The Journal of Arthroplasty Vol. 21 No. 6 Suppl. 2 2006

Postdischarge Costs in Arthroplasty Surgery

Carlos J. Lavernia, MD,* Michele R. D’Apuzzo, MD,* Victor H. Hernandez, MD,*David J. Lee, PhD,y and Mark D. Rossi, PT, PhD*

From the *Orthopaand yDepartment ofMiami, Miami, Florid

Submitted FebruBenefits or funds

the research materMercy Hospital, Art

Reprint requestInstitute at Mercy HMiami, FL 33133.

n 2006 Elsevier0883-5403/06/19doi:10.1016/j.art

Abstract: Postdischarge costs associated with primary arthroplasty surgeries have

received limited attention in the literature. Our objective was to identify the costs

incurred after discharge in primary arthroplasty and to estimate annual post-

discharge expenditures in the United States. A cohort of 136 patients who

underwent primary arthroplasty was studied. Comprehensive rehabilitation unit

(CRU) and home care (HC) costs were obtained. The National Hospital Discharge

Survey 2003 data were used to model the national discharge cost estimates. Local

patient-oriented outcome was also compared in the patients discharged to CRU vs

HC. Total costs were significantly lower in patients discharged directly to home vs

those sent to the CRU and who subsequently received HC ($2405 vs $13435, P b

.001); both patient groups experienced similar quality of life improvements. An

estimated $3.2 billion is spent annually on postsurgical rehabilitation after

arthroplasty. Postdischarge costs are significantly higher for patients going to a

CRU vs those discharged home; yet, both groups had comparable short-term

outcomes. Key words: postdischarge, costs, rehabilitation, primary arthroplasty.

n 2006 Elsevier Inc. All rights reserved.

More than 600000 primary total joint arthroplas-

ties are performed annually in the United States

[1]. Conservative projections from the American

Academy of Orthopaedic Surgeons (Chicago, IL)

estimate that about 750000 of these procedures

will be performed annually by the year 2030 [2].

These trends, combined with life expectancy

increases, as well as the desire to maintain active

pain-free lifestyles, will lead to dramatic increases

in annual joint arthroplasty surgery expenditures

in the United States.

144

edic Institute at Mercy Hospital, Miami, Florida;Epidemiology and Public Health, University ofa.ary 27, 2006; accepted May 1, 2006.were received in partial or total support of

ial described in this article from Zimmer,hritis Surgery Research Foundation.s: Carlos J. Lavernia, MD, Orthopaedicospital, 3659 S Miami Avenue, Suite 4008,

Inc. All rights reserved.06-0004$32.00/0

h.2006.05.003

Given the continued increase in the number of

arthroplasty surgeries, it will be essential to identify

cost-effective strategies that do not diminish the

quality of patient care. Furthermore, it is extremely

important to justify post–acute care services per-

taining to joint arthroplasty surgeries given that

the current cost-containment pressures within the

Medicare system will only intensify as the baby

boom generation ages. Careful documentation of

all perioperative and associated postoperative re-

habilitative costs is a necessary first step in this

process. National acute care hospital-based esti-

mates of arthroplasty expenditures are available

[3-5], but similar data on postrehabilitative dis-

charge costs are not.

Most arthroplasty patients will require some

form of rehabilitation ranging from inpatient stays

to home care (HC) services and outpatient rehabil-

itation, or a combination of both. There have been

a limited number of studies examining the efficacy

of rehabilitation before and after joint arthroplasty

surgery [6-8]. These studies did not address cost

issues within the context of comparative outcomes.

Postsurgical rehabilitation costs can be considerable

Postdischarge Costs in Arthroplasty Surgery ! Lavernia et al 145

and yet are frequently ignored when discussing the

cost of joint arthroplasty surgery [9].

National estimates for postdischarge costs associ-

ated with arthroplasty are not available; yet,

Medicare has recently instituted changes to its

reimbursement policies, which will lead to changes

in the funding available for postdischarge care (ie,

the 75% rule) [10]. These reimbursement policy

changes could have implications in terms of access

because a number of patients live alone and have

nowhere to go after surgery. Without a bbaselineQfor postdischarge expenditures, it will be difficult to

even assess the financial implications of reimburse-

ment policies because they are phased-in. Our

objective is to document the costs incurred after dis-

charge within a single surgical practice and to apply

these cost estimates to the number of arthroplasties

completed in the United States. We also compare

surgical and quality of life outcomes in patients

according to discharge status.

Methods

A cohort of 136 patients (143 procedures) from a

single surgical practice, under the direction of the

primary author, who underwent primary hip and

knee arthroplasty between January to December of

2004, was enrolled in a prospective registry study

after institutional review board approval, and

informed consent was obtained. Patient character-

istics were compared with national estimates using

weighted data from the 2003 National Hospital

Discharge Survey (NHDS) [11]. Local financial data

were obtained from 3 sources, including the

hospital cost accounting system for the compre-

hensive rehabilitation unit (CRU), and HC costs

obtained directly from the provider, and estimated

professional fees calculated using visit levels and

the 2005 Medicare fee schedule. Local data on

skilled nursing facility (SNF) costs after arthroplasty

were not available. The use of the Medicare

reimbursement rate is reasonable given that more

than 60% of arthroplasties charges in the United

States are reimbursed by this agency.

Outcome Measures

Preoperative and postoperative functional status

and quality of life scales included a Pain Visual

Analog Scale, the Western Ontario and McMaster

University Osteoarthritis Index (WOMAC) [12], and

the Short-Form 36 (SF-36) that assesses 8 domains,

including physical function, bodily pain, mental

health, social function, role limitation caused by

physical function and emotional problems, vitality

as well as general health perceptions [13,14]. The

Quality of Well-Being (QWB) Index was also

administered to assess general quality of life [15].

Postoperative measures were obtained at an average

8.6 F 3.73 SD months (range, 1-24 months).

Expenditure Definitions

Comprehensive rehabilitation unit expenditures

included both direct and indirect costs. Direct costs

included items such as devices used (ie, continuous

passive motion instrument), associated therapies

(ie, physical therapy), supplies, and medications.

Indirect costs included all support staff (ie, admin-

istration) and all ancillary services such as physical

and occupational therapies, nursing, supplies (ie,

assistive devices), and home health aides.

Estimation of National Expenditures

Total postdischarge costs included the sum of

CRU, HC, and professional fees for each patient. We

first estimated expenditures for 3 patient categories:

(1) discharged to the CRU, (2) discharged to home,

and (3) discharged to an SNF. Costs for CRU pa-

tients included CRU charges, professional fees, and

any home health care costs incurred after discharge

from the CRU. Costs for those discharged to home

were limited to HC charges only (which include

imbedded professional fees).

In our patient series, there were no SNF dis-

charges during the surveyed period. Because SNF

discharges can occur after arthroplasty, we estimat-

ed these costs based on the local average length of

stay for an arthroplasty patient at our local SNF

(20 days) multiplied by the local per diem Medicare

reimbursement rate (Fig. 1) [16].

National postdischarge costs were estimated by

applying the averages obtained for patients dis-

charged to the CRU, home, and SNF to the 2003

NHDS estimates of the number of arthroplasty

discharges in each of these 3 categories. Discharge

status was unknown for nearly 20% of the NHDS

patients. We assumed that the distribution of

unknown discharges was equal to the distribution

of known discharge subtypes to calculate a weight-

ed cost average for this subgroup.

Statistical Analyses

The SPSS software (SPSS, Chicago, Ill) was used

for the statistical analyses. Student t tests were

used to compare costs in patients discharged to the

CRU vs directed directly to HC. We calculated

preoperative and average 8.6-month postsurgical

change scores for all quality of life measures and

Fig. 1. Discharge patterns and average costs for the patients discharged to home with home health care (n = 116) and for

patients that stayed in CRU (n = 27), and estimated costs for patients discharged to an SNF. *Average costs between patients

discharged to CRU vs those discharged to home with home care were significant; yProfessional fees were included.

146 The Journal of Arthroplasty Vol. 21 No. 6 Suppl. 2 September 2006

used analysis of covariance to compare changes in

these measures in patients discharged to the CRU

vs discharged directly to home after adjustment

for age and sex. A P b.05 was considered

statistically significant.

Table 1. General Characteristics of the 2003 NHDSPatient Population and Our Cohort (n = 143)

Demographics NHDS 2003Single surgical

practice

Age (y) 66.6 72.5Female (%) 60.2 69.9RaceWhite (%) 67.5 80.4Black (%) 4.5 6.3Unknown (%) 26.3 13.3

ProceduresPrimary total hip (THA) (%) 35.5 34.3Primary total knees (TKA) (%) 64.5 65.7

Discharge dispositionCRU (%) 11.3 18.9HC (%) 45.6 81.1Skilled nurse facility (%) 22.8 0Unknown (%) 19.7 0

Results

The NHDS indicated that approximately 592664

primary arthroplasties (hips and knees) were per-

formed in 2003. Almost half (45.6%) of all primary

arthroplasties were discharged to home, 11.3%

were discharged to a CRU, 22.8% were discharged

to an SNF, 0.5% died after surgery, and 19.8% had

unknown disposition. Table 1 compares the socio-

demographic and discharge characteristics of this

national sample with those of the 136 patients who

underwent primary arthroplasty in our practice.

The national arthroplasty patient population was

younger than patients in our surgical practice

(66.6 vs 72.5 years). The typical NHDS patient was

also slightly less likely to be female (60% vs 70%).

Comparison of racial distributions is difficult given

that racial status was not available for 26% of the

NHDS sample. Similar percentages of total hip

arthroplasties (THAs) and total knee arthroplasties

(TKAs) were performed locally and nationally.

However, postdischarge patterns did vary with a

much larger proportion of local discharges going to

HC (81% vs 45.6%).

Local Costs and National Estimates

Fig. 1 shows that the average total CRU costs plus

any HC costs were significantly greater than in

patients discharged to HC ($13435 vs 2405, t = 15.1,

P b .001). The average cost per case on the CRU

for our patients was $10751 F $598, payments

made to HC averaged $2393 F $92, and the

average compensation in professional fees was

$345 F $19 (Fig. 2).

The algorithm for calculation of national reha-

bilitation costs is depicted in Fig. 2. Based on

these calculations, approximately $3.4 billion are

spent annually on rehabilitation services after

arthroplasty.

Quality of Life and Postsurgical DischargeStatus

Table 2 presents the preoperative and the aver-

age 8.6-month postoperative mean quality of life

scores for patients discharged to the CRU and

Fig. 2. Algorithm for estimation of national arthroplasty postdischarge expenditures.

Postdischarge Costs in Arthroplasty Surgery ! Lavernia et al 147

patients discharged directly to home. There were

significant improvements in WOMAC function and

corresponding reductions in pain and stiffness

reported in both discharge groups. Quality of life

as measured by the QWB and the SF-36 also tended

Table 2. Changes in Presurgical and Postsurgical QualityPatients Dischar

Outcomes

CRU

Pre Post D

QWBy 0.552 0.576 0.023VASz pain intensity 6.5 0.9 5.6SF-36Physical function 25.8 50.0 24.1Role playing 31.3 58.3 27.1Bodily pain 52.7 61.8 9.1General health 65.7 75.2 9.5Vitality 67.5 62.5 5.0Social function 67.8 59.5 8.3Role emotional 61.1 61.1 0Mental health 61.0 73.3 12.3

WOMACFunction 33.7 8.2 25.5Pain 9.7 1.9 7.8Stiffness 3.3 1.0 2.3Total 46.7 11.1 35.6

*F test compares preoperative to postoperative change in CRU vs HyQuality of Well-Being Scale.zVisual Analog Scale.§P b .05.

to show improvement after surgery in both groups,

although there was a trend toward greater

improvements for patients discharged to home

health care on many of the SF-36 measures. With

1 exception, there were no significant differences in

of Life Indicators in Patients Discharged to a CRU vsged to HC

HC

F*Pre Post D

0.560 0.590 0.030 0.87.5 1.4 6.1 0.2

23.3 59.7 36.4 1.232.4 74.0 41.6 1.255.8 70.3 14.5 1.076.2 72.9 3.3 0.057.1 65.6 8.4 1.064.1 84.9 20.8 4.7§76.6 83.8 7.2 0.267.5 76.9 9.4 0.4

35.6 10.7 24.9 0.010.7 2.3 8.4 0.03.2 1.3 1.9 0.9

49.6 14.3 35.3 0.0

C patients after adjustment for age and sex.

148 The Journal of Arthroplasty Vol. 21 No. 6 Suppl. 2 September 2006

the relative rate of improvement on the WOMAC

and quality of life measures when comparing

patients discharged to the CRU and patients dis-

charged directly to home. Age- and sex-adjusted

SF-36 social function scores did improve signifi-

cantly more in home-discharged patients relative to

the CRU-discharged patients (t = 4.7, P b .05).

Discussion

Arthritis remains the leading cause of disability

in the United States; in addition, the number of

Americans with doctor-diagnosed arthritis contin-

ues to increase [17,18]. In individuals for whom

medical therapy has failed, joint arthroplasty sur-

gery has been shown to be a cost-effective inter-

vention resulting in pain relief and improved

quality of life [19-23]. After joint arthroplasty

surgeries, most individuals undergo some form of

rehabilitation, with treatment strategies focusing

on improving range of motion, increasing strength,

and improving ability to complete activities of daily

living safely and efficiently. Moreover, other allied

health professions such as physical and occupa-

tional therapy, nursing, and social services all play

a vital role in postsurgical rehabilitation.

A number of cost-effectiveness studies have

been published on specific interventions during

the acute phase of the arthroplasty procedures.

Lavernia et al [24] demonstrated that perioperative

x-rays in joint arthroplasty generated more than a

million dollars in radiologist’s fees with no apparent

influence on patient outcomes. Kocher et al [25]

demonstrated that pathologic analysis of postsurgi-

cal total joint arthroplasty specimens did not

contribute to patient outcomes; yet, it generated

costs in excess of $233 per case for THA surgeries

and $304 per case for TKA surgeries. Several other

studies have demonstrated the cost utility of

interventions in the acute phase of the arthroplasty

procedures [24-28]. To our knowledge, no infor-

mation exists on the rehabilitation costs and their

respective cost utility.

Some studies have focused on the timing of

rehabilitation and comparison of HC and outpa-

tient therapies. Mitchell et al [8] compared the

effectiveness of HC vs outpatient hospital rehabil-

itation services in individuals before and after

unilateral knee joint arthroplasty surgery. The

authors reported that similar outcomes, which

included WOMAC, SF-36, and patient satisfaction,

were achieved in both rehabilitation groups;

however, rehabilitation was more expensive when

administered at home. In contrast to this report,

there are 2 Canadian studies that documented

similar functional outcomes between home and

hospital groups, but that home-based rehabilita-

tion was less costly [7,29]. Our cost findings are

also consistent with those reported by Tran et al

[30]. They reported that 37 individuals of 97 pa-

tients discharged to home after joint arthroplasty

surgery had an average cost of $2479 per patient.

The authors also reported that in those discharged

to a rehabilitation hospital after joint arthroplasty

surgery, 42 of 51 individuals had an average cost

of $7768 per patient. In our study, we found that

HC services were less expensive when compared

with an acute care facility by approximately $8300

per patient, findings which are consistent with

these studies.

The literature on the postdischarge costs after

arthroplasty is extremely limited. In our study,

we found that the average cost per case on CRU

was $10751, and payments made to HC aver-

aged $2393.

Differences between SNF and CRU units include

the amount of time spent in therapy per day, the

nursing-to-patient ratio (CRU 5:1, SNF 30:1 in our

institution), and the number of allowed profes-

sional visits per stay by physical medicine and

rehabilitation specialists. All of the previously

mentioned factors increase the cost of these inter-

ventions. According to our estimates, approximate-

ly $3.2 billion are spent annually in rehabilitation

services after primary arthroplasty.

Constant surgical advances (ie, minimal invasive

surgery, pain management) have lowered the

average length of stay for primary arthroplasty.

New strategies that lower the cost and use of

hospital supplies have also contributed to decreased

intrahospital costs. Despite this focus on improved

outcomes and lower in-hospital costs, orthopedic

surgeons as well as hospital administrators are

generally unaware of arthroplasty postdischarge

expenditures. Cost utility studies are needed to

assess interventions in the post–acute care setting.

Our study raises the possibility that reducing

postdischarge costs by only 10% would result in

more than $250 million dollars in savings per year.

These cost-utility studies therefore could result in

substantial savings to partially offset costs associat-

ed with the increasing number of procedures.

A drawback in our study was that we estimated

national arthroplasty costs based on financial data

from a single surgical practice. Medical cost studies

in general have noted substantial geographic var-

iation with respect to expenditures [31]. It is

therefore possible that these costs at our institution

do underestimate or overestimate costs for the

Postdischarge Costs in Arthroplasty Surgery ! Lavernia et al 149

nation as a whole. Also, the average age of our

patient population was higher than was noted in

the NHDS, although we found no correlation

between age and postdischarge expenditures. Fi-

nally, we lacked direct cost data for patients

discharged to skilled nursing home facilities. Our

national cost estimates for this type of discharge

may be imprecise.

Clearly, postrehabilitative outcomes after joint

arthroplasty surgeries must be better defined in an

effort to provide meaningful treatments that are

efficient and cost-effective. In the coming years, it

will be necessary to identify evidence-based reha-

bilitation strategies that will promote functional

recovery after knee and hip joint arthroplasty as

quickly and safely as possible while maximizing

functional abilities. It is also important to select

the least expensive option under the scenario of

equally effective rehabilitative approaches. The

only way to truly evaluate the utility and cost-

effectiveness of these postsurgical rehabilitative

strategies is to perform well-designed multicenter

randomized clinical trials. These trials are not only

critical for maximizing patient outcomes, but may

also help to at least slow the growth of annual

expenditures associated with arthroplasty.

References

1. Kurtz S, Mowat F, Ong K, et al. Prevalence of

primary and revision total hip and knee arthroplasty

in the United States from 1990 through 2002. J Bone

Joint Surg Am 2005;87:1487.

2. Frankowski JJ, Watkins-Castillo S. Primary total

knee and hip arthroplasty projections for the

U.S. population to the year 2030. www.aaos.org/

wordhtml/research/stats/TJR_Projections.pdf [Last

accessed Sep 15, 2005].

3. Antoniou J, Martineau PA, Filion KB, et al. In-

hospital cost of total hip arthroplasty in Canada and

the United States. J Bone Joint Surg Am 2004;

86-A:2435.

4. Bozic KJ, Katz P, Cisternas M, et al. Hospital resource

utilization for primary and revision total hip arthro-

plasty. J Bone Joint Surg Am 2005;87:570.

5. Lavernia CJ, Drakeford MK, Tsao AK, et al. Revision

and primary hip and knee arthroplasty. A cost

analysis. Clin Orthop Relat Res 1995;136.

6. Coyte PC, Young W, Croxford R. Costs and outcomes

associated with alternative discharge strategies fol-

lowing joint replacement surgery: analysis of an

observational study using a propensity score. J Health

Econ 2000;19:907.

7. Kramer JF, Speechley M, Bourne R, et al. Compar-

ison of clinic- and home-based rehabilitation pro-

grams after total knee arthroplasty. Clin Orthop Relat

Res 2003;225.

8. Mitchell C, Walker J, Walters S, et al. Costs and

effectiveness of pre- and post-operative home phys-

iotherapy for total knee replacement: randomized

controlled trial. J Eval Clin Pract 2005;11:283.

9. Forrest GP, Roque JM, Dawodu ST. Decreasing

length of stay after total joint arthroplasty: effect on

referrals to rehabilitation units. Arch Phys Med

Rehabil 1999;80:192.

10. Department of Health and Human Services D and

Centers for Medicare & C. Medicaid Services, CMS.

Medicare IRF Classification Requirements. CMS

Manual System; 2004. Centers for Medicare &

Medicaid Services (CMS). Inpatient Rehabilitation

Facility Prospective Payment System and The 75

Percent Rule. www.cms.hhs.gov/InpatientRehab-

FacPPS/Downloads/DLSumIRFPPS_75pcrule.pdf

[Last accessed May 26, 2006].

11. DeFrances C, Hall M, Podgomik M. 2003 National

hospital discharge survey. http://www.cdc.gov/nchs/

data/ad/ad359.pdf [Last accessed Sep 23, 2005].

12. Bellamy N, Buchanan WW, Goldsmith CH, et al.

Validation study of WOMAC, a health status instru-

ment for measuring clinically important patient

relevant outcomes to antirheumatic drug therapy in

patients with osteoarthritis of the hip or knee.

J Rheumatol 1988;15:1833.

13. Arocho R, McMillan CA, Sutton-Wallace P. Con-

struct validation of the USA-Spanish version of the

SF-36 health survey in a Cuban-American popula-

tion with benign prostatic hyperplasia. Qual Life Res

1998;7:121.

14. Lieberman JR, Dorey F, Shekelle P, et al. Outcome

after total hip arthroplasty. Comparison of a tradi-

tional disease-specific and a quality-of-life measure-

ment of outcome. J Arthroplasty 1997;12:639.

15. Kaplan R, Bush J. Health-related quality of life

measurement for evaluation research and policy

analysis. Health Psychol 1982:1.

16. Feliz M. Personal Communication, C. Lavernia,

Editor. 2005: Miami.

17. Brooks PM. Impact of osteoarthritis on individuals

and society: how much disability? Social consequen-

ces and health economic implications. Curr Opin

Rheumatol 2002;14:573.

18. Buckwalter JA, Saltzman C, Brown T. The impact of

osteoarthritis: implications for research. Clin Orthop

Relat Res 2004;(427 Suppl):S6.

19. Laupacis A, Bourne R, Rorabeck C, et al. The effect

of elective total hip replacement on health-

related quality of life. J Bone Joint Surg Am 1993;

75:1619.

20. Liang MH, Cullen KE, Larson MG, et al. Cost-

effectiveness of total joint arthroplasty in osteoar-

thritis. Arthritis Rheum 1986;29:937.

21. Liang MH, Larson M, Thompson M, et al. Costs and

outcomes in rheumatoid arthritis and osteoarthritis.

Arthritis Rheum 1984;27:522.

150 The Journal of Arthroplasty Vol. 21 No. 6 Suppl. 2 September 2006

22. Rorabeck CH, Bourne RB, Laupacis A, et al. A

double-blind study of 250 cases comparing

cemented with cementless total hip arthroplasty.

Cost-effectiveness and its impact on health-related

quality of life. Clin Orthop Relat Res 1994;156.

23. Lavernia CJ, Guzman JF, Gachupin-Garcia A. Cost

effectiveness and quality of life in knee arthroplasty.

Clin Orthop Relat Res 1997;134.

24. Lavernia CJ, Hernandez RA, Rodriguez JA. Perioper-

ative X-rays in arthroplasty surgery: outcome and

cost. J Arthroplasty 1999;14:669.

25. Kocher MS, Erens G, Thornhill TS, et al. Cost and

effectiveness of routine pathological examination of

operative specimens obtained during primary total

hip and knee replacement in patients with osteoar-

thritis. J Bone Joint Surg Am 2000;82-A:1531.

26. Healy JC, Frankforter SA, Graves BK, et al. Preoper-

ative autologous blood donation in total-hip arthro-

plasty. A cost-effectiveness analysis. Arch Pathol Lab

Med 1994;118:465.

27. Iorio R, Whang W, Healy WL, et al. The utility of

bladder catheterization in total hip arthroplasty. Clin

Orthop Relat Res 2005:148.

28. Sonnenberg FA, Gregory P, Yomtovian R, et al. The

cost-effectiveness of autologous transfusion revisited:

implications of an increased risk of bacterial infection

with allogeneic transfusion. Transfusion 1999;

39:808.

29. Williams JI, Mahomed N, Chapeskie KK. Shorter

waiting times for hip and knee replacement on the

horizon. Hosp Q 2000;4:21.

30. Tran V, Doctor D, Nelson C, et al. The cost of

postoperative home health extended care rehabili-

tation facilities for total joint replacement patients.

American Academy of Orthopaedic Surgeons An-

nual Meeting; 1999.

31. Fisher ES, Wennberg DE, Stukel TA, et al. The

implications of regional variations in Medicare

spending. Part 1: The content, quality, and accessi-

bility of care. Ann Intern Med 2003;138:273.