Development of a population survey concerning papanicolaou smear screening for cervical disease

We set out to design and evaluate a questionnaire for determining women's knowledge, attitudes, behaviors, and experience with cervical screening.

A literature search was conducted using key terms such as Pap (Papanicolaou) smear and attitudes. No questionnaire in the literature addressed all the domains of knowledge, attitudes, behavior, and experience. The expert panel method was employed, by which eight women representing the general public, experts in cervical disease, or experts in questionnaire design developed a pool of questions. Item reduction was accomplished by consensual agreement and pretesting. The questionnaire was pilot-tested for acceptability, feasibility, comprehension, test-retest reliability, and face and content validity.

Fifty-one women completed the questionnaire. The mean age was 42 ± 13 years. Of these respondents, 43% were college-educated and 28% each had a high school or university education. All women were parous (10% had one child, 31% had two, 18% had three, 33% had four, and 8% had five children). The questionnaire was completed on two occasions. Test-retest reliability was assessed using Cohen's kappa statistic, Wilcoxon's test, Kendall's tau-b, or Pearson's correlation coefficient. These were statistically significant at p < .05 for 13 of 24 of the knowledge questions, 24 of 30 experience questions, 16 of 20 attitude questions, all behavior questions, and 3 of 4 registry questions. Some questions could not be evaluated because of a lack of variance. Internal consistency was computed using Cronbach's alpha: registry (0.9), experience (0.7), knowledge (0.6), and attitude (0.2). External validity was assessed by comparing the date of a woman's last Pap smear as reported by the woman with that recorded on the physician's record. The questionnaire was rated as easily understandable by 90% of respondents and was rated as "acceptable" by 95%. Women were very knowledgeable about the purpose of Pap smears (98%) and were reasonably knowledgeable about who needs such smears (>85%). In the domain of knowledge, four questions addressed risk factors for cervical disease. In general, these questions were answered with poor accuracy (range, 37%-60%). Women believed that the experience of having a Pap smear would be improved if the staff knocked before entering the room (87%), a warm speculum was used (86%) and the physician explained the procedure (75%). Eightyeight percent were willing to participate in a cervical screening registry.

The psychometric properties of this questionnaire have been described. Two of the attitudinal questions were poorly understood by the respondents, and the items within the questions did not appear to address a single concept (low internal validity). Health care professionals could take simple measures to ensure that the experience of having a Pap smear is more agreeable.