| |
Medicare Spending For Injured Elders: Are There Opportunities For Savings?
Christine E. Bishop,
Daniel Gilden,
Jacobus Blom,
Joanna Kubisiak,
Rosemarie Hakim,
Angelina Lee and
Deborah W. Garnick
Claims for injury care provided to aged fee-for-service (FFS) beneficiaries cost Medicare more than $8 billion in 1999, almost 6 percent of Medicare claims spending for elders. More than one-fifth of aged FFS beneficiaries had an injury that resulted in a claim. Fractures, which were experienced by one in seventeen aged beneficiaries, were responsible for 67 percent of total injury claims expenses. Medicare could realize substantial savings if these injuries could be prevented; the program should consider underwriting effective prevention activities.
Injuries are a major cause of morbidity and mortality for elderly Americans. The death rate from unintentional injury almost triples with age, from around 32 per 100,000 population for adults ages 25–64 to 93.3 per 100,000 population for persons age sixty-five and older.1 Persons age seventy-five and older visited emergency departments for injury-related care at a rate of 17.5 per 100 in 2000 (the populationwide rate was 14.8 per 100).2 Using different data sources and methods, studies of medical spending have consistently found higher spending per capita for injuries to persons age sixty-five and older than for other groups.3
Here we present the first comprehensive analysis of Medicare spending for injuries among the elderly.4 The cost to the Medicare program of injury-related health services claims provides an estimate of how much Medicare might save if all injuries could be prevented. Because of the potential preventability of much injury-related morbidity, this cost makes a dollars-and-cents argument for supporting injury prevention. The analysis also suggests that Medicare claims data could be used to identify factors associated with injury for the elderly, and further analyses could allow better targeting of prevention strategies.
We based the estimates of Medicare spending for injuries in 1999 on all paid claims from a 5 percent sample of Medicare beneficiaries who were at least age sixty-five by the end of 1999 and who had at least one month of Medicare eligibility within the year. Paid claims were obtained from Medicare’s institutional claims files (inpatient, skilled nursing facility, hospice, home health, and out-patient) and professional services/supplies claims files (physician/supplier and durable medical equipment). The Barell framework, developed by the Centers for Disease Control and Prevention (CDC), uses International Classification of Diseases, Ninth Revision (ICD-9) codes to classify injuries by nature and site of injury.5 Following a modified version of this framework, Medicare claims with ICD-9 codes in the 800–995 range were used to identify beneficiaries experiencing an injury and to calculate expenditures for each category.6 Injury claims were included in the calculations only if the identifying injury diagnosis was submitted as a principal diagnosis on an institutional claim or a line-item diagnosis on a professional services/supplies claim.7
We computed the unduplicated count of beneficiaries contributing claims to each spending total and subtotal. This was necessary because many injured beneficiaries used multiple Medicare services as a result of the same injury, and some may have had several different types of injuries during 1999.8 The numerators for the population-based rates are beneficiaries with claims during the year; thus, they reflect the annual prevalence of injury claims rather than incidence of injuries, since some claims could result from continuing care for pre-1999 injuries.9
We also computed the total 1999 fee-for-service (FFS) claims spending for beneficiaries in the 5 percent sample who were not enrolled in a health maintenance organization (HMO) in January 1999, and we estimated the number of these beneficiaries from Medicare’s enrollment files. These estimates formed the denominators of our computation of injury claims expense as a proportion of total FFS spending and for our calculation of beneficiaries having an injury claim as a proportion of all FFS beneficiaries. All spending and beneficiary totals from the 5 percent sample were multiplied by 20 to estimate the total annual Medicare claims spending for injuries, by type, and beneficiaries with injury claims, by type.10
|
Medicare Spending For Injuries Of Elderly Beneficiaries
|
|---|
Spending and prevalence of injuries by age.
Claims for injury care provided to aged FFS beneficiaries cost Medicare slightly more than $8 billion in 1999, almost 6 percent of Medicare claims spending for elders (Exhibit 1 ). Slightly more than one-fifth of aged FFS Medicare beneficiaries, more than six million people, had at least one injury claim in 1999. The average expenditure per beneficiary with an injury claim was $1,272, and this increased with age (from $760 for males ages 65–74 to $1,884 for males age 85 and older). Expenditure per injured female beneficiary increased more sharply with age.
View this table:
[in this window]
[in a new window]
|
EXHIBIT 1 Medicare Injury Claims, Expenditures, And Fee-For-Service (FFS) Beneficiaries Affected, By Age And Sex, 1999
|
|
The proportion of beneficiaries with an injury claim in 1999 increased with age (Exhibit 1). Thirty-one percent of beneficiaries in the oldest age category had an injury resulting in a Medicare claim in 1999; a higher proportion of female beneficiaries than males in each age range experienced an injury.
View this table:
[in this window]
[in a new window]
|
EXHIBIT 2 Total Medicare Expenditures, Millions Of Dollars, By Injury Type, Among Aged Fee-For-Service Beneficiaries, 1999
|
|
Spending and prevalence of injuries by diagnosis class.
Fractures, the injury diagnosis category most costly to Medicare, resulted in total claims of $5.5 billion, 67 percent of all injury claims spending (Exhibits 2 and 3). The second-largest diagnosis category was open wounds ($873 million), followed by internal injuries ($503 million).11 Sprains and strains, a minor but common injury, resulted in Medicare expenditures of $323 million.
View larger version (10K):
[in this window]
[in a new window]
|
EXHIBIT 3 Proportion Of Medicare Spending For Various Major Injuries Among Aged Fee-For-Service Beneficiaries, By Injury Type, 1999
|
|
The fracture diagnosis category was the most common injury class (Exhibit 4). Six percent of all aged FFS Medicare beneficiaries incurred a claim for a fracture in 1999, representing almost 1.8 million people. The second most common injury diagnosis class was contusions and superficial injuries, followed by sprains and strains. The distribution of injury claims by diagnosis class differed for men and women. More than 7.6 percent of all female Medicare FFS beneficiaries had an injury claim for a fracture in 1999, while only 3.6 percent of their male counterparts did so. The most common treated injury among men was an open wound, followed by contusions and sprains/strains.
View larger version (17K):
[in this window]
[in a new window]
|
EXHIBIT 4 Proportion Of Aged Fee-For-Service Medicare Beneficiaries With An Injury Claim, By Diagnosis Category And Sex, 1999
|
|
A comparison of average spending by diagnosis category for men and women reveals that women had higher average Medicare spending for fractures than men had. Although fractures were also the most costly injuries among male beneficiaries, men had higher average spending than women had for internal injuries, burns, nerve injuries, and blood vessel injuries (Exhibit 5).
View larger version (16K):
[in this window]
[in a new window]
|
EXHIBIT 5 Average Claims Expenditures For Aged Fee-For-Service Medicare Beneficiaries With An Injury Claim, By Diagnosis Category And Sex, 1999
|
|
Spending and prevalence for fractures, by type.
The higher average claims spending for women experiencing fractures ($3,192 for women versus $2,824 for men) was attributable in part to women’s disproportionate number of expensive lower extremity fractures rather than to higher expenses within diagnostic categories. More than half (52 percent) of women with Medicare claims for fractures in 1999 had lower extremity fractures, while only 46 percent of their male peers had this type of fracture. The 1999 injury-related claims per beneficiary with a lower extremity fracture were costly but somewhat lower for men than for women, averaging $4,477 for men and $4,685 for women (Exhibit 6). Fractures resulting in traumatic brain injuries were also very expensive, but much more so on average for men than for women. However, this injury is rare, affecting only 6,900 men and 10,960 women in 1999 (prevalence rate, 0.06 percent of FFS beneficiaries). Fractures resulting in spinal cord injury and vertebral fractures also resulted in higher Medicare claims expense per injured male beneficiary than per injured female beneficiary.
View larger version (19K):
[in this window]
[in a new window]
|
EXHIBIT 6 Medicare Claims Expenditures For Fractures Per Injured Aged Fee-For-Service Beneficiary, By Diagnostic Category And Sex, 1999
|
|
|
Discussion And Policy Implications
|
|---|
Injuries, especially fractures, account for a substantial proportion of Medicare expenditures for persons age sixty-five and older.12 Although differing in approach and amounts, the estimates presented here generally corroborate the findings of others with respect to the magnitude of claims expenses related to fractures.13 Information on Medicare spending and prevalence for this comprehensive list of injury diagnosis categories has not been available before now. Injury surveillance has often used mortality statistics as the best available source of information on injuries. Many epidemiological studies of injuries requiring treatment but not resulting in death have relied on hospitalization rates and emergency room visits reported by participating hospitals.14 However, elders (and others) experience injuries that do not result in hospitalization and are not necessarily treated in emergency rooms. For example, large numbers of Medicare beneficiaries receive covered treatment for contusions, sprains, and strains. The sheer volume of these injuries aggregates to a meaningful expense—more than a half-billion dollars in 1999. While most of these injuries are likely minor, the Medicare claims data suggest opportunities for prevention and management that are obscured in mortality and hospitalization statistics.
Medicare spending, along with the other costs related to these injuries, could be reduced through prevention activities. Although gerontologists and other practitioners have designed and demonstrated effective interventions to reduce the risk of injury in the elderly, these interventions are not widely disseminated. A barrier to diffusion is that prevention activities often use resources and strategies that are outside the purview of the personal health care sector as it is usually defined, so that they are not covered by insurance. For example, home assessment and modifications, assistive devices for tasks of daily living, gait and strength training, and protective devices can prevent or mitigate falls; special programs for older drivers can reduce injuries to pedestrians, passengers, and drivers; patient education can increase medication safety; and smoke alarms and safer nightwear can reduce burn injuries.15 Because of the high prevalence and cost of fractures, interventions targeted to beneficiaries at high risk of fracture could have a substantial return. Medicare coverage of bone mass measurement to screen for osteoporosis was standardized in the Balanced Budget Act (BBA) of 1997, but FFS Medicare does not cover the prescription drugs, nutrition counseling, and strength training that could increase bone mass and reduce the risk of fracture.16
The cost to Medicare of health services for injured beneficiaries reported here is only part of the total cost of injuries for the elderly. These expenditures do not include the costs of long-term disability to Medicare beneficiaries who are seriously injured. Depending on the cost of prevention, information on these insured health care expenses could assist in making a purely economic case for expanding health insurance coverage for preventive services. Estimates of Medicare injury expenditures to the exclusion of costs borne by other parties (the injured beneficiary, his or her family, other payers) focus attention on costs that are the responsibility of one entity, Medicare. These potential cost offsets provide support for considering coverage and other policies to foster injury prevention, as has been done in legislation recently introduced in the Senate.17
In addition, information available in claims data concerning diagnoses, sex, age, season, residence location, and other factors could support targeting of injury prevention efforts. Better coding of injury causes on claims could support further refinements, highlighting the roles of individual characteristics, medical conditions, and environmental hazards.18 Further research could investigate factors related to injury incidence and expense using data available in Medicare claims and enrollment records. Medicare savings would be substantial if even half of spending attributable to injury could be avoided or mitigated through prevention.
Christine Bishop is a professor at the Schneider Institute for Health Policy, the Heller School, Brandeis University. Daniel Gilden is president of JEN Associates Inc. Jacobus Blom is a senior physical therapist at Gentiva Health Services. Joanna Kubisiak is a programmer/analyst at JEN Associates. Rosemarie Hakim is a social science research analyst at the Centers for Medicare and Medicaid Services. Angelina Lee is a senior programmer analyst at JEN Associates. Deborah Garnick is a professor at the Schneider Institute.
Support for this paper was provided by the Centers for Medicare and Medicaid Services (CMS) (Contract no. 500-95-0060/Task Order no. 4). The views expressed in this study are those of the authors and do not reflect those of the Department of Health and Human Services or the CMS. The authors appreciate useful input from consultants Gerry Berenholz, Douglas Kiel, and Robert Sege.
- R.N. Anderson, "Deaths: Leading Causes for 1999," National Vital Statistics Reports 49, no. 11 (Hyattsville, Md.: National Center for Health Statistics, 2001).
- L.F. McCaig and N. Ly, National Ambulatory Medical Care Survey: 2000 Emergency Department Summary, Advance Data 326 (Hyattsville, Md.: NCHS, 2002).
- D.P. Rice and E.J. MacKenzie, The Cost of Injury in the United States: A Report to Congress (San Francisco: University of California and Johns Hopkins University, 1989); and T.R. Miller and D.C. Lestina, "Patterns in U.S. Medical Expenditures and Utilization for Injury, 1987," American Journal of Public Health 86, no. 1 (1996): 89–93.[Abstract/Free Full Text]
- The Centers for Disease Control and Prevention carried out a similar analysis for Medicare claims for the year July 1991–June 1992, for fractures only. See CDC, "Incidence and Costs to Medicare of Fractures among Medicare Beneficiaries Aged Greater than or Equal to 65 Years—United States, July 1991–June 1992," Morbidity and Mortality Weekly Report 45, no. 41 (1996): 877–883.
- CDC, "The Barell Injury Diagnosis Matrix: Classification by Region of Body and Nature of Injury," 15 September 2001, www.cdc.gov/nchs/data/ice/barellmatrix.pdf (28 September 2001).
- We modified the Barell matrix to show detail of traumatic brain injury, in two Barell categories, and to differentiate lower and upper extremity fractures and sprains/strains. A small number of injury-related claims are known to be found in mixed categories outside the 800–995 ICD-9 range, but these are not included in the Barell matrix and could not be accurately identified. See Miller and Lestina, "Patterns in U.S. Medical Expenditures and Utilization for Injury;"; and D. Baugh et al., "Hospitalizations for Injury among Medicaid Children: California 1992," Health Care Financing Review 19, no. 4 (1998): 129–147.
- Medicare claims for professional services usually include multiple services, with a designated diagnosis for each line item. The file included a small number of hospital "shadow" claims with injury diagnoses for beneficiaries enrolled in Medicare HMOs. These claims cause a slight overstatement of the number of beneficiaries experiencing an injury in 1999 but do not affect spending because they are not paid claims.
- The total number of beneficiaries with an injury claim is thus less than the sum of the beneficiaries with a claim in each diagnosis category. Similarly, the unduplicated count of beneficiaries with a fracture claim in 1999 is less than the sum of beneficiaries with each type of fracture. This occurs because some beneficiaries experienced several different types of fractures during 1999.
- The results here are based on larger tables produced by the authors for this project. See "Injuries among Medicare Beneficiaries: Expenditures and Beneficiary Characteristics, 1997–1999," www.cms.hhs.gov/researchers/projects/injuriestab5-02.pdf (7 August 2002). The tables show Medicare expenditures and counts of beneficiaries with injury claims by sex and age for 1997–1999 by region, race, urban/rural location, long-term care status (community resident, long-stay Medicare home health, long-stay nursing home resident), and service category (inpatient, postacute, physician/practitioner, outpatient/ER, other).
- Reported totals include neither Medicare spending attributable to injury services included in Medicare HMO premiums, nor claims for the very small number of beneficiaries with incomplete enrollment records. Beneficiaries who do not seek Medicare-covered treatment for injuries are not included in the counts of injured beneficiaries.
- The cost and prevalence of open wounds from involuntary injury may be overstated because the diagnostic codes are sometimes used by home health agencies caring for beneficiaries with surgical wounds.
- The expenditures reported here, claims paid in one year (1999) with a specific injury diagnosis, encompass only the Medicare-insured portion of the medical costs of injuries. Additional direct medical costs include claims that also occur during the episode of care for an injury but are not labeled with an injury code as well as the cost of services not paid for by Medicare (coinsurance and deductibles and nursing home and other services paid for by Medicaid or private supplemental insurance or by beneficiaries themselves). Other costs include costs of donated family care, pain and suffering, lost earnings, and long-term consequences of loss of independence due to injury or fear of further injury. See Rice and MacKenzie, The Cost of Injury in the United States; L.C. Harlan, W.R. Harlan, and P.E. Parsons, "The Economic Impact of Injuries: A Major Source of Medical Costs," American Journal of Public Health 80, no. 4 (1990): 453–459[Abstract/Free Full Text]; Miller and Lestina, "Patterns in U.S. Medical Expenditures and Utilization for Injury";; and D.P. Rice and W. Max, "The High Cost of Injuries in the United States," American Journal of Public Health 86, no. 1 (1996): 14–15.[Free Full Text]
- Another approach is represented by the CDC estimate of $ 4.2 billion in FY 1992; see CDC, "Incidence and Costs to Medicare of Fractures." Their estimates of cost to Medicare did not aggregate claims with a fracture diagnosis but rather examined the difference in total Medicare claims for a beneficiary experiencing a fracture, comparing the six months before and after the fracture. The estimate reported for hip fracture is $15,294, much more than the mean direct Medicare claims expense for lower extremity fractures reported here ($4,638). The diagnostic category used here includes fractures of the ankle, foot, and leg, likely to be less costly than hip fractures. But our estimate is also more conservative in including only claims that are explicitly identified as injury-related. Ada Brainsky and colleagues included all direct medical care costs, instead of restricting analysis to Medicare claims, for hip fractures for 759 community-dwelling elders, finding costs of $16,300–$17,700 (1993 dollars); see A. Brainsky et al., "The Economic Cost of Hip Fractures in Community-Dwelling Older Adults: A Prospective Study," Journal of the American Geriatrics Society 45, no. 3 (1997): 281–287. In an estimate again including all medical care resources, total direct cost of fall injuries to the elderly was estimated at $22.2 billion for 1994[Medline]; see CDC, "The Costs of Fall Injuries among Older Adults," www.cdc.gov/ncipc/factsheets/fallcost.htm (7 August 2002), citing F. Englander, T.J. Hodson, and R.A. Terregrossa, "Economic Dimensions of Slip and Fall Injuries," Journal of Forensic Sciences 41, no. 5 (1996): 733–746. This estimate cannot be compared with the estimate presented here, which refers to injury diagnosis classes and includes only Medicare expenditures.[Medline]
- K.P. Quinlan et al., "Expanding the National Electronic Injury Surveillance System to Monitor All Nonfatal Injuries Treated in U.S. Hospital Emergency Departments," Annals of Emergency Medicine 34, no. 5 (1999): 637–645[Medline]; and J.A. Stevens et al., "Surveillance for Injuries and Violence among Older Adults," Morbidity and Mortality Weekly Report 48, no. SS08 (1999): 27–50.
- M. Pfeifer and H.W. Minne, "Vitamin D and Hip Fracture," Trends in Endocrinology and Metabolism 10, no. 10 (1999): 417–420[Medline]; B.A. Kumar and M.J. Parker, "Are Hip Protectors Cost Effective?" Injury 31, no. 9 (2000): 693–695[Medline]; E.K. Parra and J.A. Stevens, U.S. Fall Prevention Programs for Seniors: Selected Programs Using Home Assessment and Home Modification (Atlanta: CDC, National Center for Injury Prevention and Control, 2000); L. Rubenstein, "Hip Protectors—A Breakthrough in Fracture Prevention," New England Journal of Medicine 343, no. 21 (2000): 1562–1563[Free Full Text]; P. Vestergaard, L. Rejnmark, and L. Mosekilde, "Hip Fracture Prevention: Cost-Effective Strategies," Pharmacoeconomics 19, no. 5 (2001), Part 1: 449–468[Medline]; AARP, Older Driver Skill Assessment and Resource Guide: Creating Mobility Choices (Washington: AARP, 1992); W.J. Millar, "Older Drivers—A Complex Public Health Issue," Health Reports 11, no. 2 (1999): 59–71 (English), 67–82 (French)[Medline]; CDC, "Motor Vehicle–Related Deaths among Older Americans Fact Sheet," www.cdc.gov/ncipc/factsheets/older.htm (12 December 2000); M.B. Haselberger and B.A. Kroner, "Drug Poisoning in Older Patients: Preventative and Management Strategies," Drugs and Aging 7, no. 4 (1995): 292–297[Medline]; A.T. Elder, T. Squires, and A. Busuttil, "Fire Fatalities in Elderly People," Age and Ageing 25, no. 3 (1996): 214–216[Abstract/Free Full Text]; I. Roberts, "Smoke Alarm Use: Prevalence and Household Predictors," Injury Prevention 2, no. 4 (1996): 263–265[Abstract/Free Full Text]; C.M. Ryan et al., "A Persistent Fire Hazard for Older Adults: Cooking-Related Clothing Ignition," Journal of the American Geriatrics Society 45, no. 10 (1997): 1283–1285[Medline]; and N.J. Stiles et al., "Evaluating Fire Safety in Older Persons through Home Visits," Journal of the Kentucky Medical Association 99, no. 3 (2001): 105–110.[Medline]
- National Institutes of Health, "111. Osteoporosis Prevention, Diagnosis, and Therapy," 27–29 March 2000, consensus.nih.gov/cons/111/111_statement.htm (26 July 2002); and M. Sinaki et al., "Stronger Back Muscles Reduce the Incidence of Vertebral Fractures: A Prospective Ten-Year Follow-Up of Postmenopausal Women," Bone 30, no. 6 (2002): 836–841.[Medline]
- Elder Fall Prevention Act of 2002, S. 1922, 107th Cong., 2d sess. (7 February 2002).
- See W. Haddon Jr., "Advances in the Epidemiology of Injuries as a Basis for Public Policy," Public Health Reports 95, no. 5 (1980): 411–421[Medline]; and L. Robertson, Injury Epidemiology (New York: Oxford University Press, 1998).
       What's this?
|
Top
Study Methods
Medicare Spending For Injuries...