Outcomes for control patients referred to a pediatric asthma outreach program: an example of the Hawthorne effect

A study was designed to determine whether identification of high risk for exacerbations of asthma based on pediatrician concern, emergency department visits, or hospital stays results in a decrease of resource utilization because of referral to an asthma outreach program even if the intervention does not take place. The findings for such a group were compared with those for a group who did undergo intervention with an asthma outreach program. Fifty-six patients 1 to 14 years of age were assigned to one of two groups. The control group (those who did not undergo intervention) had consistent but not statistically significant reductions in utilization of emergency visits, hospitalizations, and dollars spent (21%, 24%, and 32%, respectively). The group who underwent intervention with the asthma outreach program had large and statistically significant decreases in the same parameters (emergency visits, 60%, P = 0.001; hospital stays, 74%, P = 0.008; dollars spent, 72%, P = 0.004). However, the apparently insignificant effect of the reductions in utilization by the control group substantially altered interpretation of the outcomes of the study. Cost savings were reduced from $11.69 per dollar spent on intervention to $6.49 per dollar spent. In before-and-after studies such as those typically conducted during continuous quality improvement projects, which typically do not have control groups, investigators need to consider control group effects when they assess the results of intervention.