Janel Hanmer, MD, PhD; Xin Lu, MS; Gary E. Rosenthal, MD; Peter Cram, MD, MBA
Disclaimer: The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs.
Grant Support: By a K24 award from National Institute of Arthritis and Musculoskeletal and Skin Diseases (AR062133; Dr. Cram); Center of Innovation Award (CIN 13-412; Drs. Cram and Rosenthal) from the Health Services and Development Service, Veterans Health Administration; Clinical and Translational Science Award (2 UL1 TR000442-06; Dr. Rosenthal) from the National Center for Advancing Translational Science; and in part by grants R01 HL085347 from National Heart, Lung, and Blood Institute and R01 AG033035 from National Institute on Aging at the National Institutes of Health.
Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M12-1977.
Reproducible Research Statement: Study protocol and statistical code: Available from Dr. Hanmer (e-mail, firstname.lastname@example.org). Data set: Available at www.hcup-us.ahrq.gov/tech_assist/centdist.jsp.
Requests for Single Reprints: Janel Hanmer, MD, PhD, University of Pittsburgh Medical Center Montefiore Hospital, Suite W93, 200 Lothrop Street, Pittsburgh, PA 15213; e-mail, email@example.com.
Current Author Addresses: Dr. Hanmer: University of Pittsburgh Medical Center Montefiore Hospital, Suite W933, 200 Lothrop Street, Pittsburgh, PA 15213.
Ms. Lu and Dr. Rosenthal: Division of General Internal Medicine, Department of Internal Medicine, University of Iowa Carver College of Medicine, 200 Hawkins Drive, Iowa City, IA 52242.
Dr. Cram: Division of General Internal Medicine, University of Toronto and University Health Network/Mount Sinai Hospital, 200 Elizabeth Street, Eaton North, Toronto, Ontario, Canada M5G 2C4.
Author Contributions: Conception and design: J. Hanmer, X. Lu, P. Cram.
Analysis and interpretation of the data: J. Hanmer, X. Lu, G.E. Rosenthal, P. Cram.
Drafting of the article: J. Hanmer, P. Cram.
Critical revision of the article for important intellectual content: J. Hanmer, X. Lu, G.E. Rosenthal, P. Cram.
Final approval of the article: J. Hanmer, X. Lu, G.E. Rosenthal, P. Cram.
Provision of study materials or patients: X. Lu.
Statistical expertise: J. Hanmer, X. Lu, P. Cram.
Obtaining of funding: P. Cram.
Administrative, technical, or logistic support: J. Hanmer, X. Lu, P. Cram.
Collection and assembly of data: J. Hanmer, X. Lu, P. Cram.
Hanmer J., Lu X., Rosenthal G., Cram P.; Insurance Status and the Transfer of Hospitalized Patients: An Observational Study. Ann Intern Med. 2014;160:81-90. doi: 10.7326/M12-1977
Download citation file:
Published: Ann Intern Med. 2014;160(2):81-90.
There is little objective evidence to support concerns that patients are transferred between hospitals based on insurance status.
To examine the relationship between patients’ insurance coverage and interhospital transfer.
Data analyzed from the 2010 Nationwide Inpatient Sample.
All patients aged 18 to 64 years discharged alive from U.S. acute care hospitals with 1 of 5 common diagnoses (biliary tract disease, chest pain, pneumonia, septicemia, and skin or subcutaneous infection).
For each diagnosis, the proportion of hospitalized patients who were transferred to another acute care hospital based on insurance coverage (private, Medicare, Medicaid, or uninsured) was compared. Logistic regression was used to estimate the odds of transfer for uninsured patients (reference category, privately insured) while patient- and hospital-level factors were adjusted for. All analyses incorporated sampling and poststratification weights.
Among 315 748 patients discharged from 1051 hospitals with any of the 5 diagnoses, the percentage of patients transferred to another acute care hospital varied from 1.3% (skin infection) to 5.1% (septicemia). In unadjusted analyses, uninsured patients were significantly less likely to be transferred for 3 diagnoses (P < 0.05). In adjusted analyses, uninsured patients were significantly less likely to be transferred than privately insured patients for 4 diagnoses: biliary tract disease (odds ratio, 0.73 [95% CI, 0.55 to 0.96]), chest pain (odds ratio, 0.63 [CI, 0.44 to 0.89]), septicemia (odds ratio, 0.76 [CI, 0.64 to 0.91]), and skin infections (odds ratio, 0.64 [CI, 0.46 to 0.89]). Women were significantly less likely to be transferred than men for all diagnoses.
This analysis relied on administrative data and lacked clinical detail.
Uninsured patients (and women) were significantly less likely to undergo interhospital transfer. Differences in transfer rates may contribute to health care disparities.
National Institutes of Health.
Jonathan M. Schwartz, M.D., M.B.A.
The Grosse Pointe Medical Group
January 22, 2014
Insured Status is a Driver of Interhospital Transfer
A patient's status as "insured" is very likely the principle key element in their likelihood to be subject to interhospital transfer, either directly from Emergency Department or following admission. Specifically, patients insured and enrolled in a managed care product, accountable care organization, or otherwise are subject to the constraints of a specifically designated network of care who present and are admitted to an "out of network" facility on an emergency or unplanned basis will be of great interest to their designated provider network for financial reasons. Both the patient and their provider network will have a financial incentive to "repatriate" the patient back to an in-network facility.
Alternatively, uninsured patients do not have either an insurer or provider network utilization management infrastructure with an incentive to enforce network boundaries.
This phenomena of interhospital transfers is likely to grow as "accountable care" (i.e. financial risk sharing care networks) becomes more pervasive. Physicians and hospitals alike will have a keen incentive to attend to their financial accountability for patient care.
Michael Pauszek, MD, FACP
Johnson Memorial Hospital
January 23, 2014
To The Editor: I read the study by Janel Hanmer titled Insurance Status and the Transfer of Hospitalized Patients, Annals 160. January 21, 2014, 81-90. Please correct me if I misinterpreted the purpose but it appeared to be a study to prove that patients without health insurance were still being dumped, however they were not. The authors then went on to suggest that patients were not being transferred to their clinical detriment because they lacked insurance. Wow! Negative dumping! Thank you for confirming that the vast majority of patients are cared for in their local community. Two factors were not addressed in the discussion, comorbidities and age. The authors did demonstrate that comorbidity was a statistically significant factor in the transferred patients. Thank you, for confirming the appropriateness of patient care. They neglected to report on the age of the variously insured groups. This is a significant omission.In my community the under 65 Medicare patients and privately insured patients would be of similar age, the Medicare group having more comorbidity. They are also the oldest age groups. The Medicaid population would span the clinical study age range, probably have some underlying disease process but would be younger than the two other groups. The uninsured in my community are the youngest patients we see. For biliary disease and skin/subcutaneous infection the study’s findings confirm that the older and sicker would be transferred for complications and procedures such as ERCP and fasciotomy. No insurance bias need be postulated, only age and underlying disease. For chest pain, older age is a risk factor. Younger patients usually have noninvasive and locally available evaluations. Higher risk, older patients are referred for invasive tests. In summary, in two of the three referenced diagnosis groups, the reported findings could be explained based upon differences in comorbidities or age in the various insured groups. In the third, chest pain, age was not addressed or reported and is a very important factor in the decision to transfer the patient. The authors avoided reporting morbidity and mortality. That would have been valuable. Were uninsured patients kept in their local hospitals to their detriment? It was not proven here. What did we really learn from this study? Hospitalized patients are not being dumped. Transfer numbers are low and appear to be related to underlying health! Perfect. To more broadly interpret the data is conjecture. We need more data not speculation. Your conclusion is unfounded and suspiciously seems preconceived. Thank you for confirming the quality of our care.Michael E Pauszek MD FACPJohnson Memorial HospitalFranklin, IndianaPotential Conflicts of Interest: None disclosed
Morehouse School of Medicine
December 7, 2014
The Dumping Syndrome
TO THE EDITOR: I read Hanmer and colleagues’ article with great interest. The relationship between a patient’s insurance coverage and inter-hospital transfers is a particularly interesting subject in light of the new Affordable Care Act and the lack of universal acceptance among the various hospital systems. The authors conclude that uninsured patients were significantly less likely to be transferred to another acute care hospital than privately insured patients after the necessary adjustments were made for patient demographics, co-morbid conditions and hospital-level factors. Further, there is an attempt in the discussion to highlight some of the limitations in their study, namely “…the complex dynamic between transferring and receiving hospitals”. The fact that there was a clear trend towards fewer transfers from large, major teaching, not-for-profit, urban hospitals is mentioned only in passing. The characteristics of the receiving hospital are integral to any analysis of the role insurance coverage may play in this context. Most Tertiary care hospitals have similar characteristics of those identified in the study as transferring less and as such would have less need to transfer patients for specialty services. Certainly the reverse – small, non-teaching, for-profit, rural hospitals often times do not have the resources or infrastructure to deal with the most complex cases, is true as well. Perhaps an interesting follow study that analyzed inter-hospital transfers between hospitals which offer the same level of service would provide more insight into the role insurance coverage and type may play in “dumping” phenomenon without the confounders of levels of service. At any rate, the interest shown in a well known but often times unspoken truth of the “tiered” hospital system, and resulting effect on patient care, is refreshing – confounders notwithstanding.
to gain full access to the content and tools.
Learn more about subscription options.
Register Now for a free account.
Results provided by:
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only