The level of participation in a family planning survey using personal interviews

In an effort to obtain baseline data on family planning use in the community and the population of women at high risk prior to implementation of a family planning program in a primary care clinic in Spain letters were sent to a sample of women requesting that they appear at the clinic for a brief interview. This work analyzes the relatively poor response to the initial letter and assesses the usefulness of this form of gathering information in a community of low educational level. 271 women aged 15-19 were randomly selected from the municipal census of Benacazon a rural community in the Basic Health Zone of Sanlucar la Mayor in Seville. Women not attending the 1st interview were sent 2nd and 3rd letters which also offered them the opportunity of changing the proposed time of the interview to an hour more convenient for them. 24.8% of the women attended after the 1st letter. 25.1% of those sent a 2nd letter and 28.9% of those sent a 3rd letter also attended. 40% did not attend after 3 letters. Reasons given by those attending after a 2nd or 3rd letter for not attending after the 1st letter were work or study in 29.9% of cases nonreceipt of the letter in 22.9% another commitment in 18.1% lack of interest in 26.5% absence in 6% and other reasons in 3.6%. Characteristics of those attending an interview after 1-3 letters and those not responding were compared using data from the municipal census. 133 of the women were illiterate or had no school attendance and another 63 had primary school only. 19.9% of the total sample were students and 10.3% worked. Differences in age educational level and employment between those attending and not attending were statistically significant but were not considered relevant. Excluding the 45.8% of nonattenders after the 1st letter who worked went to school or did not receive the letter the nonattendance was believed to be due in large part to the low educational level of the population and the embarrassing theme of the interview which was not described in the letter but was communicated by word of mouth in this small community. This type of survey was believed to be appropriate for collecting information in a rural community. It avoids the bias of clinic-based studies which do not include the nonattending population. Follow-up letters were an effective means of increasing the response rate and thus avoiding the costly home interviews that would have been required to achieve a higher response rate.