- Email: [email protected]
Trauma patients who present in a delayed fashion: a unique and challenging population
Accepted Manuscript Trauma Patients Who Present in a Delayed Fashion: a Unique and Challenging Population Mary J. Kao, Hector Nunez, BS, Sean F. Monaghan, M.D., Daithi S. Heffernan, M.D., Charles A. Adams, Jr., M.D., Stephanie N. Lueckel, M.D., Andrew H. Stephen, M.D PII:
S0022-4804(16)30420-6
DOI:
10.1016/j.jss.2016.09.037
Reference:
YJSRE 14005
To appear in:
Journal of Surgical Research
Received Date: 8 April 2016 Revised Date:
13 September 2016
Accepted Date: 21 September 2016
Please cite this article as: Kao MJ, Nunez H, Monaghan SF, Heffernan DS, Adams Jr. CA, Lueckel SN, Stephen AH, Trauma Patients Who Present in a Delayed Fashion: a Unique and Challenging Population, Journal of Surgical Research (2016), doi: 10.1016/j.jss.2016.09.037. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
REVISED: 9/13/16 Title Trauma Patients Who Present in a Delayed Fashion: a Unique and Challenging Population
RI PT
Short Title Delayed Presentation After Injury FULL NAME:
Mary J Kao, Alpert Medical School of Brown University, [email protected] (wrote initial draft of manuscript)
SC
Hector Nunez BS, Alpert Medical School at Brown University, [email protected] (edited manuscript and performed data collection, statistical analysis)
(edited manuscript)
M AN U
Sean F Monaghan, Alpert Medical School at Brown University, [email protected]
Daithi S. Heffernan MD, Alpert Medical School at Brown University, [email protected] (edited manuscript, statistical analysis)
Charles A Adams MD, Alpert Medical School at Brown University, [email protected] (edited manuscript)
(edited manuscript)
TE D
Stephanie N Lueckel MD, Alpert Medical School at Brown University, [email protected]
Andrew H Stephen MD, Alpert Medical School at Brown University, [email protected]
EP
(designed the project, research question, and hypothesis, edited manuscript)
AC C
Address for Correspondence: Andrew H Stephen MD, Division of Trauma and Surgical Critical Care, Department of Surgery Alpert Medical School of Brown University 435 APC Building, 4th Floor 593 Eddy Street, Rhode Island Hospital Providence, Rhode Island 02903 Phone: 401-444-0369 Fax: 401-444-6681 Email: [email protected] No conflicts of interests declared. This research was presented at the Academic Surgical Congress annual conference, February 2016 Authors Contributions:
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
AHS designed the project, research question, and hypothesis, MJK wrote the initial manuscript draft, HN and DSH performed the statistical analysis and compiled the tables, SM, CA, and SL contributed to the editing of the manuscript.
ACCEPTED MANUSCRIPT
Abstract: Trauma Patients Who Present in a Delayed Fashion: A Unique and Challenging Population
RI PT
Mary J. Kao, Hector Nunez BS, Sean F. Monaghan M.D., Daithi S. Heffernan M.D., Charles A. Adams Jr. M.D., Stephanie N. Lueckel M.D., Andrew H. Stephen M.D.
Alpert Medical School Of Brown University, Rhode Island Hospital, Division Of Trauma And Surgical Critical Care, Department Of Surgery, Providence, RI, USA
M AN U
SC
Introduction: A proportion of trauma patients present for evaluation in a delayed fashion after injury, likely due to a variety of medical and nonmedical reasons. There has been little investigation into the characteristics and outcomes of trauma patients who present delayed. We hypothesize that trauma patients who present in a delayed fashion are a unique population at risk of increased trauma related complications. Materials and Methods: This was a retrospective review from 2010-2015 at a Level I trauma center. Patients were termed delayed if they presented greater than 24 hours after injury. Patients admitted within 24 hours of their injury were the comparison group. Charts were reviewed for demographics, mechanism, comorbidities, complications and outcomes. A subgroup analysis was done on patients who suffered falls.
EP
TE D
Results: During the five year period, 11,705 patients were admitted. 588 patients (5%) presented greater than 24 hours after their injury. Patients in the delayed group were older (65 vs. 55 years, p<0.001) and more likely to have psychiatric comorbidities (33% vs. 24%, p = 0.0001) than the control group. They were also more likely to suffer substance withdrawal (8.9% vs. 4.1%, p<0.001) but had toxicology testing for drugs and alcohol done at significantly lower rates. Patients that presented delayed after falls were similar in age and injury severity score (ISS) but more likely to suffer substance withdrawal when compared to those with falls that presented within 24 hours. Patients with falls that presented delayed had toxicology testing at significantly lower rates than the comparison group.
AC C
Conclusion: Trauma patients that present to the hospital in a delayed fashion have unique characteristics and are more likely to suffer negative outcomes including substance withdrawal. Future goals will include exploring strategies for early intervention, such as automatic withdrawal monitoring and social work referral for all patients who present in a delayed fashion.
Keywords: delayed presentation, elderly, trauma, injury, withdrawal
ACCEPTED MANUSCRIPT
Introduction Injuries often occur in an acute fashion demanding urgent attention. Advances in trauma systems have largely focused on delivery of expedient care to minimize
RI PT
physiologic and anatomic derangements. Despite ongoing improvements in these systems, not all trauma patients present immediately following injury mechanism. A proportion of trauma patients present to the hospital hours or days later, likely due to a variety of
SC
medical and nonmedical reasons.1,2,3 Thierry et al demonstrated that 42% of farmworkers that suffered injuries either delayed seeking medical treatment or did not seek treatment
M AN U
at all.2 The authors found that the inability to work and fear of missing payments were a larger factor than pain in making decisions about seeking treatment.
Otherwise there has been little investigation regarding patients that present in a delayed fashion after injury. There is a scarcity of data about the characteristics of
TE D
patients that present delayed, their injury mechanisms, why they present delayed, and whether delayed presentation affects outcomes in injured patients. Why patients present
EP
in a delayed fashion is not well understood but some insight has emerged from limited literature. These series have looked at patients with medical illnesses and injuries and
AC C
have cited reasons including lack of awareness of the consequences of the injury, limited access to healthcare,4,5 remote location,6 social isolation,7 or unusual signs and symptoms of disease, including atypical presentations in elderly trauma victims.8 Patients with diabetes who live in rural settings or are of lower socioeconomic and insurance status have been shown to delay seeking care resulting in increased costs and complications.15,16
ACCEPTED MANUSCRIPT
Despite advances in systems to improve timeliness of trauma care a significant number of patients still present for evaluation and treatment many hours or days after their injury. It remains unclear why such delays in seeking care exist. We hypothesize
RI PT
that trauma patients who present in a delayed fashion are a unique population that have
distinct patient characteristics, injury patterns, initial evaluations and outcomes. We also seek to explain why these patients present delayed after injury and to affirm that the
SC
initial workup of these patients should follow standard trauma protocols in regards to
M AN U
initial alcohol and substance testing.
Methods
This is a retrospective review of all traumatically injured patients requiring admission to a Level I trauma center between 2010 and 2015. Institutional review board
TE D
approval and waiver of consent was obtained prior to the study. Charts were reviewed for the recorded time of injury as well as the time of presentation. Patients were categorized as "Delayed" if they presented to our hospital greater than 24 hours after their initial
EP
injury. All patients that presented to our hospital within 24 hours of their injury were categorized as "Non-Delayed". Recording time from injury to presentation at our hospital
AC C
included any potential time spent at a referral hospital.
Charts were reviewed for patient characteristics including age, gender, injury
severity score (ISS), time to presentation in hours from the injury, time spent in hours at the referral hospital, whether a blood alcohol level (BAL) was checked in the emergency room, percentage of patients that had a positive BAL of those that were drawn, mean BAL of the positive BAL tests, whether testing for illicit drugs was done, and the
ACCEPTED MANUSCRIPT
percentage of patients with the presence of illicit drugs on presentation. Outcomes studied were in hospital mortality and occurrence of alcohol/drug withdrawal.
RI PT
A more in-depth sub-group analysis was done on patients who suffered falls. In addition to the patient characteristics reviewed in the overall group we chose to look at the comorbidities of patients that suffered falls and grouped these into psychiatric,
SC
cardiovascular, pulmonary, and endocrine categories. The comorbidities that fell within these categories were as defined by our trauma registry. We also looked at injuries in the
M AN U
patients that suffered falls and grouped them by anatomic regions: traumatic brain injury(TBI), thoracic, spinal, pelvic, extremities, and intraabdominal. Complications occurring during the hospital stay were recorded including all infections and alcohol/drug withdrawal, termed "Alcohol/Drug WD". Alcohol and drug withdrawal was defined in our registry per the National Trauma Database definition: "physical symptoms noted after
TE D
suddenly stopping consumption". Additionally patients could be charted in our registry as having withdrawal if they scored higher than 9 points on Clinical Institute Withdrawal
EP
Assessment (CIWA) scales or if there was any physician documentation stating withdrawal had occurred. Infections were either clinically evident (abscess, cellulitis,
AC C
pneumonia) or culture based (UTI, catheter, pneumonia via bronchoalveolar lavage). Pneumonia was diagnosed either by clinical judgment or by bronchoalveolar lavage in intubated patients with >10,000 colony forming units (CFU)/mL. Urinalysis (UA) was sent if urinary tract infection (UTI) was suspected based on clinical judgment. If the UA was positive, culture was sent and was considered positive if there were >100,000 CFU/mL. All other infections including Surgical Site Infections (SSI) were diagnosed by CDC criteria and cultures.
ACCEPTED MANUSCRIPT
We operationally defined the term “Effect of Delay” to encompass the total time patients have suffered from their injury at home prior to receiving care and in the hospital. To calculate “Effect of delay” we calculated the total time from the occurrence of the
RI PT
injury to final hospital discharge. This was calculated as follows: “Effect of Delay” = “Time (in hours) from Injury Occurrence to Presentation to Trauma Center” plus
“Hospital Length of Stay”. For patients who were transferred from an outside facility we
SC
calculated “Time (in hours) from Injury Occurrence to Presentation to Outside hospital” plus “Time (in hours) spent at Outside Hospital” plus “Trauma Center hospital Length of
M AN U
Stay”
Patient disposition on discharge was also reviewed. The number of patients who were discharged to their “prior site” of residence or were upgraded in terms of going to a “higher care facility” were recorded and compared in each group. For example, a patient
TE D
who was in an assisted living facility at the time of injury and then was discharged back to that facility at the same level of care returned to their “prior site” or equivalent to
EP
discharge to home. If a patient was living in any facility and was discharged to their own facility requiring a higher level of care or a different facility requiring a higher level of
AC C
care that was deemed to be a “higher care facility”
We also looked at the location of injury of patients that suffered falls to assess
whether there were differences between the delayed and non-delayed groups in terms of distance to the trauma center. This was done by querying the trauma registry to determine the zip code of the injury and then calculating the distance to the zip code of the trauma center. We grouped patients’ falls into four categories: whether they occurred
ACCEPTED MANUSCRIPT
less than 10 miles, 10 to 20 miles, 20 to 30, or greater than 30 miles from the trauma center.
RI PT
Data is expressed as either mean values with standard error (SEM) or for nonparametric data expressed as medians with interquartile ranges (IQR). Data was analyzed with Chi-squared, Student’s t test or Mann-Whitney-U. Significance was set at p<0.05.
SC
SPSS Version 21 was used. Kaplan-Meier analysis was done to depict “Effect of Delay”
M AN U
Results
Over the five-year period, 11,705 patients were admitted of whom 693 patients (5.9 %) presented greater than 24 hours after their initial injury. In the delayed group, the median time to presentation was 72 hours, (IQR=48-128 hours). Patients in the delayed group, when compared to those in the non-delayed group were older (64.7 versus 55.3
TE D
years; p<0.001) more often women (46.6% versus 41.5%; p=0.04), and had lower ISS (8.7+/0.3 versus 9.5+/-0.1; p=0.04). With respect to medical comorbidities, patients in the delayed group were more likely to have underlying psychiatric diagnoses (33% vs. 24%,
EP
p = 0.0001) and endocrine comorbidities (27% vs. 23%, p = 0.04) than the non-delayed
AC C
group. Patients in the delayed group were less likely to have a BAL level checked (26.5% versus 33%; p=0.01) and were less likely to test positive for alcohol. However, comparing BAL levels among those who tested positive for alcohol, patients in the delayed group has a significantly higher BAL (226 mg/dL +/- 26 versus 179 mg/dL +/-2; p<0.01). Patients in the delayed group were less likely to be tested for drugs of abuse on presentation (19% versus 40%; p<0.01). However there was no difference in rates of testing positive among those who were tested (Table 1).
ACCEPTED MANUSCRIPT
When comparing outcomes of the delayed and non-delayed groups, there was no difference in the mortality rate between the two groups. The delayed group was more
RI PT
likely to suffer from alcohol withdrawal (8.80% versus 4%, p < 0.001). There were significant differences in the most common injury mechanisms
between the two groups. Patients in the delayed group compared to the non-delayed
SC
group demonstrated significantly higher rates of burns (7.4% versus 2.8%, p < 0.001) and falls (77% vs. 50.2%, p < 0.001), and lower rates of motorized vehicle collisions (MVC
M AN U
or MCC) (4.3% versus 14.8%; p<0.01) and assaults (2.7% versus 10.1%; p<0.01) (Table 2).
Overall, 6,114 patients (51.8% of the entire admitted trauma population over the study period) sustained a fall as the mechanism of injury. Among this fall group, 533
TE D
(7.4%) of the patients presented greater than 24 hours following the fall and constitute the Fall Delayed group (FD). In the FD group, the average time from fall to presentation was 72 hours (IQR 48-120). The remaining 5,581 patients who fell presented within 24 hours
EP
of the trauma and represent the Fall Non-Delayed (FND) group. Comparing FD to FND patients, there was no difference in respect to age or ISS.
AC C
The FD group were less likely to undergo either alcohol testing (26.6% versus 42%; p<0.01) or drug use testing (19% versus 28%; p<0.01) but were more likely to have a positive drug test if tested (68.2% versus 58%, p<0.04) (Table 3). Patients in the FD group compared to the FND group had higher rates of psychiatric (33% versus 27%; p<0.01) and pulmonary (35% versus 25%; p<0.01) comorbidities. There was no difference in rates of TBI or pelvic fractures between the FD
ACCEPTED MANUSCRIPT
and FND patients. However, FD patients had higher rates of thoracic (33% versus 16%; p<0.01), spinal (19% versus 13%; p<0.01), and intraabdominal (6% versus 2%; p<0.01) injuries, but were less likely to have sustained extremity trauma (33% versus 58%;
RI PT
p<0.01) (Table 4). With respect to fall patients who sustained thoracic trauma, 99% of
FD patients sustained rib fractures compared with only 86% in the FND group (p<0.01). Hemothorax or pneumothorax occurred in 33% of FD patients compared with 21% of
SC
FND patients (p < 0.01).
M AN U
Outcomes including mortality, hospital length of stay, ICU length of stay, rate of occurrence of an infection, and discharge location were similar between FD and FND patients. However, alcohol withdrawal was significantly more likely to occur in the FD group (9.2% vs. 4%, p<0.01). There was no significant difference in LOS in patients who developed substance withdrawal comparing FD and FND patients. FD patients were
(Table 5).
TE D
noted to have significantly greater effect of delay (11 +/-0.3 versus 7 +/-0.2 days; p<0.01)
EP
We did not find differences in the delayed and non-delayed groups in regards to location of patient falls and distance to the trauma center. This was the case across the
AC C
four distances reviewed: less than 10 miles, 10 to 20 miles, 20 to 30 miles, and greater than 30 miles from the trauma center. (Table 6) Discussion
In this review of a large trauma registry we have shown that patients who present delayed after injury had several unique patient characteristics, mechanisms and outcomes. They were older, suffered from more preinjury psychiatric and endocrine comorbidities,
ACCEPTED MANUSCRIPT
and had different trauma mechanisms compared to patients who presented more immediately after injury. Analysis of patients that suffered falls also was revealing. Injury patterns were different with twice as many thoracic injuries but far less extremity injuries
RI PT
in the delayed group. Delayed fall patients suffered alcohol and drug withdrawal at more than twice the rate of the nondelayed fall group. To our knowledge there has been
limited investigation into patients that present to the hospital many hours or days after
SC
injury.
M AN U
Patients in our delayed group presented a median of 3 days after their injuries in both the overall and fall groups. Previous studies have attempted to explain why there are delays in seeking care after injury. Thierry et al recently reported a significant number of victims of farm work injuries either underestimate the degree of injury or believe that they can self treat their injuries.2 The farm workers were also more likely to present for
TE D
evaluation if their ability to work was compromised rather than from pain citing financial concerns as the reason for this. Along these lines our patients who fell and presented
EP
delayed suffered more thoracic but less extremity injuries than the nondelayed group. While thoracic injuries can result in significant discomfort they do not lead to an
AC C
immediate sense of disability or loss of function like an extremity injury. There has been far more investigation into reasons for delayed presentations in patients with medical or psychiatric illnesses. Yousaf noted in a systematic review that fear of expressing emotions or yielding emotional and psychological control was associated with delays in seeking medical care.9 Other disease processes such as community acquired pneumonia (CAP) and acute myocardial infarction often present with atypical or more subtle findings in elderly patients leading to delays in presentation and diagnosis.10 Faverio's 2014
ACCEPTED MANUSCRIPT
review of the management of CAP in the elderly described some of these atypical signs including confusion, decreased appetite, unsteadiness, and even falls.11 We speculate that similar atypical signs and symptoms such as blunted pain responses, decreased activity,
RI PT
and decline in appetite may have occurred after injury in our patients, especially in the elderly.
SC
A number of investigators have shown reasons for delays in care including social isolation, inadequate access to care, and concerns for costs of care. Elderly patients often
M AN U
live alone and present to the hospital at the urging of a family member. They also may not have the means to physically get to the hospital as many no longer drive. In patients over 65 years old that have suffered acute myocardial infarction delays greater than 6 hours in seeking care have been ascribed to the following: lower levels of education, rural location of the patient, and assumption that symptoms were of noncardiac etiology.12
TE D
Common themes that have been described by these patients and have been shown in other investigations include living alone and a desire to not bother others or burden children. Patients with acute decompensated heart failure who present delayed have been shown to
EP
be of older age and more often male.13 We chose to look at patients that fell in more
AC C
detail as this is the most common mechanism of injury at our institution. It is also the mechanism most often associated with mortality and morbidity in the elderly. Barriers to care for those with diabetes and psychiatric comorbidities are known
to exist and may relate to the increase in time to presentation after injury in these patients in our study. Numerous patient, provider, and systems based issues have been cited as obstacles for diabetes patients. These obstacles can lead to suboptimal management and development of diabetes related complications such as infected wounds, retinopathy,
ACCEPTED MANUSCRIPT
cardiovascular disease, and diabetic ketoacidosis. Lack of health insurance has been shown to increase the risk of diabetes complications and costs.14 Also, patients who live in impoverished areas and are of lower education are seen less frequently by diabetes
RI PT
specialists and use less preventive services. These patients were more likely to report
difficulties accessing medical care in general and to have not been seen by any physician in the last year and were less likely to have health insurance. Similar access issues have
SC
been reported in diabetics that live in rural areas.15,16 Though we found a higher rate of
endocrine comorbidities which included diabetes and obesity in our delayed presentation
M AN U
patients we did not review their geographic locations or socioeconomics. Doing so may have affirmed the connection between their delay in presentation after injury and their issues with access to care for diabetes. Limitations in access have also been noted in patients with psychiatric disease. Patients with chronic mental illness often have several
TE D
associated medical comorbidities such as hypertension, diabetes, and alcohol abuse. However, they often do not receive needed procedures, medications or general care due to real or patient perceived obstacles; these include financial constraints, poor compliance
EP
with care plans, lack of awareness of available resources, and even convenience issues.17,18 Patients with psychiatric comorbidities also may face stigma and
AC C
embarrassment and may have had previous healthcare experiences that leave them hesitant to seek further care. Patients in our delayed presentation groups, overall and falls, had significantly
lower rates of alcohol and drug toxicology screening on presentation than the nondelayed groups. This may be due to the fact that these delayed injuries seem less emergent, thus leading the provider to overlook the fact that alcohol or drugs contributed to the injury.
ACCEPTED MANUSCRIPT
However, consensus recommendations that trauma centers screen for alcohol and drugs after injury have existed for decades. These recommendations are based on large amounts of data showing a strong relationship between alcohol, drug use and injury; 40 percent of
RI PT
motor vehicle and pedestrian fatalities involve alcohol use and similar findings have been shown for drug use.19 Screening can provide patients with social work consultation, brief interventions, and opportunities for outpatient programs after discharge. With adherence
SC
to these basic measures that are required for trauma level one center verification trauma
screening need to be improved.
M AN U
recidivism can be reduced. Clearly based on our study's data, rates of drug and alcohol
Delayed trauma patients in our study were more likely to suffer from alcohol/drug withdrawal during their hospital stay further supporting the need for screening on presentation to be thorough. The source of this difference in incidence of
TE D
withdrawal is unclear but may be related to the higher rates of psychiatric illness in the delayed groups noted above. It also suggests that the delayed population may need more rigorous monitoring for alcohol withdrawal symptoms to prevent further complications
EP
during their hospital stay. Alcohol withdrawal has been associated with increased
AC C
mortality in inpatients.20,21 Early and careful dosing of benzodiazepines, triage to the ICU, and surgical house staff vigilance could improve prevention or outcomes of alcohol withdrawal if screening occurred with increased frequency in trauma patients whether they present delayed or not. Limitations of our study include possible inaccurate reporting of time of injury by patients. This makes the duration of delay likely uncertain in a number of the patients. Also, by limiting our more in depth analyses to patients with falls we may be missing
ACCEPTED MANUSCRIPT
important findings in the other mechanism groups. Our psychiatric comorbidity category included dementia in addition to other forms of mental illness and we did not separate out these entities to assess their association with outcomes such as alcohol withdrawal.
RI PT
Lastly, this study is unable to account for the people who suffer injury but never report to the hospital.
SC
Conclusion
Trauma patients who presented to the hospital in a delayed fashion after their
M AN U
initial injury were older, had more frequent psychiatric and endocrine comorbidities, and more often suffered negative outcomes such as alcohol withdrawal. Yet this vulnerable group less frequently had alcohol and drug toxicology screening on presentation. It is important to assess these patients with the same standards as those who present early after
TE D
injury. This may allow them to receive more timely and focused care as it pertains to prevention and management of alcohol withdrawal. Future goals will include monitoring rates of drug and alcohol screening and exploring strategies for early intervention such as
EP
automatic alcohol withdrawal monitoring, arranging for better management of
AC C
comorbidities, and social work referral for patients who present in a delayed fashion.
ACCEPTED MANUSCRIPT
Disclosures: The authors report no proprietary or commercial interest in any product
AC C
EP
TE D
M AN U
SC
RI PT
mentioned or concept discussed in this article.
ACCEPTED MANUSCRIPT
Table 1. Demographics and Mechanisms of injury of all patients Nondelayed
N = 693
N = 11117
64.7 ±0.76
55.3 ± 0.22
<0.01
Female Gender (%)
46.6%
41.5%
<0.01
ISS* (0-75) Time to Presentation** (hrs) Time at referral hospital #(hrs) BAL age*
8.7 ±0.25
9.5 ±0.1
72 (48-128)
0.9 (0.6-2)
3.8 ±0.15
2.8 ±0.05
60.2 ±1.5
46.46 ±0.3
26.5%
33%
34%
65%
226.6 ±24.5
179 ±2
% of +BAL BAL of + tests* % Drug tested
19%
% +Drug test
70.2%
<0.01
<0.01
<0.01
SC
% of BAL drawn
<0.01
M AN U
Age* (years)
p
RI PT
Delayed
<0.01
<0.01
<0.01
40%
<0.01
65.4%
N/S
AC C
EP
TE D
*Values are Mean (SEM) **Values are Median (IQR) #Time spent at referral hospital of patients that were transferred to our hospital
ACCEPTED MANUSCRIPT
Table 2. Mechanism of injury of all patients p
Burn
7.4%
2.8%
<0.01
MVC
4.3%
14.8%
<0.01
MCC
1.0%
5.5%
<0.01
Falls
77%
50.2%
<0.01
Assault
2.7%
10.1%
<0.01
Self-Inflicted
0.1%
2.6%
<0.01
Work Related
2.3%
5.2%
<0.01
RI PT
N = 11117
SC
Nondelayed
N = 693
AC C
EP
TE D
M AN U
Delayed
ACCEPTED MANUSCRIPT
Table 3. Demographics of patients following fall Delayed
Nondelayed
p
N = 5581
68.7 ±0.26
N/S
49.5%
54.9%
0.02
9.45 ±0.28
8.97 ±0.09
72 (48-120)
0.97 (0.65-3.25)
64.4 ±1.6
60.8 ±0.55
% of BAL drawn
26.6%
42%
% of +BAL
15.5%
27.8%
% Drug tested
220.86 ±26.3 19%
194.4 ±4.2 28%
<0.01
% +Drug tests
68.2%
58%
0.03
BAL of + tests*
AC C
EP
TE D
*Values are Mean (SEM) **Values are Median (IQR)
N/S
<0.01 N/S
<0.01
<0.01
SC
ISS* Time to presentation** (hrs) BAL age*
M AN U
Female gender %
RI PT
N = 533
69.3 ±0.8
Age*
N/S
ACCEPTED MANUSCRIPT
Table 4. Comorbidities and Injuries of patients following fall Nondelayed
p
N = 533
N = 5581
33% 61% 35% 27%
27% 58% 25% 26%
<0.01 N/S <0.01 N/S
30% 33% 19% 7% 33% 6%
27% 16% 13% 6% 58% 2%
N/S <0.01 <0.01 N/S <0.01 <0.01
AC C
EP
M AN U
TE D
TBI Thoracic Spinal Pelvic Extremities Intraabdominal
SC
Injury
RI PT
Comorbidity Psychiatric Illness Cardiovascular Pulmonary Endocrine
Delayed
ACCEPTED MANUSCRIPT
Table 5. Outcomes of patients following fall
N = 533
N = 5581
5% 6.56 ±0.3 4.8 ±0.4 9%
4.7% 7.0 ±0.2 5.6 ±0.2 8.4%
N/S N/S N/S N/S
9.2%
4%
<0.01
9 ±1.7
12.04 ±1.9
N/S
11.01 ±0.3
7.04 ±0.2
<0.01
45% 48%
48% 45%
N/S N/S
RI PT
p
SC
Nondelayed
M AN U
Outcome Mortality LOS* ICU LOS* Infection rate Alcohol/Drug withdrawal LOS of pts with alcohol withdrawal Effect of Delay† Disposition Higher Care Facility Prior Site
Delayed
AC C
EP
TE D
*Values are Mean (SEM) †Effect of delay = Average of (Time from fall to presentation + Hospital LOS)
ACCEPTED MANUSCRIPT
Table 6. Distance from injury location to trauma center Nondelayed falls
N=430
N=5263
69% 20.5% 7.4% 3%
66.5% 22.9% 7.7% 3%
p N/S N/S N/S N/S
RI PT
Delayed falls
AC C
EP
TE D
M AN U
SC
Distance to trauma center (miles) ≤10 10-20 20-30 30>
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1. Kaplan-Meier analysis of patients that suffered falls. Outcome-discharge from the hospital.
ACCEPTED MANUSCRIPT
References 1. Vadivelu S, Esernio-Jenssen D, Rekate H, Naravan R, Mittler M, Schneider S. Delay in arrival to care in prepetrator identified nonaccidental head trauma:
RI PT
observations and outcomes. World Neurosurgery. 2015;84(5):1340-6.
2. Thierry A, Snipes S. Why do farmworkers delay treatment after debilitating
SC
injuries? Thematic analysis explains if, when and why farmworkers were treated for injuries. American Journal of Industrial Medicine. 2015;58(2):178-92.
M AN U
3. Wood J, Rubin D, Nance M, Christian C. Distinguishing inflicted versus accidental abdominal injuries in young children. Journal of Trauma. 2005;59(5):1203-8.
4. Marshall I, Wang Y, Crichton S, McKevitt C, Rudd A, Wolfe C. The effects of
TE D
socioeconomic status on stroke risk and outcomes. Lancet Neurology. 2015;14(12):1206
EP
5. Duron V, Monaghan S, Connolly M, Gregg S, Stephen A, Adams C, Cioffi WG, Heffernan DS. Undiagnosed medical comorbidities in the uninsured: a significant
AC C
predictor of mortality following trauma. Journal of Trauma and Acute Care Surgery. 2012;73(5):1093-8.
6. Veitch P. Anticipated response to three common injuries by rural and remote area residents. Social science and Medicine. 1995;41(5):739-45.
ACCEPTED MANUSCRIPT
7. Bowden M, Grant S, Vogel B, Prasad J. The elderly, disabled and handicapped adult burned through abuse and neglect. Burns, including thermal injury.
RI PT
1988;14(6):447-50. 8. Heffernan D, Thakkar R, Monaghan S, Ravindran R, Adams C, Kozloff M, Gregg SC, Connolly MD, Machan JT, Cioffi WG. Normal presenting vital signs are
SC
unreliable in geriatric blunt trauma victims. Journal of Trauma. 2010;69(4):81320.
M AN U
9. Yousaf O, Grunfeld E, Hunter M. A systematic review of the factors associated with delays in medical and psychological help-seeking among men. Health Psychology Review. 2015;9(2):264-76.
10. Claridge J, Crabtree T, Pelletier S, Butler K, Sawyer R, Young J. Persistent occult
TE D
hypoperfusion is associated with a significant increase in infection rate and mortality in major trauma patients. Journal of Trauma. 2000;48(1):8-14
EP
11. Faverio P, Aliberti S, Bellelli G, Suigo G, Lonni S, Pesci A, Restrepo MI. The management of community acquired pnuemonia in the elderly. European Journal
AC C
of Internal Medicine. 2014;25(4):312-9. 12. Hwang SY, Jeong MH. Cognitive factors that influence delayed decision to seek treatment among older patients with acute myocardial infarction in Korea. Eur J Cardiovasc Nurs. 2012 Jun;11(2):154-9.
ACCEPTED MANUSCRIPT
13. Darling C, Saczynski JS, McManus DD, Lessard D, Spencer FA, Goldberg RJ. Delayed hospital presentation in acute decompensated heart failure: clinical and
RI PT
patient reported factors. Heart Lung. 2013 Jul-Aug;42(4):281-6. 14. Brown DS, McBride TD. Impact of the Affordable Care Act on access to care for US adults with diabetes, 2011-2012. Prev Chronic Dis. 2015 May 7;12:E64.
SC
15. Zgibor JC, Songer, TJ. External Barriers to Diabetes Care: Addressing Personal
M AN U
and Health Systems Issues. Diabetes Spectrum. Jan 2001 vol. 14 no. 1 23-28 16. Zgibor JC, Songer TJ, Kelsey SF, Weissfeld J, Drash AL, Becker D, Orchard TJ. The association of diabetes specialist care with health care practices and glycemic control in patients with type 1 diabetes: a cross-sectional analysis from the Pittsburgh epidemiology of diabetes complications study. Diabetes Care. 2000
TE D
Apr;23(4):472-6.
17. Kisely S, Smith M, Lawrence D, Cox M, Campbell LA, Maaten S. Inequitable
EP
access for mentally ill patients to some medically necessary procedures. CMAJ 2007;176:779–784
AC C
18. McIntyre RS, Soczynska JK, Beyer JL, Woldeyohannes HO, Law CW, Miranda A, Konarski JZ, Kennedy SH. Medical comorbidity in bipolar disorder: reprioritizing unmet needs. Curr Opin Psychiatry 2007;20:406–416
19. FACS: Subcommittee on Injury Prevention and Control. Alcohol and Injury. https://www.facs.org/~/media/files/quality%20programs/trauma/alcoholinjury.ash x. Acessed February 1 2016.
ACCEPTED MANUSCRIPT
20. Tavel ME, Davidson W, Batterton TD. A critical analysis of mortality associated with delirium tremens. Review of 39 fatalities in a 9-year period. Am J Med Sci.
RI PT
1961 Jul;242:18-29 21. Yost DA. Alcohol withdrawal syndrome. Am Fam Physician. 1996
AC C
EP
TE D
M AN U
SC
Aug;54(2):657-64, 669.
S0022-4804(16)30420-6
DOI:
10.1016/j.jss.2016.09.037
Reference:
YJSRE 14005
To appear in:
Journal of Surgical Research
Received Date: 8 April 2016 Revised Date:
13 September 2016
Accepted Date: 21 September 2016
Please cite this article as: Kao MJ, Nunez H, Monaghan SF, Heffernan DS, Adams Jr. CA, Lueckel SN, Stephen AH, Trauma Patients Who Present in a Delayed Fashion: a Unique and Challenging Population, Journal of Surgical Research (2016), doi: 10.1016/j.jss.2016.09.037. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
REVISED: 9/13/16 Title Trauma Patients Who Present in a Delayed Fashion: a Unique and Challenging Population
RI PT
Short Title Delayed Presentation After Injury FULL NAME:
Mary J Kao, Alpert Medical School of Brown University, [email protected] (wrote initial draft of manuscript)
SC
Hector Nunez BS, Alpert Medical School at Brown University, [email protected] (edited manuscript and performed data collection, statistical analysis)
(edited manuscript)
M AN U
Sean F Monaghan, Alpert Medical School at Brown University, [email protected]
Daithi S. Heffernan MD, Alpert Medical School at Brown University, [email protected] (edited manuscript, statistical analysis)
Charles A Adams MD, Alpert Medical School at Brown University, [email protected] (edited manuscript)
(edited manuscript)
TE D
Stephanie N Lueckel MD, Alpert Medical School at Brown University, [email protected]
Andrew H Stephen MD, Alpert Medical School at Brown University, [email protected]
EP
(designed the project, research question, and hypothesis, edited manuscript)
AC C
Address for Correspondence: Andrew H Stephen MD, Division of Trauma and Surgical Critical Care, Department of Surgery Alpert Medical School of Brown University 435 APC Building, 4th Floor 593 Eddy Street, Rhode Island Hospital Providence, Rhode Island 02903 Phone: 401-444-0369 Fax: 401-444-6681 Email: [email protected] No conflicts of interests declared. This research was presented at the Academic Surgical Congress annual conference, February 2016 Authors Contributions:
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
AHS designed the project, research question, and hypothesis, MJK wrote the initial manuscript draft, HN and DSH performed the statistical analysis and compiled the tables, SM, CA, and SL contributed to the editing of the manuscript.
ACCEPTED MANUSCRIPT
Abstract: Trauma Patients Who Present in a Delayed Fashion: A Unique and Challenging Population
RI PT
Mary J. Kao, Hector Nunez BS, Sean F. Monaghan M.D., Daithi S. Heffernan M.D., Charles A. Adams Jr. M.D., Stephanie N. Lueckel M.D., Andrew H. Stephen M.D.
Alpert Medical School Of Brown University, Rhode Island Hospital, Division Of Trauma And Surgical Critical Care, Department Of Surgery, Providence, RI, USA
M AN U
SC
Introduction: A proportion of trauma patients present for evaluation in a delayed fashion after injury, likely due to a variety of medical and nonmedical reasons. There has been little investigation into the characteristics and outcomes of trauma patients who present delayed. We hypothesize that trauma patients who present in a delayed fashion are a unique population at risk of increased trauma related complications. Materials and Methods: This was a retrospective review from 2010-2015 at a Level I trauma center. Patients were termed delayed if they presented greater than 24 hours after injury. Patients admitted within 24 hours of their injury were the comparison group. Charts were reviewed for demographics, mechanism, comorbidities, complications and outcomes. A subgroup analysis was done on patients who suffered falls.
EP
TE D
Results: During the five year period, 11,705 patients were admitted. 588 patients (5%) presented greater than 24 hours after their injury. Patients in the delayed group were older (65 vs. 55 years, p<0.001) and more likely to have psychiatric comorbidities (33% vs. 24%, p = 0.0001) than the control group. They were also more likely to suffer substance withdrawal (8.9% vs. 4.1%, p<0.001) but had toxicology testing for drugs and alcohol done at significantly lower rates. Patients that presented delayed after falls were similar in age and injury severity score (ISS) but more likely to suffer substance withdrawal when compared to those with falls that presented within 24 hours. Patients with falls that presented delayed had toxicology testing at significantly lower rates than the comparison group.
AC C
Conclusion: Trauma patients that present to the hospital in a delayed fashion have unique characteristics and are more likely to suffer negative outcomes including substance withdrawal. Future goals will include exploring strategies for early intervention, such as automatic withdrawal monitoring and social work referral for all patients who present in a delayed fashion.
Keywords: delayed presentation, elderly, trauma, injury, withdrawal
ACCEPTED MANUSCRIPT
Introduction Injuries often occur in an acute fashion demanding urgent attention. Advances in trauma systems have largely focused on delivery of expedient care to minimize
RI PT
physiologic and anatomic derangements. Despite ongoing improvements in these systems, not all trauma patients present immediately following injury mechanism. A proportion of trauma patients present to the hospital hours or days later, likely due to a variety of
SC
medical and nonmedical reasons.1,2,3 Thierry et al demonstrated that 42% of farmworkers that suffered injuries either delayed seeking medical treatment or did not seek treatment
M AN U
at all.2 The authors found that the inability to work and fear of missing payments were a larger factor than pain in making decisions about seeking treatment.
Otherwise there has been little investigation regarding patients that present in a delayed fashion after injury. There is a scarcity of data about the characteristics of
TE D
patients that present delayed, their injury mechanisms, why they present delayed, and whether delayed presentation affects outcomes in injured patients. Why patients present
EP
in a delayed fashion is not well understood but some insight has emerged from limited literature. These series have looked at patients with medical illnesses and injuries and
AC C
have cited reasons including lack of awareness of the consequences of the injury, limited access to healthcare,4,5 remote location,6 social isolation,7 or unusual signs and symptoms of disease, including atypical presentations in elderly trauma victims.8 Patients with diabetes who live in rural settings or are of lower socioeconomic and insurance status have been shown to delay seeking care resulting in increased costs and complications.15,16
ACCEPTED MANUSCRIPT
Despite advances in systems to improve timeliness of trauma care a significant number of patients still present for evaluation and treatment many hours or days after their injury. It remains unclear why such delays in seeking care exist. We hypothesize
RI PT
that trauma patients who present in a delayed fashion are a unique population that have
distinct patient characteristics, injury patterns, initial evaluations and outcomes. We also seek to explain why these patients present delayed after injury and to affirm that the
SC
initial workup of these patients should follow standard trauma protocols in regards to
M AN U
initial alcohol and substance testing.
Methods
This is a retrospective review of all traumatically injured patients requiring admission to a Level I trauma center between 2010 and 2015. Institutional review board
TE D
approval and waiver of consent was obtained prior to the study. Charts were reviewed for the recorded time of injury as well as the time of presentation. Patients were categorized as "Delayed" if they presented to our hospital greater than 24 hours after their initial
EP
injury. All patients that presented to our hospital within 24 hours of their injury were categorized as "Non-Delayed". Recording time from injury to presentation at our hospital
AC C
included any potential time spent at a referral hospital.
Charts were reviewed for patient characteristics including age, gender, injury
severity score (ISS), time to presentation in hours from the injury, time spent in hours at the referral hospital, whether a blood alcohol level (BAL) was checked in the emergency room, percentage of patients that had a positive BAL of those that were drawn, mean BAL of the positive BAL tests, whether testing for illicit drugs was done, and the
ACCEPTED MANUSCRIPT
percentage of patients with the presence of illicit drugs on presentation. Outcomes studied were in hospital mortality and occurrence of alcohol/drug withdrawal.
RI PT
A more in-depth sub-group analysis was done on patients who suffered falls. In addition to the patient characteristics reviewed in the overall group we chose to look at the comorbidities of patients that suffered falls and grouped these into psychiatric,
SC
cardiovascular, pulmonary, and endocrine categories. The comorbidities that fell within these categories were as defined by our trauma registry. We also looked at injuries in the
M AN U
patients that suffered falls and grouped them by anatomic regions: traumatic brain injury(TBI), thoracic, spinal, pelvic, extremities, and intraabdominal. Complications occurring during the hospital stay were recorded including all infections and alcohol/drug withdrawal, termed "Alcohol/Drug WD". Alcohol and drug withdrawal was defined in our registry per the National Trauma Database definition: "physical symptoms noted after
TE D
suddenly stopping consumption". Additionally patients could be charted in our registry as having withdrawal if they scored higher than 9 points on Clinical Institute Withdrawal
EP
Assessment (CIWA) scales or if there was any physician documentation stating withdrawal had occurred. Infections were either clinically evident (abscess, cellulitis,
AC C
pneumonia) or culture based (UTI, catheter, pneumonia via bronchoalveolar lavage). Pneumonia was diagnosed either by clinical judgment or by bronchoalveolar lavage in intubated patients with >10,000 colony forming units (CFU)/mL. Urinalysis (UA) was sent if urinary tract infection (UTI) was suspected based on clinical judgment. If the UA was positive, culture was sent and was considered positive if there were >100,000 CFU/mL. All other infections including Surgical Site Infections (SSI) were diagnosed by CDC criteria and cultures.
ACCEPTED MANUSCRIPT
We operationally defined the term “Effect of Delay” to encompass the total time patients have suffered from their injury at home prior to receiving care and in the hospital. To calculate “Effect of delay” we calculated the total time from the occurrence of the
RI PT
injury to final hospital discharge. This was calculated as follows: “Effect of Delay” = “Time (in hours) from Injury Occurrence to Presentation to Trauma Center” plus
“Hospital Length of Stay”. For patients who were transferred from an outside facility we
SC
calculated “Time (in hours) from Injury Occurrence to Presentation to Outside hospital” plus “Time (in hours) spent at Outside Hospital” plus “Trauma Center hospital Length of
M AN U
Stay”
Patient disposition on discharge was also reviewed. The number of patients who were discharged to their “prior site” of residence or were upgraded in terms of going to a “higher care facility” were recorded and compared in each group. For example, a patient
TE D
who was in an assisted living facility at the time of injury and then was discharged back to that facility at the same level of care returned to their “prior site” or equivalent to
EP
discharge to home. If a patient was living in any facility and was discharged to their own facility requiring a higher level of care or a different facility requiring a higher level of
AC C
care that was deemed to be a “higher care facility”
We also looked at the location of injury of patients that suffered falls to assess
whether there were differences between the delayed and non-delayed groups in terms of distance to the trauma center. This was done by querying the trauma registry to determine the zip code of the injury and then calculating the distance to the zip code of the trauma center. We grouped patients’ falls into four categories: whether they occurred
ACCEPTED MANUSCRIPT
less than 10 miles, 10 to 20 miles, 20 to 30, or greater than 30 miles from the trauma center.
RI PT
Data is expressed as either mean values with standard error (SEM) or for nonparametric data expressed as medians with interquartile ranges (IQR). Data was analyzed with Chi-squared, Student’s t test or Mann-Whitney-U. Significance was set at p<0.05.
SC
SPSS Version 21 was used. Kaplan-Meier analysis was done to depict “Effect of Delay”
M AN U
Results
Over the five-year period, 11,705 patients were admitted of whom 693 patients (5.9 %) presented greater than 24 hours after their initial injury. In the delayed group, the median time to presentation was 72 hours, (IQR=48-128 hours). Patients in the delayed group, when compared to those in the non-delayed group were older (64.7 versus 55.3
TE D
years; p<0.001) more often women (46.6% versus 41.5%; p=0.04), and had lower ISS (8.7+/0.3 versus 9.5+/-0.1; p=0.04). With respect to medical comorbidities, patients in the delayed group were more likely to have underlying psychiatric diagnoses (33% vs. 24%,
EP
p = 0.0001) and endocrine comorbidities (27% vs. 23%, p = 0.04) than the non-delayed
AC C
group. Patients in the delayed group were less likely to have a BAL level checked (26.5% versus 33%; p=0.01) and were less likely to test positive for alcohol. However, comparing BAL levels among those who tested positive for alcohol, patients in the delayed group has a significantly higher BAL (226 mg/dL +/- 26 versus 179 mg/dL +/-2; p<0.01). Patients in the delayed group were less likely to be tested for drugs of abuse on presentation (19% versus 40%; p<0.01). However there was no difference in rates of testing positive among those who were tested (Table 1).
ACCEPTED MANUSCRIPT
When comparing outcomes of the delayed and non-delayed groups, there was no difference in the mortality rate between the two groups. The delayed group was more
RI PT
likely to suffer from alcohol withdrawal (8.80% versus 4%, p < 0.001). There were significant differences in the most common injury mechanisms
between the two groups. Patients in the delayed group compared to the non-delayed
SC
group demonstrated significantly higher rates of burns (7.4% versus 2.8%, p < 0.001) and falls (77% vs. 50.2%, p < 0.001), and lower rates of motorized vehicle collisions (MVC
M AN U
or MCC) (4.3% versus 14.8%; p<0.01) and assaults (2.7% versus 10.1%; p<0.01) (Table 2).
Overall, 6,114 patients (51.8% of the entire admitted trauma population over the study period) sustained a fall as the mechanism of injury. Among this fall group, 533
TE D
(7.4%) of the patients presented greater than 24 hours following the fall and constitute the Fall Delayed group (FD). In the FD group, the average time from fall to presentation was 72 hours (IQR 48-120). The remaining 5,581 patients who fell presented within 24 hours
EP
of the trauma and represent the Fall Non-Delayed (FND) group. Comparing FD to FND patients, there was no difference in respect to age or ISS.
AC C
The FD group were less likely to undergo either alcohol testing (26.6% versus 42%; p<0.01) or drug use testing (19% versus 28%; p<0.01) but were more likely to have a positive drug test if tested (68.2% versus 58%, p<0.04) (Table 3). Patients in the FD group compared to the FND group had higher rates of psychiatric (33% versus 27%; p<0.01) and pulmonary (35% versus 25%; p<0.01) comorbidities. There was no difference in rates of TBI or pelvic fractures between the FD
ACCEPTED MANUSCRIPT
and FND patients. However, FD patients had higher rates of thoracic (33% versus 16%; p<0.01), spinal (19% versus 13%; p<0.01), and intraabdominal (6% versus 2%; p<0.01) injuries, but were less likely to have sustained extremity trauma (33% versus 58%;
RI PT
p<0.01) (Table 4). With respect to fall patients who sustained thoracic trauma, 99% of
FD patients sustained rib fractures compared with only 86% in the FND group (p<0.01). Hemothorax or pneumothorax occurred in 33% of FD patients compared with 21% of
SC
FND patients (p < 0.01).
M AN U
Outcomes including mortality, hospital length of stay, ICU length of stay, rate of occurrence of an infection, and discharge location were similar between FD and FND patients. However, alcohol withdrawal was significantly more likely to occur in the FD group (9.2% vs. 4%, p<0.01). There was no significant difference in LOS in patients who developed substance withdrawal comparing FD and FND patients. FD patients were
(Table 5).
TE D
noted to have significantly greater effect of delay (11 +/-0.3 versus 7 +/-0.2 days; p<0.01)
EP
We did not find differences in the delayed and non-delayed groups in regards to location of patient falls and distance to the trauma center. This was the case across the
AC C
four distances reviewed: less than 10 miles, 10 to 20 miles, 20 to 30 miles, and greater than 30 miles from the trauma center. (Table 6) Discussion
In this review of a large trauma registry we have shown that patients who present delayed after injury had several unique patient characteristics, mechanisms and outcomes. They were older, suffered from more preinjury psychiatric and endocrine comorbidities,
ACCEPTED MANUSCRIPT
and had different trauma mechanisms compared to patients who presented more immediately after injury. Analysis of patients that suffered falls also was revealing. Injury patterns were different with twice as many thoracic injuries but far less extremity injuries
RI PT
in the delayed group. Delayed fall patients suffered alcohol and drug withdrawal at more than twice the rate of the nondelayed fall group. To our knowledge there has been
limited investigation into patients that present to the hospital many hours or days after
SC
injury.
M AN U
Patients in our delayed group presented a median of 3 days after their injuries in both the overall and fall groups. Previous studies have attempted to explain why there are delays in seeking care after injury. Thierry et al recently reported a significant number of victims of farm work injuries either underestimate the degree of injury or believe that they can self treat their injuries.2 The farm workers were also more likely to present for
TE D
evaluation if their ability to work was compromised rather than from pain citing financial concerns as the reason for this. Along these lines our patients who fell and presented
EP
delayed suffered more thoracic but less extremity injuries than the nondelayed group. While thoracic injuries can result in significant discomfort they do not lead to an
AC C
immediate sense of disability or loss of function like an extremity injury. There has been far more investigation into reasons for delayed presentations in patients with medical or psychiatric illnesses. Yousaf noted in a systematic review that fear of expressing emotions or yielding emotional and psychological control was associated with delays in seeking medical care.9 Other disease processes such as community acquired pneumonia (CAP) and acute myocardial infarction often present with atypical or more subtle findings in elderly patients leading to delays in presentation and diagnosis.10 Faverio's 2014
ACCEPTED MANUSCRIPT
review of the management of CAP in the elderly described some of these atypical signs including confusion, decreased appetite, unsteadiness, and even falls.11 We speculate that similar atypical signs and symptoms such as blunted pain responses, decreased activity,
RI PT
and decline in appetite may have occurred after injury in our patients, especially in the elderly.
SC
A number of investigators have shown reasons for delays in care including social isolation, inadequate access to care, and concerns for costs of care. Elderly patients often
M AN U
live alone and present to the hospital at the urging of a family member. They also may not have the means to physically get to the hospital as many no longer drive. In patients over 65 years old that have suffered acute myocardial infarction delays greater than 6 hours in seeking care have been ascribed to the following: lower levels of education, rural location of the patient, and assumption that symptoms were of noncardiac etiology.12
TE D
Common themes that have been described by these patients and have been shown in other investigations include living alone and a desire to not bother others or burden children. Patients with acute decompensated heart failure who present delayed have been shown to
EP
be of older age and more often male.13 We chose to look at patients that fell in more
AC C
detail as this is the most common mechanism of injury at our institution. It is also the mechanism most often associated with mortality and morbidity in the elderly. Barriers to care for those with diabetes and psychiatric comorbidities are known
to exist and may relate to the increase in time to presentation after injury in these patients in our study. Numerous patient, provider, and systems based issues have been cited as obstacles for diabetes patients. These obstacles can lead to suboptimal management and development of diabetes related complications such as infected wounds, retinopathy,
ACCEPTED MANUSCRIPT
cardiovascular disease, and diabetic ketoacidosis. Lack of health insurance has been shown to increase the risk of diabetes complications and costs.14 Also, patients who live in impoverished areas and are of lower education are seen less frequently by diabetes
RI PT
specialists and use less preventive services. These patients were more likely to report
difficulties accessing medical care in general and to have not been seen by any physician in the last year and were less likely to have health insurance. Similar access issues have
SC
been reported in diabetics that live in rural areas.15,16 Though we found a higher rate of
endocrine comorbidities which included diabetes and obesity in our delayed presentation
M AN U
patients we did not review their geographic locations or socioeconomics. Doing so may have affirmed the connection between their delay in presentation after injury and their issues with access to care for diabetes. Limitations in access have also been noted in patients with psychiatric disease. Patients with chronic mental illness often have several
TE D
associated medical comorbidities such as hypertension, diabetes, and alcohol abuse. However, they often do not receive needed procedures, medications or general care due to real or patient perceived obstacles; these include financial constraints, poor compliance
EP
with care plans, lack of awareness of available resources, and even convenience issues.17,18 Patients with psychiatric comorbidities also may face stigma and
AC C
embarrassment and may have had previous healthcare experiences that leave them hesitant to seek further care. Patients in our delayed presentation groups, overall and falls, had significantly
lower rates of alcohol and drug toxicology screening on presentation than the nondelayed groups. This may be due to the fact that these delayed injuries seem less emergent, thus leading the provider to overlook the fact that alcohol or drugs contributed to the injury.
ACCEPTED MANUSCRIPT
However, consensus recommendations that trauma centers screen for alcohol and drugs after injury have existed for decades. These recommendations are based on large amounts of data showing a strong relationship between alcohol, drug use and injury; 40 percent of
RI PT
motor vehicle and pedestrian fatalities involve alcohol use and similar findings have been shown for drug use.19 Screening can provide patients with social work consultation, brief interventions, and opportunities for outpatient programs after discharge. With adherence
SC
to these basic measures that are required for trauma level one center verification trauma
screening need to be improved.
M AN U
recidivism can be reduced. Clearly based on our study's data, rates of drug and alcohol
Delayed trauma patients in our study were more likely to suffer from alcohol/drug withdrawal during their hospital stay further supporting the need for screening on presentation to be thorough. The source of this difference in incidence of
TE D
withdrawal is unclear but may be related to the higher rates of psychiatric illness in the delayed groups noted above. It also suggests that the delayed population may need more rigorous monitoring for alcohol withdrawal symptoms to prevent further complications
EP
during their hospital stay. Alcohol withdrawal has been associated with increased
AC C
mortality in inpatients.20,21 Early and careful dosing of benzodiazepines, triage to the ICU, and surgical house staff vigilance could improve prevention or outcomes of alcohol withdrawal if screening occurred with increased frequency in trauma patients whether they present delayed or not. Limitations of our study include possible inaccurate reporting of time of injury by patients. This makes the duration of delay likely uncertain in a number of the patients. Also, by limiting our more in depth analyses to patients with falls we may be missing
ACCEPTED MANUSCRIPT
important findings in the other mechanism groups. Our psychiatric comorbidity category included dementia in addition to other forms of mental illness and we did not separate out these entities to assess their association with outcomes such as alcohol withdrawal.
RI PT
Lastly, this study is unable to account for the people who suffer injury but never report to the hospital.
SC
Conclusion
Trauma patients who presented to the hospital in a delayed fashion after their
M AN U
initial injury were older, had more frequent psychiatric and endocrine comorbidities, and more often suffered negative outcomes such as alcohol withdrawal. Yet this vulnerable group less frequently had alcohol and drug toxicology screening on presentation. It is important to assess these patients with the same standards as those who present early after
TE D
injury. This may allow them to receive more timely and focused care as it pertains to prevention and management of alcohol withdrawal. Future goals will include monitoring rates of drug and alcohol screening and exploring strategies for early intervention such as
EP
automatic alcohol withdrawal monitoring, arranging for better management of
AC C
comorbidities, and social work referral for patients who present in a delayed fashion.
ACCEPTED MANUSCRIPT
Disclosures: The authors report no proprietary or commercial interest in any product
AC C
EP
TE D
M AN U
SC
RI PT
mentioned or concept discussed in this article.
ACCEPTED MANUSCRIPT
Table 1. Demographics and Mechanisms of injury of all patients Nondelayed
N = 693
N = 11117
64.7 ±0.76
55.3 ± 0.22
<0.01
Female Gender (%)
46.6%
41.5%
<0.01
ISS* (0-75) Time to Presentation** (hrs) Time at referral hospital #(hrs) BAL age*
8.7 ±0.25
9.5 ±0.1
72 (48-128)
0.9 (0.6-2)
3.8 ±0.15
2.8 ±0.05
60.2 ±1.5
46.46 ±0.3
26.5%
33%
34%
65%
226.6 ±24.5
179 ±2
% of +BAL BAL of + tests* % Drug tested
19%
% +Drug test
70.2%
<0.01
<0.01
<0.01
SC
% of BAL drawn
<0.01
M AN U
Age* (years)
p
RI PT
Delayed
<0.01
<0.01
<0.01
40%
<0.01
65.4%
N/S
AC C
EP
TE D
*Values are Mean (SEM) **Values are Median (IQR) #Time spent at referral hospital of patients that were transferred to our hospital
ACCEPTED MANUSCRIPT
Table 2. Mechanism of injury of all patients p
Burn
7.4%
2.8%
<0.01
MVC
4.3%
14.8%
<0.01
MCC
1.0%
5.5%
<0.01
Falls
77%
50.2%
<0.01
Assault
2.7%
10.1%
<0.01
Self-Inflicted
0.1%
2.6%
<0.01
Work Related
2.3%
5.2%
<0.01
RI PT
N = 11117
SC
Nondelayed
N = 693
AC C
EP
TE D
M AN U
Delayed
ACCEPTED MANUSCRIPT
Table 3. Demographics of patients following fall Delayed
Nondelayed
p
N = 5581
68.7 ±0.26
N/S
49.5%
54.9%
0.02
9.45 ±0.28
8.97 ±0.09
72 (48-120)
0.97 (0.65-3.25)
64.4 ±1.6
60.8 ±0.55
% of BAL drawn
26.6%
42%
% of +BAL
15.5%
27.8%
% Drug tested
220.86 ±26.3 19%
194.4 ±4.2 28%
<0.01
% +Drug tests
68.2%
58%
0.03
BAL of + tests*
AC C
EP
TE D
*Values are Mean (SEM) **Values are Median (IQR)
N/S
<0.01 N/S
<0.01
<0.01
SC
ISS* Time to presentation** (hrs) BAL age*
M AN U
Female gender %
RI PT
N = 533
69.3 ±0.8
Age*
N/S
ACCEPTED MANUSCRIPT
Table 4. Comorbidities and Injuries of patients following fall Nondelayed
p
N = 533
N = 5581
33% 61% 35% 27%
27% 58% 25% 26%
<0.01 N/S <0.01 N/S
30% 33% 19% 7% 33% 6%
27% 16% 13% 6% 58% 2%
N/S <0.01 <0.01 N/S <0.01 <0.01
AC C
EP
M AN U
TE D
TBI Thoracic Spinal Pelvic Extremities Intraabdominal
SC
Injury
RI PT
Comorbidity Psychiatric Illness Cardiovascular Pulmonary Endocrine
Delayed
ACCEPTED MANUSCRIPT
Table 5. Outcomes of patients following fall
N = 533
N = 5581
5% 6.56 ±0.3 4.8 ±0.4 9%
4.7% 7.0 ±0.2 5.6 ±0.2 8.4%
N/S N/S N/S N/S
9.2%
4%
<0.01
9 ±1.7
12.04 ±1.9
N/S
11.01 ±0.3
7.04 ±0.2
<0.01
45% 48%
48% 45%
N/S N/S
RI PT
p
SC
Nondelayed
M AN U
Outcome Mortality LOS* ICU LOS* Infection rate Alcohol/Drug withdrawal LOS of pts with alcohol withdrawal Effect of Delay† Disposition Higher Care Facility Prior Site
Delayed
AC C
EP
TE D
*Values are Mean (SEM) †Effect of delay = Average of (Time from fall to presentation + Hospital LOS)
ACCEPTED MANUSCRIPT
Table 6. Distance from injury location to trauma center Nondelayed falls
N=430
N=5263
69% 20.5% 7.4% 3%
66.5% 22.9% 7.7% 3%
p N/S N/S N/S N/S
RI PT
Delayed falls
AC C
EP
TE D
M AN U
SC
Distance to trauma center (miles) ≤10 10-20 20-30 30>
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1. Kaplan-Meier analysis of patients that suffered falls. Outcome-discharge from the hospital.
ACCEPTED MANUSCRIPT
References 1. Vadivelu S, Esernio-Jenssen D, Rekate H, Naravan R, Mittler M, Schneider S. Delay in arrival to care in prepetrator identified nonaccidental head trauma:
RI PT
observations and outcomes. World Neurosurgery. 2015;84(5):1340-6.
2. Thierry A, Snipes S. Why do farmworkers delay treatment after debilitating
SC
injuries? Thematic analysis explains if, when and why farmworkers were treated for injuries. American Journal of Industrial Medicine. 2015;58(2):178-92.
M AN U
3. Wood J, Rubin D, Nance M, Christian C. Distinguishing inflicted versus accidental abdominal injuries in young children. Journal of Trauma. 2005;59(5):1203-8.
4. Marshall I, Wang Y, Crichton S, McKevitt C, Rudd A, Wolfe C. The effects of
TE D
socioeconomic status on stroke risk and outcomes. Lancet Neurology. 2015;14(12):1206
EP
5. Duron V, Monaghan S, Connolly M, Gregg S, Stephen A, Adams C, Cioffi WG, Heffernan DS. Undiagnosed medical comorbidities in the uninsured: a significant
AC C
predictor of mortality following trauma. Journal of Trauma and Acute Care Surgery. 2012;73(5):1093-8.
6. Veitch P. Anticipated response to three common injuries by rural and remote area residents. Social science and Medicine. 1995;41(5):739-45.
ACCEPTED MANUSCRIPT
7. Bowden M, Grant S, Vogel B, Prasad J. The elderly, disabled and handicapped adult burned through abuse and neglect. Burns, including thermal injury.
RI PT
1988;14(6):447-50. 8. Heffernan D, Thakkar R, Monaghan S, Ravindran R, Adams C, Kozloff M, Gregg SC, Connolly MD, Machan JT, Cioffi WG. Normal presenting vital signs are
SC
unreliable in geriatric blunt trauma victims. Journal of Trauma. 2010;69(4):81320.
M AN U
9. Yousaf O, Grunfeld E, Hunter M. A systematic review of the factors associated with delays in medical and psychological help-seeking among men. Health Psychology Review. 2015;9(2):264-76.
10. Claridge J, Crabtree T, Pelletier S, Butler K, Sawyer R, Young J. Persistent occult
TE D
hypoperfusion is associated with a significant increase in infection rate and mortality in major trauma patients. Journal of Trauma. 2000;48(1):8-14
EP
11. Faverio P, Aliberti S, Bellelli G, Suigo G, Lonni S, Pesci A, Restrepo MI. The management of community acquired pnuemonia in the elderly. European Journal
AC C
of Internal Medicine. 2014;25(4):312-9. 12. Hwang SY, Jeong MH. Cognitive factors that influence delayed decision to seek treatment among older patients with acute myocardial infarction in Korea. Eur J Cardiovasc Nurs. 2012 Jun;11(2):154-9.
ACCEPTED MANUSCRIPT
13. Darling C, Saczynski JS, McManus DD, Lessard D, Spencer FA, Goldberg RJ. Delayed hospital presentation in acute decompensated heart failure: clinical and
RI PT
patient reported factors. Heart Lung. 2013 Jul-Aug;42(4):281-6. 14. Brown DS, McBride TD. Impact of the Affordable Care Act on access to care for US adults with diabetes, 2011-2012. Prev Chronic Dis. 2015 May 7;12:E64.
SC
15. Zgibor JC, Songer, TJ. External Barriers to Diabetes Care: Addressing Personal
M AN U
and Health Systems Issues. Diabetes Spectrum. Jan 2001 vol. 14 no. 1 23-28 16. Zgibor JC, Songer TJ, Kelsey SF, Weissfeld J, Drash AL, Becker D, Orchard TJ. The association of diabetes specialist care with health care practices and glycemic control in patients with type 1 diabetes: a cross-sectional analysis from the Pittsburgh epidemiology of diabetes complications study. Diabetes Care. 2000
TE D
Apr;23(4):472-6.
17. Kisely S, Smith M, Lawrence D, Cox M, Campbell LA, Maaten S. Inequitable
EP
access for mentally ill patients to some medically necessary procedures. CMAJ 2007;176:779–784
AC C
18. McIntyre RS, Soczynska JK, Beyer JL, Woldeyohannes HO, Law CW, Miranda A, Konarski JZ, Kennedy SH. Medical comorbidity in bipolar disorder: reprioritizing unmet needs. Curr Opin Psychiatry 2007;20:406–416
19. FACS: Subcommittee on Injury Prevention and Control. Alcohol and Injury. https://www.facs.org/~/media/files/quality%20programs/trauma/alcoholinjury.ash x. Acessed February 1 2016.
ACCEPTED MANUSCRIPT
20. Tavel ME, Davidson W, Batterton TD. A critical analysis of mortality associated with delirium tremens. Review of 39 fatalities in a 9-year period. Am J Med Sci.
RI PT
1961 Jul;242:18-29 21. Yost DA. Alcohol withdrawal syndrome. Am Fam Physician. 1996
AC C
EP
TE D
M AN U
SC
Aug;54(2):657-64, 669.