Time for a change

The recent decreases in reimbursement for DXA studies seem to have affected and possibly decreased the quality of the studies, particularly at centers where the operator or the person reporting the data may not have undergone formal training in this important technology.

However, considering both the past and the future, I am not yet convinced that is the only problem facing DXA.

WHO presented specific limits for reporting bone mineral density (BMD) in the early 1990s, and these have not changed in almost two decades. The FRAX tool, which was developed a few years ago, has had a significant, positive effect on those who have been trained in its use, but it has been less helpful for those who have not.

The WHO criteria include the following:

BMD is considered as 2-dimensional, although the rest of the world recognizes density as a 3-D issue.
BMD is reported to three decimal places in g/cm2. This, however, is not standard practice with other laboratory measurements.
Classification of the BMD result (T-score) is reported as a negative.

WHO criteria also define T-scores in the following ways:

0 to –1.0 is normal.
–1.1 to –2.5 is low bone mass, often referred to as osteopenia.
Less than –2.5 is osteoporosis.

T-scores above +2.0 are almost always reported as normal, but this is not necessarily accurate, which further demonstrates the need for persons responsible for the DXA study to have sufficient training.

Perhaps we should revisit some of these criteria:

Is there any other set of laboratory data that uses such cutpoints?
Why not develop reference intervals as milligrams, such as whole numbers?
Why not eliminate density and report milligrams per area?
Why not elimintae T-score and Z-score?
Why have cutpoints at –2.5 SD when almost all other lab data report a range from 2 SD below and above the mean of the reference data?

Here is one scenario that provides an example of a conversation with a patient if changes were implemented.

“This is your average spine DXA value in milligrams. This is the bottom half of the range, but your risk for fracture is still quite low. Using this FRAX tool, your 10-year probability of fracture is less than 10% and it is not necessary for you to start treatment right now. We will check your DXA data again in 2 years.”

Here is a second scenario.

“This is your average hip DXA value in milligrams. This is below the range and suggests that you are at increased risk for hip fracture. Using this FRAX tool, you have more than a 5% risk for having a hip fracture over the next 10 years. It is important that you start treatment to lower that risk for fracture, and we will carefully monitor your response to the treatment once a year. The available treatments to lower fracture risk have been well studied, and each of them has been shown to reduce fracture risk.”

Unfortunately, the likelihood that our current reporting system will change is small. However, that’s not an excuse for not thinking about moving forward. My suggestions will no doubt ruffle some feathers, but my goal is to have the experts get together to try and bring DXA studies into line with most other laboratory tools in clinical medicine.