PEDIATRICS Vol. 111 No. 5 May 2003, pp. 1125
Behavior Change Counseling in the Emergency Department to Reduce Injury Risk: A Randomized, Controlled Trial
To the Editor.—The study by Johnston et al1 demonstrates that the emergency department can be a venue for behavior change counseling if the right resources are available. Adolescents continue to rely heavily on emergency services for their primary care needs.2 Unfortunately, this is not the ideal because the primary care office is the best setting to practice preventive medicine, especially in adolescents.
I have 2 comments about the study. First, the authors found that there was a positive change in bicycle helmet use and seatbelt use in the intervention group compared with the controls at both the 3- and 6-month follow-up points.1 However, it is unclear if primary care providers had seen these patients in that interim period. Follow-up visits for patients seen in the emergency department (for future removal, etc) become the primary care provider’s "teachable moment" to reinforce safety issues with patients and to provide behavior change counseling. In addition, the authors did not assess parental involvement other than for purposes of consent. This was to ensure confidentiality and honest answers on the questionnaire. However, studies show that there is a strong association between parental involvement or parental pressure and bike helmet use.3,4 It would be interesting to see if the participants who changed their behavior had parental support after their injury. These are 2 important confounding variables that were overlooked that may have influenced the relative risk reduction in the study.
Injury prevention in children and adolescents takes a multidisciplinary approach using parents, physicians (in any setting), community education programs, legislation, and school programs.
Kari M. Ketvertis, MD
UPMC St Margaret Family Practice Program
Pittsburgh, PA 15215
REFERENCES
- Johnston BD, Rivara FP, Droesch RM, Dunn C, Copass MK. Behavior change counseling in the emergency department to reduce injury risk: a randomized, controlled trial.
Pediatrics.2002; 110
:267
–274
[Abstract/Free Full Text] - Lehmann CU, Barr J, Kelly PJ. Emergency department utilization by adolescents. J Adolesc Health.1994; 15 :485 –490[CrossRef][Web of Science][Medline]
- Berg P, Westerling R. Bicycle helmet use among schoolchildren—the influence of parental involvement and children’s attitudes.
Inj Prev.2001; 7
:218
–222
[Abstract/Free Full Text] - Finch CF. Teenagers’ attitudes towards bicycle helmets three years after the introduction of mandatory wearing.
Inj Prev.1996; 2
:126
–130
[Abstract/Free Full Text]
In Reply.—
We appreciate the opportunity to respond to Dr Ketvertis’ thoughtful comments regarding our article. We, too, believe that preventive care for adolescents should take place in multiple venues. In this case, we chose the emergency department as a setting to target teens at high risk for reinjury.
As Dr Ketvertis notes, we did not measure or adjust for contact with primary care providers between initial injury and follow-up assessments. Nor did we explicitly involve parents to reinforce our behavior change messages. Although these variables might explain why some individuals within the intervention group responded to counseling while others did not, we do not believe that failure to measure or model these processes biases our conclusion that the intervention was efficacious.
The analysis plan for a study should be specified during the design of the trial.1 We relied on individual randomization to ensure that measured and unmeasured covariates were approximately equal in their distribution across treatment groups. Thus, although we did not measure contact with primary care providers, we can assume that—on average—individuals who received the intervention were as likely to see their primary care physician as were those who received usual care. Similarly, we believe that parental involvement and reinforcement should have been balanced between groups.
Of course, it may be that receipt of the intervention led to participants having more contact with primary care providers, seeking out advice about health behavior, or discussing these concerns with parents. In this case, measuring and controlling for these contacts would, in effect, adjust away an important part of an efficacious intervention.
Additional research might elucidate the importance of these cofactors in the success of behavior change counseling among adolescents. If subsequent investigators believe that these covariates are important because of their hypothesized correlation with outcome, explicit efforts to balance these factors at randomization (eg, stratification) should be considered.2
Brian D. Johnston, MD, MPH*
Frederick P. Rivara, MD, MPH*,
* Departments of Pediatrics Epidemiology
University of Washington
Seattle, WA 98104
REFERENCES
- Assmann SF, Pocock SJ, Enos LE, Kasten LE. Subgroup analysis and other (mis)uses of baseline-data in clinical trials. Lancet.2000; 355 :1064 –1069[CrossRef][Web of Science][Medline]
- Raab GM, Day S, Sales J. How to select covariates to include in the analysis of a clinical trial. Control Clin Trials.2000; 21 :330 –342[CrossRef][Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2003 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
