Sage Journals: Discover world-class research

Abstract

Background

Prisoners are a vulnerable population, and there are few contemporary studies that consider trauma patient outcomes within the prisoner population. Therefore, we sought to provide a descriptive analysis of prisoners involved in trauma and evaluate whether a healthcare disparity exists. We hypothesized that prisoners and non-prisoners have a similar risk of mortality and in-hospital complications after trauma.

Methods

The Trauma Quality Improvement Program (2015-2016) was queried for trauma patients based upon location inside or outside of prison at the time of injury. A multivariable logistic regression analysis was performed to compare these groups for risk of mortality—the primary outcome.

Results

From 593,818 trauma patients, 1115 were located in prison. Compared to non-prisoners, prisoner trauma patients had no significant difference in mortality (5.1 vs 6.0%, P = .204). However, after adjusting for covariates, prisoners had a shorter length of stay (LOS) (mean days, 6.3 vs 7.8, P < .001), shorter intensive care unit (ICU) LOS (mean days, 5.44 vs 5.89, P = .004), and fewer complications, including lower rates of drug/alcohol withdrawal (.4% vs 1.1%, P = .030), pneumonia (.5 vs 1.6%, P = .004), and urinary tract infections (.0 vs 1.1%, P < .001). Upon performing a multivariable logistic regression model, prisoner trauma patients had a similar associated risk of mortality compared to non-prisoners (OR 1.61, CI .52-4.94, P = .409).

Discussion

Our results suggest that prisoner trauma patients at least receive equivalent treatment in terms of mortality and may have better outcomes when considering some complications. Future prospective studies are needed to confirm these results and explore other factors, which impact prisoner patient outcomes.

Keywords

trauma prisoner incarceration mortality outcome

Key Takeaways

Prisoner and non-prisoner trauma patients have no differences in rate or associated risk of mortality

Prisoner trauma patients may have less in-hospital complications than non-prisoners

Prisoner trauma patients have a decreased hospital length of stay compared to non-prisoners

Introduction

The United States (US) has the highest incarceration rate worldwide, with 440 prisoners per 100,000 of the national population, and an estimated 6.6 million US citizens within the correctional system as of 2017.^1,2 In fact, this accounts for one quarter of the world’s prison population. Historically, prisoners represented a population that underwent horrific medical research experiments including those performed by the Nazis in World War II and even by the US Army in the 1970s.^3-5 At the time, an estimated 90% of all drug research used prisoners, which also included testing chemical warfare agents without the requirement of obtaining informed consent.⁶ However, throughout the 1970s to 1990s, increased public awareness ultimately led the US government to respond in the form of the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research.^6,7 Under Subpart C, additional regulations beyond the basic requirement for research with human subjects were set in place, which protected prisoners and ultimately led to a drastic decrease in studies performed on prisoners.^6,8,9 For example, it classified prisoners as a vulnerable population similar to children, and prohibits certain forms of research such as medical experimentation, cosmetic research and pharmaceutical testing.^6-9

From the 1970s to the mid-2000s, the incarcerated population has grown 5-fold.⁶ Prisons have become overcrowded, and thus prisoners’ access to services and healthcare is limited.¹⁰ In addition, many prison systems have come under fire for inadequate treatment of prisoners and are often characterized by high rates of communicable diseases, mental illness, and substance abuse.^10,11

Unfortunately, limited contemporary literature regarding outcomes for prisoner trauma patients exists. Although prisoners have a 5-times higher rate of sustaining traumatic injury compared to non-incarcerated patients, the aforementioned constraints on research have made it increasingly difficult to study. Furthermore, researchers sometimes have difficulty obtaining informed consent, such as when a patient has a debilitating condition such as traumatic brain injury where a legally authorized representative is needed for decision-making.^12-14 Therefore, we sought to provide a descriptive analysis of prisoner trauma patients and evaluate whether any healthcare disparity exist for this vulnerable population. We hypothesized that prisoner trauma patients have a similar risk of mortality and in-hospital complications compared to non-prisoner trauma patients.

Methods

This study was approved by the University of California, Irvine’s institutional review board (IRB). The IRB approved waiver of consent. We performed a retrospective analysis of the Trauma Quality Improvement Program (TQIP) between January 2015 and December 2016. TQIP began collecting the location where the trauma occurred beginning in 2015. All trauma patients ≥18 years of age sustaining injuries within prisons were identified using location codes listed in Supplemental Table.

The primary outcome studied was mortality. Secondary outcomes included total hospital length of stay (LOS), intensive care unit (ICU) LOS, ventilator days, and in-hospital complications. Variables collected included: age, gender, race, systolic blood pressure (SBP) on arrival, respiratory rate (RR) on arrival, heart rate on arrival, and total Glasgow coma scale (GCS) on arrival. Comorbidities included congestive heart failure (CHF), smoking status, chronic renal failure, cerebrovascular accident (CVA), diabetes (DM), cancer, history of myocardial infarction (MI), peripheral vascular disease (PVD), hypertension (HTN), chronic obstructive pulmonary disease (COPD), steroid use, cirrhosis, dementia, and psychiatric illness. The injury profile included the injury severity score (ISS), abbreviated injury scale (AIS), and injuries to the brain, face, spine, ribs, sternum, pelvis, extremities, intrathoracic organs, intraabdominal organs, crush injuries, and burns. Injuries were identified by the appropriate International Classification of Disease diagnosis codes (version 9).

All variables were coded as present or absent. Descriptive statistics were performed for all variables. A Student’s t-test or Mann-Whitney-U test was used to compare continuous variables and chi-square was used to compare categorical variables for bivariate analysis. Categorical data were reported as percentages, and continuous data were reported as medians with interquartile range. The magnitude of the association between predictor variables and mortality was first measured using a univariable logistic regression model. Covariates were chosen by a discussion among coauthors and review of the literature.^15-17 Covariates with statistical significance (P < .20) were included in a hierarchical multivariable logistic regression model and the adjusted risk for mortality was reported with an odds ratio (OR) and 95% confidence intervals (CIs). Covariates included in this model were prison status, GCS, AIS, and documented medical comorbidities. All P-values were two-sided, with a statistical significance level of <.05. All analyses were performed with IBM SPSS Statistics for Windows (version 24, IBM Corp., Armonk, NY).

Results

Patient Characteristics

There were 593,818 trauma patients identified, of whom 592,703 (99.8%) were non-prisoners and 1115 (.19%) were prisoners. Compared to non-prisoners, prisoners were younger (median age, 39 vs 55, P < .001) and more often male (96.0 vs 61.9%, P < .001). Prisoner trauma patients were more often black (23.4 vs 13.3%, P < .001) and Hispanic (16.1 vs 10.6%, P < .001) compared to non-prisoners. There existed no difference in median ISS (10.0 vs 10.0, P = .57) between the 2 cohorts. Compared to non-prisoner trauma patients, prisoners had a higher rate of smokers (23.0 vs 20.1%, P = .014), cirrhosis (2.2 vs 1.0%, P < .001), and psychiatric illness (19.2 vs 10.4%, P < .001). Prisoners observed lower rates of CHF (1.8 vs 3.9%, P < .001), chronic renal failure (.9 vs 1.7%, P < .001), history of CVA (.9 vs 1.7%, P = .001), DM (7.1 vs 13.8%, P < .001), prior MI (.4 vs 1.3%, P = .007), HTN (20.9 vs 36.1%, P < .001), COPD (5.1 vs 7.1%, P = .010), and dementia (.8 vs 5.5%, P < .001) (Table 1).

Table 1.

Baseline Characteristics for Non-Prisoner and Prisoner Patient Populations.

	Non-prisoner	Prisoner	—
Characteristic	(n = 592,703)	(n = 1115)	P-Value
Teaching university (%)	300 867 (50.7%)	683 (61.3%)	<.001
Demographics	—
Age, year, median (IQR)	55.0 (33, 72)	39.0 (30, 52)	<.001
Male, n (%)	366 656 (61.9%)	1070 (96.0%)	<.001
Race	—
Black (%)	74 071 (13.3%)	261 (23.4%)	<.001
White (%)	438 351 (73.8%)	655 (9.2%)	<.001
Asian (%)	12 891 (2.9%)	15 (2.2%)	<.001
Ethnicity	—
Hispanic (%)	57 283 (10.6%)	171 (16.1%)	<.001
ISS/vitals/GCS	—
Total GCS, median (IQR)	15.0 (14, 15)	15.0 (15, 15)	<.001
SBP, median (IQR)	138.0 (121, 155)	132 (118, 147)	<.001
Respiratory rate, median (IQR)	18.0 (16, 20)	18.0 (16, 20)	.002
Pulse, median (IQR)	86.0 (74, 100)	84.0 (72, 98)	.009
ISS, median (IQR)	10.0 (9, 17)	10.0 (9, 16.75)	.57
Comorbidities	—
Congestive heart failure	23 273 (3.9%)	18 (1.6%)	<.001
Chronic renal failure	9982 (1.7%)	10 (.9%)	<.001
Diabetes mellitus	82 208 (13.8%)	79 (7.1%)	<.001
Hypertension	214 504 (36.1%)	233 (20.9%)	<.001
Cirrhosis	5730 (1.0%)	25 (2.2%)	<.001
Dementia	32 844 (5.5%)	9 (.8%)	<.001
Psychiatric illness	61 591 (10.4%)	214 (19.2%)	<.001
Cerebrovascular accident	16 903 (2.8%)	14 (1.3%)	.001
History of myocardial infarction	7455 (1.3%)	4 (.4%)	.007
COPD	42 011 (7.1%)	57 (5.1%)	.01
Smoker	119 130 (20.1%)	257 (23%)	.014
Steroid use	4678 (.8%)	3 (.3%)	.05
Peripheral vascular disease	4638 (.8%)	3 (.3%)	.052
Cancer—disseminated	4008 (.7%)	1 (.1%)	.17
ADHD	4290 (.7%)	11 (1.1%)	.301

Teaching university = this represents whether the patient was treated at a university hospital as opposed to a community or non-teaching hospital as defined in the Trauma Quality Improvement Program Data Dictionary; IQR = interquartile range; ISS = injury severity score; GCS = Glasgow coma scale; SBP = systolic blood pressure; COPD = chronic obstructive pulmonary disease; ADHD = attention deficit hyperactivity disorder.

Regarding mechanism of injury, significant differences were noted when comparing the two populations. Prisoners suffered higher rates of suicide attempts (1.5 vs .8%, P = .006) and assault (6.9 vs 5.2%, P = .01). Non-prisoners suffered from a higher rate of falls (27.9 vs 8.5%, P < .001) and gunshot wounds (2.7 vs .1%, P < .001). There were no observed differences in the rates of burns, crush injuries or stabbings between the two populations. (Table 2)

Table 2.

Injury Profiles for Non-Prisoner and Prisoner Patient Populations.

	Non-prisoner	Prisoner	—
—	(n = 592,703)	(n = 1115)	P-Value
Injury Type	—
TBI	142 086 (23.9%)	100 (9.0%)	<.001
Lower extremity	114 206 (19.2%)	37 (3.3%)	<.001
Rib	94 166 (15.9%)	18 (1.6%)	<.001
Spine	79 215 (13.3%)	17 (1.5%)	<.001
Upper extremity	67 682 (11.4%)	13 (1.2%)	<.001
Lung	83 600 (14.1%)	27 (2.4%)	<.001
Pelvis	29 314 (4.9%)	3 (.3%)	<.001
Spleen	16 061 (2.7%)	11 (1.0%)	<.001
Liver	14 960 (2.5%)	3 (.3%)	<.001
Sternum	11 231 (1.9%)	1 (.1%)	<.001
Kidney	7832 (1.3%)	1 (.1%)	<.001
Face/neck	38 816 (6.5%)	48 (4.3%)	.002
Colorectal	4972 (.8%)	2 (.2%)	.016
Other genitourinary	2949 (.5%)	0 (.0%)	.018
Heart	2253 (.4%)	0 (.0%)	.039
Small intestine	5365 (.9%)	4 (.4%)	.054
Stomach	1598 (.3%)	1 (.1%)	.247
Esophagus	114 (.0%)	0 (.0%)	.603
Pancreas	521 (.1%)	0 (.0%)	.322
Injury mechanism	—
Fall	165 469 (27.9%)	95 (8.5%)	<.001
Gunshot wound	15 910 (2.7%)	1 (.1%)	<.001
Suicide	4724 (.8%)	17 (1.5%)	.006
Assault	30 865 (5.2%)	77 (6.9%)	.011
Crush	856 (.1%)	0 (.0%)	.204
Burn	644 (.1%)	0 (.0%)	.271
Stab	8135 (1.4%)	12 (1.1%)	.395

TBI = traumatic brain injury.

Primary and Secondary Outcomes

There was no significant difference in mortality between the two groups (5.1 vs 6.0%, P = .20). Prisoners had a shorter LOS (mean days, 6.3 vs 7.8, P < .001), and shorter ICU LOS (mean days, 5.44 vs 5.89, P = .004). However, prisoners had a similar amount of ventilator days (6.73 vs 6.86, P = .85) compared to non-prisoners. Prisoners had lower rates of drug/alcohol withdrawal (.4 vs 1.1%, P = .030), pneumonia (PNA) (.5 vs 1.6%, P = .004), urinary tract infections (UTI) (.0% vs 1.1%, P < .001), and other complications (8.3 vs 11.1%, P < .002), compared to non-prisoners. No significant difference existed between the two groups across all other complications studied (Tables 3 and 4).

Table 3.

In-Hospital Complications for Non-Prisoner and Prisoner Patient Populations.

	Non-prisoner	Prisoner	—
Associated Complications	(n = 592,703)	(n = 1115)	P-Value
Urinary tract infection	6417 (1.1%)	0 (.0%)	<.001
Other	66 039 (11.1%)	92 (8.3%)	.002
Pneumonia	9714 (1.6%)	6 (.5%)	.004
Drug/EtOH withdrawal	6740 (1.1%)	5 (.4%)	.03
Ventilator-associated pneumonia	3351 (.6%)	11 (1.0%)	.061
Unplanned intubation	9596 (1.6%)	11 (1.0%)	.094
Unplanned ICU	11 825 (2.0%)	15 (1.3%)	.121
Severe sepsis	3595 (.6%)	3 (.3%)	.147
Decubitus	4215 (.7%)	4 (.4%)	.162
Myocardial infarction	1836 (.3%)	1 (.1%)	.186
Extremity compartment syndrome	902 (.2%)	0 (.0%)	.192
Deep site infection	1528 (.3%)	1 (.1%)	.268
Pulmonary embolism	3571 (.6%)	4 (.4%)	.293
ARDS	4332 (.7%)	11 (1.0%)	.317
Acute kidney injury	5829 (1.0%)	8 (.7%)	.368
CLABSI	224 (.0%)	1 (.1%)	.374
CRBSI	376 (.1%)	0 (.0%)	0.4
Unplanned return to operating room	4422 (.7%)	6 (.5%)	.42
Community acquired UTI	1790 (.3%)	2 (.2%)	.456
Deep venous thrombosis	7736 (1.3%)	12 (1.1%)	.501
Osteomyelitis	218 (.0%)	0 (.0%)	.522
Cardiac arrest w/CPR	7657 (1.3%)	12 (1.1%)	.524
Cerebrovascular accident	2553 (.4%)	6 (.5%)	.584
Organ space infection	1193 (.2%)	3 (.3%)	.614
Graft flap failure	101 (.0%	0 (.0%)	.663
Superficial infection	1594 (.3%)	3 (.3%)	.999

EtOH = alcohol; ICU = intensive care unit; ARDS = acute respiratory distress syndrome; CLABSI = central line associated bloodstream infection; CRBSI = catheter-related bloodstream infection; UTI = urinary tract infection; CPR = cardiopulmonary resuscitation.

Table 4.

Outcomes for Non-Prisoner and Prisoner Patient Populations.

	Non-prisoner	Prisoner	—
Outcome	(n = 592,703)	(n = 1115)	P-Value
Total LOS days	—
Mean (CI)	7.76 (7.73, 7.78)	6.32 (5.87, 6.77)	<.001
Median (IQR)	5 (3, 9)	4 (2, 7)	<.001
Total ICU LOS	—
Mean (CI)	5.89 (5.86, 5.91)	5.44 (4.72, 6.16)	.004
Median (IQR)	3 (2, 6)	3 (2, 5)	.004
Total vent days	—
Mean (CI)	6.86 (6.81, 6.92)	6.73 (5.26, 8.19)	.848
Median (IQR)	3 (2, 9)	3 (2, 8)	.848
All mortality	35 672 (6.0%)	57 (5.1%)	.204

LOS = length of stay; CI = confidence interval; IQR = interquartile range; ICU = intensive care unit.

Risk of Mortality in Trauma Patients Who are Prisoners

After adjusting for covariates in a multivariable logistic regression model, trauma patients who were prisoners had a similar associated risk of mortality compared to non-prisoner trauma patients (OR 1.61, CI .52-4.94, P = .409). The strongest independent risk factor associated with increased mortality was cirrhosis (OR 3.77, CI 3.26-4.36, P < .001), followed by hypotension on admission (OR 2.65, CI 2.67-3.05, P < .001). The independent risk factors associated with decreased mortality included higher total GCS (OR .76, CI .76-.77, P < .001), comorbid HTN (OR .93, CI .88-.98, P < .001), and psychiatric illness (OR .71, CI .66-.77, P < .001) (Tables 5, 6).

Table 5.

Univariate Analysis for Risk of Mortality in Trauma.

Risk of Mortality	OR	CI	P-Value
Prisoners vs non-prisoners	.84	.64-1.10	.204
Severe AIS-head	6.27	6.13-6.41	<.001
Severe AIS-thorax	3.75	3.64-3.87	<.001
Severe AIS-abdomen	3.13	3.10-3.26	<.001
Hypotensive on admission	6.98	6.76-7.20	<.001
Age	1.01	1.00-1.01	<.001
ISS	1.11	1.11-1.11	<.001
Pulse rate	1.01	1.01-1.01	<.001
Respiratory rate	.95	.95-.96	<.001
Total GCS	.74	.74-.74	<.001
Comorbid CVA	1.27	1.20-1.34	<.001
Comorbid DM	1.03	1.00-1.07	.034
Comorbid history of MI	1.36	1.25-1.48	<.001
Comorbid HTN	.83	.81-.85	<.001
Comorbid COPD	1.04	1.00-1.08	.082
Comorbid steroid use	.96	.85-1.08	.472
Comorbid cirrhosis	2.80	2.61-3.02	<.001
Comorbid dementia	.97	.93-1.02	.217
Comorbid psychiatric illness	.65	.62-.68	<.001

OR = odds ratio; CI = confidence interval; AIS = abbreviated injury scale; ISS = injury severity score; GCS = Glasgow coma scale; CVA = cerebrovascular accident; DM = diabetes mellitus; MI = myocardial infarction; HTN = hypertension; COPD = chronic obstructive pulmonary disease.

Table 6.

Multivariable Analysis for Risk of Mortality in Trauma.

Risk of Mortality	OR	CI	P-Value
Prisoners vs non-prisoners	1.61	.52-4.94	.409
Severe AIS-head	1.59	1.51-1.68	<.001
Severe AIS-thorax	.80	.74-.86	<.001
Severe AIS-abdomen	1.23	1.12-1.36	<.001
Hypotensive on admission	2.65	2.63-3.05	<.001
Age	1.04	1.04-1.04	<.001
ISS	1.07	1.07-1.07	<.001
Pulse rate	1.00	1.00-1.00	<.001
Respiratory rate	1.01	1.01-1.01	<.001
Total GCS	.76	.76-.77	<.001
Comorbid CVA	1.31	1.17-1.47	<.001
Comorbid DM	1.29	1.21-1.38	<.001
Comorbid history of MI	1.45	1.23-1.70	<.001
Comorbid HTN	.93	.88-.98	.012
Comorbid COPD	1.42	1.31-1.53	<.001
Comorbid steroid use	1.32	1.05-1.67	.019
Comorbid cirrhosis	3.77	3.26-4.36	<.001
Comorbid dementia	1.13	1.02-1.24	<.001
Comorbid psychiatric illness	.71	.66-.77	<.001

Discussion

In this large national TQIP analysis, we found that trauma patients presenting from a prison make up <.2% of all trauma patients and are more often younger, male minorities, and had fewer chronic medical comorbidities. In addition, we reported a similar rate and associated risk of mortality for trauma prisoners compared to non-prisoner trauma patients. Interestingly, our data demonstrated prisoners had lower rates of certain in-hospital complications, such as drug/alcohol withdrawal, PNA, and UTI compared to non-prisoners. Consistent with these findings, prisoners experienced decreased hospital LOS compared to non-prisoners. To our knowledge, this is the first paper that directly compares prisoner trauma patient mortality and non-prisoner trauma patient mortality using a large national population.

Prisoners represent a vulnerable population due to their limited autonomy and the social stigma associated with their incarceration.^2,18 Previous studies have shown differing perceptions between the quality of treatment received by prisoners when compared to non-prisoners.^19-21 Zust et al¹⁸ found that nurses’ preconceived notions of inmates often negatively contributed to their ability to empathize with and care for their patients.¹⁸ Prisoners may also be at increased risk of traumatic injury due to poor prison conditions. For example, Haber et al discussed how shackling can lead to an inability to break falls, increases the risk of thrombosis due to movement limitations, and has a negative association with empathy amongst many care providers.² However, despite potential preconceived inadequacies regarding the quality of care of prison patients, our national TQIP analysis found that prisoner trauma patients had no difference in rate and associated risk of mortality compared to their non-prisoner counterparts. Although not specific to trauma patients, multiple studies involving non-surgical prison patients demonstrated similar outcomes between prisoner and non-prisoner medical patients.^22,23 Winter et al found no difference in mortality between inmates and non-inmates treated acutely for chest pain, and in fact demonstrated shorter time to first procedure for inmates (.46 days earlier).²² Thus, it appears while qualitative metrics may have disparities, care in terms of objective outcomes is at least similar for prisoner patients.

Hospital LOS is an important metric both for quality of care, as well as associated healthcare cost. When treating trauma patients, each additional day of excessive hospital stay is associated with a 5% increase in risk of developing independent hospital-associated medical complications.²⁴ Furthermore, the estimated cost of each day of hospitalization ranges from an average of $645-$3500 depending on state and ownership of the hospital.²⁵ Our study found a decreased LOS for prisoners compared to non-prisoners. In congruence with our findings, previous studies have also demonstrated decreased LOS for prisoner populations.^2,26 Although not specific to trauma patients, Haber et al found a significantly decreased LOS for incarcerated vs non-incarcerated patients (4 vs 5.6 days), with the largest sub-group of these patients (17.6%) being admitted for traumatic injuries.² One plausible explanation lies in the fact that the non-prisoner population often extends hospitalizations due to issues regarding discharge planning. For instance, Hwabejire et al found that only 20% of non-prisoner trauma patient discharge delays were clinical.²⁷ Looking at the remaining 80, 47% resulted because of difficulties in placement and 7% occurred due to payer-related issues. Given that prisoners have a clear discharge disposition, this may help explain their decreased LOS. Alternatively, a decreased rate of complications may be a driving factor for this discrepancy.

In support of this, we reported prisoners had significantly less drug/alcohol withdrawal, PNA, and UTIs. The decreased drug/alcohol withdrawal within the prisoner population is in spite of prior reports citing that prisoners continue to have access to, and use substances while in prison.^1,28 Although we think it is more difficult to obtain a sufficient and steady supply of drugs in prison that would allow someone to withdrawal from these substances (ie, alcohol and opioids), we are unable to quantify this statement. However, non-prisoners had nearly double the rate of DM, which could help explain the increased rate for infections in the non-prisoner trauma patient population, as DM is a well-known risk factor for infection as demonstrated by Knowlin et al who demonstrated a higher incidence in both PNA (9.2 vs 6.3%) and UTIs (11 vs 3%) among diabetic trauma burn patients compared to non-diabetic trauma burn patients.^29-32 Additionally, higher rates of COPD in the non-prisoner population may predispose them to a higher risk for PNA at baseline. Future prospective research is needed to elucidate the exact reasons for prisoners’ decreased rate of complications; however, it is reassuring that this vulnerable population appears to have fewer complications compared to the general trauma population.

Our study is susceptible to numerous limitations, such as those associated with performing a retrospective database study including missing data, reporting bias, and coding errors. Also, the populations compared in this study had numerous differences that we attempted to account for in our multivariable analysis; however, there may be other confounders that are missing within TQIP such as time to presentation from the prison or home-setting, baseline functional status, and baseline laboratory evaluations in the hospital. The database also does not allow us to identify the reason for and duration of incarceration, which may skew our data, as a prisoner for one day may have a different baseline medical status and outcomes than a prisoner of many years. Due to the fact that TQIP only includes verified trauma centers, some hospitals are not included. Injuries that are evaluated by an emergency medicine physician but not severe enough to warrant a trauma team evaluation are not included in the TQIP data set. Additionally, we assumed that traumas occurring in prison predominantly involve prisoners; however, our data simply identify the location of injury as a prison but does not specify whether the patient was a prisoner or non-prisoner (ie, correctional officer or visitor). Furthermore, we are unable to determine if any of the non-prisoners were previously prisoners, and we have an inability to account for socioeconomic differences in the populations. Since TQIP is limited to the index hospitalization, we also cannot evaluate any post-discharge and/or long-term outcome differences. Finally, while this study evaluates quantitative outcomes, qualitative outcomes cannot be measured and represent an area in need of future research. These include metrics such as patient expectations and experience which could be measured via interviews and/or focus groups as well as validated surveys such as the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) surveys. Despite these limitations, we believe this study will help foster further high-quality prospective multicenter research on this vulnerable population.

To our knowledge, this is the first large national analysis comparing outcomes of trauma patients presenting from prisons to non-prisoners. We reported that prisoners make up less than .2% of trauma patients. Supporting our hypothesis, we found no differences in the rate or associated risk for mortality between trauma patients from prisons compared to non-prisoners. Interestingly, prisoners had a shorter hospital LOS, as well as fewer complications. Future prospective studies are needed to confirm these results and explore additional areas of potential disparities such as qualitative outcomes.

Supplemental Material

sj-pdf-1-asu-10.1177_00031348221078984 – Supplemental Material for Comparative Outcomes for Trauma Patients in Prison and the General Population

Supplemental Material, sj-pdf-1-asu-10.1177_00031348221078984 for Comparative Outcomes for Trauma Patients in Prison and the General Population by AB Christian, A Grigorian, J Mo, EO Yeates, M, Dolich, TL Chin, SD Schubl, CM Kuza, M Lekawa and J Nahmias in The American Surgeon

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Bronson

Carson

. Prisoners in 2017. BJS Statisticians. Available at: https://www.bjs.gov/content/pub/pdf/p17.pdf https://www.bjs.gov/content/pub/pdf/p17.pdf (accessed September 15, 2020).

Haber

Erickson

Ranji

Ortiz

Pratt

. Acute care for patients who are incarcerated: A review. JAMA Intern Med. 2019;16. doi:10.1001/jamainternmed.2019.3881.

Schüklenk

. Protecting the vulnerable: Testing times for clinical research ethics. Soc Sci Med. 2000;51(6):969-977. doi:10.1016/s0277-9536(00)00075-7.

Nelkin

. The nazi doctors and the nuremberg code: Human rights in human experimentation. JAMA. 1993;269(9):1168-1169. doi:10.1001/jama.1993.03500090104044.

Hornblum

. They were cheap and available: Prisoners as research subjects in twentieth century America. BMJ. 1997;315(7120):1437-1441. doi:10.1136/bmj.315.7120.1437.

Gostin

. Biomedical research involving prisoners: Ethical values and legal regulation. JAMA. 2007;297(7):737-740. doi:10.1001/jama.297.7.737.

Department of Health, Education, and Welfare . National commission for the protection of human subjects of biomedical and behavioral research. The belmont report. Ethical principles and guidelines for the protection of human subjects of research. J Am Coll Dent. 2014;81(3):4-13.

Legal Information Institute. 28 . CFR Part 512. https://www.law.cornell.edu/cfr/text/28/part-512. (accessed September 15, 2020).

US Department of Health and Human Services . Protection of human subjects. 2009. https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/index.html.

10.

Murphy

. Health care in the federal bureau of prisons. Calif J Health Promot. 2005;3(2):23-37. doi:10.32398/cjhp.v3i2.1761.

11.

Olson

. Health care delivery in the texas prison system. JAMA. 2004;292(18):2212-2213. doi:10.1001/jama.292.18.2212-b.

12.

Centers for Disease Control and Prevention (CDC) . Injury surveillance in correctional facilities--Michigan, April 1994-March 1995. MMWR Morb Mortal Wkly Rep. 1996;45(3):69-72.

13.

Henning

Frangos

Simon

Pachter

Bholat

. Patterns of traumatic injury in New York City prisoners requiring hospital admission. J Correct Health Care. 2015;21(1):53-58. doi:10.1177/1078345814558046.

14.

Thorburn

. Injury monitoring in US prison systems. JAMA. 1999;282(5):430-431. doi:10.1001/jama.282.5.430.

15.

Shin

Pasechnikov

Gelmanova

, et al. Treatment outcomes in an integrated civilian and prison MDR-TB treatment program in Russia. Int J Tubercul Lung Dis. 2006;10(4):402-408.

16.

Patterson

. Incarcerating death: Mortality in US state correctional facilities, 1985-1998. Demography. 2010;47(3):587-607.

17.

Spaulding

Seals

McCallum

Perez

Brzozowski

Steenland

. Prisoner survival inside and outside of the institution: Implications for health-care planning. Am J Epidemiol. 2011;173:479-487. doi:10.1093/aje/kwq422.

18.

Zust

Busiahn

Janisch

. Nurses’ experiences caring for incarcerated patients in a perinatal unit. Issues Ment Health Nurs. 2013;34(1):25-29. doi:10.3109/01612840.2012.715234.

19.

Sarpong

Otupiri

Yeboah-Awudzi

Osei-Yeboah

Berchie

Ephraim

. An assessment of female prisoners’ perception of the accessibility of quality healthcare: A survey in the kumasi central prisons, ghana. Ann Med Health Sci Res. 2015;5(3):179-184. doi:10.4103/2141-9248.157495.

20.

Feron

Tan

Pestiaux

Lorant

. High and variable use of primary care in prison. A qualitative study to understand help-seeking behaviour. Int J Prison Health. 2008;4(3):146-155. doi:10.1080/17449200802264696.

21.

Shaw

Elger

. Improving public health by respecting autonomy: Using social science research to enfranchise vulnerable prison populations. Prev Med. 2015;74:21-23. doi:10.1016/j.ypmed.2015.01.024.

22.

Winter

. A comparison of acuity and treatment measures of inmate and noninmate hospital patients with a diagnosis of either heart disease or chest pain. J Natl Med Assoc. 2011;103(2):109-115. doi:10.1016/s0027-9684(15)30259-5.

23.

Sterling

Hofmann

Luketic

, et al.

Treatment of chronic hepatitis C virus in the virginia department of corrections: Can compliance overcome racial differences to response?

Am J Gastroenterol. 2004;99(5):866-872. doi:10.1111/j.1572-0241.2004.30310.x.

24.

Gotlib Conn

Zwaiman

DasGupta

Hales

Watamaniuk

Nathens

. Trauma patient discharge and care transition experiences: Identifying opportunities for quality improvement in trauma centres. Injury. 2018;49(1):97-103. doi:10.1016/j.injury.2017.09.028.

25.

Becker’s Hospital Review . Average Hospital Expenses Per Inpatient Day across 50 States. https://www.beckershospitalreview.com/finance/average-hospital-expenses-per-inpatient-day-across-50-states.html. (accessed September 15, 2020).

26.

Rothman

McConville

Hsia

Metzger

Ahalt

Williams

. Differences between incarcerated and non-incarcerated patients who die in community hospitals highlight the need for palliative care services for seriously ill prisoners in correctional facilities and in community hospitals: A cross-sectional study. Palliat Med. 2018;32(1):17-22. doi:10.1177/0269216317731547.

27.

Hwabejire

Kaafarani

Imam

, et al. Excessively long hospital stays after trauma are not related to the severity of illness: Let’s aim to the right target!. JAMA Surg. 2013;148(10):956-961. doi:10.1001/jamasurg.2013.2148.

28.

Rao

Mandal

Gupta

, et al. Factors affecting drug use during incarceration: A cross-sectional study of opioid-dependent persons from india. J Subst Abuse Treat. 2016;61:13-17. doi:10.1016/j.jsat.2015.08.009.

29.

Gortler

Rusyn

Godbout

Chahal

Schemitsch

Nauth

. Diabetes and healing outcomes in lower extremity fractures: A systematic review. Injury. 2018;49(2):177-183. doi:10.1016/j.injury.2017.11.006.

30.

Knowlin

Strassle

Williams

, et al. Burn injury outcomes in patients with pre-existing diabetic mellitus: Risk of hospital-acquired infections and inpatient mortality. Burns. 2018;44(2):272-279. doi:10.1016/j.burns.2017.09.022.

31.

Memmel

Kowal-Vern

Latenser

. Infections in diabetic burn patients. Diabetes Care. 2004;27(1):229-233. doi:10.2337/diacare.27.1.229.

32.

Peleg

Weerarathna

McCarthy

Davis

. Common infections in diabetes: pathogenesis, management and relationship to glycaemic control. Diabetes Metab Res Rev. 2007;23(1):3-13. doi:10.1002/dmrr.682.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.28 MB