92.14 Feasibility of Using Real-time Location Systems in Monitoring Recovery after Major Abdominal Surgery

Posted on January 26, 2017

R. Dorrell3, F. Hsu1, S. Vermillion3, C. Clark2  1Wake Forest University School Of Medicine,Public Health Sciences,Winston Salem, NC, USA 2Wake Forest University School Of Medicine,Department Of Surgery,Winston Salem, NC, USA 3Wake Forest University School Of Medicine,Winston Salem, NC, USA

Introduction:

Early mobilization after major abdominal surgery decreases postoperative complications and length of stay and has become a key component of enhanced recovery pathways.  However, patients face substantial barriers to early mobilization and objective measures of patient movement after surgery are limited.  Real-time location systems (RTLS) that are typically used for asset tracking provide a novel approach to monitoring in-hospital patterns of movement.  Since existing systems were not implemented for tracking dynamic patient movement, the current study investigated the feasibility of using RTLS to objectively track postoperative patient mobilization.

Methods:  

The real-time location system used for this study employs a meshed network of infrared and radio frequency identification sensors and detectors that sample device locations nearly every 3s resulting in over 1 million data points per day throughout an academic medical center.  RTLS tracking was evaluated systematically in three phases: 1) sensitivity and specificity of the tracking system using simulated patient scenarios, 2) retrospective passive movement analysis of patient-linked equipment (patient IV poles), and 3) prospective observational analysis of patient-attached tracking devices. 

Results:  

RTLS tracking detected simulated movement in and out of a patient room with sensitivity of 91% and specificity 100%.  Specificity decreased to 75% if time out of room was less than 3 minutes.  Sensor badge position (chest or wrist) did not change sensitivity or specificity.  All tracked patient-linked equipment (bed, IV poles, pumps, etc) were identify for 18 patients following major abdominal operations.  Movement of this equipment was retrospectively reviewed and analyzed.  For any individual patient, an IV pole may contain 4 tracking badges; therefore, the analysis was limited to a single IV pole tracking badge during the postoperative hospitalization.  Median length of stay was 6.7 days with 3386 location data points recorded.  Devices remained in the patient's room for mean duration of 684 min (SD 1216 min) per day.   Measurable patient movement on the ward was detected for only 2 patients (11%) with 1-8 out-of-room walks per day.   10 patients were prospectively monitored using wrist worn RTLS badges following major abdominal surgery. Patient movement was also recorded using patient diary, direct observation, and pedometer.  Sensitivity and specificity of RTLS patient tracking were both 100% in detecting out-of-room ambulation and correlated well with direct observation and patient-reported ambulation. 

Conclusion:

Real-time location systems are a novel technology capable of objectively and accurately monitoring patient movement and provides for an innovated approach to promoting early mobilization after major abdominal surgery.

92.13 In-Hospital Text-Paging Communication as a Surgical Quality Improvement Initiative

Posted on January 26, 2017

M. Janko1, M. G. Noujaim2, K. F. Angell1, J. Hill1, J. Kalil1, S. Steele3  3Case Western Reserve University School Of Medicine,Colorectal Surgery,Cleveland, OH, USA 1Case Western Reserve University School Of Medicine,Vascular Surgery,Cleveland, OH, USA 2University Of Massachusetts Medical School,School Of Medicine,Worcester, MA, USA

Introduction:  Medical staffs rely on paging to communicate patient safety concerns and updates in the hospital. Studies on medical wards have shown that numeric callback pages can be both disruptive and cryptic. In contrast, alphanumeric text paging using a hospital-issued pager on medicine wards has been shown to reduce disruptive pages and raise satisfaction scores among healthcare professionals. Here we report a quality improvement initiative among nurses and surgical interns involving text paging to communicate urgent and non-urgent issues on a surgical ward.

Methods:  Surgery residents recorded pre-intervention data for 1 month including average patient census, number of urgent and non-urgent pages received from surgical floors, number of text and traditional call-back pages from surgical floors, total number of pages received and major adverse events. Surgical nurses and residents completed surveys to assess pre-intervention satisfaction, responsiveness and workflow. Surgical nurses were then instructed to utilize text-paging to communicate with residents for non-urgent issues. Urgent communications continued to be communicated with traditional numeric callback pages. Paging data was again recorded for 1 month and surveys were repeated. Statistical analysis using Chi-squared and student’s t-tests were used to compare pre- and post-intervention results.

Results: After text paging implementation 40.2% of non-urgent pages sent from nurses to resident physicians were alphanumeric texts vs. only 17.9% before implementation (P < 0.0001), and there was a 19.5% reduction in the number of non-urgent numeric callback pages sent from nurses to physicians (P < 0.0001). 70% of nurses surveyed post-implementation responded alphanumeric text paging was the preferred method of contacting a physician. 70% of nurses thought text paging initiative improved efficiency in triaging and responding to pages. After implementation, 62% of nurses thought that overall communication with responding clinicians improved. Furthermore, residents reported increased satisfaction, improved workflow, and decreased educational interruptions with text paging. 

Conclusion: We successfully implemented a free web-based text paging initiative for all non-urgent pages from nurses to residents that has improved physician-nurse workflow and communication on the surgical wards without an increase in adverse events.

 

92.12 Trends in Parastomal Hernia Repair in the United States

Posted on January 26, 2017

T. Gavigan1, B. Matthews1, N. Rozario1, C. E. Reinke1  1Carolinas Medical Center,Department Of Surgery,Charlotte, NC, USA 2Carolinas Medical Center,Dickson Advanced Analytics,Charlotte, NC, USA

Introduction:  Parastomal hernia is the most common complication after stoma creation. An estimated 120,000 new stomas are created each year. Recent studies report an parastomal hernia incidence approaching 80%, with more than half requiring surgical repair. Parastomal hernias create significant morbidity, including patient discomfort, small bowel obstruction, and need for emergent surgery. Little is known about the rates of parastomal hernia repair over the last 10 years in the United States. We examined national trends in parastomal hernia repair (PHR) in this study, including annual frequency of procedure, patient characteristics, and same-admission complications.  

 

Methods: The 1998-2011 Nationwide Inpatient Sample was used to identify patients who underwent a PHR (ICD-9 PR 4642).   PHRs were classified as PHR with concurrent resiting (ICD-9 PR 4643), PHR with concurrent ostomy reversal (ICD-9 4652 or 4651), or primary PHR. Patient age, race, sex, comorbidities and type of insurance were identified. Complications, length of stay (LOS), and mortality were identified. The frequencies of patient characteristics and outcomes were calculated by year and by type of PHR and analyzed to identify trends.   

 

Results: The estimated number of annual parastomal hernia repairs increased from 4,161 to 7,646 (p=<0.01, R2=0.85) for a total of 73,659 repairs. 30% underwent a concurrent stoma reversal and 10% underwent a resiting.  The proportion of females undergoing PSHR remained steady (58%). There was an upwards trend in the proportion of privately insured patients (26%-31%, p<0.01) and the number of patients with 3 or more Elixhauser comorbidities (17%-44%, p<0.01). The frequency of reversal increased while the frequency of resiting decreased. LOS remained steady (median 6.3 days) and in-hospital mortality ranged from 1.8-3.9% annually. Mortality and emergency admission status were highest for patients who underwent primary PHR, while the distribution of number of comorbidities was not significantly different between the three groups.

 

Conclusions: The incidence of parastomal hernia repair nationwide is increasing and more than half of patients undergo primary repair.  Although the surgical focus has moved towards prevention, parastomal hernia is a persistent complication of stoma creation. Further exploration is warranted to determine if the observed increase in parastomal hernia repair is related to perceived improved techniques and outcomes, an increasing incidence of parastomal hernia, patient characteristics or other factors.  

92.11 Genitourinary Paraganglioma: An Analysis of the SEER Database (2000-2012)

Posted on January 26, 2017

S. Purnell1, A. Sidana1, M. Maruf1, C. Grant2, S. Brancato1, P. Agarwal1  1National Cancer Institute,Urologic Oncolocy Branch,Bethesda, MD, USA 2George Washington University Hospital,Urology,Washington, DC, USA

Introduction: Extra-adrenal paragangliomas (PGL) are infrequent, benign, neuroendocrine tumors arising from chromaffin cells of the autonomic nervous system. While most develop above the umbilicus, they have been reported in the genitourinary (GU) tract. Due to the paucity of literature on the rates of GU paraganglioma, our study aims to describe demographic, pathologic, and clinical characteristics of GU PGL, and compare them to non-GU sites of PGL.

Methods: Data was collected from the SEER 18 Database to compare GU and non-GU PGL diagnosed between 2000 and 2012. Chi-square and unpaired t-tests were used. Kaplan-Meier analysis and a log rank test were used to determine overall survival and statistical significance, defined as p<0.05.

Results:299 cases of PGL were retrieved and only 20 (6.7%) arose from the GU tract. 83.3% GU PGLs developed in the bladder, subsequently the kidneys/renal pelvis (16.7%), and spermatic cord (2%). Non-GU PGL developed most frequently within the endocrine system (43%). Overall, PGL was more common in men than women. The mean age at diagnosis in years was higher in non-GU than GU PGL (50.4±17.2 vs 40.8±15.6, p=0.026). GU PGL was less common in whites compared to PGL at other sites (p=0.033). The majority (50%) of GU PGL was organ confined while 5.7% of non-GU PGL was localized at diagnosis. All cases of PGL were treated with surgery. 30% of patients with GU PGL underwent LN dissection and none had radiation. There were 2 (10%) cause-specific deaths in the GU PGL groups between 2000 and 2012. 5-year overall survival was 93.3% for GU PGL versus 65.5% in non-GU PGL (p=0.062).

Conclusion:Genitourinary PGL remains rare, with low incidence (6.7% of all PGL cases) in the US population between 2000 and 2012. Also, it had high 5-year overall survival compared to PGL developing outside of the GU tract. The bladder represents the most common site of involvement and surgery is the mainstay of treatment for GU PGL. Clearer prognostic factors are needed to better elucidate PGL management in the future thus pooled studies from various institutions with detailed clinical information are needed to delineate these prognostic factors.

 

92.10 Wound Selfie: an App to Reduce Surgical Site Complications and Improve Patient-Provider Communication

Posted on January 26, 2017

G. M. Taylor1, A. L. Guzman4, J. Wilder6, A. Nakhmani5, R. T. Russell2, J. A. White3, J. H. Willig4  1University Of Alabama at Birmingham,Department Of Surgery,Birmingham, Alabama, USA 2University Of Alabama at Birmingham,Div. Of Pediatric Surgery,Birmingham, Alabama, USA 3University Of Alabama at Birmingham,Div. Of Transplantation,Birmingham, Alabama, USA 4University Of Alabama at Birmingham,Div. Of Infectious Disease,Birmingham, Alabama, USA 5University Of Alabama at Birmingham,Department Of Electrical And Computer Engineering,Birmingham, Alabama, USA 6Datascription,New York, NY, USA

Introduction: Surgical site complications (wound infection, dehiscence, and evisceration) can prolong hospital stays, contribute to readmissions, and necessitate subsequent operations.  Early identification of complications improves patient outcomes and reduces healthcare expenditures.  In our current practice, providers lack a consistent channel to monitor wound sites and communicate with patients after hospital discharge prior to scheduled follow-up.  Phone reports from patients are often insufficient, and photos are difficult to send securely.  Our goal was to create a safe and efficient platform for patients to transmit wound site photos and other clinical information to their providers. 

Methods: A multidisciplinary team including advanced practitioners, software engineers, surgeons, and infectious disease physicians was convened to leverage health informatics technology and develop a platform for communication with patients after post-operative discharge. In addition, we intend to train image recognition software and machine learning algorithms to independently identify patients with concerning wounds who needed earlier follow-up based on their wound images and supplemental patient questions.

Results:  The resulting “Wound Selfie” app for Android and iPhone iOS smartphone platforms provides a secure, HIPAA-compliant tool for providers to evaluate surgical wounds and communicate with patients. At hospital discharge, providers introduce the smartphone app to the patient and help them take an initial wound photo.  Patients submit pictures of wound sites regularly for review by the surgical team. In addition, patients answer a series of questions about symptoms related to wound problems and systemic symptoms that help providers provide timely follow-up. Providers can review patients’ wound images in the app or a desktop software interface. Digital analysis of these images will train algorithms that incorporate symptoms and risk factors for wound complications. Ultimately, the app will independently identify patients who need early post-operative intervention.

Conclusions: This app adds a new dimension to patient-provider communication.  Surgeons with limited staff and patients who may live far from providers can communicate more efficiently about their postoperative care. This app will serve as a tool to facilitate population health. For example, transplant patients, a regional population with higher rates of wound complications, can instantly send wound-site images and subjective history to their tertiary care center for review.  While further study is warranted, we are excited about the potential of new technological and interdisciplinary approaches to improve clinical outcomes for patients.

92.09 Laparotomy Trends Observed in 9,950,759 Patients Using 2009-2013 National Inpatient Sample (NIS)

Posted on January 26, 2017

M. J. Carney1, J. P. Fox1, J. M. Weissler1, J. P. Fischer1  1University Of Pennsylvania,Division Of Plastic Surgery,Philadelphia, PA, USA

Introduction:  There are between 4 to 5 million laparotomies performed annually in the United States (US) despite a distinct trend towards minimally invasive surgery. Laparoscopic surgical approaches have resulted in decreased length of stay, with no changes in short term complications, while still possessing similar oncologic outcomes to more traditional open surgery. Up to 30% of patients undergoing laparotomy will develop incisional hernia (IH), amounting a cost burden of $3.2 billion annually. Despite the paradigm shifts towards minimally invasive techniques, persistent hernia morbidity poses an equivocal disconnect warranting critical review. We aim to address these trends through analysis of the largest all-payer inpatient care database. 

Methods:  Using the 2009-2013 Nationwide Inpatient Sample, we conducted a cross-sectional review of hospital discharges of open abdominal surgery. For each event the database offered diagnostic and procedural coding (i.e., ICD-9-CM and CPT-4), as well as other socio-demographic and clinical variables. We sub-grouped the resultant discharges into the following categories: endocrine, vascular, hematologic and lymphatic, esophagus and stomach, intestine (small and large), hepatobiliary and pancreas, hernia, urology, other abdominal, gynecology, obstetrics, and transplant. To assess comorbidities, enhanced-Elixhauser algorithm was used. 

Results: Between 2009 and 2013, there were nearly 10 million discharges associated with an open abdominal surgery based on our collection of ICD-9 procedural code identifiers. Overall, there were 2,140,616 patients receiving open surgery in 2009, decreasing to 1,760,549 in 2013 (0.82% change, p<0.001). Each subgroup demonstrated a congruent decrease in open procedures except for hernia. These procedures increased from 37,325 patients in 2009 to 41,845 in 2013 (1.12% change, p=0.001). The most prevalent comorbidities within this population included uncomplicated hypertension (25.26%), chronic pulmonary diseases (13.52%), obesity (10.24%), uncomplicated diabetes (11.06%), and depression (10.72%).

Conclusion: Our large volume analysis allowed for a unique view of surgical trends, health care population dynamics, and an opportunity to use evidence-driven analytics in the understanding of IH. Previous studies have primarily focused on categorizing IH repair techniques, occurrence risk factors, and recurrence within a specific surgical field. Public health initiatives in a preventative model are paramount and encourage health care providers to implement best practice techniques at point of care. 

 

92.08 Impact of Hospital Volume on Outcomes for Laparoscopic Lysis of Adhesions for Small Bowel Obstruction

Posted on January 26, 2017

R. A. Jean1, K. M. O’Neill1, K. Pei2, K. A. Davis2  1Yale University School Of Medicine,Department Of Surgery,New Haven, CT, USA 2Yale University School Of Medicine,Department Of Surgery, Division Of General Surgery, Trauma And Surgical Critical Care,New Haven, CT, USA

Introduction:  Volume to outcome data has been studied in several complex surgical procedures, demonstrating improved outcomes at higher volume centers. Laparoscopic lysis of adhesions (LLOA) for small bowel obstruction (SBO) may result in better outcomes, but there is no information on the learning curve for this potentially complex case. This study evaluates the effect of procedural volume on length of stay, outcomes, and costs in laparoscopic lysis of adhesions for small bowel obstructions.

Methods: The Nationwide Inpatient Sample (NIS) dataset between 2000 and 2013 was queried for discharges for a diagnosis of SBO involving LLOA in adult patients. Patients with intra-abdominal malignancy and evidence of any other major surgical procedure during hospitalization were excluded. The procedural volume per hospital was calculated over the time period, and high volume hospitals were designated as those performing greater than 25 weighted LLOA per year. Patient characteristics were described by hospital volume status using stratified cluster sampling tabulation and linear regression methods. Length of stay (LOS), cost, and total charges were reported as means with standard deviation and median values. P<0.05 was considered significant.

Results:A total of 9,111 discharges were selected, which was representative of 43,567 weighted discharges nationally between 2000 and 2013. Over the study period, there has been a 450% increase in the number of LLOA performed. High volume hospitals had significantly shorter LOS (mean 4.92 days (SE 0.13); median 3.6) compared to low volume hospitals (mean 5.68 (SE 0.06); median 4.5). In multivariate analysis, high volume status was associated with a decreased LOS of 0.72 days (p <0.0001) as compared to low volume status. Other significant predictors for decreased LOS included decreased age, decreased comorbidity, and the absence of small bowel resection. There was no significant association between volume status and total charges in multivariate or univariate models but, high volume hospitals were associated with lower costs in multivariate models by approximately $984 (p=0.017). 

Conclusion:This study demonstrates that high hospital volume was associated with decreased length of stay for LLOA in SBO. Although volume was not associated with differences in total charges, there was a small decrease in hospital costs.

 

92.07 Mobile Application Design and Human Factors for Global Health Innovation

Posted on January 26, 2017

T. Schwab1,2, J. Langell1,2  1University Of Utah,Center for Medical Innovation,Salt Lake City, UT, USA 2University Of Utah,Department Of Surgery,Salt Lake City, UT, USA

Introduction:  The use of mobile technologies and software applications offers the opportunity for improvements in global healthcare delivery.  If the potential for these technologies are to be realized, human factors must be considered when designing applications, devices, and other mobile interfaces. The goal of human factors in the design process is to decrease patient safety risks and improve clinical outcomes.  The objective of this study was to determine the effectiveness and value of human factors-based application design for global health innovation.

Methods: We implemented a 4-stage development process (Figure 1) for global health mobile application design by 12 teams to develop a medical application prototype for use by developing world populations. The stepwise development process involves design specifications and market requirements identification, app software optimization, app implementation testing and user feedback, and establishing a system to promote continual improvement during the life of the application. Teams provided feedback through formal surveys and interviews to highlight key aspects of the development process which directly contributed to successful app design. Teams scored all components of the design process based on clinical outcome value.  “Significant outcome value” was defined as an overall 90 out of 100-point score, or higher.

Results: 

Five elements of the design process were identified as having significant outcome value in the medical app development process. 1) User observation: Outcomes score of 119/120.  Defined as “Understand the clinical problem and the correlated patient characteristics though user observation and validation.”  2) Safety: Outcomes score of 117/120.  Defined as “Address and ensure safety in every stage of the mobile medical application process, including development of a risk assessment model.”  3) Ease of navigation: Outcomes score of 110/120.  Defined as “A functional and well-designed navigational menu, intuitive interface, and simple multitasking functions.” 4) Anatomy: Outcomes score of 108/120.  Defined as “Human hand anatomy and capabilities consideration in the interface design process for optimal flow, function and performance.” 5) Environment: Outcomes score of 104/120.  Defined as “Identification of unique environmental factors early in the design and development process resulting in overcoming potential challenges.”

 

Conclusion: To encourage adoption of medical mobile applications in developing regions, developers must create safe, accurate and layman-friendly applications. This can be effectively accomplished through an iterative human factors–based design process involving the targeted end user.
 

92.06 Machine Learning to Identify Multigland Disease in Primary Hyperparathyroidism

Posted on January 26, 2017

J. Imbus1, R. W. Randle1, S. C. Pitt1, R. S. Sippel1, D. F. Schneider1  1University Of Wisconsin,Department Of Surgery,Madison, WI, USA

Introduction:  Most patients with primary hyperparathyroidism (PHPT) have a single adenoma (SA), but 20-25% of cases will have multigland disease (MGD). Preoperatative localization of SAs allows for a minimally invasive surgical approach, but these studies are less accurate or unnecessary in MGD. Therefore, pre-operative identification of MGD could direct the need for imaging as well as the operative approach, and potentially referral to experienced surgeons. Machine learning (ML) uses computer algorithms to build predictive models from labeled datasets. The purpose of this study is to use ML methods to predict MGD.

 

Methods:  We reviewed a prospectively managed database of patients undergoing parathyroidectomy from 2001 to 2016. Patients (age ≥ 18 years) with PHPT who underwent initial, curative resection were included. MGD was defined as > 1 gland removed. Patients with genetic syndromes, a history of lithium use, prior neck surgery, or parathyroid carcinoma were excluded. The ML platform WEKA was utilized to compare different classifiers for predicting SA vs MGD from demographic, clinical, and laboratory features. The meta-algorithm, bagging, which reduces variance by averaging probability estimates, was applied. We selected the model with the best overall accuracy and separately used cost-sensitive classifier to maximize sensitivity for MGD.  10-fold cross validation was used to evaluate accuracy.

 

Results: 2035 patients met inclusion criteria: 1522 patients had SA (75%) and 513 had MGD (25%). After testing many algorithms, we selected the rule-based algorithm, PART, for its accuracy and potential integration in a clinical decision-support tool.  Sample rules are shown in the figure.  Using PART with bagging achieved 78% accuracy; 78% recall (sensitivity), 45% specificity, 76% precision (PPV), 0.710 Area Under the Receiver Operating Characteristics curve (AUC). To maximize sensitivity of detecting MGD, the cost-sensitive classifier achieved 89% sensitivity, 0.697 AUC for MGD.  To validate the algorithm’s impact on practice, we reviewed imaging from a separate test set of 50 patients with MGD. The algorithm correctly identified 49 of these 50 patients (98%). Among these, 43 sestamibi scans and 14 ultrasounds were performed.  However, only 14 sestamibi scans and 4 ultrasounds were correct.  Eliminating the incorrect or non-localizing studies would have provided a potential cost savings of over $1200/patient.

Conclusion: Rule based ML methods can help distinguish SA from MGD early in the clinical evaluation to guide further workup including localization studies. ML can potentially save money spent on unnecessary imaging studies or guide referral to high volume surgeons who are comfortable with bilateral exploration for MGD.

92.05 Limitations of Comparing NSQIP Outcomes Over Time

Posted on January 26, 2017

E. M. Gleeson1, A. P. Johnson2, M. E. Kilbane3, H. A. Pitt4  1Drexel University College Of Medicine,Philadelphia, Pa, USA 2Thomas Jefferson University,Philadelphia, PA, USA 3Indiana University School Of Medicine,Indianapolis, IN, USA 4Temple University,Philadelpha, PA, USA

Introduction:  The National Surgical Quality Improvement Program (NSQIP) provides a unique resource to researchers at participating institutions through the de-identified national Participant Use Files (PUF).  In an effort to improve data collection, some definitions for the outcome variables have changed over time. Despite these changes, researchers often combine the PUFs across years to increase the power of their studies.  This analysis aims to determine if these definition changes have affected the rates of the outcomes recorded.

Methods:  We reviewed cases collected by NSQIP from 2005-2014.  Control charts were utilized to evaluate the association between variable definition changes and special cause variation in the occurrence rates of surgical outcomes over each admission quarter.  Mortality, known to have decreased since the institution of NSQIP without any variable changes, was used as a negative control variable.  We particularly focused on postoperative occurrences with the most dramatic variable changes: myocardial infarction (2009), bleeding requiring transfusion (2010), , and sepsis (2013).

Results: We found that the majority of postoperative occurrences decreased over time, consistent with mortality.  However, special cause variation was noted for myocardial infarction (2009), bleeding requiring transfusion (2010),  and sepsis (2013) with violation of control chart rules at the time of significant variable definition changes (Figure).

Conclusion: This study demonstrates that special cause variation in postoperative occurrences may correlate with significant variable definition changes.  This issue severely limits conclusions drawn by comparing these outcomes over time periods with significant definition changes.  Authors and readers should remain vigilant to the limitations of NSQIP data and be aware of definition changes when comparing outcomes across years.

 

92.04 Predictors of ED Visits and Readmissions Within One Year of Bariatric Surgery: A Statewide Analysis

Posted on January 26, 2017

M. C. Mora Pinzon1, D. Henkel6, R. E. Miller2, P. L. Remington1, S. N. Kothari4, J. Gould3, L. M. Funk5,6  1University Of Wisconsin,School Of Medicine And Public Health,Madison, WI, USA 2Wisconsin Department Of Health Services,Madison, WI, USA 3Medical College Of Wisconsin,Department Of Surgery,Milwaukee, WI, USA 4Gundersen Health System,Department Of Surgery,Milwaukee, WI, USA 5William S. Middleton VA Hospital,Madison, WI, USA 6University Of Wisconsin,Department Of Surgery,Madison, WI, USA

Introduction:  30-day complication and readmission rates following bariatric surgery are well reported. However, there are limited data regarding bariatric surgery readmissions and Emergency Department (ED) utilization beyond 30 days. In this study, we identified all ED visits and readmissions to any facility in Wisconsin within one year of bariatric surgery, and we examined patient and hospital characteristics associated with these visits.

Methods: Statewide hospital patient data collected by the Wisconsin Hospital Association were used to identify all obese patients >20 years old who underwent laparoscopic Roux-en-Y gastric bypass (RYGB) or laparoscopic sleeve gastrectomy (LSG) from 2011-2014. Patient demographics, complications during bariatric surgery hospitalization, and primary and secondary diagnoses were identified using the International Classification of Diseases, Ninth Revision (ICD-9). Iterative deterministic linkage was used to track individuals who subsequently received hospital care over the next year. Bivariate associations between patient/hospital factors and ED visit or readmission were examined. Factors significant at a p<0.1 were included in a multivariable logistic regression model.

Results: 5,701 procedures were identified: 70% RYGB (n=3,988), 30% LSG (n=1,713). 79% of the patients were female. The mean age was 45.7 years old (SD: 11.5). 39% of patients presented to the ED or were readmitted within one year of bariatric surgery. The frequency of ED visits during the first year ranged from 10.7% in the first 30 days to 5.7% during postoperative days 181-270. Readmission rates ranged from 4.4% in the first 30 days to 2.7% during postoperative days 91-180 (Figure). On multivariable analysis, an ED visit within 1 year of bariatric surgery was associated with younger age, female gender, RYGB (vs. sleeve), having ≥4 comorbidities, Medicare or Medicaid insurance, teaching hospital for index procedure, and experiencing a complication during the initial bariatric surgery hospitalization (all p <0.05). Readmission within one year was associated with male gender, RYGB, ≥4 comorbidities, Medicare insurance, teaching hospital, and complication during the initial bariatric surgery hospitalization (all p <0.05).

Conclusion: ED visits and hospital readmissions are more common within the first 30 days after bariatric surgery, but persist steadily throughout the first postoperative year. Quality improvement efforts focused on patients who are most likely to visit the ED or be readmitted – such as those who undergo a gastric bypass or experience a complication during their bariatric surgery admission – may improve outcomes and decrease hospital resource utilization.

92.03 Use of Mobile Health Technologies to Monitor Postoperative Recovery: Barriers and Advantages

Posted on January 26, 2017

J. S. Abelson1, E. J. Kaufman1, M. E. Charlson2, H. Yeo1,3  1Weill Cornell Medical College,Surgery,New York, NY, USA 2Weill Cornell Medical College,Integrative Medicine,New York, NY, USA 3Weill Cornell Medical College,Healthcare Policy And Research,New York, NY, USA

Introduction:  Using smartphone and mobile health (mHealth) technology to monitor recovery after surgery has the potential to improve postoperative care, prevent complications, and reduce cost. We evaluated individuals’ perception of barriers and advantages to using mHealth after surgery. 

Methods:  Data were collected by Cornell University’s Survey Research Institute. Interviews were performed across New York State. Respondents were asked the following open-ended questions: 1. What are barriers or issues you might see to using a free mobile health app after surgery to improve your care? 2. What are benefits you might see to using a free mobile health app after surgery to improve your care? Of the 800 responses, 200 responses were coded independently by 3 reviewers to develop a preliminary codebook. The subsequent 600 responses were coded independently by 2 reviewers. The codebook was refined iteratively with all changes independently verified by the 3rd reviewer; all disagreements were resolved by consensus. We used modified grounded theory to allow themes to arise from the data. 

Results: The average age of our cohort was 47 yrs (+17yrs) with an equal distribution of male and female participants. Most respondents were White (67%) while 13% identified as having Hispanic ethnicity. The most common barrier identified was protecting privacy and security of personal health data. Although less common, other areas of concern included: uncertainty about accessibility and usefulness of mHealth; preference for face-to-face interaction wtih surgeon; and high effort required by patients. There were several advantages noted by respondents; the most commonly cited advantages were the potential for mHealth to improve recovery and prevent complications as well as to strengthen communication and the relationship with surgeon. Other advantages included: increased patient knowledge and self-engagement in recovery; and saving time and money by reducing doctor visits. Several respondents identified no barriers, and a few identified no advantages. 

Conclusion: This is the first large-scale qualitative analysis to evaluate perceptions of barriers and advantages in using mHealth after surgery to monitoring recovery. The majority of participants identified a narrow range of barriers, primarily focusing on confidentiality and data security.  Nearly all participants identified a wide variety of potential advantages, ranging from improved communication with surgeons to preventing complications and saving time and money. These results indicate that while participants were in general willing to use mHealth and perceived many benefits, design and promotion of these apps should address patient concerns about data security and technology accessibility.

92.02 The Utility of Twitter in Generating High-Quality Conversations about Surgical Care

Posted on January 26, 2017

N. NAGARAJAN1,2, H. Alshaikh1, A. Nastasi1, B. Smart3, Z. Berger6, E. B. Schneider4, M. Dredze5, J. K. Canner1, N. Ahuja1  6Johns Hopkins University School Of Medicine,General Internal Medicine,Baltimore, MD, USA 1Johns Hopkins University School Of Medicine,Surgery,Baltimore, MD, USA 2Brigham And Women’s Hospital,Boston, MA, USA 3University Of Southern California,Surgery,Los Angeles, CA, USA 4Brigham And Women’s Hospital,Center For Surgery And Public Health,Boston, MA, USA 5Johns Hopkins University,Computer Science,Baltimore, MD, USA

Introduction:
There is growing interest among various stakeholders in using social media sites to discuss healthcare issues. However, little is known about how social media sites are used to discuss surgical care. There is also a lack of understanding of the types of content generated and the quality of the information shared in social media platforms about surgical care issues. We therefore sought to identify and summarize conversations on surgical care in Twitter, a popular microblogging website.  

Methods:
A comprehensive list of surgery-related hashtags was used to pull individual tweets from 3/27-4/27/2015. Four independent reviewers blindly analyzed 25 tweets to develop themes for extraction from a larger sample. The themes were broadly divided further to obtain data at the levels of the user, the tweet, the content of the tweet and personal information shared (Figure I). Standard descriptive statistical analysis and simple logistic regression analysis was used. 

Results:

In total, 17,783 tweets were pulled and 1000 from 615 unique users were randomly selected for analysis. Most users were from North America (62.4%) and non-healthcare related individuals (31.8%). Healthcare organizations generated 12.4%, and surgeons 9.5%, of tweets. Overall, 67.4% were original tweets and 79.0% contained a hyperlink (11% to healthcare and 8.7% to peer-reviewed sources).  The common areas of surgery discussed were global surgery/health systems (18.4%), followed by general surgery (15.6%). Among personal tweets (n=236), 31.1% concerned surgery on family/friends and 24.4% on the user; 61.1% discussed procedures already performed and 58.0% used positive language about their personal experience with surgical care.

Surgical news/opinion was present in 45% of tweets and 13.7% contained evidence-based information. Non-healthcare professionals were 53.5% (95% CI: 3.8%-77.5%, p=0.039) and 72.8% (95% CI: 21.1%-91.7%, p=0.017) less likely to generate a tweet that contained evidence-based information and to quote from a peer-reviewed journal, respectively, when compared to other users. 

Conclusion:

Our study demonstrates that while healthcare professionals and organizations tend to share higher quality data on surgical care on social media, non-health care related individuals largely drive the conversation. Fewer than half of all surgery-related tweets included surgical news/opinion; only 14% included evidence-based information and just 9% linked to peer-reviewed sources.  As social media outlets become important sources of actionable information, leaders in the surgical community should develop professional guidelines to maximize this versatile platform to disseminate accurate and high-quality content on surgical issues to a wide range of audiences. 

 

92.01 Vena Cava-sparing Piggyback Hepatectomy in Liver Transplant Patients with Hepatocellular Carcinoma

Posted on January 26, 2017

W. J. Bush1, C. A. Kubal1, J. A. Fridell1, B. Ekser1, R. C. Graham1, K. A. Thatch1, R. S. Mangus1  1Indiana University School Of Medicine,Transplant,Indianapolis, IN, USA

Introduction:
Liver transplant (LT) patients with hepatocellular carcinoma (HCC) are at risk for post-transplant tumor recurrence. Risk of HCC recurrence is known to be associated with the size and number of tumors present within the liver. Close proximity of tumor to major vascular structures may also increase the risk of tumor recurrence. For that reason, most surgeons employ a conventional bicaval technique, replacing the entire vena cava as part of the LT. Our center has previously published data suggesting that the vena cava-sparing piggyback (PGB) technique can be safely used without affecting clinical outcomes. This study reviews a large number of LT patients with HCC to determine long-term outcomes of using the PGB technique, as well as the impact of tumor proximity to the vena cava on recurrence rates.  

Methods:
The records of all adult patients undergoing liver transplant (LT) at a single center over a 15 year period were reviewed. Patients with HCC were extracted for further analysis. The operative records for all HCC patients were reviewed to determine if the CONV or PGB hepatectomy technique was utilized. Original computed tomography scans were reviewed to measure distance between the vena cava and the nearest tumor, and to determine which segments of the liver had tumor present. Outcomes included HCC recurrence and long term patient survival. Cox regression 10-year patient survival was calculated.

Results:
There were 1722 LT patients, and 393 were found to have HCC (23%). Among these patients, 367 (93%) underwent LT with PGB technique, while 26 had CONV hepatectomy (7%). The PGB patients were older and had an older donor age, but had lower cold and warm ischemia time. The PGB patients were more likely to have HCC in segments adjacent to the vena cava (57% vs 34%, p=0.02), but the median distance to the nearest tumor was greater for the PGB group (45 vs 28mm, p=0.06). There was no significant difference in tumor recurrence between PGB and CONV (16% vs 19%, p=0.70), nor was there a difference by Cox regression in survival at 10-years (p=0.13). Predictors of recurrence included being outside Milan criteria, and increased tumor size and number, but not tumor distance to the vena cava.

Conclusion:
These results demonstrate no significant difference in clinical outcomes between the PGB and CONV surgical techniques in LT patients with HCC. Tumor presence near the vena cava was not associated with increased risk of HCC recurrence.
 

91.20 Inguinal Hernia Repair In Octogenarians And Beyond

Posted on January 26, 2017

V. Jain1, M. S. Sultany1, T. Madni1, S. Huerta1,2  1University Of Texas Southwestern Medical Center,General Surgery,Dallas, TX, USA 2VA North Texas Health Care System,General Surgery,Dallas, TX, USA

Introduction:  Elective inguinal hernia repair (IHR) in elderly patients might still cause apprehension in general surgerons especially in the era of acceptable watchful waiting. Reports also indicate that elective rather than emergent operations should be performed for octogenarians presenting with an IH. We elected to interrogate a data base in veteran patients that controls for institution and surgeon. We hypothesize that outcomes are similar independent of age.  

Methods:  This is a retrospective review of data from a single institution and by a single surgeon. We reviewed the entire database for outcomes of IHR on all age groups between July 2005 and April 2016, which included 933 patients and over 1000 IHR.  We separated octogenarians (80-89 years) and nonagenarians (90-99yrs) (Group 1) from everyone else under 80 years (Group 2) and proceeded to compare the two groups with primary outcome studied being major complications (Inguinodynia and Recurrence) and secondary outcomes being all other morbidities/minor complications. Descriptive statistics have been used for patient demographics, X2 was used for comparison of categorical data and student t- test was used for continuous variables.

Results: Of the 933 patients, 57 (6.1 %) were octogenarians and 2 (0.2%) were nonagenarians. Mean age in group 1 was 83.68 ± 3.02 years and in group 2 was 59.28 ± 11.35 years. Both groups: Men=99.8%. Group I vs group II: bilateral IH=8.4% vs 8.5% (p=0.97), Incarcerated IH 18 % vs 6.0% (p <0.01), BMI  24.93 ± 3.36 vs. 27.06 ± 10.21 kg/m2 (p=0.06), ASA I/II 20% vs. 49.4 % (p< 0.01), ASA III/IV 79.2%, vs. 49.2% (p< 0.01). OR time 62.59 ± 21.38 min vs. 60.26 ± 23.77 min for unilateral repairs (p=0.22) and 132.8 ± 35.29 min vs. 103.4 ± 25.62 min for bilateral repair (p=0.07).  Rate of inguinodynia was 0% vs. 1.9% (p=0.27), recurrence rate was 1.7% vs 0.8% (p=0.47).  Minor complications were more common in group I (20.3% vs. 9.38% (p=0.006).  Length of stay (LOS)=1.23 ± 5.99 d vs. 0.28 ± 2.01 d (p=0.23), LOS > 2d=8.47% vs 2.4% (p=0.006).  Multivariate analysis did not identify and independent predictors of major or minor complications.  

Conclusion: There is no significant difference in the incidence of major complications for IHR in the octogenarian population.  There is a significantly higher incidence of minor complications and prolonged length of stay in the octogenarian population compared to the younger age group as observed by univariate analysis. 

 

91.19 Bending the Cost Curve for Colon Cancer Surgery: An Analysis of Nationwide Trends from 2002 to 2011

Posted on January 26, 2017

R. H. Hollis1, L. N. Wood1, M. S. Morris1, D. I. Chu1, J. S. Richman1, M. Kilgore3, M. T. Hawn2  1University Of Alabama at Birmingham,Surgery,Birmingham, Alabama, USA 2Stanford University,Surgery,Palo Alto, CA, USA 3University Of Alabama at Birmingham,Health Care Organization And Policy,Birmingham, Alabama, USA

Introduction:
Improvements in colon surgery, including the adoption of laparoscopy, are associated with reduced complications and length of stay.   Whether this has translated into reduced inpatient costs over time is unclear.

Methods:
We identified patients undergoing colectomy procedures with a diagnosis of colon cancer over years 2002-2011 in the Nationwide Inpatient Sample.  Inpatient costs in 2011 dollars were estimated using hospital charges and cost-to charge ratios.  Secondary outcomes included rates of laparoscopy, length of stay, and inpatient complications. A log-level model was used to evaluate the association between inpatient costs and year of surgery adjusting for patient, procedure, and hospital characteristics.  A separate model additionally controlling for laparoscopy, length of stay, and inpatient complications was used to evaluate the effect of these important cost mediators on temporal trends in costs.

Results:
Among 437,607 colectomies performed for cancer over one decade, the median cost for the inpatient hospitalization was $14,703 (IQR 10,779-$21,132).   From 2002 to 2011, laparoscopy use increased from 2.1% to 45.7%, and median length of stay decreased from 6 days to 5 days.   The odds of any inpatient complication in 2011 was significantly lower compared to 2002 (OR 0.79, 95%CI 0.70-0.89).  After controlling for patient, procedure, and hospital characteristics, the costs of surgery in 2011 were not significantly different from costs in 2002 (0.7 percentage point increase, 95%CI: -1.5-3.0) (figure).  When controlling for changes in laparoscopy, length of stay, and inpatient complication rates, inpatient costs were significantly higher in 2011 compared to 2002 (3.4 percentage point increase, 95%CI 1.3-5.6).

Conclusion:
Inpatient costs for patients undergoing colectomy did not significantly differ in 2011 compared to 2002.  Increased laparoscopy, decreased length of stay, and decreased complications were important mediators of costs savings, enabling stable costs over time.  These findings highlight the increased value of inpatient colectomy over time by virtue of stable costs and improved quality.
 

91.18 Patient Factors Predict Length of Stay and Readmission after Laparoscopic Fundoplication

Posted on January 26, 2017

Y. Vigneswaran1,2, K. Kuchta1, J. G. Linn1,2, S. P. Haggerty1,2, R. Joehl2, E. W. Denham1,2, M. B. Ujiki1,2  2NorthShore University HealthSystem,Surgery,Evanston, IL, USA 1University Of Chicago,Surgery,Chicago, IL, USA

Introduction:  Although a common low risk procedure, laparoscopic fundoplication for a small portion of patients can result in a complicated postoperative course.  Expected outcomes such as length of stay and unplanned readmissions have not been well studied for this procedure yet payers have decided on certain standard and expected outcomes. We hypothesize certain patient specific factors are associated with extended length of stay and unplanned 30 day readmission. The purpose of this study was to identify these risk factors from a national database and correlate them in a single institution experience.

Methods:  American College of Surgeons National Surgical Quality Improvement Program database from 2011 to 2013 was queried for elective laparoscopic fundoplication. Patient characteristics, comorbidities and operative details were used to determine predictors for complications, readmission and extended length of stay in a multivariate analysis.  These predictors were then retrospectively tested in our single institution cohort from 2009 to 2014.

Results: A total of 9,338 patients underwent laparoscopic fundoplication for reflux (41.0%) or paraesophageal hernia (59.0%).  Complications occurred in 319 (3.4%), 972 had length of stay greater than 3 days (10.4%) and 507 had unplanned 30-day readmission (5.4%).  Most common complications were urinary tract infection (27.6%) and pneumonia (26.3%).  Most common reasons for readmission were related to pulmonary (20.3%) or GI symptoms (14.4%).
From the multivariable analysis, patients at increased risk of complications were aged greater than 80 (OR 2.7, p<0.0001) and higher ASA class (OR 1.7, p <0.0001). Patients at increased risk of extended length of stay were aged greater than 80 (OR 3.9, p<0.0001), higher ASA class (OR 2.0, p <0.0001), black race (OR 1.9, p<0.0001), race listed other/unknown (OR 1.7, p <0.0001), history of pulmonary disease (OR 1.6, p=0.0001) and females (OR 1.3, p=0.004).  The odds of unplanned readmission was also significantly increased for age greater than 80 (OR 1.4, p=0.03), higher ASA class (OR 1.6, p<0.0001) and black race (OR 1.8, p<0.001).  When these predictors were used to create risk calculators and tested in our single institution cohort of 207 patients, extended length of stay had 72% sensitivity (CI: 66-78%), 45% specificity (CI 38-52%) and readmission had 71% sensitivity (95% CI: 65%-78%), 58% specificity (95% CI 51%-64%).

Conclusion: We have identified several patient dependent characteristics that are associated with increased risk of extended length of stay and unplanned 30-day readmission after laparoscopic fundoplication. We hope these results will allow for better patient counseling and patient selection by surgeons when proceeding with laparoscopic fundoplication. Additionally this data suggests outcomes of extended length of stay and 30-day readmission may not be good markers for the quality of surgical care with fundoplications, as currently used by payers.

 

91.17 Emergency Department Admission and Mortality for Inpatient Inguinal Hernia Repairs, 2009-2013

Posted on January 26, 2017

A. Mehta1, S. Hutfless2, A. B. Blair3, A. Dwarakanath4, H. T. Nguyen5  1Johns Hopkins University,School Of Medicine,Baltimore, MD, USA 2Johns Hopkins Hospital,Department Of Gastroenterology And Hepatology,Baltimore, MD, USA 3Johns Hopkins Hospital,Department Of Surgery,Baltimore, MD, USA 4Johns Hopkins Bayview Medical Center,Department Of Surgery,Baltimore, MD, USA 5Johns Hopkins Bayview Medical Center,Comprehensive Hernia Center,Baltimore, MD, USA

Introduction:  While inguinal hernias are common surgical diagnoses, minimally symptomatic patients are often not scheduled for repairs and are asked to seek medical attention if they develop symptoms. When this happens, patients commonly undergo a scheduled operation or go to an emergency department (ED) for expedited care. While emergent repairs of inguinal hernias are associated with higher mortality, little is known regarding how simply presenting through the ED impacts postoperative mortality and the patient characteristics associated with ED admission.

Methods:  We performed a retrospective analysis of the 2009–2013 Nationwide Inpatient Sample for unilateral inguinal hernia repairs. We examined inpatient care to understand the potential severity of outcomes for an otherwise elective condition. Multivariable logistic regressions adjusted for patient and hospital characteristics were used to determine how ED admission affected mortality and the predictors of ED admission. Patient and hospital characteristics included gender, race, age, payer status, comorbidities, obstruction, gangrene, recurrent hernia, hospital type, teaching institution, bed size, region, and discharge quarter.

Results: There were 116,357 inpatient hospitalizations; the majority (57%) resulted from ED admissions and 80% of ED-admitted patients had obstruction or gangrene. Overall mortality decreased from 2.03% in 2009 to 1.36% in 2013. Independent predictors of mortality included patient age (18-44: OR 0.04 [95%-CI 0.01-0.34]; 45-64: 0.27 [0.17-0.44]; ref: 65+), number of comorbidities (1: 2.79 [1.27-6.10]; 2-3: 3.94 [1.87-8.32]; 4+: 16.92 [8.16-35.12]; ref: 0) and admission through the ED (1.67 [1.21-2.29]), even after adjusting for obstruction and gangrene (Figure). Notable predictors of ED admission included black race (1.47 [1.29-1.69]), Hispanic ethnicity (1.35 [1.18-1.54]), self-pay (2.29 [1.97-2.66]) and Medicaid insurance (1.76 [1.50-2.06]), obstruction (9.77 [9.05-10.55]) and gangrene (18.24 [13.00-25.59]).

Conclusion: Inpatient inguinal hernia repairs resulting from ED admissions were predominately associated with complications necessitating urgent care and likely not from ED overutilization. However, we found that simply presenting through the ED was independently associated with a 67% higher postoperative mortality rate compared to that of a scheduled operation, even after adjusting for obstruction and gangrene. Black, Hispanic and self-pay patients were most likely to present through the ED. Our findings suggest a difference in ED utilization and in subsequent outcomes by patient race and insurance for this common surgical condition. Furthermore, additional consideration may be given for elective repairs in older patients with multiple comorbidities.

91.16 Influence of Sociodemographic Factors on Rate of Surgical Treatment in Patients with Graves’ Disease

Posted on January 26, 2017

G. A. Rubio1, T. M. Vaghaiwalla1, P. P. Parikh1, J. C. Farra1, A. R. Marcadis1, Z. F. Khan1, J. I. Lew1  1University Of Miami Leonard M. Miller School Of Medicine,DeWitt Daughtry Family Department Of Surgery,Miami, FL, USA

Introduction: Graves’ disease is the most common cause of hyperthyroidism in the United States. Management regimens include anti-thyroid medication, radioiodine ablation and thyroidectomy. Whereas patient preference and clinical features such as compressive symptoms, intolerance or intractability to medical treatment, and ophthalmopathy are known reasons for thyroidectomy, demographic and socioeconomic factors may also influence the decision for surgical treatment. This study examines the influence of these aforementioned factors in the use of thyroidectomy during hospitalizations for Graves’ disease.

Methods: A cross-sectional analysis was performed using the Nationwide Inpatient Sample (2006-2011) to identify hospitalizations for Graves’ disease. Patient demographic, socioeconomic, and clinical factors including thyroidectomy during hospitalization were assessed. Bivariate and logistic regression analyses were performed to identify characteristics independently predictive of undergoing thyroidectomy during hospitalization for Graves’ disease. Factors associated with non-elective hospitalizations were also evaluated.

Results: Of 33,279 patients admitted for Graves’ disease during the study period, 10,434 (31.4%) underwent total thyroidectomy. Majority of thyroidectomies (84.8%) were performed during elective admissions. Patients in the thyroidectomy group were younger than the non-surgical cohort (mean 40.1 vs. 42.5 years, respectively). This surgical group also had higher proportion of women (83.7% vs. 75.6%, p<0.01) and whites (59.1% vs 42.0, p<0.01) compared to the non-surgical group. Most thyroidectomy patients were covered by Medicare or private insurance (69.8% vs. 48.8%, p<0.01) with a preponderance of patients from the two highest income quartiles (50.3% vs. 38.4%, p<0.01) compared to non-surgical patients, respectively. On multivariate analysis, female sex (OR 1.52; 95% CI 1.37-1.69), white race (OR 1.27; 95% CI 1.17-1.39), Medicare/insured (OR 1.23; 95% CI 1.12-1.35), and highest income quartile (OR 1.28; 95% CI 1.14-1.45) were associated with increased odds of undergoing thyroidectomy during hospitalization for Graves’ disease. In contrast, male sex (OR 1.26; 95% CI 1.14-1.39), non-white race (OR 1.49; 95% CI 1.38-1.62), Medicaid/uninsured (OR 2.53, 95% CI 2.32-2.75), and lowest income quartile (OR 1.30; 95 CI 1.16-1.50) were associated with higher risk for emergency hospitalizations for Graves’ disease.

Conclusion: In the United States, demographic and socioeconomic characteristics may influence utilization of thyroidectomy for definitive treatment of hospitalized patients with Graves’ disease. Rate of emergent hospital admissions for Graves’ disease is also influenced by race, sex, income, and insurance status. Interventions to increase access for definitive care for Graves’ disease in these patients may lower rate of adverse outcomes and emergency healthcare utilization.

91.15 A Stitch in Time: Prevalence and Predictors of Opioid Receipt at Discharge after Traumatic Injury

Posted on January 26, 2017

M. A. Chaudhary1, A. J. Schoenfeld1, A. Ranjit1, R. Scully1, R. Chowdhury1, S. Nitzschke1,3, T. Koehlmoos2, A. H. Haider1  1Brigham And Women’s Hospital,Center For Surgery And Public Health,Boston, MA, USA 2Uniformed Services University Of The Health Sciences,Bethesda, MD, USA 3Brigham And Women’s Hospital,Trauma, Burn And Surgical Critical Care,Boston, MA, USA

Introduction: The prevalence of pain at hospital discharge among patients with traumatic injury is as high as 97%; 59% of whom experience moderate to severe pain. Inadequate pain control after trauma is associated with poor quality of life, delayed return to work and chronic pain syndrome. Opioids are considered the first line treatment for acute pain and treatment with opioids is associated with reduced likelihood of developing chronic pain. The Objectives of this study were to describe the prevalence of opioid prescription among trauma patients at hospital discharge and determine the predictors for the receipt of opioids for pain management.

Methods: The Military Health System Data Repository (MDR) was queried for TRICARE healthcare claims data from 2006 to 2014. Opioid-naive patients (18-64) admitted for traumatic injury were included in the study. Patients who died during index hospitalization or were transferred to another healthcare facility were excluded. The outcome variable was defined as at least one prescription of opioids at discharge. Logistic regression models, adjusted for patient  demographic and clinical characteristics, and environment of care were used to determine predictors of opioid prescription.

Results:Among the 27,114 patients included in the study, 14,017 (51.7%) received an opioid prescription at discharge. In risk-adjusted models, older adults (45-64y vs. 18-24y: OR= 1.38, 95% CI: 1.25-1.54), married patients (OR: 1.23, 95% CI: 1.16-1.30) and patients with higher Injury Severity Score (>9 vs. <9; OR: 1.24, 95% CI: 1.17-1.32) were associated with higher likelihood of opioid prescription. Males  (OR: 0.77, CI: 0.71-0.83), Asians vs.  non-Hispanic Whites (OR: 0.84, CI: 0.75-0.95), anxiety diagnosis (OR: 0.80, CI: 0.71-0.89) and traumatic brain injury (AIS head>3) (OR: 0.61, CI: 0.55-0.66) were associated with decreased likelihood of opioid prescription. Pre-existing comorbidities and presence of depression were not significant predictors of opioid prescription.

Conclusion:The rate of opioid prescription in trauma patients was 51.7%. Compared to the reported prevalence of pain among such patients at discharge, the rate of use of opioid for pain control at discharge seems low. Identifying factors associated with receipt of opioids at discharge, might help promote appropriate prescribing patterns among trauma patients thereby reducing incidence of chronic pain in this population.
 

« Previous Page
Next Page »