Psychometric evaluation of the French version of the questionnaire attitudes towards morphine use; a cross-sectional study in Valais, Switzerland
1 Hôpital de Sion, Avenue Grand-Champsec 80, Case Postale 736, Sion, 1951, Switzerland
2 Haute Ecole de Santé La Source, Avenue Vinet 30, Lausanne, 1004, Switzerland
3 Haute Ecole de Santé Vaud, Unité de recherche en Santé, Av. de Beaumont 21, Lausanne, 1011, Switzerland
4 Universidade Católica Portuguesa, Instituto de Ciências da Saúde, Palma de Cima, Lisboa, 1649-023, Portugal
5 Institute of Social and Preventive Medicine, University Hospital of Lausanne, route de la Corniche 10, Lausanne, 1010, Switzerland
BMC Nursing 2014, 13:1 doi:10.1186/1472-6955-13-1Published: 10 January 2014
In Switzerland, nurses are allowed to prescribe and administer morphine in emergency situations without a doctor. Still, nurses and other health professionals are often reluctant to prescribe and administer morphine for pain management in patients. No valid French-speaking instrument is available in Switzerland to assess the attitudes of nurses and other health professionals towards the prescription and administration of morphine. In this study, we evaluated the psychometric properties of the French version of the questionnaire “Attitudes towards morphine use”.
The instrument was derived from an Italian version. Forward and back translations of the questionnaire were performed. Item analysis and construct validity were assessed between April and December 2010 in a cross sectional study including five Swiss hospitals in a sample of 588 health professionals (533 nurses, mean age 38.3 ± 10.2 years). Thirty subjects participated in test-retest reliability.
The time to complete the instrument ranged between 12 and 15 minutes and neither floor nor ceiling effect were found. The initial 24-item instrument showed an intraclass correlation (ICC) of 0.69 (95% CI: 0.64 to 0.73, P < 0.001), and a Cronbach’s α of 0.700. Factor analysis led to a six-component solution explaining 52.4% of the total variance. After excluding five items, the shortened version showed an ICC of 0.74 (95% CI, 0.70 to 0.77, P < 0.001) and a Cronbach’s α of 0.741. Factor analysis led to a five-component solution explaining 54.3% of the total variance. The five components were named “risk of addiction/dependence”; “operational reasons for not using morphine”; “risk of escalation”; “other (non-dependence) risks” and “external (non-operational) reasons”. In test-retest, the shortened instrument showed an ICC of 0.797 (95% CI, 0.630 to 0.911, P < 0.001) and a Cronbach’s α of 0.797.
The 19-item shortened instrument assessing attitudes towards the prescription and administration of morphine showed adequate content and construct validity.