The trusted source for
healthcare information and
abstract & commentary
Synopsis: HIV care providers with the most experience were more likely to use more complex and newer therapies generally in keeping with recent guidelines than those with less experience.
Source: Brosgart CL, et al. Clinical experience and choice of drug therapy for human immunodeficiency virus disease. Clin Infect Dis 1999;28:14-22.
Brosgart and colleagues from the community consortium in San Francisco assessed the prescription of various drug therapies for 30 HIV-related medical conditions by HIV care providers located throughout the United States in May through June 1996. A total of 524 of 999 providers responded to the survey, although only 343 questionnaires were completed (181 providers were no longer in practice or providing care to patients with HIV). Of those who responded, 93% were physicians, most of whom were infectious disease specialists (35%), internists (34%), or family practitioners (20%). More than one-half provided primary care to more than 50 HIV-infected patients, while 20% cared for 6-20 patients and 8% had five or less. Nearly one-half (44%) worked in private practice or a private hospital.
Providers with the most experience, defined by their number of primary care HIV patients, were significantly more likely to use more complex and newer therapies generally in keeping with recent guidelines than those with less experience. More experienced providers were especially more likely to use more aggressive combination antiretroviral therapy or recently approved agents. For example, providers with more than 50 patients prescribed triple combination therapy to 35% of patients with CD4+ counts between 200-500, 78% of those with CD4+ counts between 50 and 200, and 86% of those with CD4+ counts under 50. In contrast, providers with five or fewer patients used triples in 4%, 35%, and 46% for similarly defined patient groups, respectively (P < 0.001 in all instances). Providers with fewer patients were significantly more likely to use a nucleoside analog reverse transcriptor inhibitor as monotherapy. In addition, while 75% of more experienced providers routinely used measurements of viral load, only 42% of those with five or fewer patients did so.
This more aggressive pattern of prescribing extended to several areas of prophylaxis for opportunistic infection. While patterns of use of prophylaxis against PCP were similar for all groups, and almost everyone used trimethoprim-sulfamethoxazole as first-line therapy, providers with greater numbers of patients were significantly more likely to prescribe prophylaxis for toxoplasmosis and cytomegalovirus than their counterparts with fewer patients. They were also more likely to use clarithromycin or azithromycin for Mycobacterium avium complex prophylaxis, whereas providers with fewer patients were more likely to use rifabutin.
Several earlier studies have demonstrated that experience is critical to the successful primary care of patients with HIV. In one study of 403 patients with AIDS, even modest levels of experience led to a 43% reduction in mortality.1 The median survival of patients cared for by more experienced physicians (who provided care to at least 25 patients) was 26 months compared with only 14 months for those cared for by physicians with five or fewer patients (P < 0.001).
While outcomes were not specifically examined in the study by Brosgart et al, these results provide an important link to those earlier studies. Important differences in the level of sophistication and familiarity with newer HIV therapeutics exist for providers with greater levels of experience, defined simply by the numbers of patients for whom they provide primary care. The use of more aggressive combination antiretroviral therapy, as well as prophylaxis for certain conditions such as MAC, have been shown in multiple studies to decrease the risk of opportunistic infection and prolong survival.
However, several compelling questions remain unanswered by these data, chief among them the reasons why providers who care for fewer patients in this study appeared to use older or less aggressive therapies. Numbers of patients under care does not necessarily translate to a greater knowledge base. But, providers with fewer patients obviously have less opportunity to become familiar with newer agents. No other area of primary care moves as quickly or is as complex as HIV care, especially now with 14 different antiretroviral agents presently approved for use, and one other available through expanded access.
The use of newer resistance assays, such as genotype and phenotype assays, will add greatly to this complexity, but even here experience in the interpretation of this information may matter greatly. Data from the CPCRA study showed that the use of an expert virologist’s interpretation of the genotype data in the selection of antiretroviral therapy significantly influenced the degree of HIV viral load suppression.2 At sites where the virologist’s suggested regimen was used 100% of the time, the level of plasma HIV RNA fell a mean of 1.5 logs compared with a 0.5 log reduction in patients whose disease was managed using only CD4 and HIV-RNA data without benefit of genotype data. At sites where the expert opinion was not used, however, the availability of genotype data did not provide any additional benefit in viral load suppression.
Are providers with fewer patients unaware of recent published guidelines, is the acquisition and use of newer information simply delayed, or are they more cautious in their interpretation of that information? For example, rifabutin was approved by the FDA for use in the prophylaxis of MAC in December 1992, allowing the drug to be marketed for this use. Rifabutin was subsequently recommended for the prophylaxis of disseminated MAC by an expert advisory panel in 1993,3 following which additional supportive data became available regarding the cost-benefits of its use as a prophylactic agent. On the other hand, information on the effectiveness of clarithromycin and azithromycin for MAC prophylaxis was published in abstract form in the fall of 1995 and in the winter of 1996, but the peer-reviewed journal articles establishing their efficacy were not published until July 1996,4,5 after this survey was completed. Therefore, it’s conceivable that some physicians were unaware of the newer information available only in abstract proceedings, or perhaps preferred to wait until data had been peer reviewed before altering their practices.
Although "experience," however it may be defined, is important, we should keep in mind that information can be mainstreamed too quickly, in advance of confirmatory data or important follow-up information. I knew a physician in our area whose practice was dedicated to HIV care but who lobbied hard for the expanded use of Compound Q. Prophylaxis for CMV retinitis may have been vogue for a while, at least before more active antiretroviral therapy became available, but could you make the claim that you were a better doctor because you used it? It was never clear who would benefit from this intervention, and preliminary data suggested that prophylactic ganciclovir failed to reduce the risk of CMV retinitis in patients with high plasma levels of circulating CMV DNA who were at the greatest risk for this complication.
Furthermore, saquinavir hard gel was used extensively by AIDS "experts" in our areas as soon as it became available through expanded access protocols. While this practice no doubt spared the lives of some patients, others were doomed to fail this substandard drug. The resultant emergence of cross-resistance to other agents in its same class eliminated future options for many of these patients.
Taking this thought to its extreme, while the use of complex, multidrug antiretroviral therapies has benefited thousands of patients with HIV, the exultation of the AIDS-treating community and patients alike is being confronted by the grim reality of the impossibility of maintaining "100% compliance for life," the emergence of broadly resistant strains in up to 40% of patients, and the increasing appearance of resistant strains in newly infected persons. While the timely availability of information is important, and familiarity with the myriad agents is helpful, a more cautious—or even skeptical—approach is sometimes healthier.
1. Kitahara, et al. Third Conference on Human Retroviruses and Opportunistic Infections, January 28-February 1, 1996. Washington D.C. Abstract #413.
2. Baxter JD, et al. Sixth Conference on Retroviruses and Opportunistic Infections, January 1999, Chicago. Abstract #LB 8.
3. Masur H. N Engl J Med 1993;329:898-904.
4. Pierce M, et al. N Engl J Med 1996;335:384-391.
5. Havlir DV, et al. N Engl J Med 1996;335:392-398.
a. prescribe more complex and newer therapies for HIV patients.
b. routinely use measurements of viral load.
c. are more likely to prescribe prophylaxis for toxoplasmosis and cytomegalovirus.
d. All of the above