Rate Variations in Orthopaedic Procedures: Total Shoulder Replacement, Rotator Cuff Repair, and Hemiarthroplasty
The demonstration of dramatic variations in the incidence of surgical procedures among geographical areas has been one of the major stimuli to the field of outcomes assessment.
Within the area of orthopaedic intervention, a number of low variability (e.g. hip and ankle fracture) as well as high variability procedures (e.g. spinal surgery) have been identified. However, there is considerable controversy regarding the factors which drive this variation. Specifically there has been some disagreement in the literature regarding the respective roles of physician manpower and populatiion density in driving surgical rates.
We have further explored these issues using total shoulder replacement (TSR), rotator cuff repair (RCR), and hemiarthroplasty (Hemi) as case studies. This paper documents state to state variation in the rates of these procedures and examines factors which might account for these variations, such as the density of MD’s and orthopaedic surgeons in an area (supply-side) and the socioeconomic attributes, population density, and occupational trends of the population (demand-side).
METHODOLOGY: The regional incidences of TSR, RCR, and Hemis were examined via analysis of the HCFA Medicare database (MEDPAR, 1994) and several state-wide data sets. Analysis based on MEDPAR data was restricted to only those Medicare patients over age 65, while state-wide data sets were analyzed for all age groups. All rates discussed below are age-adjusted.
RESULTS AND DISCUSSION: Analysis of MEDPAR data indicates that among those over 65 in the U.S. (32.1 million), there were 3495 TSR’s, 9821 RCR’s, and 3417 Hemis performed in 1992.
The total national incidence of each procedure in that year was 5.8 per 1,000 people over age 65 for TSR’s, 16.6 per 1,000 for RCR’s, and 5.1 per 1,000 for Hemis. Significant variations were noted among states in our sample which were not accounted for by variations in age distribution. However, population density showed a strong negative correlation with all three procedures.
For example, the incidence of TSR was significantly above the national average of 11.6 per 100,000 in Idaho and Wyoming (25.6 and 23.3 per 100,000, respectively), but below average in Massachusetts (5.36) and New York (5.97), two areas with high population densities. After controlling for the demand-side variables of age and population density, several interesting relationships emerged between the supply of physicians and the rates of surgery in each state.
Examining the total supply of all physicians in each state, no significant relationship was noted with the rates of any procedure. However, there is a positive correlation (R=.304; p=.034) between the supply of orthopaedic surgeons and the rates of TSR, but not the other procedures. This positive correlation with TSR rates was even stronger when we examined the supply of shoulder specialists (R=.388; p=.006). Again, this trend is not seen with RCTs or Hemis, suggesting that density of specialized manpower (a supply-side factor) may only have an important role in driving rates of tertiary care procedures such as TSR.
This study suggests that both supply- and demand-side issues play an important role in determining rates of these three procdures in the area of shoulder surgery. Further work is necessary to better understand the relative role of occupational trends, physician subspecialization, and styles of practice in determining the rates of these procedures.
Publication Types:
- Meeting Abstracts
Keywords:
- Age Distribution
- Arthroplasty
- Arthroplasty, Replacement
- Data Collection
- Humans
- Idaho
- Massachusetts
- Medicare
- New York
- Orthopedic Procedures
- Orthopedics
- Outcome Assessment (Health Care)
- Physicians
- United States
- Wyoming
- economics
- manpower
- methods
- surgery
- hsrmtgs
Krant JJ, Gelijns AC, Flatow EL, Heitjan DF, Arons R, Vitale MG; International Society of Technology Assessment in Health Care. Meeting.
Annu Meet Int Soc Technol Assess Health Care Int Soc Technol Assess Health Care Meet. 1997; 13: 92.
International Center for Health Outcomes and Innovation Research, New York, NY, USA.
Source: National Institutes of Health – http://gateway.nlm.nih.gov/MeetingAbstracts/ma?f=102233127.html