Person:
Bartolozzi IV, Arthur R

Profile Picture

Email Address

AA Acceptance Date

Birth Date

Research Projects

Organizational Units

Job Title

Last Name

Bartolozzi IV

First Name

Arthur R

Name

Bartolozzi IV, Arthur R

Filters

20171

Settings

Author's Publications: search.filters.isAuthorOfPublication.55dd3607-d33c-46bd-8fc8-4973c8e23932

Search Results

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Publication
    The Forest for the Trees: Using Big Data to Improve Preoperative Assessment and Risk Stratification in Pediatric Orthopedic Surgery
    (2017-05-12) Bartolozzi IV, Arthur R
    This thesis uses the Kid’s Inpatient Database to analyze surgical epidemiology, clinical outcomes, LOS, and total costs for two major pediatric surgeries. The goal is to establish clinically meaningful identifiers of risk, complexity, and variation in management both for improving preoperative assessment and understanding demographic determinants of health. Chapter 1 focused on open hip reduction for the treatment of DDH. This has become more expensive despite shorter hospital stay over time. It is also a longer, more costly experience for children who have reached walking age particularly those over 3 years old. Other patient factors including: developmental delay, ethnicity and insurance, hospital factors, and surgical management contribute to increased LOS and total charges. Despite the conclusions above a more detailed cost-analysis at an institutional level is necessary to identify how the above variables interact with each other. Chapter 2 investigated the pediatric neuromuscular population undergoing primary spinal fusions. Analyses of urinary function and anemia in addition to CCC scoring can be conducted preoperatively to determine likelihood of complications and LOS. Complications are strongly associated with both LOS and total charges and were lower at high volume centers. Additional factors including hospital size, patient race, anterior/posterior surgery, and BMP use were associated with higher charges. We presented individual prediction models for grouping patients by risk for long LOS and high cost.