Data find wide variation in BPH treatment patterns

March 1, 2005

Mumbai, India—Rates of surgical and medical therapy for BPH at the county and state levels show significant systematic variation, according to Duke University researchers, who reported their findings during a presentation at the World Congress on Endourology here.

Mumbai, India-Rates of surgical and medical therapy for BPH at the county and state levels show significant systematic variation, according to Duke University researchers, who reported their findings during a presentation at the World Congress on Endourology here.

The Duke team found a large amount of geographic variation in BPH treatment rates. They also discovered a positive association between the dispensing rates of the alpha-blocker tamsulosin (Flomax) and the rates of surgery, and a negative association between the dispensing rates of finasteride (Proscar) and surgical rates.

"We certainly didn't expect a positive relationship between surgery and tamsulosin-dispensing rates," said W. Patrick Springhart, MD, a clinical associate at Duke working under Glenn M. Preminger, MD, and David M. Albala, MD.

Dr. Springhart and colleagues conducted a cross-sectional study using data from a variety of sources, including:

They calculated medical and surgical therapy rates per 100,000 eligible men. Standard measures of variation and multivariable linear regression techniques rounded out the analysis.

"Some studies have looked at variations in prostate cancer treatment rates, but there have been no studies that have analyzed the geographical variations in BPH treatments, nor the factors related to differences in medical and surgical treatment rates," Dr. Springhart said.

The researchers derived mean annual dispensing rates of 548.6 for finasteride, 939.6 for tamsulosin, and 907.5 for doxazosin (Cardura). Dispensing rates of the three BPH drugs were positively associated with one another (p<.0001). Variations in county dispensing rates for the drugs were significantly greater than what would be expected by chance (p<.05).

"The excess variation implies that there are other influential factors, such as the rates of use of other treatment modalities, the state of residence, and the number of urologists and ambulatory surgery centers in a county," he said. "This is what we determined with our linear regressions."

Other factors affecting dispensing rates included surgery rates, managed care penetration, poverty rates, education levels, race, and state of residence (p<.05).

State of residence was also a factor in surgical rates, with men in New Jersey and New York more likely to undergo transurethral resection of the prostate than those in Florida, Pennsylvania, and North Carolina. Ambulatory surgical centers per capita, Medicare managed care penetration, and low education were all positively predictive of increased utilization of surgery (p<.05).

Overall, the mean surgical rate was 591.3. That number decreased by 1 with an 11.6 increase in finasteride use (p=.014). Conversely, tamsulosin rates were positively associated with the rate of surgery (p=.0015), while no relationship was found between doxazosin use and surgical rates.

The negative relationship between finasteride and surgical rates suggests that the two therapies are "market substitutes," while the positive association between tamsulosin and surgery implies "a complement relationship."

The studies were funded by a research grant from GlaxoSmithKline.