91.10 Economic Evaluations of Minimally Invasive HPB Resections: Nothing to CHEER About

Posted on January 31, 2019

I. Lareef1, S. Wong1,2  2Dartmouth Hitchcock,Department Of Surgery,Lebanon, NH, USA 1Geisel School of Medicine at Dartmouth,Hanover, NH, USA

Introduction:

Minimally invasive hepatopancreaticobiliary (HPB) surgery has heralded economic advantages such as reduced length of stay and this must be balanced with increased device and intraoperative costs. Therefore, the literature assessing cost effectiveness has proliferated. Reporting quality in health economics literature is variable due to the broad nature of such evaluations. The Consolidated Health Economic Evaluation Reporting Standards (CHEERS) guideline was created to improve the quality of reporting in health economics literature, and we used it to assess the HPB literature.

Methods:

A systematic review of PubMed, Web Of Science, Science Direct, and EconLit was conducted to identify articles that evaluated economic outcomes for robotic or laparoscopic oncologic HPB procedures. 2458 abstracts were reviewed and subject to the following exclusion criteria: non-peer reviewed, non-original source data (though reference lists of systematic reviews were checked), non-English language articles, no cost data reported, and non-cancer resections. A full text review of the remaining 98 articles excluded laparoscopic cholecystectomy and operations performed with non-curative intent.

Results:

A total of 44 articles were assessed for adherence to a total of 27 CHEERS checklist criteria, which are divided into the following sections: “Title and Abstract”, “Introduction”, “Methods”, “Results”, “Discussion”, and “Other”(2, 1, 16, 5, 1, and 2 items, respectively).

20/44 (45.5%) articles reported that minimally invasive surgery was more cost effective compared to the open approach, 9/44 (20.5%) reported lower cost effectiveness (higher costs), 14/44 (31.8%) reported no difference, and 1/44 (2.3%) article did not report any comparative data. The average number of total criteria met was 14.5 out of 27. Adherence to the criteria was highly variable (range 7-22). The least adhered to criteria were the inclusion of effectiveness measurement methodology and the elicitation of outcome preferences from patients. The most adhered to criteria were the inclusion of an abstract, background, objectives, and discussion. None of the articles met all of the CHEERS criteria; in fact, 21/44 (47%) did not even meet 50% of the checklist criteria. Furthermore, the average number of Methods criteria, indicative of a robust cost effectiveness evaluation, that were met was less than half of the total (7.2/16 criteria).

Conclusion:

The reporting quality around the cost effectiveness of minimally invasive HPB surgery can be improved. We found that most of these articles that claim to be cost effectiveness evaluations are, in reality, cost reporting studies. Greater adherence to CHEERS, specifically its Methods section, would improve reporting quality, as would accuracy around the classification of reporting type. Alternatively, the CHEERS criteria may be too rigid and not reflective of the quality of economic studies.

91.09 Electrolyte Analysis and Replacement: Challenging a Paradigm in Surgical Patients

Posted on January 31, 2019

K. Dammann1, M. Timmons1, C. Pierce2, E. Higdon2, M. Edelman1, A. Bernard1  1University Of Kentucky,Acute Care Surgery And Trauma, Department Of Surgery,Lexington, KY, USA 2University Of Kentucky,Pharmacy Services,Lexington, KY, USA

Introduction: Postoperative patients have increased susceptibility to alterations in electrolyte homeostasis. Although electrolytes are replaced in the critically ill, cultural norms in clinical practice today prompts the treatment of abnormal electrolytes in stable asymptomatic ward patients. Here we hypothesized that although there is no proven benefit in asymptomatic patients, electrolyte analysis and replacement is associated with significant cost. 

Methods: The study was IRB-approved. Using the electronic medical records and pharmacy database at a University academic medical center for 2016, the frequency and cost of electrolyte analysis (BMP, ion-Ca, Mg, Phos) and replacement (KCl, Mg, oral/IV Ca, oral/IV Phos) was characterized in perioperative trauma and general surgery (GS) patients. Patients without an oral diet order, with creatinine >1.4, age <16 y, admitted to ICU or with length of stay >1wk were excluded. Nursing costs were calculated as a fraction of hourly wages per each lab order or electrolyte replacement.

Results: Over the 11-month study period, forty-five of 62 patients analyzed met our criteria. Fifty-two percent were male and 48% female with an average length of stay of 4 ± 1.8 days; mean age 54 ± 14 years; and creatinine 0.67 ± 0.26. Thirty-one GS and 14 trauma patients had 421 electrolyte analysis lab orders at a cost of $2850; BMP was most frequent of these with median of (2: 1-6) per stay, accounting for 33% of lab costs. GS and trauma subjects combined received 253 doses of electrolytes at a total pharmacy cost of $928; with KCl (47), Phos (20), Ca (18), and Mg (14) percent of pharmacy costs. Mg was most frequently replaced (84% GS vs 71% trauma); followed by KCl (71% of GS vs 29% trauma); then Phos (29% GS vs 17% trauma). Electrolyte replacement was 38% more frequent in GS vs trauma subjects. Nursing costs associated with electrolyte analysis/replacement amounted to $3040.

Conclusion: There is little evidence to support electrolyte analysis and replacement in stable asymptomatic ward patients, but its prevalence and cost ($6818) in this study was substantial in a small proportion of patients. BMP’s, pharmacy charges for potassium, and nursing staff costs accounted for the most significant portion of total cost. Electrolyte replacement by house staff occurred more frequently in GS patients when compared to trauma patients managed by surgical attending providers. Considering these data, further research should determine whether these practices are truly warranted.

91.08 Wide Provider Variation in Cost for Thyroidectomy: Potential Benefits of Standardizing Practice?

Posted on January 31, 2019

B. R. Herring1, S. Jang1, Z. Aburjania1, H. Chen1  1University Of Alabama at Birmingham,Department Of Surgery, School Of Medicine,Birmingham, Alabama, USA

Introduction: Identifying provider variation in surgical costs could control rising healthcare expenditure and deliver cost-effective care. While these efforts have mostly focused on complex and expensive operations, provider-level variation in costs of thyroidectomy has not been well examined.

Methods:  We retrospectively evaluated 989 consecutive total thyroidectomies performed by 14 surgeons at our institution between September 2011 and July 2016. Data were extracted from the McKesson Business Insight program. Total length of stay and cost were evaluated using the Mann-Whitney U and the Kruskal-Wallis tests. Categorical variables were evaluated with chi-square.

Results: Median patient age was 48 years (range 8-90), 81% were females, 64% were Caucasians, and 77% were outpatients. The number of thyroidectomies performed by the 14 surgeons ranged from 4 to 635 (mean = 71). The median costs per provider varied widely from $4,390.94 to $16,754.15 (P < 0.001). The mean length of stay was 1.2d ± 8.2 with wide variation among providers (0d to 5.5d). Providers whose hospital cost exceeded the institutional mean demonstrated a significantly higher ICU admission rate, anesthesia fees, and lab costs (P < 0.001).

Conclusion: We found substantial variation in hospital cost among providers for thyroidectomy despite practicing in the same academic institution, with some surgeons spending 4x more for the same operation. Implementing institutional standards of practice could reduce variation and the costs of surgical care.
 

91.07 Validation and Extension of the Ventral Hernia Repair Cost Prediction Model

Posted on January 31, 2019

M. J. Nisiewicz1, M. Plymale2,3, D. Davenport2, S. Saleh1, T. Buckley1, Z. Hassan1, J. S. Roth2,3  1University of Kentucky,College Of Medicine,Lexington, KY, USA 2University of Kentucky,Department Of Surgery,Lexington, KY, USA 3University Of Kentucky,Division Of General Surgery,Lexington, KY, USA

Introduction:
Repair of ventral and incisional hernias (VHR) remains a costly challenge for healthcare systems. In a prior study of a single surgeon’s elective inpatient open VHR practice, a cost model was developed which predicted over 70% of hospital cost variation, and included CDC wound class, hernia defect size, age, ASA class, number of mesh pieces, and use of biologic mesh. The purpose of the current study was to evaluate the ventral hernia cost model with multiple surgeons’ elective inpatient open VHR cases and to extend to include non-elective/urgent/emergent, outpatient and laparoscopic VHR.

Methods:
With IRB approval, elective and emergent cases of open and laparoscopic VHR (CPT codes 49560, 49561, 49565, 49566, 49654, 49655, 49656, 49657) performed at a single facility by multiple surgeons from October 1, 2014 to December 31, 2017 were identified. Cases in which VHR was done as a secondary procedure were excluded. Demographics, comorbidity status, ASA class, CDC wound class, length of stay, and 30 day outcomes were obtained from the local NSQIP database. Medical record review determined hernia defect size. Hospital cost data was obtained from the hospital cost accounting system. Forward multivariable regression of log transformed costs identified independent drivers of cost (p for entry < .05, for exit > .10).

Results:
Of the 387 VHRs, 74% were open repairs, 35.4% included separation of components, and 14.7% were non-elective. Mean age was 55 years, and 52% of patients were female. The base cost for an outpatient primary small VHR without mesh implantation was $4114. Including only the open, elective VHR cases, the previously reported six-factor cost model predicted 50% of the total cost variation. With all VHRs included, ten variables were found to independently drive costs, predicting 60% of the total cost variation from the base cost. Biggest cost drivers (≥ 15% increase) were preoperative open wound (+$1207), preoperative SIRS/sepsis (+$740), hernia defect size (+$616), inpatient status (+$875), use of absorbable mesh vs. synthetic or no mesh (+$752), use of biologic mesh (+$1000), and utilization of multiple mesh pieces (+$795). Other cost drivers included age, obesity, morbid obesity, and recurrent hernia.

Conclusion:
Elective hernia repair cost variability may be predicted utilizing a six-factor model. In the broader context of all VHR repair at our institution, recurrent hernia, inpatient and non-elective surgery are greater cost drivers than wound class. Obesity, the presence of an open wound and systemic inflammation, relatively rare in the elective group due to optimization but more common in urgent/emergent cases, replaced ASA class as cost drivers.  Age, defect size, mesh type and number of meshes utilized were common to both models. A hernia cost model utilizing readily identifiable preoperative factors can be utilized to predict resource utilization.
 

91.06 Financial Burden of Respiratory Complications Following Intermediate and High-Risk Surgeries

Posted on January 31, 2019

S. M. Stokes1, B. S. Brooke2, R. E. Glasgow1, S. R. Finlayson1, T. K. Varghese3  1University Of Utah,Department Of Surgery, Division of General Surgery,Salt Lake City, UT, USA 2University Of Utah,Department Of Surgery, Division Of Vascular Surgery,Salt Lake City, UT, USA 3University Of Utah,Department Of Surgery, Division Of Cardiothoracic Surgery,Salt Lake City, UT, USA

Introduction:  Respiratory complications following high and intermediate-risk surgeries are known to have a substantial impact on a patient’s clinical post-operative course, although the resulting financial impact on the health care system is not well characterized. The objective of this study was to assess hospital costs when patient’s experience respiratory complications.

Methods:  A single institution’s database was queried to identify all patients undergoing elective intermediate (colectomy, lung resection, infra-inguinal bypass, carotid endarterectomy, and ventral hernia repair) and high-risk (esophagectomy, pancreatectomy, valve replacement, coronary bypass, and open abdominal aorta) surgeries. Urgent and emergent operations were excluded. Patient demographic, inpatient, and direct cost data were collected. The primary outcome measure was the proportional increase in total cost due to a pneumonia complication, expressed as the relative increase of the total hospital cost between pneumonia and non-pneumonia cases. Secondary outcomes were cost divisions between facility, lab, imaging, and supply costs, as well as surgical outcomes. Mixed-effects, multivariate, regression models were constructed, with patients clustered by procedure, to identify independent risk factors associated with increased hospital costs of treating pulmonary complications.

Results: We identified a total of 2,834 patients who underwent at least one of the ten selected elective surgeries. The rates of pneumonia were greater for high-risk procedures (9.3% vs. 3.8%, p < 0.001) as compared to intermediate procedures. After adjustment in regression models, the occurrence of pneumonia was found to result in an overall 1.5-fold increase in hospital costs. This cost difference was greater for intermediate-risk (1.8-fold) vs. high-risk (1.3-fold) surgeries. For high-risk procedures, pneumonia, length of stay, ASA class, and surgery length all predisposed to higher hospital costs (p<0.05 for all associations). For intermediate-risk procedures, pneumonia, failure to wean from mechanical ventilation, length of stay, Charlson-Deyo comorbidity index, and surgery length predisposed to higher hospital costs (p<0.001 for all associations). In patients who experienced a pneumonia complication, the largest contribution to costs was facility cost (33%), followed by pharmacy (19%), other services (15%), and laboratory costs (12%).

Conclusion: Our data shows that respiratory complications have a profound impact on hospital costs, especially for intermediate-risk procedures. Focusing efforts on preventing respiratory complications can have a significant impact on both patient outcomes and the financial burden to the healthcare system.  

 

91.05 The Laparoscopic Approach to Pancreaticoduodenectomy is Cost Neutral in Very HighVolume Centers

Posted on January 31, 2019

E. Eguia1, P. C. Kuo2, P. J. Sweigert1, M. H. Nelson1, G. V. Aranha1, G. Abood1, C. V. Godellas1, M. S. Baker1  1Loyola University Chicago Stritch School Of Medicine,General Surgery,Maywood, IL, USA 2University Of South Florida College Of Medicine,General Surgery,Tampa, FL, USA

Introduction:
Little is known regarding the impact of minimally invasive approaches to pancreaticoduodenectomy (PD) on the aggregate costs of care for patients undergoing PD.  

Methods:
We queried the Healthcare Cost and Utilization Project State Inpatient Database to identify patients undergoing laparoscopic (LPD) or open (OPD) pancreaticoduodenectomy in FL, MD, NY and WA between 2014 and 2016. Multivariable regression (MVR) was used to evaluate the association between surgical approach and rates of postoperative complication, overall lengths of stay (LOS) and aggregate costs of care including readmissions to 90 days following PD. Candidate variables were determined a priori using best variable subsets and included: age, gender, insurance, race/ethnicity, Charlson comorbidity index (CCI), pathology (benign vs. malignant), hospital PD volume broken to terciles  (low: <17 PD/year; moderate to high: 17<PD/year <127; very high volume: >127 PD/year) and overall LOS. 

Results:

513 (10.8%) patients underwent LPD; 4,746 (89.2%) underwent OPD. On univariate analysis, patients undergoing LPD had higher CCI (5 vs. 3, p<0.001) and higher rates of readmission (35% vs. 30%, p=0.04), malignant pathology (89% vs. 76%; p<0.001), and aggregate costs of care ($41,669 vs. $37,580, p<0.02) than those undergoing OPD.

On MVR adjusted for age, CCI, pathologic diagnosis, and hospital volume, LPD was associated with a lower risk of prolonged LOS (OR 0.77; 95% CI [0.61,0.97]) but greater risk of readmission (OR 1.24; 95% CI [1.02, 1.51). Rates of perioperative morbidity and overall LOS for patients undergoing LPD were identical to those for patients undergoing OPD.

On MVR adjusted for age, pathology, CCI, LOS, and volume, factors associated with being in the highest quartile for aggregate costs of care included: male gender (OR 1.19; 95% CI [1.04, 1.37]), CCI (OR 1.07; 95% CI [1.03, 1.11]), black race (OR1.41; 95% CI [1.12, 1.78]), Hispanic ethnicity (OR 1.90; 95% CI [1.50, 2.42]), Medicare insurance (OR 1.28 95% CI [1.05, 1.55]), readmission (OR 4:44; 95% CI [3.87, 5.09]) and low hospital volume (OR 2.46; 95% CI [1.97, 3.06]). Patients undergoing LPD in low (+$9,390; 95% CI [$2,948, $15,831]) and moderate to high (+$5,579; 95% CI [$1,783, $9,376]) volume centers had higher costs than those undergoing OPD in the same centers. In very high-volume centers, aggregate costs of care for patients undergoing LPD were identical to those undergoing OPD in the same centers (+$616; 95% CI [-$1,703, $2,936])). 

Conclusion:
Rates of morbidity and overall LOS for patients undergoing LPD are statistically identical to those undergoing OPD. At low to moderate volume centers, the laparoscopic approach to PD is associated with higher aggregate costs of care relative to OPD whereas at very high-volume centers LPD is cost neutral. This finding suggests that high volume centers are able develop efficiencies of scale that act to mitigate costs inherent to adoption of the laparoscopic approach to PD. 

91.04 Cost-Effectiveness of Sentinel Lymph Node Biopsy for Head and Neck Cutaneous Squamous Cell Carcinoma

Posted on January 31, 2019

P. Quinn1, J. Oliver1, O. Mahmoud1, R. Chokshi1  1Rutgers New Jersey Medical School,Newark, NJ, USA

Introduction:
While sentinel lymph node biopsy (SLNB) has been established as a useful tool in the staging of melanoma, its utility in cutaneous squamous cell carcinoma (CSCC) has yet to be elucidated. Analysis has shown that SLNB may have an emerging role in patients with high-risk lesions. The purpose of this study was to determine the cost-effectiveness of performing SLNB in all head and neck CSCC patients as well as by TNM staging.

Methods:
A decision model was developed to analyze costs and survival in head and neck CSCC patients based on their tumor and nodal metastasis staging and whether or not they received a SLNB. The decision model placed patients into two groups, those undergoing SLNB and those undergoing only wide local excision. Those that were found to have any positive lymph node underwent a neck dissection, then those with N2 disease were treated with radiation therapy and those with N3 disease were treated with chemotherapy and radiation based on treatment guidelines according to the National Comprehensive Cancer Network. Values for disease stage, morbidity, mortality as well as SLNB sensitivity and specificity were derived from the published literature. Costs were derived using Medicare outpatient costs (FY 2018). Survival values were calculated with the declining exponential approximation of life expectancy (DEALE) method using available data on disease-specific death rate. The average age of patients with CSCC was found to be 70, with a life expectancy of 14.3 years according to Social Security life tables. Incremental cost-effectiveness ratios (ICER) were calculated based on the change in quality adjusted life years (QALYs) and costs (US$) between the different options, with a threshold of $100,000 to determine the most cost-effective strategy. One-way and two-way sensitivity analysis was performed to validate the results.

Results:
Sensitivity and specificity of SLNB was found to be 77% and 100%. The disease incidence was 77%, 5%, and 18% for T1, T2, and T3 specifically, and the rate of node positivity was 0.4%, 12.2%, and 14.1% for each T stage. Not performing a SLNB results in 12.26 QALYs and a cost of $3,712.98. Performing a SLNB resulted in a 0.59 decrease in QALYs and an increase in cost of $1,379.58 for an ICER of -2,338.27. This trend remained the same across all tumor stages. Sensitivity analysis of tumor stage distribution, nodal stage distribution, nodal positivity rates, and SLNB sensitivity did not alter the outcome.

Conclusion:
In patients with head and neck cutaneous squamous cell carcinoma, the most cost-effective strategy is to not perform sentinel lymph node biopsies, regardless of the patient’s stage. Low rates of nodal metastasis in addition to low disease-specific death rates were the significant factors in this outcome. Increasing the sensitivity of SLNB would not impact this recommendation unless the rate of nodal metastasis were significantly higher.
 

91.03 Resource Overutilization in the Diagnosis of Lymphedema Praecox

Posted on January 31, 2019

A. A. Shah1,3, J. Roberson2, M. Petrosyan2,3, P. Guzzetta2,3  1Howard University College Of Medicine,Surgery,Washington, DC, USA 2George Washington University School Of Medicine And Health Sciences,Surgery,Washington, DC, USA 3Children’s National Medical Center,Surgery,Washington, DC, USA

Introduction: Primary lymphedema presenting in adolescence is known as lymphedema praecox.  It is a rare disorder leading to failure of the lymphatic system to drain properly, almost exclusively in the lower extremities. The ensuing limb swelling can be debilitating and associated with adverse physical and psychosocial consequences. Patients are often subjected to a myriad of diagnostic tests which often have little influence on management. The purpose of this study is to review a large cohort of patients with a rare disorder presenting to a high-volume tertiary care children’s hospital, as well as to determine the fiscal impact of testing on these patients.

Methods: A 13-year retrospective institutional review was performed of patients between the ages of 7 and 21 with a diagnosis of lymphedema praecox. Information was obtained on demographic parameters, diagnostic studies performed, pre-morbid conditions, and clinical outcomes. Descriptive analyses were performed. This study was approved by the hospital’s IRB.

Results: Data were extracted from 49 patient records. The median age was 14 (range: 7-21) years. Participants were predominantly female (n=40, 81.6%).  Thirty-one patients (64%) were African-Americans, although more than 50% of our outpatients are African-American. 19 patients had bilateral disease. In patients with unilateral disease, 16 were on the right and 14 were on the left. The diagnosis was made on clinical exam only in 14 patients. 19 patients had Doppler ultrasound (US) exams of which in 12 it was the only exam, 3 had US and MRI, 3 had US, MRI and lymphoscintigraphy, and 1 had US and lymphoscintigraphy.  11 patients had only MRI and physical exam for diagnosis.  The charges for lower extremity Doppler ultrasound, lymphoscintigraphy, and MRI with contrast were $1,715, $1,269, and $6006 respectively.

Conclusion: We believe that in the adolescent female with physical findings consistent with lymphedema praecox, diagnostic imaging should be limited to a Doppler ultrasound to rule out deep venous thrombus or a mass as the cause of the swelling. Additional imaging does not aide in the diagnosis, but does add to the expense to and inconvenience of the patient and should be avoided.

91.02 Less is More: Routine Postoperative Lab Testing in Elective General Surgery Patients

Posted on January 31, 2019

E. D. Porter1, J. L. Kelly1, A. R. Wilcox1, S. W. Trooboff1, J. D. Phillips1, C. V. Angeles1  1Dartmouth Hitchcock Medical Center,Surgery,Lebanon, NH, USA

Introduction: Overutilization of laboratory testing in hospitalized patients continues to challenge cost-effective care. Prior efforts to reduce unnecessary lab testing in the surgical population have been limited to preoperative testing and critical care patients. We sought to investigate our institution’s post-operative lab (POL) ordering practice on adult elective general surgery patients and its impact on clinical decision-making and outcomes.

Methods: Retrospective cohort study at a single academic center. All adult elective general surgery patients with a <48 hour admission from April-June 2018 were selected. Patients with a condition requiring lab testing (e.g. on coumadin) were excluded. We reviewed patient charts for demographics, comorbidities, POL orders, documentation of POL abnormalities, and 30-day complications. We compared patients receiving or not receiving POLs using student’s t-test, chi-square and one-way analysis of variance for continuous, dichotomous, and categorical covariates, respectively. Significance was set at p<0.05.

Results: In our study period, 171 patients were admitted for <48 hours after elective general surgery. Surgical procedures were classified as intestinal (n=116), solid abdominal organ (n=9), hernia (n=9), thoracic (n=22), breast (n=11), and other (n=4). Routine POLs were ordered on 89% of patients (153/171), with significant differences in frequency of ordering labs by surgery type (Figure 1). On univariate analysis, patients who did not have POLs had significantly lower BMI (p<0.004); there was no difference in age, demographics, comorbidities, blood loss, or ASA status. Of those who had POLs, 88% had an abnormality (134/153), and the abnormality was commented on in 20% of progress notes (27/134). A POL only changed care in 12% of patients (19/153). Change in care included electrolyte correction (15/19), medication held (2/19), additional test (1/19), and fluid resuscitation (1/19). White blood cell count and/or hemoglobin levels were abnormal in 84% (128/153); however, none of these abnormalities prompted a change in care. Electrolytes were abnormal in 14% of cases (21/153) and led to a change in care for 76% (16/21).  Creatinine was elevated in 3% (5/153) and led to a change in care for 80% (4/5). There was no significant difference in 30-day complication or readmission rates for those who received POLs.

Conclusion: POL testing is common among adult elective general surgery patients who have brief hospitalizations. While most labs yielded abnormal results in our cohort, these were unlikely to be commented on and even less likely to change care. Routine POL testing without clinical indication is overutilized and interventions to reduce this practice are needed.

91.01 Targeted-physician Interventions Reduce Surgical Expense and Improve Value-Driven Health Outcomes

Posted on January 31, 2019

R. Xie1, B. Lindeman1, H. Chen1, T. N. Wang1  1University Of Alabama at Birmingham,Surgery,Birmingham, Alabama, USA

Introduction:  Increasing health care costs in the U.S. have not translated to superior outcomes in comparison to other developed countries. The implementation of physician-directed intervention to reduce cost may improve value-driven health outcomes. This study aimed to evaluate the effectiveness of cost-targeted interventions to reduce surgical expense and improve care for patients undergoing total thyroidectomies.

Methods:  Two separate face-to-face interventions with individual surgeons focusing on surgical expense around thyroidectomy were implemented in two surgical services (Endocrine Surgery and Otolaryngology) at a tertiary medical center. The pre-intervention period was from Dec 2014 to Jun 2016 (19 months, 352 operations). The post-intervention period was from July 2016 to January 2018 (19 months, 360 operations). Multivariable generalized linear regression analyses and difference-in-difference analyses were conducted to compare the pre- and post-intervention outcomes including cost outcomes (total costs, fixed costs, and variable costs per thyroidectomy) and clinical outcomes (30-day readmission rate, days to readmission, and total length of stay).

Results: Patient demographics and characteristics were similar before and after the interventions. Post-intervention operative costs were significantly reduced as compared to pre-intervention costs. The percentage savings in the average total, fixed, and variable costs per surgery were 8% (from $6,571 to $6,033, p=0.03), 7% (from $4,062 to $3,781, p=0.04), and 10% (from $2,509 to $2,251, p=0.03), respectively. Additionally, the clinical outcome parameter of total length of stay improved from 1.3 days (STD: 1.9) to 1.0 day (STD: 0.2, p<0.01). Readmission rates and days to readmission were not significantly different. The effectiveness of the interventions regarding cost reductions differed between the two surgical divisions. Compared to Otolaryngology, Endocrine Surgery saved an additional average total cost of $517.30 (STD: 877.7; p=0.08), average fixed cost of $238.40 (STD: 526.9; p=0.18), and average variable cost of $279.00 (STD: 423.2; p=0.05) per surgery since the intervention.

Conclusion: Targeted-physician interventions can be an effective tool for reducing cost and improving health outcomes. The effectiveness of interventions may vary depending on the type of specialty training. Future implementation of physician interventions needs to be standardized in order to critically evaluate its effectiveness on patient outcomes.
 

90.20 Cross-Border Transfer Leads to Delays in Care for Tibia Fractures

Posted on January 31, 2019

A. Brito1, L. N. Godat1, T. W. Costantini1, J. Doucet1, A. M. Smith1, A. E. Berndtson1  1University Of California – San Diego,Surgery,San Diego, CA, USA

Introduction:  Tibia fractures are commonly seen after injury and often require operative fixation. Prior studies have shown that delays in care for treatment of tibia fractures are associated with poor functional outcomes. A sub-population of patients admitted to our trauma center are transferred from Mexico, adding barriers to prompt care for their injuries. We hypothesized that patients with tibia fractures transferred from Mexico would have delays in care and subsequently worse outcomes.

Methods:  The trauma registry of an ACS-verified level 1 trauma center was retrospectively reviewed for all tibia fractures admitted from 2010-2015. Data collection included demographics, country of injury, open vs. closed fracture, Gustilo classification, operative interventions required, complications and outcomes. Patients were then subdivided into those injured in the USA and in Mexico (MEX) and the two groups were compared.

Results: A total of 497 patients were identified, 439 from the USA and 58 transferred from MEX. MEX patients were more severely injured overall, with higher injury severity scores (Table) as well as a higher percentage of patients with AIS scores ≥ 3 for both Head (USA 14.4% vs. MEX 29.3%, p = 0.007) and Chest (USA 14.4% vs. MEX 31.0%, p = 0.004) regions. MEX patients had longer times from injury to admission, as well as increased times to both debridement and operative fixation after admission (Table). There was no difference in Gustilo classification of open fractures between groups. MEX patients were more likely to have a venous thromboembolism (USA 6.7% vs. MEX 15.6%, P = 0.003) or undergo lower extremity amputation (USA 0.91% vs. MEX 6.9%, p = 0.008).  Subgroup analysis of patients with isolated tibia fractures (other AIS < 3) still demonstrated longer times from arrival to both debridement and fixation in patients from MEX, though ISS was no longer statistically different (median ± IQR, USA 10 ± 8 vs. MEX 9 ± 4, p = 0.625).

Conclusion: Patients transferred from MEX for treatment of tibia fractures have resultant delays in time from injury to admission, but also have a longer period from admission to definitive care of their fracture.  Patients transferred from MEX also had worse outcomes, including increased rates of VTE and amputation compared to those injured in the US. Ongoing systems development is required to improve processes for transfer from Mexico to the US, when needed, in order to minimize delays in care and optimize outcomes.

 

90.19 Insurance Type Affects Short Term Outcomes After Hartmann Procedure.

Posted on January 31, 2019

M. J. Adair1, S. Markowiak1, M. Baldawi1, C. Taylor1, A. Aldhaheri1, C. Das1, W. Qu1, M. Nazzal1, S. Pannell1  1University Of Toledo Medical Center,Department Of Surgery,Toledo, OH, USA

INTRODUCTION
Hartmann’s procedure is traditionally performed to treat colonic obstruction or perforation. Using a national database, the aim of this study was to compare patients with different types of health insurance (different payer sources) for short-term postoperative outcomes after Hartmann's procedure. 

METHODS
From the Natioinal Inpatient Sample (NIS) database, patients who underwent elective or urgent Hartmann's procedure for the time period between 2008 and 2014 were included in this study. After adjustment for demographics and comorbidites, health insurance types were compared for in-hospital mortality, post-perative complications, length of hospital stay, and total hospital cost. Univariate analysis was performed using Chi-square test, two sample t-test and Mann-Whitney U test. Multiple logistic regression was utilized for the multivariate analysis.

RESULTS
A total of 9,836 patients were included in this study. The average age of  patients (mean ±SD) was 65.6±15.4 years old. Out of these patients, 4574 (46.5%) were males and 7667 (78%) were White. Overall in-hospital mortality rate was 11.9% (n=1169). The median hospital lenght of stay was 12 days and the median total hospital cost was US$104,635. Using multivariate analysis, we foud that Medicare and Medicaid patients had significantly higher mortality rate than private insurance patients (16.5% and 7.6% versus 5.9%, respectively. P<0.001). Medicare and Medicaid patients also had a longer hospital length of stay in comparison to Private insurance patients (13 days and 14 days versus 11 days, respectively. P<0.001).

CONCLUSION

Following Hartmann’s procedure, Medicare and Medicaid patients have a higher risk of mortality and longer total hospital length of stay in comparison to those with private insurance. A strong consideration of possible undiagnosed comorbidities, delayed presentation or inadequate prior management must be undertaken when treating these patients. We also recommend performing further research to fully analyze all the potential factors that can influence outcomes after Hartmann’s Procedure.

 

90.18 Mortality Related to Mass-Casualty Incidents at a Malawian Tertiary Hospital

Posted on January 31, 2019

J. Kincaid1,3, G. Mulima3, N. Rodriguez-Ormaza2, A. Charles2, R. Maine2  1Thomas Jefferson University,Surgery,Philadelphia, PA, USA 2University Of North Carolina At Chapel Hill,Surgery,Chapel Hill, NC, USA 3Kamuzu Central Hospital,Surgery,Lilongwe, Malawi

Introduction:  Mass-casualty incidents (MCI) suddenly strain a healthcare system with an influx of trauma patients. Little is known about how MCIs in low resource settings impact mortality. We aimed to determine if the resource strain from MCIs at a tertiary hospital in Malawi increased mortality for MCI patients and patients who arrived on the same day as an MCI compared to patients who presented days without MCIs.

Methods:  This is a retrospective analysis of a prospective trauma registry, from January 1, 2012 through December 31, 2016, at a tertiary hospital in Malawi. MCIs were defined as ≥ 4 trauma patients who present simultaneously to the casualty department. We conducted bivariate analysis comparing patient, mechanism of injury, and outcome characteristics by whether or not the event was an MCI. Next, we determined whether non-MCI patients presented on the same day as an MCI or on a non-MCI day and compared the same variables. Categorical variables were compared with Pearson chi-squared test or the Fisher’s exact test; continuous variables were compared using Student t-test, Wilcoxon rank sum test or the Kruskal-Wallis test by ranks, as appropriate. Multivariable analysis using a Modified Poisson regression was utilized to estimate risk ratios (RR) and 95% confidence intervals (CI). We adjusted for sex, age, primary body area injured, transfer status, nighttime presentation, vehicle-related trauma and admission year.  P-values <0.05 were statistically significant.

Results: The registry included 75,350 trauma patients; 3% (2,227) were part of an MCI and 11,365 (15%) presented on the same day as an MCI. Overall more patients who presented as part of an MCI died, 90 (4%) vs. 2,124 (2.9%), p <0.001). This difference was driven by a higher proportion of MCI patients who were dead on arrival (2.9% vs. 1.1%, p<0.001), as in-hospital mortality rates for MCI or non-MCI traumas did not differ statistically (4.1% vs. 3.7%, p=0.671). However, trauma patients who were not a part of an MCI but presented to the ED the same day as an MCI had higher in-hospital mortality than patients who presented on days without an MCI (7.0% vs. 5.4% vs. 5.6%, p=0.015).  When compared to non-MCI trauma patients presenting on a non-MCI day, being part of an MCI increased the risk of in-hospital mortality by 19% (RR=1.19, 95%CI: 0.98-1.44, p=0.0821).

Conclusion: MCIs presented frequently to this Malawian tertiary hospital, which stressed the hospital’s limited capacity. The higher in-hospital mortality of trauma patients not involved in MCI but who presented the same day as an MCI points to the strain on the limited resources resulting in poorer patient outcomes when the hospital suffers the stress of an MCI. Both improved capacity for treating trauma patients at the central hospital and district hospitals coupled with improved triage protocols could decrease inappropriate transfers of trauma patients, which contributes to overwhelming the central hospital.
 

90.17 Factors Influencing Delays In Care For Patients With Peritonitis At A Rwanda Referral Hospital

Posted on January 31, 2019

M. Munyaneza1,2, S. Jayaraman3, F. Ntirenganya1,2, J. Rickard2,4  1University of Rwanda,College Of Medicine And Health Sciences, Department Of Surgery,Kigali, XX, Rwanda 2University Teaching Hospital of Kigali,Department Of Surgery,Kigali, XX, Rwanda 3Virginia Commonwealth University,Department Of Surgery,Richmond, VA, USA 4University Of Minnesota,Department Of Surgery,Minneapolis, MN, USA

Introduction:  

Peritonitis is a common surgical emergency with a high mortality rate. Prompt recognition and surgical treatment is the mainstay of therapy. In Rwanda, patients often present critically ill with delayed presentation. A better understanding of patient care prior to arrival at the referral hospital is needed to identify areas for improvement. The aim of this study was to describe delays in Rwandan patients presenting to a referral hospital with peritonitis.

Methods:  

This is a cross-sectional observational study of patients with peritonitis admitted to the department of surgery at a tertiary referral hospital in Rwanda. Data was collected on demographics, clinical course and patient delays. Patient delays were characterized according to the Three Delays Model. Factors related to delays in seeking care included consultation of traditional healers, understanding the need for medical attention, perceptions and acceptance of surgery and the healthcare system. Factors related to delays in reaching care included travel time, cost, and availability. Data entry and analysis was done using Google Form software.

Results: 

Over an 8-month time period, 54 patients with peritonitis were admitted to the referral hospital with peritonitis. Most (n=37, 68%) patients were male.  For education, 20 (37%) patients had attended only primary school and 15 (28%) never went to school. A large number (48%) were unemployed and most (n=45, 83%) patients used community-based health insurance. For most patients (n=44, 81%) the monthly income was less than 10,000 Rwandan Francs (11 U.S. Dollars).  The average duration of symptoms prior to presentation at the referral hospital was 48 hours.

A large number (n=37, 69%) of patients consulted a traditional healer prior to presentation at the healthcare system. Most (n=29, 53%) patients travelled more than 2 hours to reach a health facility. A large number (n=39, 72%) reported prior good experience with health system and believed that surgery could cure abdominal pain.

From the health center to the district hospital, most (n=36, 66%) patients travelled more than 10km. The cost of transportation ranged between 5000-10000 RWF (5-11 U.S. Dollars) for most of them, and 52% of patients arrived at the district hospital between 24 – 48 hours after the onset of abdominal pain. After arrival at the referral hospital, almost all (98%) patients were operated.

 

Conclusion: 

In this study, factors that were influencing seeking and reaching care were associated with sociodemographic characteristics, health-seeking behaviors, the cost of care, and travelling time. These findings may highlight points of interest to conduct a community-based survey, to understand better factors associated with delays in seeking and reaching care for patients with peritonitis.
 

90.16 Applying Lean Healthcare in Lean Settings: Early Results of a Pilot Program

Posted on January 31, 2019

P. K. Rao1, A. J. Cunningham1, M. C. Boulos1, D. Kenron1, P. Mshelbwala2, E. Ameh3, S. Krishnaswami1  1Oregon Health And Science University,Department Of Surgery,Portland, OR, USA 2University of Abuja-Teaching Hospital,Department Of Surgery,Gwagalada, FCT, Nigeria 3National Hospital,Department Of Surgery,Abuja, FCT, Nigeria

Introduction: Lean healthcare methodology is frequently utilized in high income settings to maximize capacity and operational efficiency during process improvement (PI) initiatives.  However, the utility of PI for healthcare in low and middle income countries (LMIC) has not been well studied. Operating theaters in LMIC are often characterized by high cancellation rates and delays resulting from suboptimal theater space, prolonged turnover times, and limited surgical workforce.  In order to study the applicability of lean methodology in LMIC, a comprehensive pilot program was developed in 2017 to promote sustainable operating theater efficiency at two hospitals in Abuja, Nigeria.

Methods: Perioperative committees were established at both institutions, a primary-tertiary center and a quaternary referral center, to evaluate current processes, identify problems therein, and compile a list of priorities.  A physician champion and a PI specialist in conjunction with local physician-partners held a workshop to address these priorities in December 2017 as part of an ongoing collaboration. The workshop was designed to teach practical applications of PI methodology, including process mapping, value stream thinking, and root cause analysis to nurses, surgeons, anesthesiologists, and administrators. Pre- and post-workshop surveys were administered to assess perceived benefit, and compared with a chi-square test of independence.

Results: In total, 42 individuals attended the PI workshop. 26/42 (62%) completed a pre-workshop survey, and 31/42 (74%) completed the post-workshop survey.  Pre-workshop, 10 respondents (38.5%) believed a process improvement workshop would be valuable for them as individual providers, and 11 (42.3%) saw its value for the perioperative committee.  After the workshop, all 31 respondents reported the workshop as valuable both personally and for the perioperative committee (p < 0.001), and all reported that PI methodology could benefit the institution overall. Workshop components identified as most valuable were development of quality improvement tools (52%), and fostering of team culture (36%). The most frequently listed barrier to PI implementation was the institutions’ ability to sustainably apply the concepts learned (40%).  Outcomes from the workshop led to development of block time utilization measurements, optimal staffing and avoidable-delay dashboards, and workflow diagrams to track trends and improve perioperative care.

Conclusion:
Lean methodology may be more applicable in lean settings than previously recognized.  All respondents noted that PI techniques have potential to improve operational efficiency. This could be of even greater relative benefit in such severely resource-constrained environments.  Interval measurement of outcome data is planned at 1 year. Sustainability will be facilitated by telementoring, and future efforts include expansion beyond the perioperative setting to inpatient wards and outpatient clinics.

90.15 Epidemiology and Perioperative Mortality of Exploratory Laparotomy at a Referral Center in Ghana

Posted on January 31, 2019

B. S. Hendriksen1, L. K. Keeney1, D. J. Morrell1, X. Candela2, P. Ssentongo3, J. S. Oh1, C. S. Hollenbeak3, T. E. Arkorful4, E. K. Marfo4, F. Amponsah4  1Penn State Health Milton S. Hershey Medical Center,Hershey, PA, USA 2Penn State University College Of Medicine,Hershey, PA, USA 3The Pennsylvania State University,University Park, PA, USA 4Eastern Regional Hospital,Koforidua, EASTERN REGION, Ghana

Introduction: Exploratory laparotomy represents one of the most common operations performed at rural surgical referral centers throughout Ghana. Late disease presentation combined with a frequent lack of pre-operative imaging makes these operations challenging. In order to identify areas for future quality improvement efforts, we aimed to assess the epidemiology of exploratory laparotomy and to investigate perioperative mortality as a benchmark quality measure.

Methods: Surgical logbooks were queried at a regional referral hospital in Eastern Region, Ghana to identify cases of exploratory laparotomy from July 2017 through June 2018. The logbooks allowed for corroboration of patient data in the electronic medical record. Logistic regression was used to identify predictors of perioperative mortality.

Results: 
The study included 286 adult and 60 pediatric operations. Appendicitis (29%), obstruction (26%), perforated peptic ulcer disease (15%) and major abdominal trauma (6%) were the most common diagnoses in adults. Appendicitis (40%), intussusception (17%), major abdominal trauma (10%), and typhoid ileitis (7%) were the most common in children. Males accounted for 65% of cases. Only 60 % of operations were covered by national health insurance. The overall perioperative mortality rate was 11.5% (12.6% adults; 6.7% pediatric). 60% of mortalities were referrals from outside hospitals and the mortality rate for referrals was 13.5%. Mortality had 13 times greater odds with perforated peptic ulcer disease (OR 13.1, p=0.025) and 12 times greater odds with trauma (OR 11.7, p=0.042). Female gender (OR 0.3, p=0.016) and the national health insurance (OR 0.4, p=0.031) were protective variables. Individuals 60 and older (OR 3.3, p=0.016) had higher mortality.

Conclusion: 
Perforated peptic ulcer disease and major abdominal trauma carry significant risk of mortality and represent high impact areas for quality improvement. Efforts to improve national healthcare coverage and care for patients requiring surgical referral could decrease surgical mortalities.

 

90.14 Mortality Following Trauma Exploratory Laparotomy in Sub-Saharan Africa

Posted on January 31, 2019

L. N. Purcell1, A. N. Yohann1, R. N. Maine1, T. N. Reid1, C. Mabedi2, A. Charles1  1University Of North Carolina At Chapel Hill,General Surgery,Chapel Hill, NC, USA 2Kamuzu Central Hospital,General Surgery,Lilongwe, LILONGWE, Malawi

Introduction: Trauma is a leading cause of morbidity and mortality, particularly in those 15 to 45 years old.  Over 90% of trauma mortality occurs in low- and middle-income countries (LMICs), especially in sub-Saharan Africa. Head injury is the main driver of trauma mortality, specifically in the pre-hospital setting. For patients presenting with torso injury, mortality is potentially preventable if bleeding, particularly from solid organ injury, is controlled expeditiously. We therefore sought to determine the risk of mortality in trauma patients requiring laparotomy in Malawi.

Methods:  This is a retrospective analysis of prospectively collected data at Kamuzu Central Hospital from 2008 – 2017 of admitted patients with torso trauma. Data variables include basic demographics, injury severity and characteristics, surgical intervention, and mortality outcome. Bivariate analysis was performed for covariates based on exploratory laparotomy status. A Poisson regression analysis was performed to estimate risk of mortality after trauma laparotomy controlling for pertinent covariates (injury severity, night time and weekend penetration, injury mechanism, time from injury to presentation).

Results: Over the study period, there were 120,573 trauma patients. Of the 20,522 (17%) patients admitted, 6,474 (31.6%) had torso trauma. Of these, 341 (5.3%) had exploratory laparotomies. Exploratory laparotomy had a male and blunt injury mechanism preponderance of 73.3% and 92.8%, respectively. The crude mortality for patient undergoing exploratory laparotomy versus non-operative management was 9.5% and 6.6 %, respectively. There was an 6.8% overall mortality for torso trauma. Following Poisson regression analysis, the incidence risk ratio for mortality following exploratory laparotomy after controlling for covariates was 3.74 (CI 2.06 -6.78, p <0.001).

Conclusion: After adjusting for injury severity, there is a greater than three-fold increased risk of mortality following trauma exploratory laparotomy. This may be attributable to limited availability of allogenic blood transfusion, inadequate perioperative resuscitation, in-hospital delays to operative intervention including limited access to the operating room, and delays in providers’ decision to perform operative intervention. Trauma protocols are imperative in low-resource settings to optimize timely and appropriate operative management of torso trauma.

 

90.13 Comparison of the Incidence of Wilms’ Tumor across Global Regions

Posted on January 31, 2019

M. E. Cunningham1,2, T. D. Klug2, J. G. Nuchtern1,2, B. J. Naik-Mathuria1,2  1Baylor College of Medicine,Pediatric Surgery,Houston, TX, USA 2Texas Children’s Hospital,Pediatric Surgery,Houston, TX, USA

Introduction:
Wilms’ tumor, also known as nephroblastoma, accounts for more than 90% of kidney neoplasms and 6% of all cancers in children worldwide. Survival after diagnosis and treatment is excellent in most developed countries (>90%), but underdeveloped countries throughout Africa, Asia, and Latin America continue to struggle with detection and treatment resulting in a large discrepancy in survival. The International Agency for Research on Cancer (IARC) has been collecting incidence data since the 1970s and recently released its third volume. The purpose of this study was to compare the incidence of Wilms’ tumor across countries at a global level in order to identify at-risk populations.

Methods:
The World Health Organization (WHO) International Incidence of Childhood Cancer Volume III data set, derived from independent country-based data banks, was queried to identify the incidence of Wilms’ tumor (ages 0-14 years). When multiple registries were available for a single country, the most comprehensive registry was used. The age-specific rate (ASR) per million was compared between developed, transition, and developing countries. Descriptive statistics and independent-sample Kruskal-Wallis Test were utilized.

Results:
Data was available from 75 countries spanning six global regions. The initiation of data collection ranged from 1982-2010 and was last updated between 2006-2014. The median global incidence of Wilms tumor was 7.7 [IQR 5.4-9.1] ASR/million (Figure 1) with a median male to female distribution of 0.9:1. Diagnosis was more common in children aged 0-4 years (median 15.1[IQR 11.8-18.7] ASR/million) compared to 5-9 years (4.2[2.9-5.1]) and 10-14 years (0.7[0.4-1.1]), respectively (p<0.01). Incidence ranged from 2.8 ASR/million in Thailand to 21.2 ASR/million in Mali and was higher in developed (8.9[8.4-9.6] ASR/million) and transition (9.9[9-10.5] ASR/million) countries compared to developing countries (6.1[4.9-7.4] ASR/million) (p<0.01). Of the 6 global regions, the highest incidence was in North America (9.2[9.1-9.2] ASR/million), followed by Europe (9.1[8.6-9.7] ASR/million) and Oceania (8.5[8.2-8.7] ASR/million).

Conclusion:
Wilms’ tumor is a common cancer among children worldwide. It is reported more often in developed and transitional countries; however, this may reflect incomplete data reporting from the developing world. Continued consistent data collection is needed for further elucidation of trends and allocation of resources.
 

90.12 A Systematic Review of Delays and Barriers to the Care of Colorectal Cancer in LMIC

Posted on January 31, 2019

N. R. Brand1, L. Qu2, A. Chao3, A. Ilbawi4  1University Of California – San Francisco,Surgery,San Francisco, CA, USA 2Monash University,Faculty Of Medicine, Nursing & Health Sciences,Victoria, Australia 3National Cancer Institute,Center For Global Health,Bethesda, MD, USA 4World Health Organization,Management Of Noncommunicable Diseases Unit, Department For Management Of Noncommunicable Diseases, Disability, Violence And Injury Prevention,Geneva, Switzerland

Introduction:
Of the 746,000 colorectal cancer (CRC) diagnoses made each year, the majority occur in high-income countries (HIC), while over 50% of deaths occur in low- and middle-income countries (LMIC).  Stage of disease at diagnosis is a significant prognosticator of survival and the higher rates of advanced stage diagnoses made in LMIC may contribute to the difference in death rates between HIC and LMIC.  This review focuses on delays and barriers to CRC diagnoses of patients in LMIC, where CRC incidence is increasing.

Methods:
We conducted a systematic review of peer reviewed literature published on these topics in LMIC.  Inclusion criteria for our systematic review was any full text article that addressed barriers to care or delays in early diagnosis of CRC that was conducted in LMIC.   Studies were required to contain any of the following: (i) defined or reported delay intervals in the diagnosis of symptomatic CRC or (ii) reported predictive factors or barriers that delayed early diagnosis of symptomatic CRC.

Results:
Of the 10,193 abstracts screened, 9 studies met inclusion criteria. All 9 studies were conducted in middle-income countries.  Five studies assessed the intervals along the pathway from symptom onset to cancer treatment, and significant delays were identified along all stages of the cancer care continuum.  All 5 studies identified that the greatest delay occurred prior to disease diagnosis.  Of the 4 studies that assessed individual intervals of CRC diagnosis; 2 (50%) found the greatest delay occurred during the interval between symptom onset and presentation to the healthcare system, and 2 (50%) found the greatest delay occurred between first presentation to the healthcare system and cancer diagnosis.  Six studies assessed barriers to cancer care, and 4 studies assessed knowledge of CRC. All studies found low levels of knowledge of CRC as a disease, its risk factors, or how it is diagnosed, in both the general population and among healthcare workers.

Conclusion:
Despite the  increasing burden of CRC in LMIC, there is little published research on delays to CRC diagnosis and treatment or the barriers that cause them in resource-limited settings. Our review demonstrates significant delays throughout the entire process of cancer diagnosis and treatment and identifies the period prior to CRC diagnosis as the most vulnerable to delays.  In addition, we have identified low levels of knowledge about CRC in both the general population and healthcare workers.  Our study highlights the tremendous need for research and action to reduce CRC morbidity and mortality in LMIC.

90.11 Spatial Accessibility to Colorectal Surgeons in the State of Illinois

Posted on January 31, 2019

J. Day1, W. E. Zahnd2, V. Poola1, J. Rakinic1, S. Ganai1,2  1Southern Illinois University School Of Medicine,Surgery,Springfield, IL, USA 2Southern Illinois University School Of Medicine,Population Science And Policy,Springfield, IL, USA

Introduction:   Disparities in screening, incidence, and mortality for colorectal cancer (CRC) include race/ethnicity, geography, insurance, and other socioeconomic factors. Access disparities can cause treatment delays that impact outcomes. Shortages of gastroenterologists, surgeons, and radiation oncologists have been identified in rural areas. We hypothesized that disparities in spatial access to specialty-trained colorectal surgeons exist between rural and urban Illinois.  

Methods: Data on colorectal surgeon location in Illinois and surrounding border locales were abstracted from the 2014 American Society of Colon and Rectal Surgeons directory with addresses geocoded. Data on population characteristics (age, education, median household income, and race/ethnicity) were obtained from the American Community Survey at the zip code tabulation area (ZCTA) level. Rurality was approximated using the University of Washington’s approximations of the United States Department of Agriculture’s Rural-Urban Commuting Area (RUCA) codes. Network Analyst tool in ARCGIS was used to calculate travel time. Independent t-test and ANOVA were performed to evaluate differences in travel time to the nearest colorectal surgeon by rurality. Choropleth maps were created to display travel time by ZCTA.  

Results: Over half of individuals in isolated-rural (54.1%) and small rural (51.1%) locales in Illinois live more than 60 minutes away from a colorectal surgeon. For those who live in large rural areas, a large proportion (69.4%) live >30-minutes from a colorectal surgeon. In contrast, the majority who live in urban areas (88.8%) had <30 minutes travel time. Overall, mean travel time to a colorectal surgeon was 43.6±27.9 (SD) minutes.  Mean travel time to the nearest colorectal surgeon was significantly greater from rural areas [60.30 ±24.40 minutes (SD)] compared to urban areas [29.10±22.29 minutes; p<0.001). Travel time to the nearest colorectal surgeon also differed across the 4-group RUCA rurality scale (p<0.001), with the greatest travel time for isolated (65.57±22.08 min) and small rural areas (66.88±24.97 min), and shorter travel time for large rural (47.24±19.28 min) and urban areas (29.10±22.29 min). 

Conclusion: The rural population of Illinois experiences a greater burden of travel time for access to colorectal surgeons. The results support the hypothesis that a difference exists in access to colorectal cancer care in rural Illinois via geographic proximity to a colorectal surgeon. Further analysis including correlation to incidence and mortality and the role of complementary providers will be necessary to properly assess access to CRC care needs in Illinois.

« Previous Page
Next Page »