Daniel M. Keller, PhD

October 18, 2011

October 18, 2011 (Belgrade, Serbia) — In the first study to evaluate the feasibility of rapid HIV testing by primary care physicians in private practice in France, physicians had trouble drawing blood and preparing the sample, and complained of time constraints. At this point, these factors appear to be significant barriers to rapid testing in physicians offices, Raphael Gauthier, MD, from the Department of General Medicine at the University of Paris, France, told delegates here at the 13th European AIDS Conference of the European AIDS Clinical Society.

He explained that rapid HIV testing is new in France, that training is needed to ensure quality control, and that an improved test with a simplified sampling method could possibly overcome the barriers. Generally, three quarters of the 5 million HIV screening tests are performed by private laboratories with a physician's prescription. About 6700 new infections are diagnosed each year (a yield of about 0.13%).

"Despite this screening activity, the proportion of people unaware they that are infected with HIV remains high, and a third of the newly diagnosed patients access care [with a] late presentation of their HIV disease," Dr. Gauthier said. He estimates that there are 50,000 people in France who are unaware that they are seropositive.

French HIV testing guidelines updated in 2008 recommended routine testing, and on the basis of many studies from the United States, rapid HIV testing has been encouraged for every primary care venue in France. However, it was not clear whether such a plan would be feasible in private practices in Europe.

Thus, Dr. Gauthier and colleagues designed a study to determine the feasibility and acceptability of screening with a rapid HIV test in French private physicians offices and to identify any difficulties with the procedure.

This prospective study was performed in physicians' offices over 30 working days from June to October 2010. The physicians were affiliated with HIV healthcare networks, but were not working in sexually transmitted infection or genitourinary medicine clinics. All were trained in the use of the rapid test, and all adults tested were covered by medical insurance.

Patients waiting to see a physician were informed that they could have a rapid HIV test. Those accepting the offer or consenting after a physician recommendation were tested with the Vikia HIV 1/2 test, using finger-stick whole blood. The blood was collected in a capillary tube and tested in a disposable chromatographic device; the result was available in 30 minutes.

Physicians (84 general practice, 6 dermatology, 1 gynecology) recruited 383 patients (48.8% female) with a mean age of 36.2 years (range, 18 to 86 years), and 382 patients completed the test, for an acceptance rate of 99.7%. About two thirds of the tests were done at the patient's request; the rest were done at the physician's recommendation.

Of 378 test results, 92.1% were negative and 7.9% were invalid; 41.9% of doctors had difficulty with the test. Of those having difficulties, "the main problem was the blood sampling technique," Dr. Gauthier said. Physicians reported various difficulties: "the inefficacy of the microlancet to draw enough blood, blood not coming easily into the capillary tube, and air formation or blood coagulation in the capillary." Almost 92% of the problems that physicians reported involved blood sampling.

Thirty test results were invalid, 90% of which were associated with physicians' difficulties in performing the tests.

"We believe that those invalid tests correspond to the extreme cases of blood sampling difficulty, where the physicians could not obtain the 75 μL of blood necessary to perform the test," Dr. Gauthier explained. "Those results came from 19 physicians. The demographic data and participation data are similar to the rest of the investigator population, but we observed that two thirds of them attended the training session in Paris. So we can assume that this training session failed in some way to provide them a complete technical competency."

Seventy-two physicians filled out a poststudy questionnaire, and about 40% said they would not continue to use the rapid test in their daily practices, citing blood sampling issues (43.1%), time constraints (12.5%), and various other barriers.

Session chair Georg Behrens, MD, professor of immunology at Hanover Medical School in Germany, told Medscape Medical News that "this study shows that physicians are a main barrier to HIV testing."

He said that more rapid tests might not overcome this barrier because testing is an extra effort in a general practice. "The immediate feedback is that you need a lot of tests to get 1 positive one. I think that many [general practitioners] will give up over time, rather than using even a shorter test system," he said. In essence, the yield and reward are just not there to justify doing the tests.

Dr. Behrens pointed out that patients are interested in rapid testing because of the quick result. He suggested that a better alternative is to contact the patient by telephone: "I think it's a good way to have a connection...to get [the patient] back into the clinic and then discuss the results." Such a system would require accurate contact information for the patient and assigning someone to do the follow-up.

Dr. Behrens said he thinks a better alternative to universal rapid testing in physicians' offices, with potentially higher yields, is to "go into high-risk groups and other settings."

The study received grant support from Gilead and Vikia HIV 1/2 rapid test, a product of bioMérieux. Dr. Gauthier and Dr. Behrens have disclosed no relevant financial relationships.

13th European AIDS Conference of the European AIDS Clinical Society (EACS): Abstract PS8/4. Presented October 14, 2011.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.