Vol. 219, No. 3S, September 2014 Surgical Forum Abstracts S101

CONCLUSIONS: Disparities exist in access to urgent surgicalcare according to social and economic factors. These delays

result in increased rates of postoperative complications, lengthof stay, cost, and in many cases mortality. Efforts must bemade to address these differences and provide equitable carefor all patients.

Outcomes and Discharge Destination after EmergentColectomy in Older AdultsZaid Abdelsattar, MD, Andrew A Gonzalez, MD, JD, MPH,Samantha Hendren, MD, MPH, FACS, Scott E Regenbogen, MD,FACS, Sandra L Wong, MD, MS, FACSUniversity of Michigan, Ann Arbor, MI

INTRODUCTION: Colon resections in elderly patients are oftendone in emergent settings. While known to be riskier, there are

limited data regarding postoperative disposition. We comparerisk-adjusted outcomes and discharge destinations between emer-gent and elective colectomy, and determine the hospital-level vari-ation in patient disposition.

METHODS: We evaluated Medicare beneficiaries who under-went elective and emergent colectomy between 2008 and2010. We estimated patient- and hospital-level risk-adjusted

rates of morbidity, 30-day mortality, 90-day mortality anddischarge-to-home (with/without home health services vs otherfacilities) using hierarchical logistic regression. Hospitals per-

forming >25 emergent colectomies in the study period weredivided into even quintiles based on their risk-adjusteddischarge-to-home rates.

RESULTS: Of 361,238 cases, 117,146 (32.4%) were performedon an emergent basis at 2,966 hospitals. Compared to electivecases, emergency colectomy was associated with markedly worse

morbidity rates of 40.3% vs 21.6%, 30-day mortality rates of11.8% vs 4.4% and 90-day mortality rates of 17.7% vs 7.2%.Only 43.5% of patients surviving an emergent resection were dis-

charged-to-home vs 76.3% after elective cases (all p<0.001).There was wide variation in risk-adjusted discharge-to-home ratesafter emergent colectomies across hospital quintiles (18%to 61%).

CONCLUSIONS: Over half of Medicare beneficiaries are dis-charged to non-home destinations after emergent colectomy.This rate varies widely between hospitals, which may suggest

room for improvement. In addition, this high-risk patient popula-tion is twice as likely to develop complications and is at nearly athree-fold increased risk for perioperative mortality compared to

elective colectomy.

Injury Prevention Programs against Distracted Driving:Are They Effective?Bardiya Zangbar, MD, Bellal Joseph, MD, FACS,Viraj Pandit, MD, Narong Kulvatunyou, MD, FACS,Sandeep S Bains, Andrew L Tang, MD, FACS,

Terence O’Keeffe, MBchB, FACS, Donald J Green, MD, FACS,Randall S Friese, MD, FACS, Peter M Rhee, MD, FACSUniversity of Arizona, Tucson, AZ

INTRODUCTION: Distracted driving (talking and/or texting) is agrowing public safety problem with increasing incidence amongadult drivers. The aim of this study was to identify the incidenceof distracted driving (DD) among health care providers and to

create awareness against DD.

METHODS: We performed a 4-phase prospective interventionalamong healthcare providers.Phase one: 1 week pre-intervention

observation,phase two:1 week intervention(email survey, pam-phlets and banners), phase three:1 week post-intervention obser-vation, and phase 4:1 week 6 month post-intervention

observation. Observations were performed outside employeeparking garage at time intervals:6.30-8.30am, 4.30-5.30pm, and6.30-7.30pm. Hospital employees were identified with:badges

and scrubs,employees exiting through employee gate, and parkingpass on the car. Outcome measure was incidence of DD pre, post,and 6 months post intervention.

RESULTS: 15,316 observations (Pre:6,639, Post:4,220, 6 monthsPost:4,557) and 520 survey responses were collected.35.5% respon-dents admitted to DD and 4.5% respondents were involved in anaccident due to DD.77% respondents felt more informed after the

survey and 91% respondents supported a state legislation againstDD.There was a significant reduction in DD in each time interval(p¼0.001) of observation between the pre and post intervention.On

sub-analysis, there was a significant decrease in talking (p¼0.001)and texting (p¼0.01) while driving post intervention as comparedto that of pre-intervention.The reduction in the incidence of DD

post intervention was sustained even on the follow up at 6 months.

CONCLUSIONS: There was 50% reduction in the incidence ofdistracted driving post-intervention which remained low even at

6 month follow up. Implementation of a naitonal injury preventioncampaign can be effective.

Outcomes Improvement is not Continuous Along theLearning Curve for PancreaticoduodenectomyTaylor M Coe, BS, Samuel E Wilson, MD, FACS,Mark A Talamini, MD, FACS, David C Chang, PhD, MPH, MBAUniversity of California, San Diego, San Diego, CA, Universityof California, Irvine, Irvine, CA, Stony Brook Medicine, Stony

Brook, NY

INTRODUCTION: Most studies on learning curves for complex

surgery have been based on single-institutional series or are inferredindirectly from volume-outcome relationships. It is unknownwhether those findings are generalizable. The goal of this study is

to describe mortality rates associated with cumulative surgical expe-rience among a large group of hospitals.

METHODS: Observational study of a statewide in-patient data-

base. Analysis included hospitals that began performing pancreati-coduodenectomy within the time window captured by the

Cumulativeexperience

Mortalityrate (%)

No. ofcases

No. ofhospitals

1st-10th 9.6 677 142

11th-20th 5.77 260 33

21st-30th 7.87 127 16

31st-40th 14.63 82 10

41st-50th 8.16 49 6

51st-60th 3.23 31 4

61st-70th 0 21 3

71st-80th 0 10 1

S102 Surgical Forum Abstracts J Am Coll Surg

California Office of Statewide Health Planning and Development

database (1996-2010). Cases across all years were numberedsequentially within each hospital. The same sequential series (e.g.first 10 cases, 11th through 20th cases) were identified and

aggregated across hospitals. The main outcome measure was in-hospital mortality.

RESULTS: A total of 1,264 cases in 142 hospitals were analyzed.

The average mortality rate for the first ten aggregated cases was9.6%. This mortality rate improved for subsequent cases, reaching5.8% for the 11th-20th cases. However, the mortality rates then

increased after the 20th case, reaching 14.6% by the 31st-40thcases before falling to 0% by the 61st-70th cases.

CONCLUSIONS: Initial improvement in surgical outcomes rela-tive to cumulative surgical experience is not sustained. Factorsother than surgical experience may affect outcomes, such as

expanded patient selection criteria or a false sense of confidence.Future studies should focus on identifying these confounders. Vig-ilance regarding outcomes should be maintained even after initial

outcomes improve.