Currently Viewing:
The American Journal of Managed Care November 2013
Opioid Analgesic Treated Chronic Pain Patients at Risk for Problematic Use
Joseph Tkacz, MS; Jacqueline Pesa, PhD, MPH; Lien Vo, PharmD, MPH; Peter G. Kardel, MA; Hyong Un, MD; Joseph R. Volpicelli, MD, PhD; and Charles Ruetsch, PhD
Safety and Effectiveness of Mail Order Pharmacy Use in Diabetes
Julie A. Schmittdiel, PhD; Andrew J. Karter, PhD; Wendy T. Dyer, MS; James Chan, PharmD, PhD; and O. Kenrik Duru, MD, MSHS
Depression Self-Management Assistance Using Automated Telephonic Assessments and Social Support
John D. Piette, MSc, PhD; James E. Aikens, PhD; Ranak Trivedi, PhD; Diana Parrish, MSW; Connie Standiford, MD; Nicolle S. Marinec, MPH; Dana Striplin, MHSA; and Steven J. Bernstein, MD, MPH
Creating Peer Groups for Assessing and Comparing Nursing Home Performance
Margaret M. Byrne, PhD; Christina Daw, PhD; Ken Pietz, PhD; Brian Reis, BE; and Laura A. Petersen, MD, MPH
Currently Reading
Upcoding Emergency Admissions for Non-Life-Threatening Injuries to Children
Zachary Pruitt, MHA; and Etienne Pracht, PhD
Using Health Outcomes to Validate Access Quality Measures
Julia C. Prentice, PhD; Michael L. Davies, MD; and Steven D. Pizer, PhD
Collecting Mortality Data to Drive Real-Time Improvement in Suicide Prevention
Brian K. Ahmedani, PhD; M. Justin Coffey, MD; and C. Edward Coffey, MD

Upcoding Emergency Admissions for Non-Life-Threatening Injuries to Children

Zachary Pruitt, MHA; and Etienne Pracht, PhD
For-profit status was found to influence the probability of upcoding for inpatient cases involving non-life-threatening injuries with implications for Medicaid and other insurers.
Objectives: To assess the influence of investorowned for-profit (IOFP) status on upcoding pediatric inpatient admissions for inconsequential injuries as emergency when urgent or elective would be more suitable.

Study Design: Using Florida inpatient discharge data for children 15 years and younger during 2001 to 2010, we examined injuries originating from the emergency departments (EDs) resultingin 1 overnight stay. Only non–life-threatening injuries were included. We assessed the probability of emergency categorization (vs urgent/elective) of admissions at IOFP hospitals compared with other types of hospitals (public, not for profit).

Methods: Logistic regression was used to explore the probability that hospital admission following non–life-threatening injury to a child was classified as an emergency on the billing claim. The model controlled for age, race, sex, Hispanic ethnicity, trauma center status, insurance type and status, number of injuries, and market competition conditions.

Results: For those patients satisfying the inclusion criteria (n = 8694), about 68% of the time hospitals classified the admissions as emergent. The model provides strong statistical evidence that IOFP hospitals had a higher probability (odds ratio = 1.1) of reporting emergency priorities for children admitted to the hospital from the ED, holding all other variables constant.

Conclusions: Upcoding by IOFP hospitals may be a consequence of payer payment practices, utilization management policies, and local market dynamics. Florida Medicaid regulators and managed care organizations should examine their policies to identify inefficiencies associated with pediatric patients admitted for non–life-threatening injuries.

Am J Manag Care. 2013;19(11):917-924
For-profit status was found to influence the probability of upcoding for inpatient cases involving non–life-threatening injuries with implications for Medicaid and other insurers.
  • Payer payment practices and policies affect the probability of upcoding.

  • Differences based on payer type (commercial vs public) point to the influence of utilization management policies that address the potential for upcoding.

  • Florida Medicaid regulators and managed care organizations should examine their policies to identify inefficiencies associated with pediatric patients admitted for non–lifethreatening injuries.
Improper billing aimed at maximizing revenue, whether intentional or not, creates inefficiencies that could be identified and prevented. This study examines a specific type of upcoding related to classification of injured pediatric patients upon inpatient admission following emergency department (ED) evaluation. This study asked how the type of hospital ownership affects admission categorization (emergent, urgent, elective) for non–life-threatening injuries for children admitted to Florida hospitals from the ED from 2001 to 2010. We hypothesized the presence of a statistically significant greater probability of emergency categorization of inpatient admissions (vs urgent/elective) at investor-owned, for-profit (IOFP) facilities versus other types of hospitals (public, not for profit [NFP]).

Cost-control regulations and management practices have addressed wasteful healthcare provider billing practices to some extent, but opportunities for improvement remain.1,2 Providers retain considerable leeway in preparing claims from medical charts.3 For example, ambiguity in payment policies may lead to inpatient admissions for inconsequential injuries being categorized as “emergency” when “urgent” or “elective” would be more suitable to the patient’s condition. This type of upcoding is similar to substituting patient Diagnosis-Related Groups that qualify for higher rates of reimbursement without any associated change in  illness severity or treatment intensity, a practice that has been studied by Silverman and Skinner3 and Danfy.4 Categorizing minor  injuries asemergency for hospital admissions from the ED has 2 potential financial impacts on the hospital in case of Medicaid-covered patients.5 First, concerning all age groups, the emergency classification is necessary to circumvent prior authorization  requirements, therefore reducing administrative costs and consequently increasing residual revenue. The second reason applies to the adult population and concerns lifting the $1500 cap associated with procedures performed in the ED before inpatient admission. However, there are alternative relevant criteria that must be met for the cap to be lifted in association with a surgical procedure, one of which is  classification as an emergency. While the latter reason does not affect pediatric patientsdirectly, it may impact general institutional procedures when classifying patients’ conditions as either urgent or emergent.

Prior analysis of the Florida Agency for Health Care Administration (AHCA) inpatient discharge data analysis revealed that injury, poisonings, toxic effects, and other external cause diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 800-999) were the second-most common reason for hospitalization in the pediatric population. (The most common reason for admission in the inpatient pediatric population concerned diagnoses related to the respiratory system.) We selected injury for 3 reasons. First, the severity measure used in the study, the International Classification Injury Severity Score (ICISS), is well established and available for the study population for the past 20 years. Next, the injury diagnoses we used lent themselves more easily to the notion of non–life-threatening admissions. While the ICISS main criterion, mortality, could have been  determined for other types of admissions (eg, respiratory diagnoses), it was necessary to account for severity for those  hospitalizations thatwere not associated with mortality. The count of injuries has been documented in the literature as a valid proxy for apparent severity.6 Finally, it was deemed important to distinguish among hospitals’ treatment capabilities. The relevant experience and specialization in treating injury diagnoses are accounted for through the Florida trauma center (TC) certification variable in the model.

Inconsequential injuries resulting in admissions are highly prevalent.6 In other words, many of these admissions, while credible, do not result in death for children and may not require emergency prioritization upon admission. Despite their relatively mild conditions, the hospitals claimed that the patients required “immediate medical intervention as a result of severe, life-threatening or potentially disabling condition.”7

Investor-owned, for-profit hospitals have been found to be more likely to maximize profits8 and increase revenues through billing processes than their NFP counterparts, a practice perhaps related to IOFP hospitals’ abilities to implement more effective managerial and financial strategies.9 In addition,  the studies that specifically addressed upcoding billing practices found that  for-profit hospitals are willing to risk regulatory investigation and subsequent reputation damages in favor of profit obtained by upcoding.3,4 These findings are consistent with the altruism model of hospital behavior theorized by Newhouse.10 Pressures on IOFP management, including investors’ expectation for return on investment and the requirement to pay taxes11 ordifferences in the incentives provided to managers,12 may be at the core of differences in billing behaviors.


We analyzed the inpatient discharge data set from the Florida AHCA for 2001 to 2010. The data contained patient information related to diagnosis (primary and up to 30 other diagnoses) and demographics (age, sex, race, Hispanic ethnicity). Each observation includes a hospital identifier, which was used to derive facility-specific information such as hospital ownership type (IOFP, NFP, or public) and TC certification status. Finally, the inpatient data set also indicates the payer type (eg, commercial, Medicaid, KidCare, or uninsured).

The outcome variable of interest was defined as dichotomous and indicated the type/priority assigned to the admission. Value options for this variable included emergency (the patient requires immediate medical intervention as a result of a severe, life-threatening, or potentially disabling condition) versus urgent (the patient requires attention for the care and treatment of a physical or mental disorder) or elective (the patient’s condition permits adequate time to schedule the availability of a suitable accommodation).7 Newborn admissions were excluded. These admission priority codes are entered by the hospitals on the billing claim (Inpatient Uniform Billing-04, form locator-14).

The study population was defined as children aged 0 to 15 years who were admitted to the hospital through the ED with a primary diagnoses in the ICD-9-CM code range of 800 to 959, excluding late effects of injuries, poisonings, and toxic effects, and effects of foreign bodies entering through an orifice. In addition, the population was limited to those children with non–life-threatening injuries. To define non–life-threatening injuries, this study used the ICISS.6,13,14 This technique measures the proportion of patients who survive after admission with a specific combination of ICD-9-CM codes, and only those diagnoses in which all patients survived (ICISS = 1) were included. The ICISS values were calculated using survival risk ratios dating back to 1991, indicating at least a 10-year history of zero mortality associated with the injuries. The inclusion criterion of an ICISS of 1 accounts for the observed risk associated with the injuries. To eliminate bias from unobserved characteristics that might indicate higher injury severity not captured by the ICISS methodology, all inpatient episodes lasting longer than a single overnight stay were also omitted from the analysis. Based on these inclusion and exclusion criteria, the data set contained 8694 pediatric patients in Florida categorized as emergent or urgent/elective upon hospital admission from the ED.


Given the dichotomous nature of the dependent variable, a logistic regression was used to predict the probability that a patient admitted to the hospital from the ED with non–life-threatening injuries would be classified as emergency as opposed to urgent or elective. Model variables may be conceptually divided into 5 categories: injury type and count (fractures, skull and spinal cord injuries [SSCIs], traumatic brain injury [TBI], vascular injury, thorax injury, and burns); hospital types (ownership, trauma certification); payer types and status (uninsured, commercial, Medicaid, KidCare);patient geography and demographics (distance to hospital, age, race, sex, and Hispanic ethnicity); and market structure (hospital and health maintenance organization [HMO] concentration).

The study population included pediatric patients admitted to the hospital from the ED due to injury, but with an ICISS of 1, indicating zero associated mortality. Therefore, the measure of severity used in our analysis contained no variation. Nonetheless, variation in severity not measured by the ICISS value was expected to influence the probability of classifying patients as emergent as opposed to urgent or elective. To capture the relative severity of study injuries not associated with mortality, the model included the number of individual injuries, as defined above. A positive relationship between emergency status and injury count was hypothesized. The remaining criteria, specifying the types of injury, indicate whether the patient was admitted with TBI, SSCI (other than TBI), a fracture (other than TBI or SSCI), a vascular injury, an injury to the thorax, or a burn. The International Classification of Diseases, Ninth Revision, Clinical Modification  ranges for these injuries are discussed in more detail by Pracht and colleagues.6 Patients admitted with TBI were used as the control group to examine the impact of injury type.

Also, during the study period, the Florida Department of Health designated 22 hospitals as TCs, including 2 pediatriconly TCs. There is strong statistical evidence of improved mortality for children with severe injuries when treatment is provided in designated TCs versus nontrauma centers.6 However, the advantage associated with Florida TCs that treat children with non–life-threatening injuries is not well understood. We included TC status in the model to control for the greater experience, both pertaining to institutional operation and to potential medical expertise and proficiency in recognizing potential mortality risks. These qualities may allow TCs to more accurately assess severity and, by extension, reduce the probability of admitting patients for non–life-threatening injuries. Because 20 of the 22 TCs are NFP hospitals, the inclusion of this variable was important to keep its hypothesized influence from incorrectly getting attributed to NFP hospitals in general, therefore biasing the IOFP treatment variable in the model.

Copyright AJMC 2006-2019 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
Welcome the the new and improved, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up