Opioid Prescription Practices at Discharge and 30-Day Returns in Children with Sickle Cell Disease and Pain

Background: Acute pain episodes in children with sickle cell disease (SCD) represent a leading cause of readmissions. Lack of evidence-based guidelines contributes to variability in the management of children with SCD after hospitalization for acute pain. However, the impact of treatment variability on readmissions in this population has not been well studied. We examined prescription practices at time of discharge in children with SCD and acute pain to determine their impact on incidence of 30-day Emergency Department (ED) revisits and readmissions. We hypothesize that prescription of scheduled opioids after discharge for acute pain episodes is associated with lower 30-day ED revisits and readmissions when compared to the prescription of as needed or no opioids.

Methods: We performed a single-institution, retrospective study of all patients with SCD aged 7 to 21 who were hospitalized or discharged from the ED with a diagnosis of acute pain episode from June 2009 to May 2014. Patients with ≥ 8 hospitalizations in a single year were excluded from the analysis. We reviewed demographic, treatment and discharge data from each encounter along with 30-day returns, defined as ED revisits and readmissions within 30 days of discharge. Between-groups comparisons for continuous variables were performed using the Mann-Whitney U test and categorical variables using the Chi-Square or Fisher's Exact test where appropriate. Independent predictors of 30-day returns were evaluated using multivariable logistic regression (IBM, SPSS v22).

Results: We reviewed a total of 290 encounters (n=110, ED discharge; n=180, hospital discharge) in 97 patients (51% male, median 11.9 years old) during the 5-year period. Patients had hemoglobin SS or S/ß⁰ thalassemia in 209/290 (72%) encounters and were on hydroxyurea in 104/290 (36%). For hospitalizations (median length of stay 4 days), patients were treated with parenteral opioids in 159/180 (88%) encounters, most commonly by patient-controlled analgesia. Patients were prescribed opioids at the time of discharge in 259/290 (89%) encounters, more commonly at discharge from the hospital versus ED (96 vs. 79%, p < 0.01). The remaining patients were either prescribed non-steroidal anti-inflammatory drugs (NSAIDs) or a prescription was not given. Compared to as needed or no therapy, scheduled opioids were more frequently prescribed when discharged from the hospital versus ED (61 vs. 19%, p < 0.01). In total, 56/290 (19%) encounters resulted in 30-day ED revisits or readmissions after discharge from the hospital or ED. Compared to hospital discharges, discharge from the ED for acute pain was associated with a higher incidence of 30-day returns (OR = 2.7 [95% CI 1.5, 4.8], p < 0.01). Median number of days to return was also shorter for patients discharged from the ED versus hospital (4 vs. 11.5 days, p < 0.01). In general, we found no association between prescribed opioid frequency and incidence of 30-day returns. Prescription of scheduled opioids was similar between encounters that did or did not result in a 30-day ED revisit or readmission (46 vs. 48%, p = 0.8). Using multivariable logistic regression, we examined if other factors were associated with an increased incidence of 30-day returns. After discharge from the ED, the prescription of NSAIDs only, without opioids, was independently associated with a higher incidence of 30-day ED revisits but not readmissions. Neither 30-day ED revisits nor readmissions was affected by age, sex, genotype, or hydroxyurea use. After discharge from the hospital, none of these factors, including hospital length of stay or use of patient-controlled analgesia, were independently associated with 30-day ED revisits or readmissions.

Conclusions: Variability exists in opioid prescription practices after discharge from the ED or hospital in children with SCD and acute pain episodes. Prescribed opioid frequency did not impact overall 30-day returns after discharge from the ED or hospital. After discharge from the ED, however, prescription of NSAIDs only, without opioids, was an independent predictor of higher 30-day ED revisits. Formalized studies to better understand factors that influence 30-day returns in children with SCD and acute pain are needed. Standardized approaches to outpatient opioid management after discharge and their impact on 30-day returns are also warranted in this population.