| Popis: |
Introduction. A pilot study on the feasibility of the establishment of a serumbank of a representative sample of the Dutch general population was carried out. The aims of this so-called 'PIENTER-project' were to estimate: 1. age-specific immunity of the general population against childhood diseases of the Netherlands Immunization Programme, 2. the incidence of infectious diseases with a frequent subclinical course 3, the prevalence of serum-determinants of other illnesses. This report describes the results of the non-response study which was performed to obtain insight in the reasons for nonparticipation and to test the representativity. Design. Nonresponse study as part of a cross-sectional population survey. Methods. 510 persons in the age-strata (0, 1-4, 5-9 to 75-79 years) were randomly selected out of four municipalities in het Province of Utrecht. The participants have been asked to fill in a questionnaire and to visit a local health service to give some blood. The nonparticipants were contacted by telephone (or mail). In this nonresponsestudy data were obtained on the reasons for nonparticipation, willingness to visit additional hours to give blood. Information on religion, vaccination history, self-perception on health status, country of nationality and level of education were collected with a short questionnaire. Data obtained from the registries of the municipalities were age, sex, nationality and marital status. Part of the eligible persons received a written reminder. Differences between participants and nonparticipants were tested in logistic regression analyses. To get information on the immunity of nonparticipants and participants, the immunity against hepatitis A and measles was weighted by the variables in the logistic regression model. Results. Four groups of (non)participants were distinguished: 1. initial participants (n=714, 34%) ; 2. additional participants who visited the additional hours of the local health service to give blood (n=113, 6%) ; 3. partial nonparticipants from whom questionnaire data were obtained (n=667, 27%) ; 4. absolute nonparticipants from whom information on data out of the registry of the municipality and some times the reason for nonparticipation were available (n=546, 27%). 1021 of the 1326 (77%) nonparticipants were reached by telephone ; 75% of them were reached in two, 95% in five attempts. Information was collected on 58.9% of nonparticipants. The reason for nonparticipation were very diverse. 36% of the reasons given were considered impressionable. The logistic regression analyses showed that childeren aged 0-4 years, men, single persons, persons who didn't receive a reminder by mail, persons with low education level, persons with bad perception of health status were more frequent nonparticipants. Persons who belong to a religion from which it is known that vaccination is refused participated less frequent, particularly when they had indeed not participated in the Netherlands Immunization Programme. Persons aged 5-14 years participated more frequent. The estimation of the immunity against hepatitis A and measles were not biased by the selection due to nonparticipation. Conclusion. It is possible to get information on nonparticipants by telephone. The non-participants gave usefull information on possible adaptions to increase the participation rate in further surveys. A telephone reminder turned out to have a possitive influence on the participation rate. Although nonresponse selection existed, the estimation on the immunity against hepatitis A and measles were not biased by this nonparticipation selection. The results of the nonresponse study can be used to correct the measured seroprevalence of other diseases by selective nonparticipation. |
