Studies show benefits of early HIV care

Early drug use lowers nondrug costs

Three studies presented recently at scientific meetings make the argument that early drug intervention reduces the average monthly cost of care for people with HIV. All three studies indicate that medical costs for people with HIV treated with early drug intervention stabilize over time. And all three attempt to determine the most effective way to manage HIV costs and influence public policy associated with medical care for people with HIV.

Marta E. Urdaneta, PhD, manager in the outcomes and research management division of Merck and Company in West Point, PA, participated in two of the studies. Urdaneta used data collected by Clinical Partners, a San Francisco-based company that provides data support and HIV case management services to managed care organizations nationwide.

The first study found that average monthly costs of care for people with HIV in managed care stabilized from 1995 to 1997, the period in which protease inhibitors and combination antiretroviral therapy became widely adopted in medical practice.

"We found that between 1995 and 1997, quarterly mortality rates dropped from almost 5% to less than 1%, and average monthly costs of patient care stabilized at about $1,400 per patient among persons with HIV enrolled in several managed care plans in California and Texas," says Diane Lapins, MPH, RN, BSN, MSA, vice president for strategic services at Clinical Partners.

The patient population in this study included HIV-positive non-Medicaid and non-MediCal adult males and females who were enrolled in capitated, non-capitated, and preferred provider health plans managed by Clinical Partners from Jan. 1, 1995, through Dec. 31, 1997. During this period, the number of patients enrolled each month ranged from 474 to 723. "The range changed because we looked at the number of people enrolled each month," explains Urdaneta, who presented the findings at the recent Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) in San Diego.

During the course of the study, the average monthly per patient cost of treating a person with HIV stabilized in the managed care population. "Although the introduction of triple-combination therapy led to a rise in patient-per-month drug costs from $200 to $1,100, this increase in drug costs was accompanied by a corresponding decrease in average monthly non-drug costs from almost $1,300 to about $200, including hospital, professional, laboratory, and home care services," Lapins notes. "In fact, the overall average monthly cost of treating HIV patients held stable at about $1,300 over the last 14 months of the 36-month study."

Healthy patients cost less

The second study from Merck’s outcomes research and management division found that the average monthly costs for treating persons with HIV in managed care is lower for patients with undetectable viral loads than for those with detectable ones. "We defined undetectable viral load as less than 400 copies/ml," explains Urdaneta, who presented the findings at the recent annual meeting of the Infectious Diseases Society of America (IDSA) in Denver.

"This analysis shows that the average monthly HIV treatment costs per HIV member — including drug, non-drug, and total costs — were lower for patients with undetectable viral loads," says Lapins. "In addition, increases in average monthly HIV patient drug costs per quarter were offset by decreases in average monthly non-drug costs per quarter in both groups. In fact, monthly total costs per patient for the HIV patient population studied were stable over the study period."

The main difference between the two studies was the focus on cost implications of viral suppression, Urdaneta points out. The study focused on the period from the first quarter of 1996 through the last quarter of 1997 for the simple reason that viral load information was not widely available for patients in the Clinical Partners data set before 1996, she says, adding that patients in the second study ranged from 556 to 723. "Obviously, patients moved between groups as their viral load status changed. They were classified as suppressed or not suppressed each quarter and average monthly drug costs were calculated each quarter."

Study findings include the following:

• The proportion of patients with viral levels suppressed to undetectable levels rose from 6% in the first quarter of 1996 to 56% in the fourth quarter of 1997.

• Overall monthly costs per patient with undetectable viral loads were consistently lower ($1,067) than costs for patients who did not achieve undetectable viral loads ($1,634).

"Not surprisingly, care for patients with undetectable viral loads is less costly than for those with higher viral loads in a managed care setting," Lapins says.

"The second study also underscores the need to get patients on drugs with a specific goal of suppression if we really want to keep costs in check," adds Urdaneta. "By demonstrating that all costs are significantly reduced for patients who are suppressed, we hope we can gain more support to get patients on drugs earlier."

"As many managed care organizations tend to view drug budgets and drug costs independently of other operational costs, these data underscore the need for viewing the per patient cost of HIV care as a whole," Lapins says. "It is important to recognize that HIV care encompasses many other components of care, such as indirect costs to community services, employers, friends, and families."

Persuading managed care organizations that early drug intervention reduces average overall monthly costs for members with HIV is challenging. Persuading state and federal governments that early drug interventions is cost-effective is even more daunting.

Research presented at the recent 12th World AIDS Conference in Geneva, Switzerland, used a mathematical model to demonstrate that providing early access to AIDS treatment for the poor and uninsured would be cost-effective, or at least cost-neutral.

Current Medicaid eligibility rules only extend health care coverage to persons with HIV who have suffered an AIDS-defining illness or who have a CD4 count below 200, notes Gary Rose, JD, director of public policy for the Title II Community AIDS National Network and the Treatment Access Expansion Project in Washington, DC. "The state of medical science is such that there is a possibility with early, aggressive treatment that persons with HIV may never get AIDS," he says. "The problem with that is about 25% of persons with HIV have no insurance whatsoever, and the Health Care Financing Administration says they have to meet requirements of full-blown AIDS before qualifying for Medicaid coverage."

Cost of survival benefit a bargain

The study, conducted by pharmacoeconomists using established simulation techniques and data from existing clinical trials, drew several conclusions:

• Early intervention with AIDS drug cocktails delays the progression of AIDS.

• Providing early treatment, when patients have CD4 counts between 200 and 500 and have never experienced an AIDS-defining event, results in prolonged survival and is cost-effective when compared with costs of delaying treatment until patients have CD4s count of less than 200 or have experienced an AIDS-defining event.

• By five years, 15.3% of patients receiving early drug intervention are expected to develop an AIDS-defining event, compared with 25.4% if treatment is delayed.

• By 10 years, 40% of patients receiving early with drug intervention are expected to develop an AIDS-defining event, compared with 48.8% if treatment is delayed.

"At the most basic level, the research shows that patients receiving early, aggressive treatment experience increased life span," says Rose. "Moreover, the overall cost of providing patients with this survival benefit is negligible."

"This study provides an important addition to any discussion on early treatment for persons with HIV," says John C. Hornberger, MD, MS, director of health economics for Roche Global Pharmacoeconomic Research and a clinical associate professor of medicine in the division of general internal medicine at Stanford University in Stanford, CA, who conducted the study. "There is little argument that the clinical guidelines for treatment of persons with HIV, including early treatment with drug therapies, make sense in terms of efficacy, but we wanted to prove that they also make sense economically."

In fact, the simulation model found that the costs associated with delaying the onset of the symptoms of AIDS only increases overall medical care costs over the first five years by $241 per patient per year, he says.

The next step is to take Hornberger’s simulation model and do it one better, Rose says. "We want to look at all the costs associated with treating early vs. treating late on every single number we could get our hands on, including productivity levels, prison, unemployment. We want to beta test the pharmacoeconomic model using state Medicaid recipients to see if we can make a real-world argument that early treatment has a positive cost impact," he says, adding that several states have been approached about participating in the beta test.

"If the model works, and the intervention proved cost-effective, there’s no reason why the same model can’t be played with and used to show effectiveness of early intervention and prevention for other disease," he adds.