factor for many diseases, including lung cancer and cardiovascular diseases. The earlier an individual starts to smoke, the greater the risks. Smoking habits are more common among adolescents from lower social classes and are associated with other unhealthy behaviour such as use of alcohol and narcotics (SOU 2001b, Bremberg 2002). Young people tend to continue their smoking habits in adulthood, which makes them a primary target group for marketing campaigns. These factors also make them an important group in terms of preventive measures and projects to stop smoking.

Swedish adolescents smoke very little in an international perspective. In fact, Swedish boys smoke least in Europe (Public Health Report 2001).

Technical criteria • Useability The indicator is based on self-reports and is used in different health questionnaires. It is not included in routine reports in whole populations but is directed toward representative selections of adolescents, e.g. selected schools and at registration for military service. • Robustness The indicator has been used nationally and internationally for many years. It has been evaluated and is judged to give a good picture of actual conditions (WHO 2000). • Understandability The indicator is easy to understand and to delimit.

112 Data sources Swedish Council for Information on Alcohol and other Drugs, CAN Danielsson M, Marklund U. Skolungdomars hälsovanor (Adolescent schoolchildren’s health habits) 1997/1998. National Institute of Public Health, Stockholm 2001

Data availability to the municipalities National data are not currently available on the municipal level, but many municipalities make their own surveys. As of 2000, information on smoking habits has been available from compulsory military service examinations (90 percent of the country’s 18-19-year- old boys). It is possible to localise these data to the region and the size of the places where the boys have grown up and to the boys’ employment and foreign background. The indicator has been excluded because data are not available on the municipal level.

Key references Bremberg, S. (2002). Social inequities in health in Swedish children and adolescents – a review). National Institute of Public Health, Stockholm

CAN (2002 a). Drogutvecklingen i Sverige. (Development of drug abuse in Sweden) Centralförbundet för alkohol- och narkotikauppplysning, Stockholm

CAN (2002 b). Skolelevers drogvanor 2002. (Drug abuse among schoolchildren 2002). Centralförbundet för alkohol- och narkotikaupplysning, Stockholm

Rigby, M., Köhler, L. (2002). Child Health Indicators of Life and Development (CHILD). European Commission, Luxembourg

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm

SOU (2001b). Barns och ungdomars välfärd. Antologi. Kommittén Välfärdsbokslut. (Welfare of children and young people. An antology). The National Committee of Welfare Balance Sheet Fritzes, Stockholm

WHO (2000c). Health and Health Behaviour among Young People, Copenhagen

Indicator Children’s and adolescents’ alcohol habits

Operational definition Proportion of children in the 9th grade who report intensive consumption of alcohol some time during a month or more frequently, per 100 boys and girls and total, and according to socioeconomic group

Background Alcohol is a risk factor for many health problems of a physical, mental and and social nature. Consumption is also closely associated with injuries and violent behaviour. Intensive consumption (intoxicated once a month or more often) is a considerably larger risk factor than moderate drinking.

113 Swedish young people generally consume less alcohol than young people on the Continent, but they drink great quantities on each occasion (“binge drinking”), particularly boys.

Consumption has also increased in recent years (CAN 2002). It is of particular concern that the number of adolescents treated in hospital for alcohol poisoning increased dramatically during the 1990s (National Board of Health and Welfare 2003). Consumption habits that have been laid in adolescent years have a tendency to continue in adulthood (Pape 1996, Public Health Report 2001).

Frequent use of alcohol is more common among adolescents of lower social class and is associated with other unhealthy behaviour such as use of tobacco and narcotics (Children’s and Adolescents’ Welfare 2001, Bremberg 2002, CAN 2002).

Preventive measures have a better effect if they are started early and in a local context.

Technical criteria • Useability The indicator is based on self-reports and is used in different health questionnaires. It is not included in routine reports in the entire population but is oriented toward a selection of adolescents in selected schools and toward boys registering for military duty. • Robustness The indicator has been used nationally and internationally for many years. It has been evaluated and judged and is considered to give a good picture of actual conditions (WHO 2000). • Understandability The indicator is easy to understand and delimit.

Data sources Swedish National Institute of Public Health Swedish Council for Information on Alcohol and other Drugs, CAN

Data availability to the municipalities Information on use of alcohol among schoolchildren and military servicemen is based on anonymous questionnaires and is not available on the municipal level, although many municipalities carry out their own surveys. Data from investigations of boys registering for military service (90 percent of the country’s 18-19-year-olds) can be localised to the region of residence and place where they have grown up and to boys’ employment and immigrant background. The indicator has been excluded because data are not available on the municipal level.

Key references Bremberg, S. (2002). Social inequities in health in Swedish children and adolescents – a review). National Institute of Public Health, Stockholm.

CAN (2002 a). Drogutvecklingen i Sverige. (Development of drug abuse in Sweden) Centralförbundet för alkohol- och narkotikauppplysning, Stockholm

CAN (2002 b). Skolelevers drogvanor 2002. (Drug abuse among schoolchildren 2002). Centralförbundet för alkohol- och narkotikaupplysning, Stockholm

114 Pape, H. & Hammer, T. (1996). How does young people's alcohol consumption change during the transition to early adulthood? A longitudinal study of changes at aggregate and individual level. Addiction, 91, 1345-1357.

Rigby, M., Köhler, L. (2002). Child Health Indicators of Life and Development (CHILD). European Commission, Luxembourg

Socialstyrelsen (2001b). Folkhälsorapport 2001. (Public Health Report 2001). Socialstyrelsen, Stockholm.

Socialstyrelsen (2003). Folkhälsa och sociala förhållanden 2002. (Public health and social conditions 2002). Socialstyrelsen, Stockholm.

SOU (2001b). Barns och ungdomars välfärd. Antologi. Kommittén Välfärdsbokslut. (Welfare of children and young people. An antology). The National Committee of Welfare Balance Sheet Fritzes, Stockholm.

WHO (2000c). Health and Health Behaviour among Young People. Copenhagen

Domain D. Service, support and health policies

Indicator Safe community

Operational definition The municipality has been named by the Swedish National Institute of Public Health/ The Rescue Services Agency as a safe community for the year in question.

Background A number of municipalities have focused on children’s welfare in a common project, Local Balance Sheet for Welfare (Balance Sheet for Welfare 1999), run by the Swedish Association of Local Authorities, the Federation of Swedish County Councils and the Swedish National Institute of Public Health. Work is now going on to develop a database of municipal base facts as a foundation for public health work, where attention is given to some areas dealing with children (National Institute of Public Health, PM May 2002). This base register includes “safe community”, where the municipality fulfils six criteria set up by WHO meaning that the municipality works in a structured way across sectors to prevent injuries in all risk groups. The criteria are now fulfilled by 14 of the country’s municipalities. The long term goal is that all municipalities will become safe and secure. The indicator is based on policy and only indirectly reflects children’s health. Long term, systematic and coordinated preventive efforts have however been shown to be a future recipe for reducing injuries in the population, for children and others (Sylwander 2001).

Technical criteria • Useability The indicator has already been developed and is used in a register that is being constructed for all municipalities. It covers all age groups, thus not

115 specifically children, but gives a good picture of the municipality’s efforts to prevent injuries, which also benefits children, an often high priority group. Administration, evaluation and follow-up of the register are now being done at the Rescue Services Agency in cooperation with the Child Safety Delegation. • Robustness A large number of studies have shown that conscious and structured preventive efforts in the municipality lead to a decreasing number of injuries in the population. Coordinated, inter-sectoral preventive efforts are a factor behind Sweden’s prominent position in injury prevention. The criteria for achieving the title of “safe community” are relatively vague and not entirely easy to define unambiguously, but work is being done to clarify them. • Understandability The criteria for the indicator need to be clarified and specified to make them easy to understand and be used in the municipalities.

Data sources Swedish National Institute of Public Health’s database (www.fhi.se/fakta/personskad6.asp) Rescue Services Agency.

Data availability to the municipalities Construction of the database is ongoing and the register is managed by the Rescue Services Agency. At this time (November 2002), 14 municipalities fulfil the criteria. The indicator has been excluded because the concept is vaguely defined and unclear in terms of its continuity and because the evidence for a direct relation to children’s health is weak.

Key references Folkhälsoinstitutet (1999). Institute of Public Health (1999) Välfärdsbokslut, (Balance sheet of welfare). National Institute of Public Health, Stockholm

Karolinska institutet (2002). Safe Communities, Stockholm (www.phs.ki.se/csp)

Sylwander, L. (2001). Child Accident Prevention. A Swedish success story. In: Protection Prevention Promotion. Development and future of Child Health Services (ed. G. Norvenius, L. Köhler, G. Wennergren, & J. Johansson). Nordic School of Public Health, Göteborg.

Indicators for international Safe Communities, according to WHO - A safe communicty shall have: 1. An infrastructure based on participation and cooperation, led by an inter-sectoral group that is responsible for promoting safety in its municipality; 2. A long term, permanent programme covering both sexes and all ages, environments and situations; 3. A programme directed at high risk groups and environments and a programme to promote safety in vulnerable groups; 4. A programme that documents the frequency and causes of injuries; 5. A method of evaluation that allows a judgement of its programmes and process and the effect of changes; 6. Running participation in national and international Safe Community networks)

116 Indicator Allergy-adapted municipality

Operational definition The municipality has been named an allergy-adapted municipality for the year by the Swedish National Institute of Public Health.

Background Allergies are the fastest growing health problem in Sweden and in many other Western countries. They are currently the most common cause of chronic disease among children and adolescents. The exact cause of the illness and its increase is not known, but conditions during childhood and the local environment are probably important.

Work is going on in Local Balance Sheets for Welfare (Balance Sheet for Welfare 1999), a cooperative project by the Swedish Association of Local Authorities, the Federation of County Councils and the Swedish National Institute of Public Health, to develop a database of municipal base facts as a foundation for public health work, where a few areas are also important for children (Swedish National Institute of Public Health, PM May 2002) (www.fhi.se/fakta/allergi10.asp). This base register includes “allergy-adapted municipality”, which means that the municipality has taken a political decision to work to prevent allergies, that there is an allergy committee with broad representation and an action plan for the work. At this time, 38 municipalities fulfil these criteria (www.fhi.se October 2002). The goal is that all the municipalities in the country will be allergy adapted.

The indicator is based on policy and reflects children’s health only in an indirect way. Long term, systematic, coordinated efforts for prevention and protection have however been judged to be important for the population, for children and others. (Nationella folkhälsokommitten 2000).

Technical criteria • Useability The indicator has already been defined and is used in a register that is being constructed for all municipalities. It covers all age groups, that is, not specifically children, but gives an overview of the municipality’s efforts to prevent allergies, which will also benefit children. • Robustness Systematic and structured work in the municipality leads to a greater awareness of allergies and their treatment and is intended to give greater opportunities for prevention. The criteria are formulated in a vague way, however, and are open to broad interpretations. There is no evidence as to the extent to which the work has in fact reduced allergies or their consequences. • Understandability The criteria in the indicator must be clarified and specified in order to become easy to understand and use in the municipalities.

Data sources Swedish National Institute of Public Health’s database (www.fhi.se/fakta/allergi10.asp)

117 Data availability to the municipalities Work is ongoing to construct the database. At this time (October2002) 38 municipalities fulfil the criteria. The indicator has been excluded because the criteria are not clear and the consequences for children’s health are doubtful.

Key references Folkhälsoinstitutet (1999). Institute of Public Health (1999) Välfärdsbokslut, (Balance sheet of welfare) National Institute of Public Health, Stockholm

Nationella folkhälsokommittén (2000). Hälsa på lika villkor- nationella mål för folkhälsan. Slutbetänkande. (Health and equity. National goals for the citizen’s health). Ministry of Health, Stockholm.

Indicator Action programme against bullying in schools

Operational definition Proportion of schoolchildren who attend schools with a written, defined, established action programme, with follow up, to work against bullying, per 100 of all schoolchildren

Background Bullying in schools is a well known risk factor for problems with wellbeing. Children who are bullied more often consider suicide, have physical and mental disorders and perform poorly in school. According to a few studies, the frequency of bullying has remained largely unchanged; 18 percent of boys 11-15 years report that they have been bullied at least one time and 2-3 percent report that they have been bullied every week. The frequency among girls is 12 percent and 0-2 percent, respectively. The figures were the same four years earlier (Public Health Report 2001). Other studies show an increase. For example, bullying in all the Nordic countries increased between 1984 and 1996, most among children between 7 and 12 years of age, but even among children 2- 6 years of age (Nordhagen 2000). Interestingly enough, it is found that children who bully other children run a greater risk of being bullied themselves as compared with children who do not bully. All studies have shown that children who are bullied come to a greater extent from socially, economically and culturally vulnerable families. Several studies indicate that children with functional disorders and handicapped children are at greater risk (Nordhagen 2000; Janson 2001).

The Swedish National Agency for Education, as the national authority is responsible for this area, has followed developments and identified bullying and other offensive treatment in schools as a high priority area for countermeasures (Swedish National Agency for Education 1999, 2000). Well tested and effective measures are available, both in Sweden and in other countries. A written action programme against bullying is a first step toward solving the problem and should exist in all schools, according to the Swedish National Agency for Education.

118 Technical criteria • Useability The indicator has been proposed to be included with the same definition in the European system for health surveillance of children on the national level. • Robustness Established, written and evaluated action programmes in a school are no guarantee that bullying does not take place, but they facilitate the management of the problem, both for prevention and for intervention. They also show children and parents that the school takes the problem seriously. • Understandability The indicator is a policy indicator that is easy to understand and collect.

Data sources Swedish National Agency for Education

Data availability to the municipalities The Swedish National Agency for Education follows the development in the municipalities but comparative data are not yet available in all municipalities.

Key references Janson, S. (2001). En rapport om kroppslig bestraffning och annan misshandel i Sverige i slutet av 1900-talet.(Physical punishment and other abuse in Sweden at the end of the 1990’s). Fritzes, Stockholm

Nordhagen, R. (2000). Mobbning och mobbare. (Bullying and being bullied)In: Det är bra men kan bli ännu bättre. En studie av barns hälsa och välfärd i de fem nordiska länderna från 1984 till 1996 (It is well, but could be better. A study of children’s health and weellbeing in the five Nordic countries from 1984 to 1996) (ed. L. Köhler). Nordic School of Public Health, Göteborg

Skolverket (1999). Nationella kvalitetsgranskningar 1999. (National quality assessments 1999) Skolverket, Stockholm

Socialstyrelsen (2001a). (Yearbook of health services 2001). Socialstyrelsen, Stockholm

Appendix 3 Final indicators

Domain B Health status and well-being B1. Proportion of children 0-17 years hospitalised for external injuries Domain C. Health determinants (Risk- and protective factors) C1. Proportion of children exposed to tobacco while foetuses C2. Proportion of children with birthweight under 2 500 g C3. Proportion of children breastfed at 4 months C4. Proportion of children vaccinated against morbilli, mumps and rubella (MPR) C5. Proportion of teenage girls with abortion

119 Save the Children fights for children’s right. We deliver immediate and lasting improvements to children’s lives worldwide.

107 88 Stockholm SE-Sweden Visiting address: Landsvägen 39, Sundbyberg Phone: + 46 8 698 90 00 Fax: +46 8 698 90 10 [email protected]