Cost Effectiveness of MRSA Screening & DecolonizationJoseph A. Bosco, MD, Vice Chair of Clinical AffairsJames Slover, MD, MS, Associate Professor, Orthopaedic SurgeonLorraine Hutzler, Quality Project Manager

The National Comparative Effectiveness Summit, Washington D.C.11/6/2012

DIVISION OF QUALITY & PATIENT SAFETY

BACKGROUND & PROCEDURE COSTS

• Arthroplasty and spinal fusion procedures are common and projected to grow rapidly over the next 25 years.

• Approximately 658,000 primary joint knee arthroplasties performed in the U.S. (2011)

• 100,00 revisions (2011)• Demand projected to rise to 4 million primary arthroplasties and 375,000 revisions by 2030

• Spinal fusion procedures had an exponential increase of 73% between 1997 and 2005

• Infection rates remain significant:• hip and knee ( 1.0%-2.0%)• Spine (2.0%- 3.0%)

• The proportion of revisions due to infection projected to rise for the next 25 years• Economic burden of infections is expected to exceed 50% of inpatient resources for revision arthoplasties in 2016 for total hip and in 2030 for total knees

2

Epidemiology

3

SPINE

TOTAL JOINT

Kurtz et al. JBJS, 2007

Deyo et al. Spine, 2005

RESOURCE UTILIZATION

4

Bozic et al. JBJS, 2005

RESOURCE UTILIZATION

5

Calderone et al. Ortho clin North Am, 1996

Projected Costs

• Cost per infected case: 75K• Numbers of infected TJR’s: 10,000

• Assuming infection rate of 1.0% and 1M TJR’s

• Total cost of treating the infected TJR’s : 750M

• In 2020 the projected cost will be1.5B

6

MRSA and Surgical Site Infections

• Staphylococcus aureus is a major pathogen among orthopaedic SSI• Accounts for 50% of SSIs in U.S. and British hospitals

• Colonization with S. aureus has been identified as a risk factor for SSI among orthopaedic patients

• Decolonization has been shown as a way to reduce MRSA SSI risk among colonized patients preparing for surgery

7

The Non Financial Costs: PUBLIC REPORTING (PERCEPTIONS)• Public reporting requirement for certain health-care associated infections (HACs)

• In 32 states reporting of HACs is mandated• 16 states use data collected from adverse event reporting systems for both regulatory and quality improvement purposes

• 31 states track at least one Medicare HAC

• Patients preparing to undergo elective surgery are encouraged to evaluate providers based on outcomes ie. infection

• Insurers, Medicare/Medicaid are likely to select participating hospitals or provide better reimbursement to institutions showing best practices and outcomes

8

COST OF NEGATIVE PUBLICITY

9

STATE TRACKING OF MEDICARE HACs

10

STATE TRACKING OF MEDICARE HACs

11

NYUHJD MRSA SCREENING & TREATMENT PROTOCOL• S. aureus screening

program was instituted for patients undergoing primary hip or knee arthroplasty at our institution

• All patients going through our preadmission testing participated in the screening

• Patients were given a prescription for mupirocin treatment and received a nasal culture preoperatively

• Screening and treatment regimen as shown on the right

12

SURGICAL SITE INFECTION REDUCTION

• In our patient population, S. aureus decolonization led to a 13% decrease in deep surgical site infections

• These findings did not reach statistical significance

• They represented a positive trend towards the efficacy of a decolonization program in decreasing infections

• We performed a power analysis and determined that a sample size of 57,604 patients in each group would be required for statistical significance given the low rate of infections in total joint surgery

13

MRSA HOSPITAL ACQUIRED INFECTION REDUCTION

• Our decolonization program resulted in a 30.8% decrease in the rate of MRSA HAI’s at our institution

• The rate of positive MRSA cultures was 1.23 per 1000 inpatient days in the 15 month period preceding the initiation of our protocol

• This rate decreased to 0.83 in the 24 months after initiation of our protocol(p=0.02)

• We cannot attribute the decrease in MRSA HAI’s to a overall decrease in MRSA prevalence in the community, as the rate of MRSA HAI’s at our affiliated Medical center located 1 mile away remained constant during the study period

14

MRSA HOSPITAL ACQUIRED INFECTION REDUCTION

15

MRSA HOSPITAL ACQUIRED INFECTION REDUCTION

16

MRSA HOSPITAL ACQUIRED INFECTION REDUCTION

17

LENGTH OF DECOLONIZATION

18

•10.8% of the 5,638 patients who completed our decolonization program returned to pre-admission testing within a year of the index culture and were recultured

•These repeat visits included scheduled staged procedures or a rescheduled surgery.

•We divided the results of the repeat cultures into three groups:Group 1 –persistent nasal decolonizationGroup 2-recolonization or failed decolonizationGroup 3- newly colonized

MRSA decolonization persisted in 61.5% of patients at a mean of 159 days.MRSA decolonization persisted in 72.2% of patients at a mean of 155 days.Overall 70.1% of patients with initial positive cultures remained decolonized.

LENGTH OF DECOLONIZATION

19

•MRSA decolonization persisted in 61.5% of patients at a mean of 159 days.•MRSA decolonization persisted in 72.2% of patients at a mean of 155 days.•Overall 70.1% of patients with initial positive cultures remained decolonized.

PATIENT SATISFACTION WITH MRSA PROTOCOL

20

•146 patients undergoing total joint arthroplasty or spine surgery were enrolled in the study

•The S. aureus eradication regimen was introduced to patients at our Pre-Admission Testing clinic

•Patients were instructed to purchase a chlorhexidine gluconate (CHG) containing soap and were prescribed a five-day course of intranasal mupirocin ointment (MO)

•Nasal cultures were performed preoperatively

PATIENT SATISFACTION WITH MRSA PROTOCOL

21

•On the day of surgery, hospital staff assessed patient compliance with the pre-operative S. aureus reduction regimen and obtained culture results

•Post-operatively, patients were given anonymous surveys assessing OOP expenses and attitudes towards the eradication regimen

•The survey evaluated:•Patient awareness/concern for infection•Ease of compliance with standardized S. aureus reduction protocol•Financial burden of treatment

•Compliance items were rated on a 4-point Likert type scale ranging from very easy to very hard

•Completed surveys were collected prior to discharge

PATIENT SATISFACTION WITH MRSA PROTOCOL

22

Surveys Distributed 146

Surveys returned 100 (68%)

Demographics

Age: Mean (range) 55 (19-85)

Gender* Male Female

48/100 (48%) 50/100 (50%)

* 2 patients did not specify gender

SSI prevention Initiative: Patient Attitude and Compliance

PATIENT SATISFACTION WITH MRSA PROTOCOL

23

PAT attendance

Attended PAT and received protocol 85/100 (85%)

Did not attend PAT 14/100 (14%)Unable to recall PAT attendance 1/100 (1%)

Followed protocol for MO 69/85 (81%)Followed protocol for CHG 76/85 (89%)

PATIENT SATISFACTION WITH MRSA PROTOCOL

24

MO Ease of Compliance

Followed Protocol 69/85 (81%)

Out of pocket expense 37/69 (54%) $25 ($2-$115)

Hard or very hard to purchase 9/69 (13%)

Difficult to locate 4/69 (6%)

CHG Ease of Compliance

Follow Protocol 76/85 (89%)Easy or very easy to use 71/76 (93%)

Yes No

Concerned about SSIs 46/100 (46%) 54/100 (54%)

COST EFFECTIVENESS-DECISION MODEL

• A Markov decision model was used to evaluate the cost of the S. aureus screening program.

• The decision model depicts the pathway followed by patients in the screening program.

• Patients transition along the appropriate arm of the decision tree according to the likelihood of each event, determined by the probability of each event in our patient cohort or from published literature. In addition, cost was assigned to every test, treatment, and primary and revision procedure within the model.

25

MODEL PROBABILITIES• Probability of patient compliance and positive nasal cultures for MRSA/MSSA taken

directly from our experience with our patient cohort

• Each patient was called the night before surgery and asked whether they had completed the mupirocin treatment regimen to assess compliance

• Probability for infection set at 1.5% for the base case

26

COSTS

• Average cost of a primary joint arthroplasty $15,000• Average cost of a high-risk spine surgery $50,000• Average cost of a septic joint revision $70,000• Average cost of treating a spine infection $100,000

27

COST OF TREATING INFECTED TJA

28

• The graph demonstrates the results of a 2- way sensitivity analysis

• Areas in green indicate profiles where the screening program is cost saving

• Areas in blue indicate areas where it is cost saving to not institute the screening program.

COST OF TREATING INFECTED SPINE

29

• The graph demonstrates the results of a 2- way sensitivity analysis

• Areas in green indicate profiles where the screening program is cost saving

• Areas in blue indicate areas where it is cost saving to not institute the screening program

COST OF TREATING INFECTIONS

30

•A modest reduction in the surgical site infection rate = high cost savings

•This is due to the low cost of screening and decolonization and high cost of treating infected joint replacements

RESULTS

• The larger the impact of the screening program on the number of joint and spine infections and the higher the cost of treating an infected joint, the more likely the screening program is to be cost saving.

• If the cost of treating an infected hip or knee arthroplasty = the cost of a primary knee arthroplasty ($15,000) the screening program needs to have a 35% reduction in the revision rate or a relative revision rate of 65% for patients in the screening program to be cost saving.

• For spine patients the reduction to make the program cost saving is 10% if the cost of treating a spine infection = the average cost of a primary spine surgery ($50,000).

31

RESULTS

32

In Simple Terms:

• Cost per decolonizing one patient: $110.00• Cost for decolonizing 1000 patients: 110K• If you prevent 2 infections in the group of 1000

• Savings of 150K

33

CONCLUSIONS

•Currently the costs of the program are shared between the hospital and the patient/insurer

•The cost of screening and the addition of vancomycin for surgical prophylaxis of MRSA-positive patients is borne by the hospital

•The cost of the decolonization regimen is the patient or insurer’s responsibility

34

CONCLUSIONS

•An alternative strategy for screening and decolonization is to use molecular testing with rapid turnaround time and prescribe decolonization regimen to only those patients who are SA positive

•The hospital would have increased screening costs but the decolonization medication would only be needed by 30% of patients

35

CONCLUSIONS

•The Markov decision analysis model demonstrates that universal SA screening and decolonization for hip and knee arthroplasty and spinal fusion patients needs to result in only a modest reduction in the surgical site infection rate to be cost saving

36

THANK YOU!

37