oversight

Medicare: Trends in Beneficiaries Served and Hospital Resources Used in Implantable Medical Device Procedures

Published by the Government Accountability Office on 2012-05-14.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

United States Government Accountability Office
Washington, DC 20548



           May 14, 2012
           The Honorable Orrin Hatch
           Ranking Member
           Committee on Finance
           United States Senate

           Subject: Medicare: Trends in Beneficiaries Served and Hospital Resources Used in
                   Implantable Medical Device Procedures

           Dear Senator Hatch:

           The use of implantable medical devices (IMD) among Medicare beneficiaries is
           widely recognized as a way to prolong and improve the quality of life for patients that
           receive them. 1 In 2009, about 1.6 million IMD procedures were performed on
           beneficiaries under traditional, fee-for-service Medicare at a cost of roughly
           $20 billion. Orthopedic and cardiac implantations—the most common IMD
           procedures provided to beneficiaries—accounted for nearly all IMD-related Medicare
           spending in that year. With beneficiaries expected to live longer and innovations in
           IMD technology, the use of orthopedic and cardiac IMDs is likely to continue to have
           important implications for hospital services paid for by Medicare.

           The number of hospital admissions for IMD procedures, the duration of hospital
           stays, and the location to which patients are discharged are influenced by such
           factors as age and health status. In that light, you expressed interest in obtaining
           descriptive information about changes in the demographics of Medicare
           beneficiaries undergoing major IMD procedures and their use of hospital and
           postacute care resources. In this report, we examined three trends for Medicare
           beneficiaries who received orthopedic or cardiac IMDs: (1) hospital admission rates,
           by age and health status; (2) hospital lengths of stay, by health status; and
           (3) discharge disposition following admission for these procedures, by health status.

           Our review of orthopedic IMD procedures focused on those related to knees, hips,
           shoulders, and lumbar fusions. We defined knee, hip, and shoulder replacement
           procedures as stays where the procedure was a primary elective new total knee,
           total hip, or total shoulder replacement. We defined lumbar fusion procedures as
           stays where the procedure was a primary elective initial lumbar or lumbosacral


           1
            We define IMDs as artificial devices implanted entirely within the body that are intended to remain
           there permanently. However, some of these devices have a limit to their effective life span and will
           require replacement.



                                                                   GAO-12-583R Implantable Medical Devices
fusion with a posterior technique. 2, 3 Our review of cardiac IMDs focused on
procedures related to certain devices used to treat blocked coronary arteries—drug-
eluting stents—or heart rhythm problems—automatic implantable cardioverter
defibrillators (AICD) and dual-chamber pacemakers. The orthopedic IMD procedures
we studied are nearly always performed in the hospital inpatient setting. However,
procedures involving drug-eluting stents, AICDs, and dual-chamber pacemakers can
be performed in either inpatient or outpatient settings. We focused only on inpatient
trends in the use of cardiac IMDs. Further research on topics such as readmission
rates 4 and revisions 5 would be needed to understand the full impact of orthopedic
and cardiac IMD use patterns.

To examine trends for Medicare beneficiaries who received orthopedic or cardiac
IMDs, we obtained hospital discharge data on individuals age 65 and over from the
Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS)
files from 2003 through 2009. 6, 7 To calculate trends in Medicare hospital admission
rates, we divided the number of inpatient IMD procedures performed on these
individuals by the total number of Medicare Part B beneficiaries age 65 or over
during the same period, as reported by the Centers for Medicare & Medicaid
Services (CMS). 8 We sorted beneficiaries into four age cohorts—65 to 69, 70 to 74,
75 to 79, and 80 or older—and categorized beneficiary health status as good or fair,
poor, or very poor based on the patient’s condition at admission and other factors. 9
2
 Lumbar and lumbosacral fusions are those that involve certain vertebrae in the lower region of the
spine. Posterior fusions refer to how the surgeon approaches the spine—through the lower back.
3
 We excluded (1) partial joint replacements and procedures involving more than one joint,
(2) replacement surgeries needed when the effective performance of some devices declines,
(3) lumbar fusion procedures that did not use the posterior technique, and (4) other types of spinal
fusions, such as those related to the cervical spine.
4
 Research has found an increase in 30- and 90-day all-cause readmission rates for total hip
replacement patients in recent years. Xueya Cai, et al., “Clinical Characteristics and Outcomes of
Medicare Patients Undergoing Total Hip Arthroplasty, 1991-2008,” Journal of the American Medical
Association 305, no. 15 (2011): 1560-1567.
5
 Revisions, procedures that replace part or all of an IMD, accounted for 8.9 percent of all orthopedic
IMD procedure Medicare expenditures in 2004 and 11.0 percent in 2009, increasing from about
$0.5 billion to about $1 billion. Some revisions may reflect device recalls. From 2005 through 2009,
orthopedic and cardiovascular devices constituted 12 and 15 percent of medical device recalls,
respectively. See GAO, Medical Devices: FDA Should Enhance Its Oversight of Recalls, GAO-11-468
(Washington, D.C.: June 14, 2011).
6
 Medicare covers virtually all of the population age 65 and over. We defined Medicare admissions as
those of individuals who are in that age cohort.
7
 The NIS is designed as a representative 20 percent sample of all hospitals. It contains hospital
discharge data provided by states that participate in HCUP. In 2009, 44 states provided data from
about 1,000 hospitals. NIS data do not include outpatient procedures.
8
 We define Medicare admission rates as admissions per 10,000 Medicare Part B beneficiaries.
Among other things, Medicare Part B covers the physicians’ services used in IMD procedures.
9
 Across all procedure types, we assigned health status using several patient variables, including a
beneficiary’s principal and secondary diagnosis, procedures codes, age, sex, and discharge
disposition. Patient demographics, such as secondary diagnoses, can be risk factors for in-hospital
complications and mortality. For example, research on patients undergoing bilateral total knee
arthroplasty has shown that the presence of congestive heart failure and pulmonary hypertension
have been associated with increased risk for adverse outcome. See Ya-Lin Chiu, et al., “Bilateral
Total Knee Arthroplasty: Risk Factors for Major Morbidity and Mortality,” Anesthesia & Analgesia,
July 13, 2011.



2                                                      GAO-12-583R Implantable Medical Devices
To analyze trends in average lengths of stay, we excluded beneficiaries with hospital
stays of zero days and outliers with exceedingly long stays. To examine trends in
discharge disposition, we stratified beneficiaries into those discharged to home or
self-care, to home for home health care, and to inpatient rehabilitative facilities, such
as a skilled nursing facility (SNF) or an inpatient rehabilitation facility (IRF). 10 In
addition, we reviewed relevant journal articles and CMS regulations. We determined
that the data we used were sufficiently reliable for the purposes of our analysis by
performing appropriate electronic data checks.

We conducted this performance audit from March 2011 to April 2012 in accordance
with generally accepted government auditing standards. Those standards require
that we plan and perform the audit to obtain sufficient, appropriate evidence to
provide a reasonable basis for our findings and conclusions based on our audit
objectives. We believe that the evidence obtained provides a reasonable basis for
our findings and conclusions based on our audit objectives.

Results in Brief

Overall, orthopedic IMD admission rates were substantially higher in 2009 compared
with 2003, while admission rate patterns among cardiac IMDs were mixed.
Admission rates rose for each of the orthopedic IMDs in our study, with knee
replacement rates growing 6.7 percent per year. The picture for inpatient cardiac
IMD procedures was more mixed; admission rates for dual-chamber pacemakers
decreased steadily while rates for AICDs and drug-eluting stents increased through
2006 and generally declined thereafter, in part reflecting a shift of surgeries to the
outpatient setting. While the proportion of both orthopedic and cardiac IMD
beneficiaries in poor or very poor health grew throughout our period of study, this
trend was far more evident for cardiac IMD beneficiaries after 2007.

Even with the increase in admissions of IMD beneficiaries in poorer health, overall
lengths of stay for the IMD procedures we studied generally did not rise. Average
lengths of stay for orthopedic IMD beneficiaries decreased from 2003 through 2009,
while the lengths of stay for cardiac IMD beneficiaries fell through 2007 but
increased thereafter. For all orthopedic IMD procedures in our study, lengths of stay
declined during the period for beneficiaries in all reported health status groups. From
2003 through 2007, the average length of stay decreased among cardiac IMD
beneficiaries in each health status group. From 2007 to 2009, average lengths of
stay patterns varied by health status and specific cardiac IMD procedure.

Poorer health status and reductions in lengths of stay for inpatient IMD beneficiaries
were not accompanied by an increase in discharges to rehabilitative facilities.
Rather, the proportion of orthopedic IMD beneficiaries discharged to home health
care increased substantially while the proportion discharged to a skilled nursing or
rehabilitation facility dropped sharply. The discharge disposition pattern for cardiac
IMD beneficiaries remained relatively stable throughout the study period, with a large
majority of cardiac IMD beneficiaries discharged to home or self-care.

10
  Discharge disposition indicates the postacute care, if any, a beneficiary received directly after
discharge. However, beneficiaries can receive a series of postacute care services in various settings.



3                                                      GAO-12-583R Implantable Medical Devices
Background

IMD Device Descriptions

In 2009, procedures related to knees, hips, shoulders, and the spine accounted for
98 percent of Medicare’s orthopedic IMD expenditures. Typically, hip, knee, and
shoulder implants have a variety of components and are made up of different
materials, which may be configured in various ways to make the total device. For
example, for a hip replacement with four different components, there are several
configurations and materials (metal, plastic, and ceramic) that can be used, as well
as different ways to secure the implant (cemented in place or fitted into the bone
with new bone growth to hold the implant in place). Lumbar fusion surgeries may
involve many different IMDs; some of the most common IMDs used in lumbar
fusions include screws, rods, cages, and bone morphogenetic protein (BMP). 11

In 2009, procedures involving pacemakers, AICDs, and stents represented most
Medicare spending on cardiac IMD procedures. A pacemaker monitors a patient’s
underlying heart rhythm and delivers an electrical pulse to cause the heart to beat at
the desired rate. An AICD is similar to the pacemaker in design, but it is capable of
delivering a higher energy electrical pulse—called a defibrillation shock—to correct
more serious, rapid, and sustained heart rhythm irregularities. A coronary stent is a
wire mesh tube used to prop open a blocked coronary artery. Drug-eluting stents are
coated with drugs that slowly release and are intended to help keep the artery open.

Medicare IMD Spending and Overall Lengths of Stay

In Medicare’s traditional fee-for-service program, spending on IMD procedures
performed in the inpatient and outpatient settings has grown at the same rate as
spending for other hospital services. 12 As we previously reported, from 2004 through
2009, expenditures for IMD procedures rose from about $16 billion to about
$20 billion, an increase of 4.3 percent per year—a rate equal to that of Medicare
spending for other hospital care. Spending on orthopedic IMD procedures performed
grew substantially faster than that for cardiac IMD procedures. From 2004 through
2009, Medicare hospital expenditures related to orthopedic IMD devices increased
8.1 percent per year, while expenditures related to cardiac IMD procedures
increased 1.2 percent yearly. 13

A factor that contributes to Medicare inpatient spending is beneficiaries’ length of
stay. From 2003 through 2009, average hospital lengths of stay declined for
Medicare beneficiaries overall. For instance, the average annual decrease in length
of stay ranged from 0.6 percent to 7.0 percent for 10 hospital inpatient services that

11
  Screws and rods are used to hold the spine still to aid the fusion process. Cages placed between
two vertebrae and BMP, a synthetic bone-forming protein, are often used together to promote fusion
in lumbar fusion surgeries.
12
 In 2011, about three quarters of all beneficiaries were in fee-for-service Medicare and the rest were
enrolled in private health plans under the Medicare Advantage program.
13
  See GAO, Medicare: Lack of Price Transparency May Hamper Hospitals’ Ability to Be Prudent
Purchasers of Implantable Medical Devices, GAO-12-126 (Washington, D.C.: Jan. 13, 2012).



4                                                      GAO-12-583R Implantable Medical Devices
ranked among those with the highest number of non-IMD elective admissions in
2009. The proportion of beneficiaries admitted with poor or very poor health also
increased for each of those 10 services from 2003 to 2009.

Discharge Disposition after IMD Procedures

After receiving an orthopedic or cardiac IMD, Medicare beneficiaries can be
discharged home or to one of several postacute care settings. Numerous factors
such as age, functional status, and whether the beneficiary lives alone can affect the
decision about where a beneficiary is discharged. Those discharged to home or self-
care require minimal postacute care or only need to receive services, such as
physical therapy, on an outpatient basis. Other beneficiaries who are discharged
home but require a higher, or more intense, level of postacute care may receive
home health care services, such as intermittent skilled nursing and physical therapy.
Orthopedic and cardiac IMD beneficiaries may also be discharged to inpatient
facilities such as SNFs or IRFs for rehabilitation services. In general, the cost of
postacute care is more expensive for individuals discharged to SNFs and IRFs than
to home, with or without home health care or outpatient rehabilitation services.

Inpatient Admission Rates Increased Consistently across Orthopedic IMDs,
but Varied across Cardiac IMDs; IMD Procedures for Beneficiaries in Poor or
Very Poor Health Were Increasingly Common, a Trend More Pronounced
Among Cardiac IMD Beneficiaries

Admission Rates for Orthopedic IMD Procedures Increased Steadily

•   Admission rates for knee, hip, and shoulder replacements and lumbar fusion
    procedures increased steadily overall among Medicare beneficiaries from 2003
    through 2009. (See fig. 1.)

    •   The admission rate for knee replacements, by far the most common
        orthopedic procedure of those we studied, rose substantially from 2003
        through 2009. The admission rate increased from 59 to 87 per 10,000
        beneficiaries, an average annual increase of 6.7 percent.

    •   The admission rate for hip replacements, the second most common
        orthopedic procedure studied, grew moderately over the study period. The
        admission rate increased from 29 to 37 per 10,000 beneficiaries, an average
        annual increase of 4.1 percent.

    •   Although shoulder replacements and lumbar fusions were far less common
        than knee or hip replacements, their admission rates grew most rapidly during
        the time period, with shoulder replacement and lumbar fusion admissions
        growing at annual rates of 20.1 percent and 11.0 percent, respectively.




5                                             GAO-12-583R Implantable Medical Devices
Figure 1: Medicare Admission Rates for Orthopedic Implantable Medical Devices, 2003-2009




Note: We define Medicare admissions as those of individuals who are age 65 and over.


•    Increases in knee replacement surgeries have been attributed to changes in
     medical practice, enhanced awareness of the benefits of knee replacements,
     increased patient satisfaction rates, and an increasing prevalence of
     osteoarthritis, which in turn may be related to an increase in obesity rates. 14

     •   According to the National Institutes of Health, 85 percent of beneficiaries who
         undergo knee replacement surgery are satisfied with the results. 15

     •   The rate of obesity among Medicare beneficiaries who received a knee
         replacement was higher than those who received the other IMD procedures
         studied; they also experienced the largest increase in obesity rates over the
         time period. 16
14
  C. Mehrotra, P. Remington, T. Naimi, W. Washington, and R. Miller, “Trends in Total Knee
Replacement Surgeries and Implications for Public Health, 1990–2000,” Public Health Reports 120
(2005): 278-282. S. Kurtz, F. Mowat, K. Ong, N. Chan, et. al, “Prevalence of Primary and Revision
Total Hip and Knee Arthroplasty in the United States from 1990 through 2002,” Journal of Bone and
Joint Surgery; July 2005; 87, 7; ProQuest Medical Library, pg. 1487.
15
  Patient satisfaction rates are even greater for hip and shoulder replacements at 90 percent and
97 percent, respectively. See: E. Fisher, J. Bell, I. Tomek, A. Esty, and D. Goodman, “Trends and
Regional Variation in Hip, Knee, and Shoulder Replacement,” Dartmouth Atlas Surgery Report
(2010).
16
 Obesity appears to be undercoded in hospital data given the much higher prevalence in the general
population. See http://www.hcup-us.ahrq.gov/reports/statbriefs/sb20.jsp. Accessed on March 23,
2012.


6                                                                GAO-12-583R Implantable Medical Devices
•    The nearly twofold increase in the admission rate for lumbar fusions may
     exemplify the role that new technology plays in IMD utilization. For example, from
     2003 to 2009, the proportion of lumbar fusions performed with BMP, a relatively
     new technology, increased from approximately 7 percent to 40 percent of all such
     surgeries. 17

Changes in Inpatient Admission Rates for Cardiac IMD Procedures Differed by
Device

•    The trends in cardiac IMD inpatient admission rates were mixed over the study
     period. (See fig. 2.)

     •   Inpatient admission rates for drug-eluting stents increased rapidly from 2003
         through 2006, declined sharply from 2006 through 2007, and remained flat
         thereafter. Rates for AICDs also increased through 2006 and then declined,
         although both the increase and subsequent decline were much more
         moderate.

     •   In contrast, the rate of inpatient dual-chamber pacemaker admissions
         declined slowly and steadily.




17
 Since BMP has not been approved for use in posterior lumbar fusion, growth in usage during this
period is off label.



7                                                    GAO-12-583R Implantable Medical Devices
Figure 2: Medicare Admission Rates for Cardiac Implantable Medical Devices, 2003–2009




Note: We define Medicare admissions as those of individuals who are age 65 and over.


•    Decreases in the rates of cardiac IMD admissions were associated with
     beneficiaries receiving cardiac IMD procedures in the outpatient rather than the
     inpatient setting. Had the pattern of care not changed after 2006, it is likely that
     more beneficiaries would have been admitted to a hospital for a cardiac IMD
     procedure.

     •   As we previously reported, Medicare claims data indicated a general shift of
         cardiac IMD procedures from the inpatient to the outpatient setting from 2004
         through 2009, with the largest growth in outpatient cardiac IMD procedures
         occurring from 2007 through 2009. 18

     •   This trend coincided with Medicare Recovery Audit contractors collecting
         overpayments for certain inpatient cardiac IMD procedures that could have
         been performed in the outpatient setting, possibly prompting other hospitals to
         change their admission patterns. 19 Generally, Medicare pays hospitals a
         relatively lower rate for the same procedure when it is delivered in the
         outpatient rather than the inpatient setting.


18
 GAO, Medicare: Lack of Price Transparency May Hamper Hospitals’ Ability to Be Prudent
Purchasers of Implantable Medical Devices, GAO-12-126 (Washington, D.C.: Jan. 13, 2012).
19
 Recovery audit contractors conduct postpayment reviews to identify overpayments and
underpayments and recoup any overpayments they identify.



8                                                                GAO-12-583R Implantable Medical Devices
•     In addition to the general shift to the outpatient setting, the significant decrease in
      inpatient drug-eluting stent admission rates from 2006 to 2007 may have resulted
      from a shift to the use of bare metal stents or a decline in overall stent utilization.

Admission Rates Rose across All Age Groups for All Orthopedic IMD Procedures
and Drug-Eluting Stents but Fell for AICDs and Dual-Chamber Pacemakers

•     From 2003 through 2009, admission rates rose for all four orthopedic IMD
      procedures among every beneficiary age group. (See table 1.)

      •      Rates for knee and hip replacements increased most rapidly for the youngest
             Medicare beneficiaries.

      •      In contrast, older Medicare beneficiaries exhibited the fastest growth in
             shoulder replacements and lumbar fusion procedures.

Table 1: Average Annual Percentage Growth in Medicare Beneficiary Admission Rates for Orthopedic
IMD Procedures, by Age, 2003-2009

                                                                                  Growth in admissions
                                            Total knee                           Total hip                 Total shoulder
 Age                                      replacement                        replacement                     replacement    Lumbar fusion
 65 to 69                                                7.9                                5.9                      16.5            12.2
 70 to 74                                                6.6                                3.7                      23.2            11.3
 75 to 79                                                6.8                                4.7                      26.0            11.2
 80 or older                                             5.8                                3.2                      26.0            15.2
Source: GAO analysis of Healthcare Cost and Utilization Project Nationwide Inpatient Sample data.

Note: We define Medicare admissions as those of individuals who are age 65 and over.


•     Across all age groups, inpatient admission rates for drug-eluting stents
      increased, while AICD and dual-chamber pacemaker rates declined during our
      study period. (See table 2.)

Table 2: Average Annual Percentage Growth in Medicare Beneficiary Admission Rates for Cardiac IMD
Procedures, by Age, 2003-2009

                                                                                   Growth in admissions
 Age                                     Drug-eluting stent                                    AICD                Dual-chamber pacemaker
 65 to 69                                                      10.1                                 -3.3                             -1.3
 70 to 74                                                      10.5                                 -5.9                             -3.0
 75 to 79                                                      11.5                                 -2.5                             -1.9
 80 or older                                                   13.1                                 -4.1                             -0.9
Source: GAO analysis of Healthcare Cost and Utilization Project Nationwide Inpatient Sample data.

Note: We define Medicare admissions as those of individuals who are age 65 and over.




9                                                                                       GAO-12-583R Implantable Medical Devices
Growing Share of IMD Beneficiaries Were Admitted in Poor or Very Poor Health, a
Trend More Pronounced among Inpatient Cardiac IMD Beneficiaries

•    We found moderate increases in the proportion of orthopedic IMD beneficiaries
     who were in poor health from 2003 through 2009. (See fig. 3.)

     •   For the four orthopedic IMD procedures studied, the increase in the
         proportion of beneficiaries in poor health ranged from 2.9 to 3.9 percentage
         points from 2003 through 2009.

Figure 3: Proportion of Inpatient Orthopedic IMD Beneficiaries in Poor Health, by Type of Procedure,
2003-2009




Note: We define Medicare admissions as those of individuals who are age 65 and over.


•    The proportion of orthopedic IMD beneficiaries in very poor health increased
     slightly but remained relatively low—roughly 1 percent—throughout the study
     period.

•    We found an increase in the proportion of beneficiaries who were admitted in
     poor health for inpatient cardiac IMD procedures from 2003 through 2009. (See
     fig. 4.)

     •   The percent of cardiac IMD beneficiaries in poor health rose for every type of
         device procedure. This was particularly evident for drug-eluting stent and
         dual-chamber pacemaker recipients from 2007 forward.




10                                                               GAO-12-583R Implantable Medical Devices
Figure 4: Proportion of Inpatient Cardiac IMD Beneficiaries in Poor Health, by Type of Procedure, 2003-
2009




Note: We define Medicare admissions as those of individuals who are age 65 and over.


•    We also found an increase in beneficiaries in very poor health being admitted for
     each of the cardiac IMD devices from 2003 through 2009. (See fig. 5.)

     •   Again, the proportion of beneficiaries admitted with very poor health
         increased more rapidly from 2007 forward.

     •   The increase was particularly dramatic for AICDs; the share of beneficiaries
         that received an AICD who were in very poor health nearly doubled from 2007
         through 2009.




11                                                               GAO-12-583R Implantable Medical Devices
Figure 5: Proportion of Inpatient Cardiac IMD Beneficiaries with Very Poor Health, by Type of Procedure,
2003-2009




Note: We define Medicare admissions as those of individuals who are age 65 and over.


•    The migration of cardiac IMD beneficiaries to the hospital outpatient setting most
     likely removed healthier beneficiaries from the inpatient population, leaving a
     greater proportion of beneficiaries in the inpatient setting with poor or very poor
     health.

Lengths of Stay for Orthopedic IMD Beneficiaries Fell Steadily; Stays for
Cardiac IMD Beneficiaries Grew After 2007, Reflecting a Marked Decline in
Patient Health Status

Lengths of Stay Consistently Declined for Orthopedic IMD Beneficiaries

•    When comparing 2003 and 2009 data, we found substantial decreases in the
     average length of stay for all orthopedic IMD beneficiaries in our study. (See
     fig. 6.)

     •   For example, the length of stay for knee replacement beneficiaries fell from
         4.02 to 3.38 days, or 2.8 percent per year. This represented a reduction of
         64 days per 100 hospital admissions.

     •   The rates of decline in lengths of stay for these procedures were similar to
         those of non-IMD elective procedures.




12                                                               GAO-12-583R Implantable Medical Devices
Figure 6: Beneficiary Average Length of Stay, by Orthopedic IMD Procedure, 2003-2009




Note: Lengths of stay are those of individuals who are age 65 and over.


Cardiac IMD Beneficiaries’ Lengths of Stay Generally Declined through 2007, but
Increased Afterward

•    The average length of stay of beneficiaries admitted for cardiac IMD procedures
     generally declined from 2003 through 2007 but increased thereafter. (See fig. 7.)

     •    For example, the length of stay for those receiving drug-eluting stents fell
          1.0 percent annually from 2003 through 2007 and increased at an annual rate
          of 5.5 percent during the last 2 years of our study period.

•    The more recent increases in length of stay for cardiac IMD beneficiaries were
     associated with a change in the mix of patients receiving these procedures. As
     cardiac IMD surgeries shifted to the outpatient setting, more of the remaining
     inpatient beneficiaries were in poor or very poor health.




13                                                                 GAO-12-583R Implantable Medical Devices
Figure 7: Beneficiary Average Length of Stay, by Cardiac IMD Procedure, 2003-2009




Note: Lengths of stay are those of individuals who are age 65 and over.


Lengths of Stay Differed Substantially by IMD Beneficiaries’ Health Status; Stays
Declined across Health Status Groups

•    For both orthopedic and cardiac IMD procedures, beneficiaries’ lengths of stay
     differed substantially by health status across all years studied. Hospital stays
     were generally 1 to 3 days longer for beneficiaries in poor health compared with
     those in good or fair health.

•    For all IMD procedures in our study, lengths of stay declined during the period for
     beneficiaries in all reported health status groups, falling most dramatically for
     IMD beneficiaries in poor health.

•    From 2003 to 2009, for each of the four types of orthopedic IMDs,

     •    The average length of stay for orthopedic IMD beneficiaries decreased
          considerably across all health status groups.

     •    Because the rate of decline in the average length of stay was greater for IMD
          beneficiaries in poor health compared with those in good or fair health, the
          differences in lengths of stay narrowed significantly by 2009. (See fig. 8.)




14                                                                 GAO-12-583R Implantable Medical Devices
Figure 8: Average Length of Stay for Orthopedic IMD Beneficiaries, by Procedure and Health Status,
2003-2009




Note: Lengths of stay are those of individuals who are age 65 and over. The average lengths of stay for orthopedic IMD
beneficiaries in very poor health are not shown because relatively few beneficiaries were in this category.


•    For the period studied, the average IMD beneficiary length of stay generally
     declined for all cardiac procedures and in all health status groups. (See fig. 9.)

     •    From 2003 through 2007, the average length of stay for inpatient cardiac IMD
          beneficiaries declined for all health status groups.

     •    From 2007 through 2009, the average length of stay for those in good or fair
          health increased but generally decreased for those in poor or very poor
          health.




15                                                                 GAO-12-583R Implantable Medical Devices
Figure 9: Average Length of Stay for Cardiac IMD Beneficiaries, by Procedure and Health Status, 2003-
2009




Note: Lengths of stay are those of individuals who are age 65 and over.


Orthopedic IMD Beneficiaries Were Increasingly Discharged to Home Health
Care Rather Than Rehabilitative Facilities, while Cardiac IMD Beneficiaries’
Discharge Disposition Pattern Was Relatively Unchanged

Orthopedic Beneficiaries Were Increasingly Discharged to Home Health Care

•    From 2003 through 2009, an increasing percentage of inpatient orthopedic IMD
     beneficiaries were discharged to home health care. (See fig.10.)

     •    For example, the share of beneficiaries receiving knee or hip replacements
          who were discharged to home health care grew from 22 to 35 percent and 20
          to 34 percent, respectively.

     •    At the same time, there was generally a notable reduction in the proportion
          that was discharged to a rehabilitative facility.




16                                                                 GAO-12-583R Implantable Medical Devices
•     The shift from rehabilitative facilities to home health care has the potential to
      lower Medicare expenditures. In 2008, the estimated overall average Medicare
      payment for SNF and IRF stays were $8,910 and $16,649, respectively, whereas
      the average payment for a home health episode of care was $2,800. 20

Figure 10: Beneficiaries’ Discharge Disposition Following Orthopedic IMD Admissions, by Procedure,
2003-2009




Note: Rehabilitative facilities include inpatient rehabilitation facilities, skilled nursing facilities, and other facilities that provide
rehabilitative care. Discharge dispositions are those of individuals who are age 65 and over.


•     The migration of orthopedic IMD beneficiaries to home health care discharges
      was most evident among those in good or fair health but was also observable for
      those in poor health. (See table 3.)




20
    Other factors could also affect the cost of postacute care, such as readmissions.



17                                                                          GAO-12-583R Implantable Medical Devices
Table 3: Discharge Disposition of Orthopedic IMD Beneficiaries, by Health Status, 2003-2009

                                                                                           Percentage of beneficiaries
                           Discharge
 Health status             disposition                             2003           2004              2005   2006   2007   2008        2009
 Good or fair              Rehabilitative facility                     53             54              51     47     44      40          40
 health                    Home health care                            21             25              27     31     32      35          34
                           Home or self-care                           25             20              22     22     23      25          25
 Poor health               Rehabilitative facility                     66             67              64     65     61      59          60
                           Home health care                            15             18              20     21     23      24          24
                           Home or self-care                           16             13              14     12     14      15          13
Source: GAO analysis of Healthcare Cost and Utilization Project Nationwide Inpatient Sample data.

Note: Rehabilitative facilities include inpatient rehabilitation facilities, skilled nursing facilities, and other facilities that provide
rehabilitative care. Disposition data for orthopedic IMD beneficiaries in very poor health are not shown because relatively few
beneficiaries were in this category. Discharge dispositions are those of individuals who are age 65 and over.


•     The Medicare Payment Advisory Commission cited CMS actions when
      describing the shift in orthopedic IMD beneficiaries’ discharge disposition from
      rehabilitation facilities to home health care. 21

      •      In 2004, CMS revised its list of conditions for determining the medical need of
             patients for inpatient rehabilitation services, recognizing only certain
             categories of patients with knee or hip replacements. 22

      •      From 2005 to 2008, CMS Medicare Recovery Audit Contractors found
             medically unnecessary services performed in IRFs following joint replacement
             surgery. This may have further reduced the amount of IRF admissions related
             to joint replacements.

Changes in Health Status for Inpatient Cardiac IMD Beneficiaries Were Not
Associated with a Substantial Increase in Use of Postacute Care

•     During our study period, the share of inpatient cardiac IMD beneficiaries
      discharged to a rehabilitative facility remained relatively stable. (See table 4.)

      •      Between 2003 and 2007, the share of inpatient cardiac IMD beneficiaries
             discharged to home or self-care remained at roughly 86 percent.

      •      After their general decline in health status since 2007, the share of inpatient
             cardiac IMD beneficiaries discharged to home or self-care began to fall
             slightly.
21
 See MedPAC, Report to the Congress: Medicare Payment Policy (Washington, D.C.: March 2011),
28, accessed October 19, 2011, http://www.medpac.gov/documents/Mar11_EntireReport.pdf.
22
  In order for an IRF to be paid under the IRF prospective payment system instead of the acute care
hospital inpatient prospective payment system CMS requires that 75 percent of the facility’s
beneficiaries have one or more qualifying medical conditions. CMS revised the 75 percent rule by
requiring that beneficiaries who receive knee or hip replacements must have undergone bilateral
 joint surgery, be extremely obese, or be 85 years or older at the time of admission to the IRF. See
69 Fed. Reg. 25752, 25775 (May 7, 2004) (relevant provisions currently codified at 42 C.F.R.
§ 412.29(b)(2)(xiii) (2011)) .



18                                                                                      GAO-12-583R Implantable Medical Devices
Table 4: Discharge Disposition of Cardiac IMD Beneficiaries, 2003-2009

                                                                                     Percentage of beneficiaries
 Discharge disposition                                     2003             2004            2005        2006   2007     2008   2009
 Home or self-care                                             85              86               86        87       85     84     83
 Home health care                                                7               7                  7     6        7      9      9
 Rehabilitative facility                                         7               6                  5     5        7      7      7
Source: GAO analysis of Healthcare Cost and Utilization Project Nationwide Inpatient Sample data.

Note: Discharge dispositions are those of individuals who are age 65 and over.


•     The share of cardiac IMD beneficiaries who were discharged to a rehabilitative
      facility also remained relatively stable by health status.

Agency Comments

We obtained comments on a draft of this report from the Department of Health and
Human Services. The agency responded that it had no general comments and
provided technical comments, which we incorporated as appropriate.

                                                                          –––––

As we agreed with your office, unless you publicly announce the contents of this
report earlier, we plan no further distribution of it until 30 days from its date. We are
sending copies of this report to the Secretary of Health and Human Services. The
report will also be available at no charge on our website at http://www.gao.gov.

If you or your staff have any questions about this report, please contact me at
(202) 512-7114 or cosgrovej@gao.gov. Contact points for our Offices of
Congressional Relations and Public Affairs may be found on the last page of this
report. Individuals making key contributions to this report include Rosamond Katz,
Assistant Director; Luis Serna III; and Brian O’Donnell. Zhi Boon also provided
valuable assistance.

Sincerely yours,




James Cosgrove
Director, Health Care




(290938)



19                                                                                      GAO-12-583R Implantable Medical Devices
This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.
                      The Government Accountability Office, the audit, evaluation, and
GAO’s Mission         investigative arm of Congress, exists to support Congress in meeting its
                      constitutional responsibilities and to help improve the performance and
                      accountability of the federal government for the American people. GAO
                      examines the use of public funds; evaluates federal programs and
                      policies; and provides analyses, recommendations, and other assistance
                      to help Congress make informed oversight, policy, and funding decisions.
                      GAO’s commitment to good government is reflected in its core values of
                      accountability, integrity, and reliability.

                      The fastest and easiest way to obtain copies of GAO documents at no
Obtaining Copies of   cost is through GAO’s website (www.gao.gov). Each weekday afternoon,
GAO Reports and       GAO posts on its website newly released reports, testimony, and
                      correspondence. To have GAO e-mail you a list of newly posted products,
Testimony             go to www.gao.gov and select “E-mail Updates.”

Order by Phone        The price of each GAO publication reflects GAO’s actual cost of
                      production and distribution and depends on the number of pages in the
                      publication and whether the publication is printed in color or black and
                      white. Pricing and ordering information is posted on GAO’s website,
                      http://www.gao.gov/ordering.htm.
                      Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
                      TDD (202) 512-2537.
                      Orders may be paid for using American Express, Discover Card,
                      MasterCard, Visa, check, or money order. Call for additional information.
                      Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Connect with GAO      Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts.
                      Visit GAO on the web at www.gao.gov.
                      Contact:
To Report Fraud,
Waste, and Abuse in   Website: www.gao.gov/fraudnet/fraudnet.htm
                      E-mail: fraudnet@gao.gov
Federal Programs      Automated answering system: (800) 424-5454 or (202) 512-7470

                      Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 512-
Congressional         4400, U.S. Government Accountability Office, 441 G Street NW, Room
Relations             7125, Washington, DC 20548

                      Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
Public Affairs        U.S. Government Accountability Office, 441 G Street NW, Room 7149
                      Washington, DC 20548




                        Please Print on Recycled Paper.