Recall of drug utilization depends on subtle structural questionnaire characteristics
Objective To test the hypothesis that questionnaire organization affects the recall of antimalarial drugs utilization. Setting Maputo, Mozambique. Methods Treatment of last malaria episode was assessed using two alternative versions (A and B) of a questionnaire, which differed only by the order each proposed drug was presented in the response options (version A: quinine and most frequently used drugs presented first; version B: less frequently used drugs first and quinine at the end). Questionnaires A or B were randomly assigned to each of 32 classes in a private University in Maputo, Mozambique. Within each classroom all subjects received the same questionnaire version, and a similar number of participants fulfilled questionnaires A (n = 249) and B (n = 255). Main outcome measures Recall of the antimalarial drugs utilization in a previous malaria episode. Results Mefloquine and clindamycin were not used by any subject in the last malaria episode. The overall recall of quinine utilization was higher with questionnaire A (19.5% vs. 11.6%, P = 0.006) and the use of artemisinine/artesunate was reported more frequently by subjects inquired with questionnaire B (16.5% vs. 7.3%, P = 0.012). When considering subjects reporting more than one malaria episode, the recall of quinine utilization was higher with questionnaire A (20.0% vs. 6.4%, P = 0.004), and the use of artemisinine/artesunate was reported more frequently by subjects inquired with questionnaire B (18.3% vs. 8.8%, P = 0.069). No differences were observed among those having had only one malaria episode, neither for quinine (22.2% vs. 20.0%, P = 0.807) or artemisinine/artesunate (5.0% vs. 6.7%, P = 0.701). Conclusion The structure of the questionnaire used to collect self-reported information about antimalarial treatments influences the recall, even when close ended questions asking for specific drugs are used. Among subjects having been treated for malaria more than once, the first antimalarial drugs being asked were more likely to be selected.
KeywordsData collection Drug utilization Epidemiologic methods Malaria Mozambique Questionnaire
This study was conducted with the support of the ISCTEM (Higher Institute of Sciences and Technology from Mozambique).
Conflicts of interest The authors declare that there are no conflicts of interest to disclose.
- 11.International Travel and Health: situation as on 1 January 2005. Geneva: World Health Organization; 2005. ISBN: 92 4 158036 4.Google Scholar
- 15.Dean AG, Dean JA, Coulombier D, Burton AH, Brendel KA, Smith DC. Epi Info, version 6.04d: a word processing, database, and statistics for epidemiology in microcomputers. Atlanta, USA: Centers for Disease Control and Prevention, Geneva, Switzerland: World Health Organization; 1996.Google Scholar
- 16.Bland M. Cluster randomised trials in the medical literature [document on the Internet]. 2003 [updated 2005 Jan 12; cited 01/07/2007]; Available from: http://www-users.york.ac.uk/∼mb55/talks/clusml.htm.