Major Finding: The use of unrecommended molecular tests for lower genital tract infections increased total molecular test spending by 29%, compared with the cost of recommended molecular tests alone.
Data Source: Laboratory claims from 82,443 patients in a large national insurance database from the year 2008.
Disclosures: Dr. Eckert said that she had no relevant financial conflicts to disclose.
SANTA FE, N.M. — Molecular diagnostic tests appear to be overused for the diagnosis of lower genital tract infections in women, results from an analysis of national practice patterns showed.
In fact, use of unrecommended tests increased total molecular test spending by 29%, compared with the cost of recommended molecular tests alone.
Vaginitis is a common clinical problem in which accurate diagnosis “can be challenging,” Dr. Linda O'Neal Eckert said during the meeting “In fact, many of us make a living by trying to accurately diagnose challenging cases of vaginitis. The availability of molecular testing for female genital tract infections is definitely increasing. Their use may offer appeal, partly because there's a perception of accuracy when you can get the result back and hope that that might determine the etiology of the symptoms, and also because using these tests does not require a microscope.”
Dr. Eckert of the University of Washington, Seattle, and her associates conducted a cross-sectional study of laboratory claims within a large national insurance database for the year 2008. “This database represents 3.5 million commercially insured patients,” she said. “More than 500 laboratories are in this database, including both commercial and hospital laboratories.”
The researchers used ICD-9 codes to select women who presented for a first evaluation of vaginal and cervical infections, and then identified molecular tests performed to detect infections from laboratory CPT codes billed on the same visit. They used published guidelines (N. Engl. J. Med 2006;355:1244-52; ACOG Practice Bulletin No. 72, May 2006, reaffirmed 2008) to classify molecular tests as either recommended (Chlamydia trachomatis, Neisseria gonorrhoeae, Trichomonas vaginalis, herpes simplex virus) or not recommended (Candida species and subspecies, Gardnerella vaginalis, staphylococcus, streptococcus, enterococcus, cytomegalovirus, and others) for use in this setting.
Of 211,933 women in the database, 82,443 met criteria for inclusion in the study. Of these patients, 61,132 (74%) had recommended tests, 17,934 (22%) had both recommended and nonrecommended tests, and 3,377 (4%) had only nonrecommended tests performed at their initial visits.
The most common molecular test classified as not recommended that was ordered was for the detection of an agent not otherwise specified by amplified DNA probe method. Dr. Eckert reported that this test was performed 15,526 times at an average cost of $21.60 per test, for a total amount spent of $335,290.
The second most common test performed, a direct DNA probe to search for G. vaginalis, was performed 14,698 times at an average cost of $21.30 per test, for a total amount spent of $313,298. The third most common test performed, a direct DNA probe for Candida species and subspecies, was performed 14,630 times at an average cost of $21.37 per test, for a total amount spent of $312,707.
Overall, a total of $6,328,168 was spent on molecular testing, Dr. Eckert said. Of this total, the cost of recommended tests amounted to $4,816,407 (76.1%), whereas the cost of nonrecommended tests amounted to $1,408,270 million (22.3%). The use of nonspecified molecular tests accounted for the remaining 1.6%, or $103,491.
The researchers determined that the average cost of recommended testing was $61 per patient visit, whereas the average cost of additional unrecommended testing was $66 per visit. In the aggregate, use of unrecommended tests increased total molecular test spending by 29%, compared with the cost of recommend testing alone.
One of the meeting attendees, Dr. Harold C. Wiesenfeld, director of the division of reproductive infectious diseases in the University of Pittsburgh, said that some of his patients who aren't covered by insurance are getting bills approaching $1,400 for vaginitis panels. “There is no data that any clinical outcome is improved based on these tests,” he commented.
Dr. Eckert noted that one of the labs studied accounted for the vast majority of nonrecommended molecular testing. “There is a significant variation that occurs between laboratories for the use of recommended vs. nonrecommended molecular tests,” she said.
The use of unrecommended tests increased total molecular test spending by 29%.
Source DR. ECKERT