Tuesday 14th July      Wednesday 15th July      Thursday 16th July      Friday 17th July     

Friday 17th July, 13:00 - 14:30 Room: HT-102

Health inequalities between health survey participants and non-participants
Convenor	Dr Hanna Tolonen (National Institute for Health and Welfare, Helsinki, Finland )

Session Details

Health surveys, especially health examination surveys where physical measurements are conducted and biological samples are collected, are important data sources about health of the population. This information can be used for evidence-based policy making, planning and evaluation of prevention programmes as well as for research.

The participation rates of health surveys have declined in past decades similarly to other surveys. There is also evidence that health and risk factor profiles of survey participants and non-participants differ. Previous studies have shown that mortality of non-participants is twice as high as that of participants, and for example smoking related mortality is three times higher among non-participants. These observed health inequalities between health survey participants and non-participants imply that survey non-participants have more diseases and worse health behaviors, such as smoking, than participants. Better understanding of these differences is needed, to ensure that our interpretation of survey results for evidence-based policy making and research are accurate.

The aim of this session is to present research findings about the differences in health and risk factor profiles of survey participants and non-participants, with special focus on health surveys.

Paper Details

1. Non-respondents' health vs. repondents' health in Slovenian health surveys
Mrs Darja Lavtar (National Institute of Public Health (NIJZ), Slovenia)
Mrs Tina Zupanič (National Institute of Public Health (NIJZ), Slovenia)

In the last decade Slovenian National Institute of Public Health (NIJZ) implemented various national health surveys. In three surveys among adult population (EHIS 2007, EHSIS 2012, EHIS 2014) that were implemented by face-to-face or mixed mode, the responses on three specific health questions among non-respondents who refused to participate in the whole survey, but were willing to give answers to three questions of Minimum European Health Module, were investigated in order to determine whether the answers of non-respondents differentiate from answers of survey respondents. These comparisons from national health surveys will be presented in this article.

2. Health survey non-participants have more hospitalizations during the survey period than participants
Dr Hanna Tolonen (National Institute for Health and Welfare, Finland)
Mr Kennet Harald (National Institute for Health and Welfare, Finland)
Dr Satu Männistö (National Institute for Health and Welfare, Finland)

In health examinations surveys (HES), physical measurements are conducted and biological samples collected additional to questionnaire information. In Finland, the National FINRISK study has been conducted among adult population every 5 years since 1972. In these analyses surveys, from 1992 till 2012 were used. The survey samples were linked to the national Health Care Register to obtain hospitalization periods for both survey participants and non-participants. In all surveys, non-participants had significantly more hospitalizations during the 3-month survey period than participants in all surveys from 1992 to 2012.

3. Strategies to analyze representativeness of net sample in health examination surveys. Analyses from the "German Health Interview and Examination Survey for Adults" DEGS1
Dr Antje Goesswald (Robert Koch Institute)
Mr Robert Hoffmann (Robert Koch Institute)
Mr Panagiotis Kamtsiuris (Robert Koch Institute)

Willingness to participate in health interview and examination surveys (HES) is decreasing in many countries. Responders and non-responders are known to differ in socio-demographic characteristics. Systematic differences in health behaviour and health status are of special interest in order to assess the representativeness of health survey results, but generally only a little amount of information about non-responders is available. Therefore different strategies of analysis have to be chosen to calculate the quality of the net sample. Three methods of analysis will be presented based on data of the German Health Interview and Examination Survey for Adults (DEGS1).

4. Determinants of second-stage recruitment in the Belgian Health Interview Survey
Mr Stefaan Demarest (Scientific Institute of Public Health)
Dr Jean Tafforeau (Scientific Institute of Public Health)
Dr Johan Van Der Heyden (Scientific Institute of Public Health)

In this study multivariate logistic regression was used to assess the socio-demographic, socio-economic and health determinants associated with a second-stage recruitment for a dental health examination survey (recruitment rate: 35%) among households that previously participated in a health interview survey. Only age (of the households’ reference partner) and educational level were predictive for the consent to participate in the examination study. Second stage recruitment for a similar survey reduces the financial burden to conduct a survey, enables to study the characteristics of non-responders, but does not yield in higher consent-to-participate levels.