Abstract
Objective
To evaluate the potential influence of acute intoxication and dependence to alcohol on extra-utilization of health care resources by ethnic minority trauma patients in a level I trauma center.
Methods
We analyzed the data of 1,493 patients enrolled in a study that evaluated the effectiveness of brief alcohol intervention among ethnic minority trauma patients. The database included detailed demographic, injury-related and drinking-related characteristics (including acute intoxication and alcohol dependency status). Patients were categorized into the following groups: non-intoxicate/non-dependent (NI/ND), non-intoxicated/dependent (NI/D), intoxicated/non-dependent (I/ND) and intoxicated/dependent (I/D). We compared utilization of several diagnostic and therapeutic procedures among these four categories. We placed special emphasis on ethnicity as a potential effect modifier.
Results
Relative to NI/ND trauma patients, I/ND patients (relative risk (RR): 1.8, 95% CI: 1.2-2.8) and I/D patients (RR: 2.4, 95% CI: 1.6-3.6) had significantly higher chance of being evaluated by abdominal ultrasound during the first 24 hours of hospital arrival. Similar pattern was observed for head CT scan (with the corresponding RRs of 2.1 and 2.6, respectively). Chance of admission to the intensive care unit (ICU) was not associated with intoxication/dependence status. Length of hospital stay was negatively associated with drinking status with the shortest length of stay for I/D. Including ethnicity in the models, did not change the results and conclusions.
Discussion
Acute intoxication and dependence to alcohol are both associated with more frequent utilization of selected health care resources and the utilization pattern was not associated with patient ethnicity. This emphasizes on the importance of routine screening for drinking problems among all trauma patients, regardless of their blood alcohol level in the ED.
Introduction
Studies have consistently demonstrated that risky drinking 33 is strongly associated with a higher frequency of emergency department (ED) visits and hospitalization 10, 12, 29, 30, 32. Alcohol is involved in approximately half of all trauma admissions 17, 34, 36, 40, 41 in Level 1 and 2 trauma centers. Yet, studies have seldom evaluated the potential influence of acute intoxication and dependence to alcohol on over-utilization of health care resources by trauma patients.
Acute intoxication and medical evaluations in the ED
Acute intoxication often complicates the medical work-up required to rule out a potential undetected injury 32. Often times, the patient either does not remember or is not able to give a good medical history or cooperate with the trauma team during medical examination 32. Therefore, physicians cannot always rely on the patient's response to medical evaluations. As a result, even in a non-complicated injury, physicians often utilize more advanced techniques in order to rule out the presence of potentially unidentified injuries 32.
In addition to an increased number of diagnostic procedures conducted at the time of ED arrival, monitoring response to treatment is also more challenging. Even in the absence of abnormal findings in primary medical evaluations, intoxicated patients are often hospitalized for 24 hour observation and they might have a higher chance of being admitted to the ICU 32. Moreover, for those under observation in surgical wards or ICUs, medical teams often cannot rely on serial physical exams for detecting covert injuries. As a result, more advanced techniques such as serial CT scans or more invasive procedures such as insertion of an intracranial pressure monitor might be required 32.
Alcohol dependence and trauma care
Studies have consistently demonstrated that multiple organ damages due to long term alcohol overconsumption, the pattern that is frequently seen among alcohol dependents, could negatively influence body's response to injury 22. In addition, the sudden cessation of alcohol consumption might predispose alcohol dependents to an alcohol withdrawal syndrome (AWS) or even delirium tremens 6, 23 which can complicate the treatment process. Unfortunately, in spite of the high prevalence of AWS among trauma patients 27, 44 a significant proportion of the patients are undiagnosed 8. If alcohol use disorders are not diagnosed, the patient might go through unnecessary and costly evaluations that might postpone timely management of the disease 27, 43, 44. Even when alcohol dependence is diagnosed, the patient might require extra care to cope with the signs and symptoms of alcohol withdrawal 6, 8. This often requires longer length of hospital stay 27, 44 which further predisposes the patient to complications such as hospital acquired pneumonia 6, 7, 14, 15.
Ethnicity and utilization of health care resources
Several studies have demonstrated that ethnicity is significantly associated with disparities in receiving health care in the United States 1, 4, 11. However, while injuries are the leading cause of mortality and morbidity in individuals younger than 45 years old 16, few studies have focused on health care disparities among trauma patients. These few studies have mainly examined traumatic brain injuries, because of their high mortality and substantial morbidity 3, 9, 20, 21, 28, 42 and have often considered utilization of rehabilitation services as the main outcome of interest 42. As a result of this approach, potential differences in utilization of resources during hospitalization have not received the attention they deserve. In addition, there is no study that has evaluated the influence of acute or chronic alcohol problems on utilization of resources by hospitalized ethnic minority trauma patients.
In this study, we evaluated if high blood alcohol concentration (BAC) at the time of hospital arrival and dependence to alcohol, are associated with higher utilization of health care resources for Whites, Hispanics and Blacks in an urban level I trauma center.
Methods
The methodology in this trial has been described in detail elsewhere 38. Briefly, the “Multidisciplinary Approach to Reduce Injury and Alcohol Use (MARIA)” was designed to evaluate the effectiveness of a brief intervention (BI) in reducing alcohol consumption and injury recidivism among Whites, Blacks and Hispanics admitted to a Level I trauma center. Since the main objective of the study was comparison of effectiveness of BI among minorities, Blacks and Hispanics were oversampled.
In this trial, trained research assistants screened trauma patients based on the following four criteria: a) BAC level in the ED, b) self-reported drinking within 6 hours prior to injury, c) positive on one or more items of the CAGE questionnaire in the last year 13 and d) drinking pattern beyond drinking limits guidelines (four drinks per occasion for women and five for men) 35. CAGE is a four item questionnaire that asks patients about their feeling in regards to: the necessity of cutting (C in CAGE) the number of drinks; getting annoyed (A in CAGE) because of other people's criticism of their drinking behavior; guilt (G in CAGE) because of inappropriate dinking; drinking alcohol early in the morning as an eye opener (E in CAGE) 13. A total of 1,493 trauma patients were included in the study. The baseline interview collected comprehensive data on demographic characteristics, risk perception and history of risky behaviors, history of alcohol and non-alcohol related injuries, alcohol consumption and physical, social and legal problems associated with alcohol or drug use, including indicators of alcohol or drug abuse and dependence.
Ethnic identification: This was based on subjects' responses to four questions. First participants were asked if they were Hispanic, Spanish or Latino. Those who answered yes were asked about their original Spanish-speaking country. The third question asked about the race (White, Black, Asian, Native Hawaiian or Pacific Islander, American Indian or Alaska Native) of the participant. If more than one race was mentioned, participants were then asked about the race group that described them the best.
Alcohol dependence status was assessed at baseline using the Composite International Diagnostic Interview (CIDI)2. Drug use and dependence was assessed at baseline using the World Health Organizations Composite International Diagnostic Interview Short Form (CIDI-SF) 26. Identification of the mechanism and intentional nature of the injury was based on subjects' self-report about the way the injury occurred. Severity of injury was measured using the Injury Severity Score (ISS) 5.
Based on the BAC in the ED and dependence status, we categorized patients into four categories: non-intoxicated (BAC<0.08%) /non-dependent; non-intoxicated/dependent; intoxicated (BAC≥0.08%)/non-dependent; intoxicated/dependent.
Identification of the utilized resources was based on the reported International Classification of Disease (ICD-9) procedure codes reflected in the trauma registry. Since many different procedures could be reported for each individual, we decided to focus on the utilization of the following resources, which were most commonly reported for the MARIA patients based on our preliminary analysis of the data: abdominal ultrasonography (88.76); CT scan of head (87.03), abdomen (87.41) and thorax (88.01); urinary catheter (59.8) and endotracheal intubation (96.04).
Our preliminary analyses of the data demonstrated that on average less than 3% of the patients received a procedure more than two times. Therefore, for each procedure code, we created a dummy variable that was equal to “1” if a patient had utilized the resource ≥ 1 time; “0” otherwise. To evaluate if the risk of ICU admission was associated with intoxication/dependence status, we created a binary variable that was equal to “1” if a patient was admitted to the ICU; “0” otherwise.
Often, the sedative influences of alcohol disappear up to 24 hours after consumption. Therefore, we did not expect any significant difference in resource utilization because of acute intoxication after the first 24 hours of hospital arrival. However, patients with dependence to alcohol might experience signs and symptoms of alcohol withdrawal several hours after admission, because of lack of access to alcohol. As a result, we hypothesized that resource utilization in these patients would increase after the first 24 hours of hospital arrival. To evaluate these hypotheses, we created two sets of outcome variables that addressed resource utilization during or after the first 24 hours.
Finally, we added two continuous outcomes, namely length of ICU and hospital stay, to the list of the outcomes of interest evaluated in this study.
Data analysis approach
To evaluate the association between the exposures of interest and the binary outcomes before and after the 24 hour cutoff, we adopted a logistic regression approach. However, it has been demonstrated that in case-control or cross sectional studies with common outcomes (more than 10% frequency), the odds ratio (OR) might not be a good representative for the actual relative risk 18, 31, 37. Since some outcomes such as abdominal ultrasound and head CT scan were commonly reported for the patients in MARIA project, we used log-binomial regression model to directly calculate the risk ratios (RRs), rather than using logistic regression model and calculating ORs 18, 31, 37. We adjusted our estimates for the following variables: age, sex, ethnicity, insurance status, mechanism and severity of injury, and drug use/dependence status. Since one of the hypotheses of the study was evaluation of the modification effect of ethnicity on the association between intoxication/dependence status and resource utilization, we repeated the same analysis mentioned above, but added the interaction term between the exposure and ethnicity to the model.
Finally, we used linear regression models to evaluate the association between intoxication/dependence status and length of hospital or ICU stay (continuous outcomes) and adjusted our estimates for the same covariates mentioned above. However, since length of hospital and ICU stay were skewed, we log transformed these two variables. By adding the interaction term between intoxication/dependence status and ethnicity, we also evaluated the potential modification effect of ethnicity.
Results
Demographic characteristics of the study population
Whites, Hispanics and Blacks comprised 45%, 36% and 19% of the patients, respectively. Table 1 summarizes main demographic and alcohol-related characteristics of the patients for each ethnic group. In general, Hispanics were younger (mean age 29 years old) than Whites or Blacks and a higher proportion of them were male (89% vs. 78% among Whites and 81% among Blacks). Hispanics also had less education. While approximately 80% of the Whites and Blacks had at least some high school education, 65% of the Hispanics had less than high school education.
Table 1.
Distribution of the demographic and alcohol/drug-related characteristics of the trauma patients enrolled in the MARIA project *
| White (N=668) |
Hispanic (N=537) |
Black (N=288) |
|
|---|---|---|---|
| Mean age (years) * | 35 | 29 | 38 |
| Male (%)* | 78 | 89 | 81 |
| Marital status (%)* | |||
| Single, never married | 43 | 47 | 47 |
| Married, living with lifetime partner | 27 | 33 | 24 |
| Separated or not living with spouse | 31 | 19 | 29 |
| Education status (%)* | |||
| Less than high school | 23 | 65 | 26 |
| Some high school | 37 | 22 | 52 |
| More than High school | 39 | 13 | 21 |
| Insured (%)* | 46 | 24 | 33 |
| Alcohol consumption pattern (%)* | |||
| Non-intoxicated/non-dependent | 36 | 28 | 36 |
| Non-intoxicated/dependent | 13 | 17 | 16 |
| Intoxicated/non-dependent | 24 | 24 | 24 |
| Intoxicated/dependent | 27 | 31 | 24 |
| Drug consumption pattern (%)* | |||
| Neither user nor dependent | 50 | 61 | 49 |
| User, but not dependent | 37 | 30 | 40 |
| Dependent | 13 | 10 | 11 |
| Mechanism of injury (%)* | |||
| Motor Vehicle Collision | 40 | 38 | 32 |
| Assault | 6 | 12 | 11 |
| Falls | 20 | 14 | 18 |
| Motorcycle collision | 14 | 4 | 3 |
| Gunshot wounds | 3 | 9 | 9 |
| Stab Wound | 3 | 6 | 11 |
| Others | 14 | 17 | 16 |
p value for all comparisons <0.05
Table 1 also summarizes alcohol intoxication/dependence and drug use/dependence status for each ethnic group. As presented, a smaller proportion of the Hispanics were non-intoxicated/non-dependent (28%) in comparison to Whites (36%) or Blacks (36%). In addition, being intoxicated/dependent was more common among Hispanics, relative to other ethnic groups (31% vs. 27% in Whites or 24% in Blacks) (p<0.05).
Most commonly utilized resources
On average, 4.9 (SD: 5.6) ICD-9 procedure codes were recorded for non-intoxicated/non-dependent patients, 5.2 (SD: 6.0) for non-intoxicated/dependent patients, 5.7 (SD: 6.3) for intoxicated/non-dependent patients and 5.6 (SD: 6.5) for intoxicated/dependent. Table 2 demonstrates that among Hispanics significantly higher number of ICD-9 procedure codes were reported for intoxicated/dependent patients, relative to non-intoxicated/non-dependent individuals (β= 1.95, 95% CI: 0.57- 3.33, p=0.006), even after adjustment for age, gender, insurance status, injury severity, ED assessment, mechanism of injury, and drug use.
Table 2.
Linear regression analysis evaluating the association between the number of reported ICD-9 procedure codes and alcohol intoxication/dependence status *
| Regression coefficient (95% CI) | |||
|---|---|---|---|
| White (N=668) |
Hispanic (N=537) |
Black (N=288) |
|
| Non-intoxicated/non-dependent | Reference | Reference | Reference |
| Non-intoxicated/dependent | 1.00 (-0.58, 2.58) |
1.37 (-0.26, 3.01) |
.76 (-0.93, 2.44) |
| Intoxicated/non-dependent | 0.50 (-0.70, 1.71) |
1.18 (-0.31, 2.67) |
0.55 (-0.91, 2.00) |
| Intoxicated/dependent | .62 (-0.57, 1.80) |
1.95 (0.58, 3.33) |
.58 (-0.81, 1.98) |
Adjusted for age, gender, insurance status, injury severity, ED assessment, mechanism of injury, and drug use
Table 3 summarizes the frequency of the most commonly reported procedures for each ethnic group during the first 24 hours of hospital arrival and after that. As presented, during the first 24 hours, the use of abdominal ultrasound, CT scan of head, abdomen, and thorax and urinary catheter were significantly more common among White patients relative to Hispanics and Blacks. However, such associations were only observed for head CT-scan after the first 24 hours of hospital arrival (Table 3).
Table 3.
Proportion (%) of the MARIA patients who utilized each resource at least one time
| Whites | Hispanics | Blacks | |
|---|---|---|---|
| First 24 hours | |||
| Abdomen Ultrasound * | 59.6 | 48.2 | 37.9 |
| Head CT scan * | 50.2 | 39.1 | 29.5 |
| Abdomen CT scan * | 20.5 | 16.0 | 11.5 |
| Thorax CT scan * | 11.2 | 8.2 | 2.8 |
| Urinary Catheter * | 22.9 | 21.4 | 14.6 |
| Endotracheal Intubation | 14.1 | 16.4 | 11.8 |
| After 24 hours | |||
| Abdomen Ultrasound | 2.4 | 1.1 | 1.4 |
| Head CT scan * | 8.4 | 4.8 | 2.1 |
| Abdomen CT scan | 4.0 | 2.2 | 2.1 |
| Thorax CT scan | 3.4 | 1.7 | 2.1 |
| Urinary Catheter | 1.7 | 1.5 | 1.0 |
| Endotracheal Intubation | 5.4 | 4.3 | 3.1 |
Chi square p value comparing different ethnic groups < 0.01
Resource utilization during the first 24 hours of hospital arrival
In order to evaluate if intoxication/dependence status was associated with the utilization of each resource mentioned in Table 3, we conducted six log binomial analyses. The six models considered use (vs. lack of use) of each resource during the first 24 hours of hospital arrival as the binary outcome of interest, and alcohol intoxication/dependence status as the main exposure. In these models (Table 4) we adjusted our estimates for age, gender, ethnicity, insurance status, type and mechanism of injury, injury severity score, and drug use/dependence status. Intoxicated/non-dependent patients (RR: 1.79, 95% CI: 1.15-2.80) and intoxicated/dependent patients (RR: 2.41, 95% CI: 1.60-3.60) had significantly higher chance of being evaluated by abdominal ultrasound during the first 24 hours of hospital arrival, relative to non-intoxicated/non-dependent trauma patients (Table 4). While mechanism and severity of injury were associated with the outcome, ethnicity, insurance status and drug use/dependence were not. Similar patterns were observed for head CT scan and urinary tract catheterization during the first 24 hours of ED arrival. Alcohol intoxication/dependence status was not significantly associated with other outcomes in this analysis.
Table 4.
Results of the log binomial regression analysis evaluating the association between acute and chronic drinking problems and utilized resources during the first 24 hours of ED arrival *
| Risk Ratio (95% Confidence interval) | ||||||
|---|---|---|---|---|---|---|
| Abdomen US | Head CT scan | Abdomen CT scan | Thorax CT scan | Endotracheal Intubation | Urinary Catheter | |
| Non-intoxicated/non-dependent | Reference | Reference | Reference | Reference | Reference | Reference |
| Non-intoxicated/dependent | 1.37 (0.83-2.24) |
1.35 (0.84-2.17) |
1.04 (0.60-1.81) |
0.67 (0.28-1.58) |
1.42 (0.76-2.64) |
1.41 (0.67-2.31) |
| Intoxicated/non-dependent | 1.79 (1.15-2.80) |
2.11 (1.35-3.28) |
0.96 (0.61-1.51) |
1.14 (0.57-2.26) |
0.96 (0.57-1.54) |
1.61 (1.12-2.41) |
| Intoxicated/dependent | 2.41 (1.60-3.60) |
2.64 (1.74-4.00) |
1.07 (0.69-1.67) |
1.18 (0.67-2.07) |
1.49 (0.94-2.37) |
1.77 (1.22-2.73) |
Adjusted for age, gender, ethnicity, mechanism and severity of injury, insurance status and drug use/dependence status
The models with interaction term between intoxication/dependence and ethnicity did not demonstrate any significant modification effect for ethnicity (results are not presented).
Resource utilization after the first 24 hours of hospital arrival
There was no statistically significant association between alcohol intoxication/dependence status and resource utilization after the first 24 hour of hospital arrival.
ICU admission
Using the same log-binomial analysis approach mentioned above and after adjustment for the same variables, we found that ICU admission was not significantly associated with intoxication/dependence status. In addition, we were not able to detect any significant modification effect for ethnicity.
Length of ICU stay
Results of a linear regression analysis adjusted for age, gender, ethnicity, insurance status, mechanism of injury, ISS and drug use/disorder status demonstrated that log transformed length of ICU stay, for patients who were admitted to ICU, was not significantly associated with intoxication/dependence status (Table 5). The model with the interaction term for intoxication/dependence and ethnicity did not demonstrate a significant modification effect for ethnicity.
Table 5.
Linear regression analysis, evaluating the association between intoxication/dependence status and length of ICU and hospital stay *
| Regression coefficient (95% CI) | ||
|---|---|---|
| Log length of ICU stay** | Log length of Hospital stay | |
| Non-intoxicated/non-dependent | Reference | Reference |
| Non-intoxicated/dependent | 0.28 (-0.9, 0.67) | -0.12 (-0.25, 0.01) |
| Intoxicated/non-dependent | 0.08 (-0.25, 0.41) | -0.13 (-0.24, -0.01) 1 |
| Intoxicated/dependent | 0.20 (-0.14, 0.55) | -0.21 (-0.31, -0.11) 2 |
Adjusted for age, gender, ethnicity, mechanism and severity of injury, insurance status and drug use/dependence status
Only patients admitted to ICU are included in the analysis (n=348)
p < 0.05,
p < 0.001
Length of hospital stay
Intoxicated/non-dependent patients and intoxicated/dependent patients had significantly shorter log transformed length of hospital stay, relative to non-intoxicated/non-dependent individuals (Table 5). Similar to length of ICU stay, length of hospital stay was not significantly influenced by modification effect of ethnicity either.
Discussion
Our study demonstrated that intoxication/dependence status was significantly associated with higher chance of utilizing abdominal ultrasound, head CT scan and urinary catheter during the first 24 hours of hospital arrival. In addition, significantly higher number of ICD-9 procedure codes was reported for Hispanics in comparison to Whites or Blacks, although utilization of the most commonly used procedures was not different among different ethnic groups.
Before discussing the main findings, we would like to review the limitations of the study. First, since patients should be able to respond to the study questionnaire, seriously injured patients and especially patients with severe traumatic brain injury were not included in the study. Although these patients generally comprise a small proportion of injured patients, they consume significant proportion of the utilized resources 16. Second, in spite of the large number of the subjects in this study, patients were recruited from a large urban level 1 trauma center. Therefore, generalizability of the results to smaller hospitals might be limited.
In spite of these limitations, our study had several strengths. First, we evaluated the influence of drinking problems beyond alcohol intoxication, on utilized resources in a level I trauma center. A few previous studies that have addressed the issue, have only focused on the potential influence of acute intoxication on resource utilization19, 24, 25. For example, Jurkovich and colleagues reported that intoxicated patients were more likely to be intubated in the field or emergency department (relative risk (RR) = 1.3, 95% (CI)= 1.1-1.5) or require intracranial pressure monitor (RR = 1.4, 95% CI = 1.1-1.8) 25 as compared to non-intoxicated patients. Lack of association between intoxication status and these procedures in our study could be partially explained by lower severity of injury in our study.
In addition, while Jurkovich and colleagues wisely compared the potential influence of acute intoxication on early (first 24 hours of hospital arrival) and late trauma fatality 24, none of the previous studies have evaluated utilization of the resources during these time periods. This approach gave us a unique opportunity to differentiate between potential over utilization of resources mainly because of acute intoxication or alcohol dependence.
The first significant finding of the study was the association between intoxication/dependence and the number of the reported ICD-9 procedure codes for Hispanics and not for other ethnic groups. Lack of access to underlying health conditions of the Hispanic patients does not allow us to attribute the larger number of the reported ICD-9 codes to the more frequent co-existing health problems in this ethnic group. However, a potential language barrier between patients and physicians might play a role. In other words, potential communication problems encourage physicians to more frequently utilize procedures that would have not been used if the language barrier had not existed.
Another significant finding of the study was the association between intoxication and dependence status and higher utilization rate of abdominal ultrasound, head CT and urinary tract catheter during the first 24 hours of hospital arrival. In addition, intoxication and dependence had distinct influence on utilization rate. In other words, intoxicated/dependent patients had the highest chance for utilizing these resources, followed by intoxicated/non-dependents and non-intoxicated/dependent patients. We didn't observe any significant association between the exposures of interest and utilization of other resources and especially not after the 24 hours. Although the small number of the subjects that have utilized other resources might have restricted us to detect small but significant differences among exposure groups, the RRs very close to one and the CIs that were not very wide weaken this potential explanation. Mechanism of injury (e.g. motor vehicle collision, violence), the injured body organ (e.g. head injury), availability and feasibility of use of a service such as abdominal ultrasound and the potential influence of acute or chronic alcohol consumption on the body response to injury are probably among the most important factors that determine the use or frequency of use of different resources. Future studies should focus on the relative importance of these variables on resource utilization.
We also found that intoxication/dependence status was not associated with the probability of being admitted in the ICU. Unfortunately, inadequacies of the existing data limit our ability to interpret these findings. However, one potential explanation is that studies have demonstrated that severity of injury is one of the most important predictors for ICU admission among injured patients (57). In other words, ISS is in the causal pathway between an injury and chance of ICU admission. Therefore, adjustment for ISS decreases the strength of the positive association between the injury and ICU admission.
In regards to the length of hospital stay, in general, intoxicated patients have a higher chance of being hospitalized than non-intoxicated patients 32. In other words, a patient could be treated and released from the ED had s/he not been under the influence of alcohol. As a result, as soon as the sedative influences of alcohol disappear, the patient gets discharged. This issue might show itself as a negative association between length of hospital stay and intoxication status. Another potential explanation for shorter length of hospital stay among alcohol dependents could be the excessive desire of these patients to consume alcohol while they are hospitalized. Since hospitals do not provide alcohol and patients experience the undesirable consequences of alcohol withdrawal, they prefer to leave hospital even against medical advice. In addition, physicians might be more inclined to discharge alcohol dependents as soon as their medical conditions allow to prevent potential problems that their misbehavior can cause for other patients and medical staff.
Finally, our study did not demonstrate any significant difference in utilization of the services based on the ethnicity. In other words, adjusting for other variables, White, Hispanic and Black trauma patients in our trauma center had equal opportunity to utilize the existing resources. This was in contrast to some other studies that reported disparity in utilization of health care services among trauma patients 3, 9, 20, 21, 28, 42. However, in comparison of our results with the results of published literature, we have to take the following issues into consideration. First, we focused on utilization of the resources during hospitalization. However, others have mainly focused on utilization of rehabilitative services after hospital discharge 3, 9, 20, 21, 28, 42. As a result, the settings are not comparable. Second, most of other studies have used National Trauma Data Bank (NTDB) which is a conglomerate of more than 600 small and large hospitals 3, 9, 20, 21, 28, 42. While using data from several hospitals could be extremely valuable, analysis of mutli-center databases is challenging. Roudsari and colleagues demonstrated how not taking into account the clustering of patients within trauma centers during analysis or the high proportion of missing data in the NTDB could affect the results and interpretation of a study 39.
In conclusion, our analyses demonstrated that acute intoxication and dependence to alcohol are associated with overutilization of certain health care resources. This over utilization was not influenced by patients' ethnicity.
Acknowledgments
This study was supported by grants R01- AA017497 and R01- AA015439 from the National Institute on Alcohol Abuse and Alcoholism and National Institutes of Health
Footnotes
Publisher's Disclaimer: 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 citable 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.
References
- 1.National Healthcare Disparities Report. Rockville, MD: Agency for Healthcare Research and Quality; 2003. [Google Scholar]
- 2.Andrews G, Peters L. The psychometric properties of the composite international diagnostic interview. Social Psychiatry and Psychiatric Epidemiology. 1998;33:80–8. doi: 10.1007/s001270050026. [DOI] [PubMed] [Google Scholar]
- 3.Arango-Lasprilla JC, Rosenthal M, Deluca J, et al. Functional outcomes from inpatient rehabilitation after traumatic brain injury: how do Hispanics fare? Arch Phys Med Rehabil. 2007;88:11–8. doi: 10.1016/j.apmr.2006.10.029. [DOI] [PubMed] [Google Scholar]
- 4.Bach PB, Pham HH, Schrag D, et al. Primary care physicians who treat blacks and whites. N Engl J Med. 2004;351:575–84. doi: 10.1056/NEJMsa040609. [DOI] [PubMed] [Google Scholar]
- 5.Baker SP, O'Neill B, Haddon W, Jr, Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. Journal of Trauma. 1974;14:187–96. [PubMed] [Google Scholar]
- 6.Bard MR, Goettler CE, Toschloge EA, et al. Alcohol withdrawal syndrome: Turning minor injuries into a major problem. Journal of Trauma-Injury Infection and Critical Care. 2006;61:1441–5. doi: 10.1097/01.ta.0000245981.22931.43. [DOI] [PubMed] [Google Scholar]
- 7.Bercault N, Boulain T. Mortality rate attributable to ventilator-associated nosocomial pneumonia in an adult intensive care unit: A prospective case-control study. Critical Care Medicine. 2001;29:2303–9. doi: 10.1097/00003246-200112000-00012. [DOI] [PubMed] [Google Scholar]
- 8.Bostwick JM, Seaman JS. Hospitalized patients and alcohol: who is being missed? Gen Hosp Psychiatry. 2004;26:59–62. doi: 10.1016/j.genhosppsych.2003.08.007. [DOI] [PubMed] [Google Scholar]
- 9.Bowman SM, Martin DP, Sharar SR, Zimmerman FJ. Racial disparities in outcomes of persons with moderate to severe traumatic brain injury. Med Care. 2007;45:686–90. doi: 10.1097/MLR.0b013e31803dcdf3. [DOI] [PubMed] [Google Scholar]
- 10.Calle PA, Damen J, De Paepe P, et al. A survey on alcohol and illicit drug abuse among emergency department patients. Acta Clinica Belgica. 2006;61:188–95. doi: 10.1179/acb.2006.033. [DOI] [PubMed] [Google Scholar]
- 11.Collins K, Hall A, Euhaus C. U.S. Minority Health: A Chartbook. New York: Commonwealth Fund; 1999. [Google Scholar]
- 12.D'Onofrio G, Becker B, Woolard RH. The impact of alcohol, tobacco, and other drug use and abuse in the emergency department. Emerg Med Clin North Am. 2006;24:925–67. doi: 10.1016/j.emc.2006.06.008. [DOI] [PubMed] [Google Scholar]
- 13.Ewing JA. Detecting alcoholism. The CAGE questionnaire. JAMA : the journal of the American Medical Association. 1984;252:1905–7. doi: 10.1001/jama.252.14.1905. [DOI] [PubMed] [Google Scholar]
- 14.Fernandezsola J, Junque A, Estruch R, et al. High alcohol intake as a risk and prognostic factor for community-acquired pneumonia. Archives of Internal Medicine. 1995;155:1649–54. doi: 10.1001/archinte.1995.00430150137014. [DOI] [PubMed] [Google Scholar]
- 15.Fine MJ, Smith MA, Carson CA, et al. Prognosis and outcomes of patients with community-acquired pneumonia - A meta-analysis. Jama-Journal of the American Medical Association. 1996;275:134–41. [Google Scholar]
- 16.Finkelstein EA, Corso PS, Miller TR. Incidence and Economic Burden of Injuries in the United States. New York: Oxford University Press; 2006. [Google Scholar]
- 17.Gentilello LM, Ebel BE, Wickizer TM, et al. Alcohol interventions for trauma patients treated in emergency departments and hospitals: a cost benefit analysis. Ann Surg. 2005;241:541–50. doi: 10.1097/01.sla.0000157133.80396.1c. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Greenland S. Interpretation and choice of effect measures in epidemiologic analyses. Am J Epidemiol. 1987;125:761–8. doi: 10.1093/oxfordjournals.aje.a114593. [DOI] [PubMed] [Google Scholar]
- 19.Gurney JG, Rivara FP, Mueller BA, et al. The effects of alcohol intoxication on the initial treatment and hospital course of patients with acute brain injury. Journal of Trauma. 1992;33:709–13. doi: 10.1097/00005373-199211000-00020. [DOI] [PubMed] [Google Scholar]
- 20.Hart T, Whyte J, Polansky M, et al. Community outcomes following traumatic brain injury: impact of race and preinjury status. J Head Trauma Rehabil. 2005;20:158–72. doi: 10.1097/00001199-200503000-00004. [DOI] [PubMed] [Google Scholar]
- 21.Howard I, Joseph JG, Natale JE. Pediatric traumatic brain injury: do racial/ethnic disparities exist in brain injury severity, mortality, or medical disposition? Ethn Dis. 2005;15:S5-51–6. [PubMed] [Google Scholar]
- 22.Hungerford DW, Pollock DA. Alcohol Problems Among Emergency Department Patients: Proceedings of a Research Conference on Identification and Intervention Atlanta (GA): National Center for Injury Prevention and Control. Centers for Disease Control and Prevention. 2002 [Google Scholar]
- 23.Jenkins DH. Substance abuse and withdrawal in the intensive care unit. Contemporary issues. Surg Clin North Am. 2000;80:1033–53. doi: 10.1016/s0039-6109(05)70112-2. [DOI] [PubMed] [Google Scholar]
- 24.Jurkovich GJ, Rivara FP, Gurney JG, et al. The effect of acute alcohol intoxication and chronic alcohol abuse on outcome from trauma. JAMA. 1993;270:51–6. [PubMed] [Google Scholar]
- 25.Jurkovich GJ, Rivara FP, Gurney JG, et al. Effects of alcohol intoxication on the initial assessment of trauma patients. Annals of Emergency Medicine. 1992;21:704–8. doi: 10.1016/s0196-0644(05)82783-0. [DOI] [PubMed] [Google Scholar]
- 26.Kessler RC, Andrews G, Mroczek D, et al. The world health organization composite international diagnostic interview short-form (CIDI-SF) International Journal of Methods in Psychiatric Research. 1998;7:171–85. doi: 10.1002/mpr.168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Marik P, Mohedin B. Alcohol-related admissions to an inner city hospital intensive care unit. Alcohol Alcohol. 1996;31:393–6. doi: 10.1093/oxfordjournals.alcalc.a008168. [DOI] [PubMed] [Google Scholar]
- 28.Marquez de la Plata C, Hewlitt M, de Oliveira A, et al. Ethnic differences in rehabilitation placement and outcome after TBI. J Head Trauma Rehabil. 2007;22:113–21. doi: 10.1097/01.HTR.0000265099.29436.56. [DOI] [PubMed] [Google Scholar]
- 29.McCaig LF, Nawar EW. National Hospital Ambulatory Medical Care Survey: 2004 Emergency Department Summary. Advanced Data. 2006;372 [PubMed] [Google Scholar]
- 30.McDonald AJ, Wang N, Camargo CA. US emergency department visits for alcohol-related diseases and injuries between 1992 and 2000. Archives of Internal Medicine. 2004;164:531–7. doi: 10.1001/archinte.164.5.531. [DOI] [PubMed] [Google Scholar]
- 31.McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003;157:940–3. doi: 10.1093/aje/kwg074. [DOI] [PubMed] [Google Scholar]
- 32.Moore EE. Alcohol and trauma: the perfect storm. Journal of Trauma. 2005;59:S53–6. doi: 10.1097/01.ta.0000174868.13616.67. discussion S67-75. [DOI] [PubMed] [Google Scholar]
- 33.National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much A CLINICIAN'S GUIDE. Updated 2005. U.S. DEPARTMENT OF HEALTH & HUMAN SERVICES, National Institutes of Health, National Institute on Alcohol Abuse and Alcoholism; 2005. [Google Scholar]
- 34.National Institute on Alcohol Abuse and Alcoholism. The physician's guide to helping patients with alcohol problems (NIH Publication No.953769) Washington, DC: National Institute on Alcohol Abuse and Alcoholism, National Institute of Health; 1995. [Google Scholar]
- 35.NIAAA. The physician's guide to helping patients with alcohol problems (NIH Publication No.953769) Washington, DC: National Institute on Alcohol Abuse and Alcoholism, National Institute of Health; 1995. [Google Scholar]
- 36.Parry-Jones BL, Vaughan FL, Miles Cox W. Traumatic brain injury and substance misuse: a systematic review of prevalence and outcomes research (1994-2004) Neuropsychological Rehabilitation. 2006;16:537–60. doi: 10.1080/09602010500231875. [DOI] [PubMed] [Google Scholar]
- 37.Robbins AS, Chao SY, Fonseca VP. What's the relative risk? A method to directly estimate risk ratios in cohort studies of common outcomes. Ann Epidemiol. 2002;12:452–4. doi: 10.1016/s1047-2797(01)00278-2. [DOI] [PubMed] [Google Scholar]
- 38.Roudsari B, Caetano C, Frankowski R, Field C. Do minority trauma patients respond differently to brief alcohol intervention in trauma centers? Ann Emerg Medicine. 2009;54:285–93. doi: 10.1016/j.annemergmed.2009.01.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Roudsari B, Field CA, C R. Clustered and missing data in the US National Trauma Data Bank: Implications for analysis. Inj Prev. 2008;14:96–100. doi: 10.1136/ip.2007.017129. [DOI] [PubMed] [Google Scholar]
- 40.Saitz R, Palfai TP, Freedner N, et al. Alcohol screening and brief intervention online for college students. Alcoholism-Clinical and Experimental Research. 2005;29:109A-A. [Google Scholar]
- 41.Schwan R, Allen JP. Unhealthy alcohol use. N Engl J Med. 2005;19:2139–40. doi: 10.1056/NEJM200505193522021. [DOI] [PubMed] [Google Scholar]
- 42.Shafi S, de la Plata CM, Diaz-Arrastia R, et al. Ethnic disparities exist in trauma care. J Trauma. 2007;63:1138–42. doi: 10.1097/TA.0b013e3181568cd4. [DOI] [PubMed] [Google Scholar]
- 43.Spies CD, Dubisz N, Neumann T, et al. Therapy of alcohol withdrawal syndrome in intensive care unit patients following trauma: Results of a prospective, randomized trial. Crit Care Med. 1996;24:414–22. doi: 10.1097/00003246-199603000-00009. [DOI] [PubMed] [Google Scholar]
- 44.Spies CD, Neuner B, Neumann T, et al. Intercurrent complications in chronic alcoholic men admitted to the intensive care unit following trauma. Intensive Care Med. 1996;22:286–93. doi: 10.1007/BF01700448. [DOI] [PubMed] [Google Scholar]