Hospital admissions related to mental disorders in U.S. Army soldiers in Iraq and Afghanistan.

We conducted a retrospective study of 473,964 U.S. Army soldiers deployed to Iraq and Afghanistan through December 2004 using deployment and admission records. We categorized mental disorder diagnoses using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) and identified attempted suicide/ self-inflicted injuries using ICD-9-CM diagnosis codes E950-E959. We estimated and evaluated relative risks (RR) using Poisson regression models. Analysis found 1,948 psychiatric hospitalizations of deployed soldiers. The most common mental problems were mood, adjustment, and anxiety disorders (including post-traumatic stress disorder [PTSD]). RR of mental disorders ranged from 1.6 to 3 for females and 2 to 6 for enlisted soldiers compared to their counterparts. Younger soldiers had 30-60% higher substance abuse disorders. Combat units in Iraq demonstrated higher risk of any mental disorder and anxiety problems compared to combat support units. Younger women had the highest incidence of attempted suicide/ self-inflicted injuries. Further mental disorders surveillance is recommended.

U.S. Army disease and nonbattle injury model, refined in Afghanistan and Iraq.

Previous analysis of Operation Desert Shield/Operation Desert Storm data yielded a disease and nonbattle injury (DNBI) model using distinct 95th percentile daily admission rates during the three phases of a war-fighting operation to predict medical requirements. This study refines the model with data from Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF). Inpatient health care records of U.S. Army soldiers deployed to OEF and OIF who were admitted with DNBI diagnoses were analyzed. DNBI admission rates for OEF and OIF were compared with rates for Operation Desert Shield/Operation Desert Storm. DNBI admission rates for OEF and OIF were lower than those for Operation Desert Shield/Operation Desert Storm. Rates among the phases of OIF were distinctly different. DNBI admission rates have been reduced during recent deployments. The concepts of the original model based on Operation Desert Shield/Operation Desert Storm data were validated by experiences during OEF and OIF. Continuous surveillance of DNBI admission rates is recommended.

Comparison of Operation Iraqi Freedom and Patient Workload Generator injury distributions.

The Patient Workload Generator (PATGEN) simulation model is an important part of the Army Medical Department force requirement planning. The current version of the PATGEN model is based on historical major combat operations such as World War II and the Korean War. The purpose of this study was to determine whether there was a significant difference between injury distributions used in the PATGEN model and the injury distributions based on the data from Operation Iraqi Freedom (OIF). To make the comparison feasible, the PATGEN injury categories were created using the Barell Injury Matrix. Analyses were performed using two independent OIF data sources, the Joint Theater Trauma Registry and the Standard Inpatient Data Record. Based on X2 test results, both analyses suggest a significant difference between PATGEN and OIF injury distributions. A major concern is the underestimation by PATGEN of battle injuries with multiple wounds. The findings support future use of data-driven diagnosis-based injury distributions for current operations and new more flexible simulation models that will allow for changes in injury probability distributions.

The challenge of mapping between two medical coding systems.

OBJECTIVE: Deployable medical systems patient conditions (PCs) designate groups of patients with similar medical conditions and, therefore, similar treatment requirements. PCs are used by the U.S. military to estimate field medical resources needed in combat operations. Information associated with each of the 389 PCs is based on subject matter expert opinion, instead of direct derivation from standard medical codes. Currently, no mechanisms exist to tie current or historical medical data to PCs. Our study objective was to determine whether reliable conversion between PC codes and International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis codes is possible. METHODS: Data were analyzed for three professional coders assigning all applicable ICD-9-CM diagnosis codes to each PC code. Inter-rater reliability was measured by using Cohen's K statistic and percent agreement. Methods were developed to calculate kappa statistics when multiple responses could be selected from many possible categories. RESULTS: Overall, we found moderate support for the possibility of reliable conversion between PCs and ICD-9-CM diagnoses (mean kappa = 0.61). CONCLUSION: Current PCs should be modified into a system that is verifiable with real data.

Status of trauma care in U.S. Army hospitals.

OBJECTIVE: This study documents the recent trends and current state of inpatient trauma care in U.S. Army hospitals. METHODS: Inpatient trauma cases from Army hospitals worldwide from October 1988 through April 2001 were analyzed. Facilities included 3 Certified Trauma Centers (CTCs), 7 non-CTC Army Medical Centers, and 42 Army Community Hospitals. Logistic regression identified mortality risk factors. RESULTS: Overall, the Army treated 166,124 trauma cases, with a mortality rate of 0.8% (trend of 0.66% to 1.18% in fiscal years 1989-2000, p < 0.0001). The number of Army hospitals decreased by 44% and the number of trauma cases decreased by nearly 75%. Injury severity, patient age, hospital trauma volume, beneficiary category, hospital type, and a resource intensity measure were all significantly associated with the probability of death. CONCLUSIONS: The overall trauma mortality rate at Army hospitals during the study period was lower than that reported for civilian trauma centers. However, changes in patient profiles, increased average severity, and decreased trauma volume might have contributed to a 13% increase in mortality rates at CTCs.

A disease and non-battle injury model based on Persian Gulf War admission rates.

BACKGROUND: Military planners must ensure adequate medical care for deployed troops-including care for disease and non-battle injuries (DNBI). This study develops a heuristic model with the three distinct phases of a warfighting operation (build-up, ground combat, post-combat) to assist in predicting DNBI incidence during warfighting deployments. METHODS: Inpatient healthcare records of soldiers deployed to the Persian Gulf War who were admitted with DNBI diagnoses were analyzed. DNBI admission rates for the three phases of the operation were examined and compared to rates for US Army Forces Command (FORSCOM) posts in the US. RESULTS: DNBI admission rates among the phases were distinctly different. The operation's overall rate and 95th percentile daily rate were less than the FORSCOM FY 1990 annual rate. CONCLUSIONS: The level of combat must be considered. The traditional use of average or overall rates should be abandoned when forecasting DNBI rates. Medical support projections should use separate 95th percentile DNBI admission rates for each of the phases.

The implications of TRICARE on medical readiness.

The objective of this study was to compare two alternative sources of replacement personnel for a medical treatment facility experiencing personnel loss due to a deployment. The two replacement strategies included the reserve component option and the TRICARE internal resource-sharing option. A hypothetical scenario was used as a mechanism for the analysis, and three key variables were considered: effectiveness, feasibility, and operational expense. From the perspective of effectiveness, the TRICARE strategy demonstrated an ability to provide a slightly larger percentage of the requested replacement personnel. With regard to feasibility, both strategies were feasible in that both could provide replacements for the duration of the 270-day deployment and within an established 180-day report date. Operational expense was a decisive factor with the reserve component option significantly less costly than the TRICARE alternative. Weaknesses and strengths of each option were identified and discussed, and alternatives were recommended.