Development and Validation of a Knowledge, Attitude and Practice Questionnaire on Antibiotic Use in Arabic and French Languages in Lebanon

Abstract

Objectives: Validated knowledge-attitude-practice (KAP) questionnaires are essential to design and evaluate intervention programs on antibiotic use. Recently, we validated the first KAP questionnaire on antibiotics in Spain. Cross-cultural adaptation and validation of research tools increase their universal usefulness. Here, we aimed to validate the questionnaire in a developing country with different socioeconomic characteristics from that of Spain.

Methods: We translated the previously developed KAP-questionnaire into Arabic and French, tailored it and then validated it in adult population in Lebanon. The item content validity index (I-CVI), scale content validity index (S-CVI/Ave) and modified Kappa (k*) were calculated. The construct validity of the questionnaire was evaluated using confirmatory factorial analysis (CFA, N = 1460) and its reliability was assessed using intraclass correlation coefficients (ICC, N = 100) and Cronbach's alpha statistic.

Results: ICV-I (>0.78), k* (equal to ICV-I for all items) and S-CVI/Ave (≥0.95) confirmed the questionnaire content validity. Pilot testing (N = 40) and face validity showed the understandability of the questionnaire by the population. Test-retest reliability analysis (N = 100) yielded ICC ≥ 0.59 for all knowledge and attitude items, showing the capacity of the questionnaire to generate reproducible results. CFA evidenced adequate fit of the chosen model, thus establishing the construct validity of the questionnaire (root mean squared error approximation = 0.053, standardized root mean square residual = 0.045, comparative fit index = 0.92 and Tucker-Lewis index = 0.90). The questionnaire showed an acceptable internal reliability (Cronbach's alpha = 0.62) and was highly accepted in Lebanon (response rate = 96% and item response rates ≥ 94%).

Conclusions: The validity of the KAP-questionnaire on antibiotics in Arabic and French was demonstrated in Lebanon.

Keywords: Arabic; French; Lebanon; antibiotics; knowledge–attitude–practice; questionnaire; validation.

Figure 1

Representation of the knowledge and attitudes model selected by CFA analysis (Model 4). The Figure represents each of the three factors (knowledge, personal attitudes towards antibiotics and patient attitudes towards healthcare provider) with its respective standardized item loadings and item residuals. Double-sided arrows represent correlations between the variables. Single headed arrows represent the correlation of each item with its corresponding factor(s).