<p><strong>CONTEXT: </strong>Pediatric oncology patients often receive prolonged courses of opioids, which can result in constipation.</p>

<p><strong>OBJECTIVES: </strong>Comparing patients who received senna matched with similar patients who received other oral bowel medications, determine the subsequent risk of "problematic constipation," assessed as the occurrence of the surrogate markers of receiving an enema, escalation of oral bowel medications, and abdominal radiographic imaging.</p>

<p><strong>METHODS: </strong>This was a retrospective cohort study of hospitalized pediatric oncology patients less than 21&nbsp;years of age in 78 children's and adult hospitals between 2006 and 2011 who were started on seven consecutive days or more of opioid therapy and were started on an oral bowel medication within the first two days of opioid therapy. Clinically detailed administrative data were used from the Pediatric Health Information System and the Premier Perspective Database. After performing propensity score matching of similar patients who started senna and who started a different oral bowel medication, Cox regression modeling was used to compare the subsequent hazard of the surrogate markers.</p>

<p><strong>RESULTS: </strong>The final matched sample of 586 patients averaged 11.5&nbsp;years of age (range 0-20&nbsp;years); 41.8% (n&nbsp;=&nbsp;245) had blood cancer, 50.3% (n&nbsp;=&nbsp;295) had solid tumor cancer, and 7.9% (n&nbsp;=&nbsp;46) had brain cancer. Initiating senna therapy within two days of starting the prolonged opioid course, compared with initiating another oral bowel medication, was significantly associated with a lower hazard during the ensuing five days for receipt of an enema (hazard ratio [HR], 0.31; 95% CI, 0.11-0.91) or undergoing abdominal radiographic imaging (HR, 0.74; 95% CI, 0.55-0.98), was marginally associated with a lower hazard of oral bowel medicine escalation (HR, 0.78; 95% CI, 0.59-1.03), and overall was significantly associated with a lower hazard of the composite end point of problematic constipation (HR, 0.70; 95% CI, 0.56-0.88).</p>

<p><strong>CONCLUSION: </strong>Initiating senna therapy, compared with other oral bowel medications, diminishes the subsequent risk of surrogate markers of problematic constipation in this population.</p>