Date of Award

January 2018

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Saral Mehra

Second Advisor

John Geibel



Rance J.T. Fujiwara and Saral Mehra. Section of Otolaryngology, Department of Surgery, Yale University, School of Medicine, New Haven, CT.

Recognition of variations in costs and complication rates may uncover avenues to decrease health expenditures and improve value of care. The primary aims of this study were two-fold: 1) to measure variations in costs and surgical complication rates for patients undergoing fibula free flaps or major salivary gland cancer resections, and 2) to investigate any relationship between costs and complication rates with hospital surgical volume.

This study is a retrospective, cross-sectional analysis using the Healthcare Cost and Utilization Project National Inpatient Sample from 2001 to 2011. For evaluation of hospital costs, univariate and multivariate analyses were conducted using the non-parametric Kruskal-Wallis H test and a generalized linear model, respectively, assuming a gamma distribution for the outcome variable. For evaluation of complications, univariate and multivariate analyses were conducted using chi-square and binary logistic regression, respectively.

For fibula free flaps, a total of 504 met inclusion criteria, 106 (21.0%) of whom were treated at high-volume hospitals, defined as the 95th percentile among hospitals (>5 cases per year). High-volume institutions were associated with a $10,617.04 (95% confidence interval [CI] 2,308.27–18,295.80, p=0.01) decrease in costs compared to low-volume institutions. The Northeast ($20,965,86 [CI 1,1241.58–30,690.14], p<0.001) and West ($9,075.30 [CI 1079.07–17,071.53], p=0.03) had decreased costs relative to the South. High-volume hospitals had significantly decreased surgical complication rates (OR 0.43 [CI 0.24-0.76], p=0.004), though no geographic variations in complications were observed.

For salivary gland cancers, a total of 4,290 patients with median age 68 years (interquartile range, 54-78) were included. The majority of patients were treated at academic institutions (71.5%), and 727 (16.9%) were treated at high-volume hospitals with >10 cases annually. High-volume institutions were associated with a $1,401.46 (CI 553.80-2249.10, p=0.001) increase in costs relative to very-low volume institutions. Both the West ($1280.92 [CI 573.92-1987.91], p<0.001) and Northeast ($3,197.15 [CI 2,555.85-3,838.44], p<0.001) were associated with decreased costs relative to the South. On multivariate analysis, no significant differences in rates of surgical complications or hospital costs were noted across geographic regions or by institutional volume.

In conclusion, significant cost variations exist among patients undergoing fibula free flaps or resections for major salivary gland malignancies. High-institutional volume was associated with decreased costs and improved postoperative outcomes for patients undergoing post-ablative fibula free flap reconstruction but paradoxically increased cost for those with major salivary gland cancers. Given these findings and that significant geographic variation existed without associated regional differences in complication rates, a multitude of unmeasured variables or differences in patterns of care exist. In the context of rising healthcare costs, further efforts should be conducted towards identifying opportunities to decrease costs and improve value of care.


This thesis is restricted to Yale network users only. It will be made publicly available on 06/25/2100