Psychometric properties of the Blood-borne Virus Transmission Risk Assessment Questionnaire, BBV-TRAQ

FRANÇAIS :
Les propriétés psychométriques du BBV-TRAQ sont évaluées sur un échantillon de 209 usagers actuels de drogues injectables. Les résultats montrent une fiabilité élevée (interne, de test-retest et entre observateurs) et que cet outil à une couverture suffisante pour évaluer les pratiques à risques liées à l'injection, aux pratiques sexuelles et à la pénétration dermique.
ENGLISH :
Aims: To develop a standard measure of blood-borne virus transmission risk behaviour, and examine the underlying psychometric properties. Design: The Blood-borne Virus Transmission Risk Assessment Questionnaire (BBV-TRAQ) was developed over three consecutive phases of the original BBV- TRAQ study in adherence to classical scale development procedures, culminating in the recruitment of a development sample of current injecting drug users via convenience and snowball sampling. Setting: Needle and syringe programmes (NSPs), medical clinics, alcohol/drug agencies, peer-based and outreach organizations across inner and outer metropolitan Melbourne. Participants: Two hundred and nine current injecting drug users. The mean age was 27 years, 68% were male, 65% unemployed, 36% with prison history and 25%, in methadone maintenance. Measurements: BBV-TRAQ items cover specific injecting, sexual and skin penetration risk practices. BBV-TRAQ characteristics were assessed via measures of internal and test-retest reliability; collateral validation: and principal components analyses. Findings: The BBV-TRAQ has satisfactory psychometric properties. Internal (a=0.87), test-retest (r=0.84) and inter-observer reliability results were high, suggesting that the instrument provides a reliable measure of BBV risk behaviour and is reliable over time and across interviewers. A principal components analysis with varimax rotation produced a parsimonious factor solution despite modest communality, and indicated that three factors (injecting, sex and skin penetration/hygiene risks) are required to describe BBV risk behaviour. Conclusions: The BBV-TRAQ is reliable and represents the first risk assessment tool to incorporate sufficient coverage of injecting, sex and other skin penetration risk practices to be considered truly content valid. The questionnaire is indicated for use in addictions research, clinical, peer education and BBV risk behaviour surveillance settings. (Author' s abstract)