Why there are fewer images being made

David Lee from GE Healthcare and Frank Levy from MIT have published a thoughtful article in Health Affairs entitled "The Sharp Slowdown In Growth Of Medical Imaging: An Early Analysis Suggests Combination Of Policies Was The Cause." The authors noticed that, well before implementation of bundled or global payments, the growth in usage of certain radiological modalities had moderated (see chart above).  They were curious why.

The abstract:

The growth in the use of advanced imaging for Medicare beneficiaries decelerated in 2006 and 2007, ending a decade of growth that had exceeded 6 percent annually. The slowdown raises three questions. Did the slowdown in growth of imaging under Medicare persist and extend to the non-Medicare insured? What factors caused the slowdown? Was the slowdown good or bad for patients? Using claims file data and interviews with health care professionals, we found that the growth of imaging use among both Medicare beneficiaries and the non-Medicare insured slowed to 1–3 percent per year through 2009. One by-product of this deceleration in imaging growth was a weaker market for radiologists, who until recently could demand top salaries. The expansion of prior authorization, increased cost sharing [i.e., with patients], and other policies appear to have contributed to the slowdown. A meaningful fraction of the reduction in use involved imaging studies previously identified as having unproven medical value. What has occurred in the imaging field suggests incentive-based cost control measures can be a useful complement to comparative effectiveness research when a procedure’s ultimate clinical benefit is uncertain. 

The hypothesis:

We hypothesize that prior authorization policies, higher deductibles, and lower reimbursements worked to offset strong nonmedical incentives, such as physicians’ fear of malpractice litigation or a desire to generate revenue to order imaging studies. Furthermore, we speculate that the slowdown may have included a meaningful proportion of procedures with marginal or unproven medical value, as discussed below. If our hypothesis is correct, what has occurred in the imaging field is evidence that reducing nonmedical incentives to perform a procedure is a useful cost-control strategy, where a procedure’s ex ante clinical benefit is uncertain and clinical guidelines are hard to write.

The conclusions:

Logic suggests that the growth in use of advanced imaging would have slowed eventually, but interviews and available evidence point to several policies that slowed the growth in utilization beyond any exhaustion of trend.

These authors identify these as prior authorization, increased cost sharing, reimbursement reductions in the deficit reduction act of 2005, and fear of radiation.