Association Between Treatment by Fraud and Abuse Perpetrators and Health Outcomes Among Medicare Beneficiaries | Geriatrics | JAMA Internal Medicine | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Adjusted and Unadjusted 3-Year, All-Cause Mortality Rates Among Medicare Beneficiaries by Type of Fraud and Abuse Exposure in 2013
Adjusted and Unadjusted 3-Year, All-Cause Mortality Rates Among Medicare Beneficiaries by Type of Fraud and Abuse Exposure in 2013

Analysis included all beneficiaries in a random 5% sample of Medicare claims data who were not treated by a known fraud and abuse perpetrator (FAP) in 2012 and were treated by 1 or more known FAPs in 2013 or by 1 or more non-FAP health care professionals within a random sample selected from counties with at least 1 known FAP in 2013. Adjusted mortality rates are based on regression models of mortality on any FAP exposure in 2013, demographic characteristics, indicators for each month of enrollment in the Medicare Fee-for-Service program in 2012-2013, and county indicators. Error bars show 95% CIs. Additional model results are found in eTable 1 in the Supplement.

Figure 2.  Adjusted and Unadjusted Rates of Emergency Hospitalization Among Medicare Beneficiaries by Type of Fraud and Abuse Exposure in 2013
Adjusted and Unadjusted Rates of Emergency Hospitalization Among Medicare Beneficiaries by Type of Fraud and Abuse Exposure in 2013

Analysis compared all beneficiaries in a 5% random sample of Medicare claims data who were not treated by a known FAP in 2012 and were treated either by 1 or more known FAPs in 2013 or by a health care professional from a random sample of non-FAPs practicing in counties with at least 1 FAP in 2013. Adjusted emergency hospitalization rates are based on regression models of mortality including any FAP exposure in 2013, demographic characteristics, indicators for each month of enrollment in the Medicare Fee-for-Service program in 2012-2013, and county indicators. Error bars show 95% CIs. Additional model results are found in eTable 2 in the Supplement.

Table 1.  Characteristics of Sample Health Care Professionals by FAP Statusa
Characteristics of Sample Health Care Professionals by FAP Statusa
Table 2.  Characteristics of Medicare Beneficiaries by New Exposure to Fraud and Abuse in 2013 Statusa,b
Characteristics of Medicare Beneficiaries by New Exposure to Fraud and Abuse in 2013 Statusa,b
Table 3.  Percentage Point Differences in Medicare Beneficiaries’ 3-Year, All-Cause Mortality and Emergency Hospitalization Rates Associated With Exposure to FAPs in 2013a
Percentage Point Differences in Medicare Beneficiaries’ 3-Year, All-Cause Mortality and Emergency Hospitalization Rates Associated With Exposure to FAPs in 2013a
1.
Centers for Medicare & Medicaid Services; Medicare Learning Network.  Medicare Fraud & Abuse: Prevent, Detect, Report. Baltimore, MD: US Dept of Health & Human Services; February 2019. ICN MLN4649244. https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNProducts/Downloads/Fraud-Abuse-MLN4649244.pdf. Accessed September 30, 2019.
2.
Office of the Inspector General, Department of Health and Human Services.  12-2018 Updated LEIE Database. Washington, DC: US Dept of Health & Human Services; 2018.
3.
Berwick  DM, Hackbarth  AD.  Eliminating waste in US health care.  JAMA. 2012;307(14):1513-1516. doi:10.1001/jama.2012.362PubMedGoogle ScholarCrossref
4.
Goldberg  SR, Lindquist  SC.  Fighting costly health care fraud.  J Corp Account Finance. 2005;16(4):29-34. doi:10.1002/jcaf.20115Google ScholarCrossref
5.
Kalb  PE.  Health care fraud and abuse.  JAMA. 1999;282(12):1163-1168. doi:10.1001/jama.282.12.1163PubMedGoogle ScholarCrossref
6.
Morris  L.  Combating fraud in health care: an essential component of any cost containment strategy.  Health Aff (Millwood). 2009;28(5):1351-1356. doi:10.1377/hlthaff.28.5.1351PubMedGoogle ScholarCrossref
7.
Burton  BR.  A Demographic Portrait of Physicians Sanctioned by the Federal Government in the United States [dissertation]. Irvine: University of California; 2016. https://escholarship.org/uc/item/1022m81j. Accessed September 13, 2019.
8.
McDaniels  AK. Owings Mills imaging firm owner guilty of fraud. Baltimore Sun. 2016:A3.
9.
Rhodes  D. Doctor pleads guilty in fraud case. Chicago Tribune. September 11, 2014:33-35.
10.
Stewart  CS, Carreyrou  J. Government joins suits against cardiologist. Wall Street Journal. January 6, 2015:A3.
11.
Nicholas  LH, Segal  J, Hanson  C, Zhang  K, Eisenberg  MD.  Medicare beneficiaries’ exposure to fraud and abuse perpetrators.  Health Aff (Millwood). 2019;38(5):788-793. doi:10.1377/hlthaff.2018.05149PubMedGoogle ScholarCrossref
12.
Centers for Medicare & Medicaid Services. The Health Care Fraud and Abuse Control Program protects consumers and taxpayers by combating health care fraud. https://www.cms.gov/newsroom/fact-sheets/health-care-fraud-and-abuse-control-program-protects-consumers-and-taxpayers-combating-health-care-0. Published January 18, 2017. Accessed September 13, 2019.
13.
Elixhauser  A, Steiner  C, Harris  DR, Coffey  RM.  Comorbidity measures for use with administrative data.  Med Care. 1998;36(1):8-27. doi:10.1097/00005650-199801000-00004PubMedGoogle ScholarCrossref
14.
Centers for Medicare & Medicaid Services. Medicare provider utilization and payment data: physician and other supplier. CMS.gov website. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Physician-and-Other-Supplier.html. Updated July 12, 2019. Accessed August 1, 2019.
15.
Centers for Medicare & Medicaid Services. Medicare Fee-For-Service 2014 improper payments report. https://www.cms.gov/Research-Statistics-Data-and-Systems/Monitoring-Programs/Medicare-FFS-Compliance-Programs/CERT/Downloads/MedicareFeeforService2014ImproperPaymentsReport.pdf. Published 2015. Accessed September 13, 2019.
16.
Office of Public Affairs, US Department of Justice. Justice Department recovers over $4.7 billion from False Claims Act cases in fiscal year 2016: third highest annual recovery in FCA history. https://www.justice.gov/opa/pr/justice-department-recovers-over-47-billion-false-claims-act-cases-fiscal-year-2016. Published December 14, 2016. Accessed September 13, 2019.
17.
Office of Public Affairs, US Department of Justice. National health care fraud takedown results in charges against over 412 individuals responsible for $1.3 billion in fraud losses. https://www.justice.gov/opa/pr/national-health-care-fraud-takedown-results-charges-against-over-412-individuals-responsible. Published July 13, 2017. Accessed September 13, 2019.
18.
Department of Health & Human Services; Department of Justice.  Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2016. Washington, DC: Office of the Inspector General; 2017.
19.
Medicare fraud whack-a-mole: false health care claims are huge—and spreading. The Economist. https://www.economist.com/united-states/2010/08/12/whack-a-mole. Published August 12, 2010. Accessed September 23, 2019.
20.
Bach  PB, Pham  HH, Schrag  D, Tate  RC, Hargraves  JL.  Primary care physicians who treat blacks and whites.  N Engl J Med. 2004;351(6):575-584. doi:10.1056/NEJMsa040609PubMedGoogle ScholarCrossref
21.
Culler  SD, Schieb  L, Casper  M, Nwaise  I, Yoon  PW.  Is there an association between quality of in-hospital cardiac care and proportion of low-income patients?  Med Care. 2010;48(3):273-278. doi:10.1097/MLR.0b013e3181c161baPubMedGoogle ScholarCrossref
22.
Dimick  J, Ruhter  J, Sarrazin  MV, Birkmeyer  JD.  Black patients more likely than whites to undergo surgery at low-quality hospitals in segregated regions.  Health Aff (Millwood). 2013;32(6):1046-1053. doi:10.1377/hlthaff.2011.1365PubMedGoogle ScholarCrossref
23.
Grabowski  DC.  The admission of blacks to high-deficiency nursing homes.  Med Care. 2004;42(5):456-464. doi:10.1097/01.mlr.0000124307.17380.dfPubMedGoogle ScholarCrossref
24.
Hasnain-Wynia  R, Baker  DW, Nerenz  D,  et al.  Disparities in health care are driven by where minority patients seek care: examination of the Hospital Quality Alliance measures.  Arch Intern Med. 2007;167(12):1233-1239. doi:10.1001/archinte.167.12.1233PubMedGoogle ScholarCrossref
25.
Jha  AK, Orav  EJ, Zheng  J, Epstein  AM.  The characteristics and performance of hospitals that care for elderly Hispanic Americans.  Health Aff (Millwood). 2008;27(2):528-537. doi:10.1377/hlthaff.27.2.528PubMedGoogle ScholarCrossref
26.
Rui  P, Okeyode  T. National Ambulatory Medical Care Survey: 2015 state and national summary tables. https://www.cdc.gov/nchs/data/ahcd/namcs_summary/2015_namcs_web_tables.pdf. Published 2015. Accessed September 14, 2019.
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Health Care Policy and Law
    October 28, 2019

    Association Between Treatment by Fraud and Abuse Perpetrators and Health Outcomes Among Medicare Beneficiaries

    Author Affiliations
    • 1Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
    • 2Johns Hopkins University School of Medicine, Baltimore, Maryland
    JAMA Intern Med. 2020;180(1):62-69. doi:10.1001/jamainternmed.2019.4771
    Key Points

    Question  Is receiving treatment from fraud and abuse perpetrators (FAPs) associated with adverse health events among Medicare beneficiaries?

    Findings  In this cross-sectional study of 8204 Medicare beneficiaries who received health care services from FAPs in 2013 and 296 298 who received care from non-FAPS, the 3-year, risk-adjusted mortality was 3.3 to 4.6 percentage points (13% to 23%) higher, and rates of emergency hospitalization in 2013 were 3.2 to 9.3 percentage points (11% to 30%) higher for FAP-treated patients compared with patients treated by nonperpetrators.

    Meaning  This study’s findings suggest that avoidance of Medicare FAPs may be associated with improvements in beneficiary health and longevity.

    Abstract

    Importance  Fraud and abuse contribute to unnecessary spending in the Medicare program, and federal agencies have prioritized fund recovery and the exclusion of health care practitioners who violate policy. However, the human costs of fraud and abuse in terms of patient health are unknown.

    Objective  To assess whether Medicare beneficiaries’ receipt of health care services from fraud and abuse perpetrators (FAPs) is associated with worse health outcomes.

    Design, Setting, and Participants  Retrospective cross-sectional study comparing mortality and emergency hospitalization rates of 8204 patients treated by an FAP with those among patients treated by a randomly selected non-FAP in 2013. Known FAPs were identified from the December 2018 List of Excluded Individuals/Entities (LEIE) published by the Office of the Inspector General in the Department of Health and Human Services. Patients were identified in a 5% sample of Medicare claims data and were enrolled in the Fee-for-Service program.

    Exposures  Treatment by a health care professional subsequently excluded from Medicare for fraud, patient harm, or a revoked license.

    Main Outcomes and Measures  All-cause mortality between 2013 and 2015 and 2013 emergency hospitalizations.

    Results  A total of 8204 Medicare beneficiaries in the study sample (mean [SD] age, 69.2 [14.2] years; 58.2% female, and 23.0% nonwhite) saw an FAP for the first time in 2013. Of these, 5054 (61.6%) were treated by fraud perpetrators, 1157 (14.1%) by patient harm perpetrators, and 1193 (24.3%) by revoked license perpetrators. Compared with 296 298 beneficiaries treated by non-FAPs (mean [SD] age, 71.1 [12.4] years; 58.6% female, and 16.5% nonwhite), beneficiaries exposed to an FAP were more likely to be eligible for both Medicare and Medicaid (34.7% [2845 of 8204] vs 21.9% [64 989 of 296 298]; P < .001) and more likely to be disabled at an age younger than 65 years (27.2% [2231 of 8204] vs 18.6% [55 168 of 296 298]; P < .001). All FAP exposures were associated with higher mortality and emergency hospitalization rates after risk adjustment and propensity score weighting: for mortality, exposures to fraud FAPs were associated with an increase of 4.58 percentage points (95% CI, 2.02-7.13; P < .001); to patient harm FAPs, with an increase of 3.34 percentage points (95% CI, 1.40-5.27; P = .001); and to revoked license FAPs, with an increase of 3.33 percentage points (95% CI, 1.58-5.09; P < .001). Increases were similar for emergency hospitalization rates: for fraud FAP exposures, 3.24 percentage points (95% CI, 0.01-6.46; P = .049); for patient harm FAP exposures, 9.34 percentage points (95% CI, 6.02-12.65; P < .001); and for revoked license FAP exposures, 9.28 percentage points (95% CI, 6.43-12.13; P < .001).

    Conclusions and Relevance  This study’s findings suggest that receiving medical care from FAPs may be associated with significantly higher rates of all-cause mortality and emergency hospitalization after risk adjustment. Identifying and permanently removing FAPs from the Medicare program may be associated with improved beneficiary health in addition to financial savings.

    Introduction

    Medicare fraud and abuse, defined broadly by the Centers for Medicare & Medicaid Services as “any practice, either directly or indirectly, that results in unnecessary costs to the Medicare program,”1 includes illicit activities such as billing for unnecessary health care services or for services never provided, providing medically inappropriate services, practicing without a license, and unlawfully distributing controlled substances. In 2018, more than 47 000 health care professionals, including physicians, nurses, and aides, were barred from participating in federal health care insurance programs because of fraudulent or abusive activities.2 These activities have been estimated to cost Medicare between $30 billion and $140 billion annually.3-6

    Although Medicare fraud and abuse are typically viewed by policy makers as financial problems, fraud and abuse perpetrators (FAPs) engage in a number of activities that could also harm patient health. A recent study found that 46% of the physicians excluded from federal health care insurance programs were excluded primarily for compromising patient health or safety.7 In addition, many crimes prosecuted as fraud because they involve billing Medicare for unnecessary or unsanctioned services also jeopardize patients’ well-being. Recent fraud cases included some resulting in the patient’s death after untrained workers read radiographs and failed to detect lethal findings; distribution of opioids and unsafe or counterfeit medications; and provision of lucrative but medically contraindicated procedures.8-10

    Fraud and abuse perpetrators disproportionately treat minority, disabled, and low-income Medicare beneficiaries, which may contribute to health outcome disparities; however, to our knowledge, no previous study has investigated the possibility of an association between exposure to FAPs and health care outcomes.11 In this study, we estimated the direct patient harm associated with health care services provided by professionals who are subsequently found to be FAPs. Using Medicare claims data to link health care professionals who had been excluded from Medicare and Medicaid programs to the Medicare beneficiaries they treated prior to their exclusion, we compared 3-year, all-cause mortality and emergency hospitalizations in this group of beneficiaries with 3-year, all-cause mortality and emergency hospitalizations among beneficiaries who were treated by non-FAPs. The study was approved by the institutional review board of the National Bureau of Economic Research. Because we used deidentified patient data, informed patient consent was not required.

    Methods
    Fraud and Abuse Perpetrators

    We identified known FAPs from the December 2018 List of Excluded Individuals/Entities (LEIE) published by the Office of the Inspector General in the Department of Health and Human Services. This list includes all health care professionals, facilities, and employees excluded from federal health care insurance programs from January 2012 to December 2018.2 We restricted our sample of FAPs to the 1364 health care professionals who billed Medicare under their own National Provider Identifier number prior to their exclusion for one or more of the following offenses classified as fraud and abuse: fraud (including fraud, theft, embezzlement, breach of fiduciary responsibility, and financial crimes); patient harm or improper drug dispensing (crimes related to health care delivery or meeting mental, physical, or emotional patient needs, including through controlled substance distribution); and practicing with a license that was revoked or suspended over concerns about competence, performance, or integrity.11,12 We included health care professionals whose crimes triggered either mandatory or permissive exclusion from Medicare. The LEIE lists only the crime against the federal programs, although there may be additional harms to beneficiaries. Many FAPs excluded for fraud were billing for illicitly acquired or unnecessary prescription drugs, using untrained and poorly paid staff to deliver medical care, and cycling patients unnecessarily through hospitals and nursing homes to maximize revenue. We did not include in our study categories of professionals, such as facility administrators, home health aides, and health care business owners, who cannot be matched to individual patients.

    Study Population

    We performed a retrospective cohort study using Medicare claims data from a 5% sample of Medicare beneficiaries enrolled in the Fee-for-Service (FFS) program with at least 1 claim in 2012 and 2013. We first used Medicare Carrier Claims to identify FFS beneficiaries treated at least once by a subsequently excluded FAP to quantify the magnitude of FAP exposure. Since it takes time for fraud and abuse to be detected and adjudicated, we matched health care professionals from the LEIE to earlier years of Medicare claims filed when they were still treating patients.

    To assess the association between exposure to FAPs and patient outcomes, we constructed a cohort of patients with at least 1 Medicare claim filed in 2013 and no claims filed in 2012 by a subsequently excluded FAP. This process provided a baseline year of data (2012) without FAP contact, allowing us to characterize the patients’ comorbidities. We compared this cohort with a group of beneficiaries (comparison group) who were treated by a clinician from a randomly selected pool of 25 000 clinicians who billed Medicare in 2013 and were not on the LEIE as of December 2018. To restrict our comparison group to Medicare beneficiaries who had the opportunity to be treated by an FAP, we drew our random sample from clinicians who practiced in counties with 1 or more practicing FAPs. Since beneficiaries treated by FAPs are seeking medical care, we reasoned that they might be sicker than the typical person enrolled in Medicare. Thus, we included in our comparison group only patients who also had 1 or more Medicare claims filed in 2013. We excluded beneficiaries from the comparison group if they were also treated by an FAP.

    Mortality, Emergency Hospitalizations, and Control Variables

    The primary outcome of interest was all-cause mortality between January 2013 and December 2015. We also compared the 2 cohorts’ rates of emergency hospitalization (ie, inpatient admission from the emergency department) in 2013 to assess whether FAP-exposed beneficiaries were more likely than non–FAP-exposed beneficiaries to experience serious health events that could lead to death. We controlled for factors that could alter mortality beyond treatment by an FAP and that might be confounders. We identified comorbidity indicators from diagnosis codes appearing in the 2012 Carrier Claims data according to the method of Elixhauser et al.13 We also constructed a categorical variable for the number of claims the beneficiary had in 2012, with categories defined by the quartile of the total number of claims since beneficiaries with a larger number of visits to a health care professional in 2012 would have encountered additional opportunities to have an FAP become part of their care team in 2013. We used the Medicare Denominator Files from 2012 to 2015 to identify each beneficiary’s age, sex, race, and vital status at the end of 2015; monthly enrollment status in Medicare Advantage; and 2012 characteristics, including dual (Medicare/Medicaid) eligibility, Social Security Disability Insurance receipt, end-stage renal disease status, and county of residence. We excluded beneficiaries younger than 18 years.

    Statistical Analysis

    We used ordinary least squares regression to estimate the associations between treatment by a subsequently excluded FAP in 2013, mortality between 2013 and 2015, and 2013 emergency hospitalizations. Our regression models included control variables for beneficiary age, sex, and nonwhite race; indicators for disability before age 65 years, dual eligibility, and end-stage renal disease status; the quartile of the number of claims in 2012; and interactions between age and sex, sex and race, disability and race, and comorbidities assessed from 2012 claims. In all models, we included county indicators and a set of indicator variables for each month from January 2012 to December 2013 indicating whether the beneficiary was enrolled in Medicare Part B. With the exception of 2013 FFS enrollment indicators, all covariates were measured in 2012 prior to potential FAP exposure. We estimated separate models for each of the 3 types of FAP exposure (fraud, patient harm, and licensing violation), with the comparison group including all Medicare beneficiaries in the analysis sample who were not treated by any type of FAP in 2013.

    In light of the strong sociodemographic differences between patients treated by FAPs and those treated by non-FAPs, we weighted all observations by the propensity to be treated by each type of FAP. We generated propensity scores by performing logistic regressions of 2013 FAP exposure on all of the control variables from our regression models, beyond county and measures of patient health and use of health care services in 2012, so that we could use the same weights in models comparing 2012 and 2013 health outcomes.

    We first estimated models with a single exposure variable capturing any treatment by an FAP in 2013. For a second set of regressions, we added an indicator for beneficiaries with 2 or more visits with any FAP to test the hypothesis that more frequent contact with an FAP is associated with a higher rate of harm. To assess whether the severity of the violation committed by the FAP was associated with patient health outcomes, we also compared FAPs whose crime was severe enough to trigger a mandatory exclusion with those whose violations left them subject to permissive exclusion by the Department of Health and Human Services. We considered a 2-sided P value of less than .05 to indicate statistical significance and clustered the SEs at the level of the sampling FAP or non-FAP. All data management and analysis were performed in Stata, version 14 (StataCorp LLC).

    We assessed the sensitivity of our results to the use of other outcome measures, including mortality between 2014 and 2015 only, alternative criteria for identifying the non–FAP-treated comparison group, excluding from both patient groups all beneficiaries enrolled in any Medicare Advantage plan in 2012 or 2013, including beneficiaries who were treated by 1 type of FAP in the unexposed group for the analyses of other FAP types, and logistic regression for the dichotomous outcome measure (see the eAppendix in the Supplement).

    Because FAP exposure might have been nonrandom, we conducted a falsification test to see whether there were differences in patient health and use of health care services across the 2 groups of beneficiaries in 2012, the year before they received services from an FAP or non-FAP health care professional. Specifically, we compared emergency hospitalizations and number of health care visits in 2012. If differences in health were responsible for observed differences in post–FAP exposure outcomes, those same health differences should be observable in pre–FAP exposure data as well.

    Results

    In our 5% sample, we identified 22 782 FFS-enrolled Medicare beneficiaries who were treated by 1272 subsequently excluded FAPs in 2012 and 18 355 who were treated by 1039 FAPs in 2013. Extrapolating from these results, we estimated that 455 640 Medicare beneficiaries were treated by FAPs in 2012 and 367 100 in 2013, with approximately 639 160 unique beneficiaries exposed over the 2-year period. Of the patients treated by FAPs, 5054 (61.6%) were treated by fraud perpetrators, 1157 (14.1%) by patient harm perpetrators, and 1193 (24.3%) by revoked license perpetrators. Fraud FAPs in our sample billed Medicare for more patient visits in 2013 than other health care professionals in our sample and had the highest mean billing per beneficiary. Compared with non-FAPs, all FAP types were more likely to be generalists (Table 1).

    After applying the sample restrictions, our cohort for the mortality and emergency hospitalization regression analyses included 8204 Medicare beneficiaries treated by 836 FAPs. Beneficiaries had a mean (SD) age of 69.2 (14.2) years; this cohort was 58.2% female (n = 4775) and 23.0% nonwhite (n = 1887) (Table 2). Our reference cohort included 296 298 beneficiaries who were not treated by known FAPs in 2012 or 2013. This group had a mean (SD) age of 71.1 (12.4) years, was 58.6% female [n = 173 631], and was 16.5% nonwhite (n = 48 889). Compared with beneficiaries treated by non-FAPs, those treated by FAPs were markedly more likely to be dually (Medicare/Medicaid) eligible (2845 [34.7%] vs 64 989 [21.9%]; P < .001) and disabled with age younger than 65 years (2231 [27.2%] vs 55 168 [18.6%]; P < .001). After propensity score weighting, there were no statistically significant differences in the sociodemographic characteristics of beneficiaries treated by FAPs vs those treated by non-FAPs (Table 2).

    Raw mortality and emergency hospitalization rates were considerably higher among those treated by FAPs (2013-2015 mortality, 25.0% vs 19.6% for those treated by non-FAPs; P < .001; 2013 emergency hospitalizations, 38.3% vs 30.1%; P < .001). Adjustments for sociodemographic differences and comorbidities identified in 2012 attenuated but did not eliminate the differences in mortality and emergency hospitalizations associated with FAP treatment (Figure 1 and eTable 1 in the Supplement as well as Figure 2 and eTable 2 in the Supplement). Mortality among patients exposed to FAP fraud was 17.3% higher (percentage point increase, 4.58; 95% CI, 2.02-7.13; P < .001); among those exposed to patient harm, 13.7% higher (percentage point increase, 3.34; 95% CI, 1.40-5.27; P = .001); and among those exposed to license violations, 14.8% higher (percentage point increase, 3.33; 95% CI, 1.58-5.09; P < .001) (Figure 1). Mortality differences were more pronounced after contact with FAPs excluded for acts with mandatory vs permissive exclusion. We observed a similar association with emergency hospitalizations in 2013: percentage point differences of 3.24 (95% CI, 0.01-6.46; P = .049) after exposure to health care professionals later excluded for fraud, 9.34 (95% CI, 6.02-12.65; P < .001) after exposure to those excluded for patient harm, and 9.28 (95% CI, 6.43-12.13; P < .001) after exposure to those excluded because their licenses were revoked (Figure 2).

    Having multiple FAP exposures was not associated with additional differences in mortality beyond those observed among beneficiaries with a single exposure, but it was associated with lower rates of emergency hospitalization for patients of FAPs excluded for fraud (Table 3). These results might reflect financial crimes or the inability of sicker beneficiaries to return to an FAP.

    Our results were not sensitive to alterations in model specifications, control variables, or definitions of FAP and non-FAP exposures. Our estimates of association between FAP exposure and subsequent mortality and emergency hospitalization rates were consistent across all of the robustness checks previously described (eTable 3 in the Supplement). In addition, we found no association between beneficiary exposure in 2013 to health care professionals later excluded for fraud or a revoked license and 2012 emergency hospitalization and fewer 2012 claims for fraud- and patient harm–exposed beneficiaries, suggesting that the post–FAP-exposure mortality and emergency hospitalization results are not advanced by unmeasured differences in beneficiaries’ pre–FAP exposure health (eTable 4 in the Supplement). Although 2012 emergency hospitalization rates were higher for beneficiaries exposed in 2013 to FAPs excluded for patient harm, the difference was much smaller in magnitude before than after FAP exposure (2.9 percentage points vs 9.35 percentage points, P < .001).

    Based on our mortality models, we estimate that exposure to known FAPs may have contributed to an additional 6700 premature deaths among Medicare beneficiaries enrolled in the FFS program in 2013. These deaths represent a loss of 98 500 life-years (95% CI, 44 000-153 500 life-years), with an estimated value of $4.9 billion to $14.8 billion, using a minimum valuation of $50 000 per life-year to a maximum valuation of $150 000 (eTable 5 in the Supplement).

    Discussion

    The results of this quantitative assessment of patient harms associated with Medicare fraud and abuse suggest that Medicare beneficiaries treated by known FAPs may be more likely to die and more likely to experience emergency hospitalization within 3 years after this exposure than a comparison group of beneficiaries who did not have contact with FAPs. Fraud and abuse perpetrators treated nearly 1 million Medicare beneficiaries annually in recent years. The exposed beneficiaries were more likely to be nonwhite, dually eligible, and disabled than those treated by FAPs, although these demographic differences did not explain the mortality gap.

    There are 2 possible interpretations of our results. The first is that FAPs engage in activities that directly jeopardize patient health and lead to premature mortality among the patients they treat. This explanation is consistent with numerous Federal Bureau of Investigation and Department of Justice reports of FAP crimes that investigators directly linked to patient deaths, such as using untrained staff to perform medical duties, falsifying medical records with incorrect diagnoses, cycling patients across facilities to maximize reimbursement, and illegally distributing prescription medication and counterfeit chemotherapy.15-19

    However, our study was observational, and we cannot determine whether the observed mortality differences are attributable to FAP behavior, unobserved characteristics of Medicare beneficiaries treated by FAPs, or a combination of these factors. Although we used propensity score weights that minimized differences in observable characteristics across FAP-exposed and nonexposed beneficiaries and restricted our comparisons to beneficiaries living in geographic proximity to FAPs, there may be additional unobserved differences between the 2 groups that explain the mortality differences. For example, low-income patients with food insecurity might seek out FAPs who provide financial kickbacks for the use of their Medicare number for sham billing purposes or who buy back diabetic test strips for resale. In this scenario, the mortality differences we found could reflect sick and vulnerable Medicare beneficiaries clustering around FAPs who involve beneficiaries in illicit activities. The negligible differences in 2012 emergency hospitalizations and office visits suggest that any unobserved factor that explains the differences in postexposure health would have had to emerge around the same time as the FAP exposure, which is unlikely.

    Both interpretations are concerning, particularly given the known actions of FAPs that directly compromised patient health and safety. Our findings suggest that the benefits to the patient and to society of removing FAPs from practice may be greater than previously appreciated given the possibility of averting hospitalizations and deaths. Furthermore, coordination between the Centers for Medicare & Medicaid Services and investigative agencies could lead to strategies to identify Medicare beneficiaries requiring additional medical or social service intervention.

    Our study contributes to a growing body of work highlighting the association of patients’ choice of health care professionals with disparities in health outcomes.20-25 Previous research has shown that racial and ethnic minority populations and low-income patients are more likely to use lower-performing health care professionals, hospitals, and nursing homes across a range of quality metrics. The results of our study suggest that nonwhite, dually eligible, and disabled Medicare beneficiaries also are more likely to be treated by health care professionals whose actions lead to Medicare exclusion and possibly more likely to be harmed by this exposure.

    Limitations

    Our study has several limitations. We were able to classify health care professionals as FAPs only if they had been successfully identified and excluded from Medicare programs. Thus, our sample of non-FAP health care professionals likely includes some FAPs who had not yet been detected. Because their patients would have been included in our comparison group, our results might be biased against an association between FAP exposure and mortality, resulting in a lower bound estimate. Second, our classification of FAP types relies on the LEIE designation. Although this designation reflects the underlying conduct of health care professionals, it may also be a result of prosecutorial priorities. Third, our patient sample was limited to Medicare FFS beneficiaries, although FAPs likely treat patients with other sources of insurance as well. However, FFS patients accounted for nearly 30% of outpatient visits in the United States in 2015, and our results would still represent nontrivial harms associated with fraud and abuse if no other patients were treated by FAPs.26 Last, additional harms may have been caused by nonclinical staff who also were excluded for fraud and abuse but who could not be directly linked to the patients in their care. Thus, our results represent a lower bound estimate of the health outcomes associated with exposure to fraud and abuse.

    The observational nature of our study precludes the identification of causal effects of FAP exposure. However, our baseline data suggest that both groups of beneficiaries had similar use of health care services prior to FAP exposure. Risk-adjusted mortality rates were highest among beneficiaries with multiple FAP visits, and emergency hospitalization rates sharply increased following FAP exposure.

    Conclusions

    This study’s findings suggest that receiving medical care from a health care professional subsequently excluded from Medicare may be associated with significantly higher rates of all-cause mortality and emergency hospitalization. Identifying and permanently removing FAPs from the Medicare program may be associated with improved beneficiary health in addition to financial savings.

    Back to top
    Article Information

    Accepted for Publication: August 27, 2019.

    Corresponding Author: Lauren Hersch Nicholas, PhD, MPP, Johns Hopkins Bloomberg School of Public Health, 624 N Broadway, Room 450, Baltimore, MD 21205 (lauren.nicholas@jhu.edu).

    Published Online: October 28, 2019. doi:10.1001/jamainternmed.2019.4771

    Author Contributions: Drs Nicholas and Hanson had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: All authors.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Nicholas, Hanson.

    Critical revision of the manuscript for important intellectual content: Nicholas, Segal, Eisenberg.

    Statistical analysis: Nicholas, Hanson, Eisenberg.

    Obtained funding: Nicholas.

    Administrative, technical, or material support: Nicholas.

    Supervision: Nicholas, Segal.

    Conflict of Interest Disclosures: Drs Nicholas, Hanson, and Eisenberg reported receiving a grant from the Social Security Administration (SSA) Disability Research Consortium, and Dr Nicholas reported receiving a grant from the National Institute on Aging. No other disclosures were reported.

    Funding/Support: This research was partially supported by grant 5-DRC12000002-06 from the SSA to the National Bureau of Economic Research as a participant in the SSA Disability Research Consortium.

    Role of the Funder/Sponsor: The SSA had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Disclaimer: The findings and conclusions expressed are solely those of the authors and do not represent the views of the SSA, any other agency of the federal government, or the National Bureau of Economic Research.

    References
    1.
    Centers for Medicare & Medicaid Services; Medicare Learning Network.  Medicare Fraud & Abuse: Prevent, Detect, Report. Baltimore, MD: US Dept of Health & Human Services; February 2019. ICN MLN4649244. https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNProducts/Downloads/Fraud-Abuse-MLN4649244.pdf. Accessed September 30, 2019.
    2.
    Office of the Inspector General, Department of Health and Human Services.  12-2018 Updated LEIE Database. Washington, DC: US Dept of Health & Human Services; 2018.
    3.
    Berwick  DM, Hackbarth  AD.  Eliminating waste in US health care.  JAMA. 2012;307(14):1513-1516. doi:10.1001/jama.2012.362PubMedGoogle ScholarCrossref
    4.
    Goldberg  SR, Lindquist  SC.  Fighting costly health care fraud.  J Corp Account Finance. 2005;16(4):29-34. doi:10.1002/jcaf.20115Google ScholarCrossref
    5.
    Kalb  PE.  Health care fraud and abuse.  JAMA. 1999;282(12):1163-1168. doi:10.1001/jama.282.12.1163PubMedGoogle ScholarCrossref
    6.
    Morris  L.  Combating fraud in health care: an essential component of any cost containment strategy.  Health Aff (Millwood). 2009;28(5):1351-1356. doi:10.1377/hlthaff.28.5.1351PubMedGoogle ScholarCrossref
    7.
    Burton  BR.  A Demographic Portrait of Physicians Sanctioned by the Federal Government in the United States [dissertation]. Irvine: University of California; 2016. https://escholarship.org/uc/item/1022m81j. Accessed September 13, 2019.
    8.
    McDaniels  AK. Owings Mills imaging firm owner guilty of fraud. Baltimore Sun. 2016:A3.
    9.
    Rhodes  D. Doctor pleads guilty in fraud case. Chicago Tribune. September 11, 2014:33-35.
    10.
    Stewart  CS, Carreyrou  J. Government joins suits against cardiologist. Wall Street Journal. January 6, 2015:A3.
    11.
    Nicholas  LH, Segal  J, Hanson  C, Zhang  K, Eisenberg  MD.  Medicare beneficiaries’ exposure to fraud and abuse perpetrators.  Health Aff (Millwood). 2019;38(5):788-793. doi:10.1377/hlthaff.2018.05149PubMedGoogle ScholarCrossref
    12.
    Centers for Medicare & Medicaid Services. The Health Care Fraud and Abuse Control Program protects consumers and taxpayers by combating health care fraud. https://www.cms.gov/newsroom/fact-sheets/health-care-fraud-and-abuse-control-program-protects-consumers-and-taxpayers-combating-health-care-0. Published January 18, 2017. Accessed September 13, 2019.
    13.
    Elixhauser  A, Steiner  C, Harris  DR, Coffey  RM.  Comorbidity measures for use with administrative data.  Med Care. 1998;36(1):8-27. doi:10.1097/00005650-199801000-00004PubMedGoogle ScholarCrossref
    14.
    Centers for Medicare & Medicaid Services. Medicare provider utilization and payment data: physician and other supplier. CMS.gov website. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Physician-and-Other-Supplier.html. Updated July 12, 2019. Accessed August 1, 2019.
    15.
    Centers for Medicare & Medicaid Services. Medicare Fee-For-Service 2014 improper payments report. https://www.cms.gov/Research-Statistics-Data-and-Systems/Monitoring-Programs/Medicare-FFS-Compliance-Programs/CERT/Downloads/MedicareFeeforService2014ImproperPaymentsReport.pdf. Published 2015. Accessed September 13, 2019.
    16.
    Office of Public Affairs, US Department of Justice. Justice Department recovers over $4.7 billion from False Claims Act cases in fiscal year 2016: third highest annual recovery in FCA history. https://www.justice.gov/opa/pr/justice-department-recovers-over-47-billion-false-claims-act-cases-fiscal-year-2016. Published December 14, 2016. Accessed September 13, 2019.
    17.
    Office of Public Affairs, US Department of Justice. National health care fraud takedown results in charges against over 412 individuals responsible for $1.3 billion in fraud losses. https://www.justice.gov/opa/pr/national-health-care-fraud-takedown-results-charges-against-over-412-individuals-responsible. Published July 13, 2017. Accessed September 13, 2019.
    18.
    Department of Health & Human Services; Department of Justice.  Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2016. Washington, DC: Office of the Inspector General; 2017.
    19.
    Medicare fraud whack-a-mole: false health care claims are huge—and spreading. The Economist. https://www.economist.com/united-states/2010/08/12/whack-a-mole. Published August 12, 2010. Accessed September 23, 2019.
    20.
    Bach  PB, Pham  HH, Schrag  D, Tate  RC, Hargraves  JL.  Primary care physicians who treat blacks and whites.  N Engl J Med. 2004;351(6):575-584. doi:10.1056/NEJMsa040609PubMedGoogle ScholarCrossref
    21.
    Culler  SD, Schieb  L, Casper  M, Nwaise  I, Yoon  PW.  Is there an association between quality of in-hospital cardiac care and proportion of low-income patients?  Med Care. 2010;48(3):273-278. doi:10.1097/MLR.0b013e3181c161baPubMedGoogle ScholarCrossref
    22.
    Dimick  J, Ruhter  J, Sarrazin  MV, Birkmeyer  JD.  Black patients more likely than whites to undergo surgery at low-quality hospitals in segregated regions.  Health Aff (Millwood). 2013;32(6):1046-1053. doi:10.1377/hlthaff.2011.1365PubMedGoogle ScholarCrossref
    23.
    Grabowski  DC.  The admission of blacks to high-deficiency nursing homes.  Med Care. 2004;42(5):456-464. doi:10.1097/01.mlr.0000124307.17380.dfPubMedGoogle ScholarCrossref
    24.
    Hasnain-Wynia  R, Baker  DW, Nerenz  D,  et al.  Disparities in health care are driven by where minority patients seek care: examination of the Hospital Quality Alliance measures.  Arch Intern Med. 2007;167(12):1233-1239. doi:10.1001/archinte.167.12.1233PubMedGoogle ScholarCrossref
    25.
    Jha  AK, Orav  EJ, Zheng  J, Epstein  AM.  The characteristics and performance of hospitals that care for elderly Hispanic Americans.  Health Aff (Millwood). 2008;27(2):528-537. doi:10.1377/hlthaff.27.2.528PubMedGoogle ScholarCrossref
    26.
    Rui  P, Okeyode  T. National Ambulatory Medical Care Survey: 2015 state and national summary tables. https://www.cdc.gov/nchs/data/ahcd/namcs_summary/2015_namcs_web_tables.pdf. Published 2015. Accessed September 14, 2019.
    ×