THE COST-EFFECTIVENESS OF AN AUDIT / FEEDBACK ADHERENCE PROGRAM TO PROMOTE UTILIZATION OF ACTIVE PHYSICAL THERAPY INTERVENTION STRATEGIES FOR LOW BACK PAIN

   Purpose:  The purpose of this model was to determine the cost-effectiveness of adherence program designed to improve utilization of active physical therapy interventions for low back pain (LBP) patients.  Clinical practice guidelines have long favored active versus passive physical therapy interventions for LBP. The literature also suggests a cost-savings associated with active intervention strategies.  Despite these findings, guideline compliance is generally poor creating a demand for effective adherence programs to reduce widespread practice variation.  The cost-effectiveness of implementing such an adherence program however has not been established. 

   Methods:  A decision analytic model was used to compare the cost-effectiveness of implementing an educational/audit based guideline adherence program to a usual care approach without an adherence strategy.  Probabilities for adherence and transitioning among health states were obtained from the literature and related preliminary data.  In addition, costs and benefits were acquired from the literature as well as the University of Pittsburgh Medical Center (UPMC) Health Plan.  The analysis was conducted from a payer perspective over a two year time frame.  Discounting was not applied because of the limited time horizon.  In the base case analysis the adherence program was assumed to cost $8,500 per 1000 patients enrolled. One-way, two-way and probabilistic sensitivity analyses were also performed. 

   Results: Implementing an adherence program cost $706 less per patient while gaining 0.11 QALY, dominating the alternative strategy of not implementing an adherence program.  In one-way sensitivity analyses, results were most sensitive to the probability of guideline adherence with or without the adherence program.  Two-way sensitivity analysis of the probability of adherence given a program versus no program showed that only an 8% absolute increase in the probability of adherence with the program (base case 42%) would be required for the intervention strategy to be cost saving.  The program remained cost saving if total program costs were < $43,000/1000 patients.  Finally, probabilistic sensitivity analysis, varying all parameters simultaneously over distributions, found the adherence program cost saving in 91% of the 10,000 model iterations.

   Conclusions:  An adherence program that promotes active therapeutic interventions is both less costly and more effective than a usual care rehabilitation approach without an adherence program.   Health plan payers should consider adopting this approach as a cost-effective strategy to reduce clinical practice variation and improve service utilization.