Prompting discussions of youth violence using electronic previsit questionnaires in primary care: a cluster randomized trial

Abstract

Objective: Youth violence (YV) screening during primary care visits is not routinely performed. Electronic previsit questionnaires (PVQs) are viewed favorably by adolescents and can prompt disclosure and discussion of sensitive health topics. This study aimed to determine the efficacy of an electronic PVQ in prompting YV discussions.

Methods: A 4-month cluster-randomized controlled trial was conducted in a large urban academic primary care clinic. The clinic's 4 practice groups were randomized to intervention or control assignment. A consecutive sample of adolescents aged 13 to 19 years presenting for annual visits were recruited. Participants completed an electronic PVQ (TickiT) either with (intervention) or without (control) YV questions. PVQ results were delivered to physicians before the visit. The frequency of YV discussions was measured using exit surveys of adolescents. Patients who reported YV discussion rated the helpfulness of the discussion. Multilevel mixed effect logistic regression was conducted to compare likelihood of YV discussion between intervention and control groups.

Results: A total of 183 adolescents (90% of eligible) participated. Overall, 30% of adolescents reported some YV involvement. Sixty-five percent of the intervention group and 42% of the control group reported discussing YV during their visit. Thirty-one percent of adolescents in the intervention group who disclosed YV involvement reported not having a YV discussion. The intervention group had 2.6 (95% confidence interval 1.2-5.6) times the odds of discussing YV. Sixty-six percent of adolescents who discussed YV with their doctor rated the discussion as very helpful.

Conclusions: An electronic PVQ with items related to YV is acceptable and feasible, and it significantly improves frequency of patient-provider YV discussion.

Keywords: doctor–patient communication; health risk behaviors; youth violence.