[NATAP] Depression & Attitude: Overlooked in HIV

[Guest guest]

http://www.natap.org

Depression: The Overlooked Ogre in HIV Care?

A report from the XVI International AIDS Conference

Toronto, August 13-18, 2006

Mark Mascolini

It is much more important to know what sort of patient has the disease than what sort of disease the patient has.-- Osler

That dose of wisdom from Osler, bedside Brahmin and dean of medical aphorisms, heads an intriguing article on psychological state, CD4 trends, and death in US women with HIV infection [1]. The punchline in this 773-woman analysis is that a positive mood, a favorable health outlook, and "finding meaning" despite infection with a remorseless virus foretell a lower risk of HIV-related death.

Measuring mood remains a nebulous enterprise streaked with opaque or open-ended terms like "psychological resources," "life satisfaction," and "positive affect." Toting T cells and viral particles in plasma seems crystalline in contrast. Perhaps as a result, meshing mood and the rock-hard endpoints of HIV disease--CD4 counts, symptoms, death--may strike some as a dubious enterprise.

But in this still-early chapter of the epidemic's history, over a century past Osler's prime, tell-tale studies have started hinting that an HIV-infected person's psyche--"what sort of patient has the disease"--may prove the most-ignored malleable variable in progression of this indurate infection.

Two of those studies came to light at the XVI International AIDS Conference in Toronto [2,3] and one appeared shortly thereafter in AIDS [1]. Here's what they found:

Does depression keep consistent company?

Depression can go untreated in people with HIV infection because clinicians may fail even to ask about a person's mood, focusing instead on important and easier-to-measure metrics--viral load, CD4 count, resistance, lipids, and glucose. In that way depression shares company with other amorphous morbidities like insomnia, fatigue, and peripheral neuropathy. At the AIDS Conference and colleagues from the University of South Carolina and the University of Arizona offered evidence that depression shares more than diagnostic ambiguity with these three problems [2]. These researchers proposed that depression, insomnia, fatigue, and neuropathy form a "symptom cluster" in people with HIV infection.

Symptom clusters, explained, consist of three or more concurrent symptoms. Research on this phenomenon, mostly in people with cancer or mental illness, indicates that simultaneous symptoms multiply the distress a person feels. In other words their combined impact is more than additive.

The South Carolina-Arizona team studied 134 women living in the rural southeastern United States. Most of these women were single (80%) and black (84%), and most had an annual income below $5000 (56%). Nearly half of the women (47%) never finished high school, and only 25% had a full- or part-time job. So this study group represents a distinct, fairly homogeneous population.

The women answered questions that rated them for depression, insomnia, peripheral neuropathy, fatigue, and quality of life:

o Depression: A 20-item self-report measuring depressed mood, feelings or guilt and worthlessness, feelings of helplessness and hopelessness, psychomotor retardation, loss of appetite, and sleeplessness.

o Insomnia: A 19-item self-report measuring subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, using sleeping pills, and daytime dysfunction.

o Fatigue and peripheral neuropathy: A questionnaire measuring the frequency and intensity of lack of energy and numbness or tingling.

o Quality of life: A 28-item scale gauging physical status, family and friends, emotional status, financial state, spiritual well being, peace of mind, and overall life satisfaction.

On a scale of 0 to 60, scores for depression ranged from 16 to 56 and averaged 18.3. Average scores for all four symptoms lowered quality of life measured at the time of the interview, as estimated 3 months earlier, and as anticipated 3 months in the future (Table). When combined as a single factor (Table), depression, insomnia, peripheral neuropathy, and fatigue had a much greater--and highly significant--impact on quality of life (P < 0.0001).

Correlation of four symptoms--separately and together--with quality of life

and confederates believe their findings indicate that depression, insomnia, peripheral neuropathy, and fatigue cluster together as symptoms, at least for this population of poor US women with HIV infection. They urge clinicians to appreciate that these symptoms may occur together and suggest that interventions addressing these symptoms simultaneously may prove helpful.

Depression, adherence, and SSRIs

Bangsberg from the University of California, San Francisco, a top-drawer expert on antiretroviral adherence, calls depression the most-overlooked and easiest-to-treat hurdle to adherence. Researchers at the Kaiser Permanente system in northern California confirmed the first part of that equation, but perhaps not the second [3]. They produced evidence that treating depression with selective serotonin reuptake inhibitors (SSRIs such as ?? and Lexapro??) appears to improve adherence in adults taking antiretrovirals. But HIV-infected people using the popular antidepressants had significantly worse 12-month virologic and CD4 responses than people without depression.

Horberg and coworkers from Kaiser Permanente in Oakland, California, combed records of 3431 HIV-infected Kaiser clients split into four groups--1053 with a depression diagnosis who never used SSRIs, 348 with a depression diagnosis who took the drugs, 69 who took SSRIs with no depression diagnosis on their charts, and 1961 who never got diagnosed with depression and never took SSRIs.

Average ages of the four groups lay in the low 40s when follow-up began, they all had detectable viral loads, and average CD4s varied from 287 in the depression-no-SSRI group to 342 in the depression-SSRI group, 318 in the no-depression-SSRI group, and 226 in the control group without depression or SSRI experience. Most people in this cohort, 83%, were men.

In a multivariate analysis factoring in age, gender, type of antiretroviral regimen, temporal trend, and viral load and CD4 count when antiretroviral therapy began, people with depression who did not use SSRIs had an average continuous adherence (figured by pharmacy refill records) 3% below the control group, a significant difference (P = 0.04). People who took SSRIs for depression, on the other hand, had an average 2% higher adherence than controls, a nonsignificant difference. People who took SSRIs without a formal depression diagnosis also had an adherence level equivalent to that of the control group. (Horberg figures this last group actually had depression that did not get recorded or they were taking SSRIs to help them stop smoking.)

A separate multivariate analysis found that depressed people not taking SSRIs were 18% less likely than controls to have better than 90% adherence, again a significant difference (P = 0.03). But once more the two SSRI-taking groups did not differ from controls in this measure of adherence.

When Horberg looked at how many people had a viral load under 500 copies a year after starting antiretrovirals, his multivariate analysis determined that the two groups with a depression diagnosis had a significantly lower chance of reaching a sub-500 load than the control group, and depressed people taking SSRIs did worse than those not taking the antidepressants. People with depression but not taking SSRIs had a 25% lower chance of getting below 500 copies (P = 0.01), and people with depression and taking SSRIs had a 60% lower chance (P < 0.001).

Reconciling that finding with the better adherence in depressed clients taking SSRIs than in depressed clients not taking these antidepressants may not be possible in this kind of analysis. Horberg suggested "selection bias" may have weighed against the SSRI-using group; in other words taking SSRIs may mean they had more severe depression than those not taking SSRIs. But that does not explain why SSRI takers controlled HIV worse than the depression-without-SSRI group, despite their apparently better adherence. Compared with the control group, treatment-induced CD4 gains also proved worse in the depression-with-SSRI group than in the depression-without-SSRI group.

Horberg suggested that randomized trials may afford a better appraisal of how SSRIs affect adherence and therapeutic response. But running such a trial could be hard because clinicians may not want to deny SSRI therapy to people who seem to need it.

Does mind matter when it comes to mortality?

How important is it for people with HIV to beat depression? It could be a matter of life or death, according to results of a study involving 773 women in Baltimore, the Bronx, Detroit, and Providence [1]. This thoughtful 5-year analysis yielded evidence that a positive frame of mind not only slows the CD4-cell slide, but also boosts the chance of longer survival with HIV.

As in the two depression studies reported at the Toronto meeting, this research evaluated a distinct subset of US residents with HIV, so the results may not apply to everyone. And the results rely on answers to questions about nonclinical concepts like hope, "sense of control," and "positive life changes." But if these HIV Epidemiologic Research Study (HERS) investigators are right, their findings should give clinicians ample motivation to tend to their patients' mental health.

The HERS team recruited HIV-infected women without an AIDS diagnosis from 1993 through 1995 and saw them at follow-up visits through March 2000, well into the era of potent antiretroviral combinations. Over that span only 91 women (12%) stopped coming back for visits. Researchers surveyed the women to rate them for "psychological resources," defined in three ways:

o Positive affect. This score relied on answers to six questions on positive emotions (such as hope) felt during the last 6 months.

o Positive HIV expectancy. Seven questions sized up women's expectations about their disease and sense of control over it.

o Finding meaning. Five questions weighed positive attitudes women had about their HIV infection, for example, allowing them to spend more time with family.

Statisticians calculated median scores for each "resource" and gave women 1 point every time they scored at or above the group median. Health workers also logged data on viral load, CD4 counts, HIV-related symptoms, antiretroviral therapy, prophylaxis for opportunistic infection, illicit drug use, alcohol use, cigarette use, depressive symptoms, antidepressant therapy, and sociodemographic characteristics.

The women's ages ranged from 19 to 55 years (average 35.5 years), 60% were black, 20% Latina, and 20% white or another race or ethnicity. Most of them, 82%, had no job, 72% had a monthly income below $1000, 65% relied on public assistance, and 45% never finished high school. When they entered the study, the women had an average CD4 count of 431, and 82% had a viral load under 10,000 copies. When the study began 380 women (49%) had more than one "psychological resource" and another 257 (33%) had one; 429 women (58%) reported "substantial" depressive symptoms.

During the 5-year study, 106 women (14%) died. The death rate tracked significantly with the number of "resources" a women claimed--17% for women with no resources, 14% for those with one, 10% for those with two, and 6% for those with three (P = 0.003). Among women who counted fewer than 200 CD4 cells when they signed up for the study, 50% with no resources died, compared with 16% of those with three resources (P = 0.02).

Statistical analysis that factored in CD4 count, viral load, antiretroviral use, opportunistic infection prophylaxis, symptoms, antidepressant use, and age determined that women with no psychological resources had a 3.5 times higher risk of dying than women with three resources. Results did not change in an analysis considering time of study entry (to account for availability of potent antiretroviral combinations). Nor did results differ in an analysis limited to HIV-related deaths.

Higher CD4 counts in women with more psychological resources probably explain their lower risk of dying. Statistical analysis indicated that only psychological resources and clinical factors such as entry viral load and CD4 count correlated with CD4 changes during follow-up. A statistical modeling accounting for clinical and sociodemographic traits determined that psychological resources upon entering the study significantly predicted CD4 drops over time: Women with more resources lost CD4 cells more slowly.

The HERS team postulates that preservation of CD4 cells in women with a more positive frame of mind may reflect a proposed adverse impact of stress on the immune system [4]. Aside from this possible direct effect on CD4s, the researchers speculate that psychological resources could have an indirect impact on immunity via exercise, diet, or consistency of healthcare.

Better antiretroviral adherence in women with a positive mind set probably does not explain these findings. One third of the women in this study took no antiretrovirals. Among 282 women taking two or more antiretrovirals after mid-1999, psychological resources did not correlate with better adherence measured as skipped doses yesterday, skipped doses in the last 3 days, and number of drug holidays in the past 6 months.

The HERS investigators acknowledge that both HIV disease and mental health are complex, multifactorial processes that cannot be precisely defined by a set of biomarkers or questionnaire answers. A sunny mood does not guarantee a longer life with HIV, but it may offer subtle advantages that can never be adequately measured. Or, as the HERS team concludes, "it is not a question of mind over matter, but it is important to recognize that mind does matter."

Mark Mascolini writes about HIV infection (markmascolini@...).

References

1. Ickovics JR, Milan S, Boland R, et al. Psychological resources protect health: 5-year survival and immune function among HIV-infected women from four US cities. AIDS 2006;20:1851-1860 (http://www.natap.org/2006/HIV/091206_02.htm).

2. KD, Moneyham L, Tavakoli A, et al. Depression, insomnia, peripheral neuropathy, and fatigue: a cluster of HIV-related symptoms. XVI International AIDS Conference. August 13-18, 2006. Toronto. Abstract MOPE0080.

3. Horberg M, Silverberg M, Hurley L, et al. Effects of depression and SSRI use on adherence and HIV clinical outcomes in patients taking combination antiretroviral therapy (cART). XVI International AIDS Conference. August 13-18, 2006. Toronto. Abstract TUPE0117.

4. Danner DD, Snowdon DA, Friesen WV. Positive emotions in early life and longevity: findings from the nun study. J Personality Soc Psychol 2001;80:804-813.