HIV care cascade at Kaiser Permanente varies by sex and age

Though there has been improving performance of healthcare delivery at each point of the HIV care cascade, from linkage to care through viral suppression, "success varies significantly by age and gender, even in an integrated care system with equal access to care,” Michael Horberg of the Mid-Atlantic Permanente Research Institute reported earlier this month at IDWeek 2014 in Philadelphia, United States.

Horberg presented the results of an analysis evaluating the care received over a three-year period by all people with HIV enrolled in the Kaiser Permanente (KP) integrated healthcare system, which operates in nine US states and the District of Columbia.

The study found that during the first two years of the study, the percentage of women who were linked to care after diagnosis was significantly lower compared with men, even though their retention once in care was higher. Likewise, a statistically lower percentage of women were taking antiretroviral therapy (ART) and maintaining viral suppression. However, these differences were not observed in 2012, the final year of the study – suggesting that quality improvement initiatives may be addressing gender differences.

Meanwhile, even though there was no difference in linkage to care after receiving an HIV diagnosis by age bracket (13-34, 35-54, and 55+), the increase in age was associated with significantly better retention, higher rates of ART prescription, and viral suppression.

Background

Many studies have looked at the effects of individual interventions that aim to increase the number of people at risk of HIV who get tested, or the number of people who achieve full viral suppression on treatment. Health systems, however, are best judged by how they manage patients over the continuum of care – which actually requires that a number of interventions work well. When they do not, people fall through the cracks and are lost on the "HIV care cascade".

For instance, in the Centers for Disease Control and Prevention's (CDC) HIV in the US: Stages of Care (July 2012), it was estimated that out of the nearly 1,800,000 people living with HIV in the US, less than 80% had been diagnosed, and among these only around 60% had been linked to care. Once they are linked to care, people with HIV need to be retained in care, and the CDC estimated that only 40% were. Those who are ready for treatment need to be prescribed ART, get their prescriptions filled, and adhere to treatment, all of which occur at successively lower percentages. If people do not take ART, they cannot achieve the final goal of the health system response – viral suppression. In fact, the CDC estimated that only 328,000 had undetectable viral load on ART – or 28% of all people living with HIV in the country.

This analysis demonstrated that there is a great deal of attrition over the cascade of care. However, part of the reason for this attrition has to do with poor access to care, and research has previously shown that people living with HIV who are enrolled in a comprehensive medical service such as KP have much better outcomes – with as many as 61% achieving viral suppression – and that systems' results appeared to be improving somewhat each year.

Other research has suggested that there might be differences based on age and gender, so Horberg's team decided to evaluate whether there were differences in the KP system. They analysed data from the shared electronic health records used by all KP providers, which includes all pharmacy, laboratory, radiology, and patient visit data.

The study

Electronic data were collected from 2010, 2011, and 2012 for all members age 13 years or older with at least eight months of KP membership in a year who were diagnosed with HIV – a total of 18,270 people in 2012.

Data were collected on the following components of the HIV care cascade:

• Linkage to care: defined as a visit and CD4 cell count within 90 days of being identified as HIV positive for newly diagnosed individuals or those newly enrolled in KP and at least one medical visit during the year for established clients.
• Retention in care: defined as at least two medical visits >60 days apart
• Filled ART: filled at least three or more months of DHHS-defined combination antiretroviral regimens.
• Viral suppression: HIV RNA <200 copies/ml measured during the last year.

Care cascades were then developed and evaluated by gender (87% of participants were male and 13% were female) and by age (11% were <35 years, 57% were age 35-54, and 33% were age 55+). The cascades were created by year, but were also compared by year, with differences assessed both over time and by subgroup.

Results

As noted above, a significantly higher percentage of men than women were linked to care, filled their ART prescriptions, and achieved viral suppression in 2010 and 2011. However, it should be noted that while statistically significant, the differences in percentages were not very profound. For instance, 96% of the women were linked to care in 2010 versus 97% of the men. But there was attrition over the successive steps of the cascade, so that while 62% of the men were virally suppressed in 2010, only 56% of the women were.

In 2012, however, 96% of the men were linked to care versus 97% of the women. Again, there was attrition along the cascade, so that by the final step, 67% of the men were virally suppressed compared to 65% of the women – a difference that did not reach statistical significance.

When looking at the age groups, after linkage to care there was still attrition over each step of the cascade, but differences by age were more marked when it came to retention in care over time, filing ART prescriptions, and achieving viral suppression.

For instance, in 2010, 77% of people younger than 35 years, 77% of those between 35-44, and 82% of those over age 55 were retained in care. There was more attrition in the youngest group when it came to ART prescriptions and viral suppression, so that only 49% of those below 35 years, 61% of those between 35-44, and 66% of those over age 55 had viral loads below 200 copies/ml.

Again, outcomes improved a bit – that is, attrition over the care cascade decreased – each year, so that by 2012, 57% of those below 35 years, 65% of those between 35-44, and 72% of those over 55 had suppressed viral loads.

The annual improvement by age and by gender may have been related to the use of multidisciplinary care teams and electronic health records shared by all care team members, the use of continual quality measurement and quality improvement initiatives, and improving medication adherence rates (for example, in 2012, 66% of participants achieved >90% adherence).

But there continue to be gaps and differences that persist – particularly according to age. This suggests that outreach programmes targeting young people may be warranted, according to Horberg.