Date of Award

January 2026

Document Type

Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Jonathan N. Grauer

Abstract

Background:Adolescent idiopathic scoliosis (AIS) is a common condition for which posterior scoliosis fusion (PSF) may be considered. Adequate pain management during the early postoperative days is an important consideration following such intervention. In the pediatric orthopaedic literature, disparities in narcotic administration have been reported in the postoperative outpatient setting, and after presentation for long-bone fractures in the emergency department. Given the invasive nature of PSF, the question of whether inpatient postoperative pain management is associated with similar disparities is of potential interest.

Research Aims:The primary aim of the current thesis was to identify clinical and non-clinical factors associated with variations in inpatient opioid use following corrective surgery for AIS. The secondary aim was to characterize recent trends in inpatient opioid and non-opioid analgesic use. The final aim was to explore the strengths and weaknesses of the newly evolved Epic Cosmos database (from which data were abstracted for the current research) in broader context of orthopaedic/spine-related outcomes research. Hypothesis: It was hypothesized that a combination of clinical factors (such as age, sex, levels fused, and length of stay [LOS]) and non-clinical factors (such as race, ethnicity, preferred language, primary insurance, social vulnerability index [SVI], geographic region, and hospital setting [community, academic, or pediatric]) would be associated with variations in postoperative inpatient opioid use. Additionally, it was hypothesized that opioid use would decrease, with a corresponding increase in non-opioid analgesic use, over the years of the study period.

Methods:Patients were identified in the Epic Cosmos dataset from 2017-2025 using International Classification of Disease (ICD) codes for AIS and Current Procedural Terminology (CPT) codes for PSF. Patients aged 10-20 at the time of surgery were included in the study. Patients were excluded if they underwent fusion to pelvis, if the length of their hospitalization exceeded 14 days, or if their record did not include medication information. Patient demographics were extracted from the dataset. Medication records were used to calculate the average Morphine Equivalent Daily Dose (MEDD) received by each patient during their postoperative hospital stay. Multiple linear regression was used to model the relationship between clinical and non-clinical factors and MEDD. Statistical significance was set at p<0.05. Yearly trends in opioid dosing were using the average MEDD per patient per year. The percentage of patients receiving postoperative acetaminophen (oral and intravenous), ketorolac, and gabapentinoids (gabapentin and pregabalin) was also recorded annually, and trends in non-opioid analgesic use were evaluated over the same time period.

Results:In total, 11,343 AIS patients undergoing PSF were identified in the dataset. Patients were mean  standard deviation (SD) 14.4  2.2 years old at the time of surgery and predominantly female (69.4%). The majority of patients identified as White, Non-Hispanic or Latino (56.6%) and spoke English as their preferred language (95.6%). Higher MEDD was associated with older age (+1.25 per year) and longer LOS (+0.54 per day). Lower MEDD was associated with male sex (-2.54, relative to female sex). For race/ethnicity, relative to White, Non-Hispanic or Latino patients, lower MEDD was found for White, Hispanic or Latino (-2.85), Black or African American (-3.05), Asian (-5.56), American Indian or Alaskan Native (-7.71), and Other/Multiracial (-1.09) patients. Lower MEDD was also associated with preferred language other than English (-2.22, relative to English). Lower MEDD was also associated with insurance status (relative to Commercial insurance, Medicaid [-1.68]), geographic region (relative to West, Midwest [-3.50], South [-4.08], and Northeast [-8.84] regions), and hospital setting (relative to Community hospitals, academic [-2.45] and pediatric [-2.26] hospitals) (p<0.05 for all). Between 2017-2025, average MEDD decreased from 29.3 to 20.4. The percentage of patients receiving postoperative intravenous (IV) acetaminophen increased from 37% to 51% and the percentage receiving ketorolac increased from 62% to 79%. Gabapentinoids, after reaching a peak of 69% in 2020, decreased to 48% in 2025.

Statement of scientific impact and relevance for communities of interest:The observed association between MEDD and non-clinical factors like patient race, ethnicity, and preferred language highlights the importance of developing standardized pain management pathways to help minimize disparities in narcotic administration. Additional research is needed to evaluate whether the observed differences in opioid prescriptions are associated with differences in postoperative pain scores and/or patient satisfaction. The decrease in opioid dosing and increase in non-opioid analgesic use suggests that multimodal protocols have been adopted at many hospitals for the treatment of postoperative pain. These trends reflect national efforts to avoid overprescribing narcotics, while still providing adequate pain management, in light of the ongoing opioid epidemic. Finally, this thesis demonstrates the utility of the new Epic Cosmos database for orthopaedic/spine-related research. The primary strengths of the database included the presence of demographic variables and the inclusion of inpatient medication records. By addressing gaps left by existing national databases, Cosmos has the potential to play an important role in orthopaedics clinical outcomes research in the future.

Comments

This thesis is restricted to Yale network users only. It will be made publicly available on 06/01/2029

Download

Share

COinS