As antiretroviral (ARV) treatment has become more and more effective at suppressing the virus while also becoming decreasingly toxic the health prospects of people with HIV have steadily improved over the past two decades. Life expectancy, by various estimations, is approaching normal among those on successful treatment for the virus. Nevertheless, HIV-positive individuals maintain a higher burden of numerous chronic health conditions compared with the general population.

Cancer, cardiovascular disease and liver diseases have become the top causes of death among people with the virus.

Publishing their findings in The Lancet HIV, a team of U.S. and Canadian researchers sought to estimate the degree to which a host of modifiable risk factors may drive the development of the following four major diseases among HIV-positive individuals: non-AIDS-defining cancers (NADCs), heart attack, end-stage liver disease (ESLD) and end-stage renal (kidney) disease (ESRD).

The results of their paper effectively provide a road map—centered on addressing these risk factors—for improving the health of those living with the virus as they age.

Hunting for drivers of disease:


The paper’s authors analyzed data regarding people with HIV who participated in the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD). This included people who received care for the virus in academic and community-based outpatient clinical cohorts in the United States and Canada between 2000 and 2014. The investigators assembled four cohorts, one to analyze the effects of modifiable risk factors on diagnoses of each of the four major diseases.

The cohort that focused on NADCs included 1,405 people with and 60,095 without such a cancer diagnosis. Specifically, the researchers looked at an individual’s first diagnosis of NADC, a category that excludes cervical cancer, non–Hodgkin lymphoma and Kaposi sarcoma. For heart attack, they looked at 347 people who had been diagnosed with that particular cardiovascular health event and 29,168 people who had not. The ESLD cohort included 387 people diagnosed with the condition and 34,657 without it. And in the ESRD group, 255 people had the severe kidney disease while 35,365 did not.

By looking at modifiable risk factors among the members of these cohorts, the study authors sought to estimate the “proportion of disease cases over a specified time that would be prevented following elimination of the exposure, assuming the exposure is causal.” They acknowledged that researchers have not necessarily proved a causal relationship between the factors this study identified as contributors to disease diagnoses.

In each of the four cohorts, about 17 percent of the members were older than 50 upon entering NA-ACCORD, about half were nonwhite and about 80 percent were men.

The median follow-up time in the cohorts ranged between 3.1 years in the ESLD group and 3.7 years in the NADC group.

In all four cohorts, the subgroups of people who were diagnosed with each of the four diseases, compared with the subgroups who were not diagnosed, had a higher prevalence of a history of smoking, HBV, HCV, low CD4 count (below 200) and AIDS diagnosis (a factor the study authors counted as separate from low CD4 count, which otherwise triggers an AIDS diagnosis).

Because the study considered former and current smokers as members of one category—ever smokers—and compared them with never smokers, it may have underestimated tobacco’s effects on disease diagnosis. It is possible, if not probable, that by quitting cigarettes, those who formerly smoked, compared with current smokers, lowered their risk of the diseases considered for this study and that this health benefit dragged down current smoking’s apparent contribution to such negative health outcomes.

Non-AIDS-defining cancers:


Among the 1,405 NADCs, 230 (16 percent) were lung, 225 (16 percent) were anal, 167 (12 percent) were prostate, 96 (7 percent) were Hodgkin lymphoma, 90 (6 percent) were liver, 83 (6 percent) were oral cavity and pharynx, 82 (6 percent) were breast, 69 (5 percent) were melanoma and 65 (5 percent) were colon and rectum cancers. The remaining cancer categories each made up less than 5 percent of the total.

Twenty-four percent of NADCs, the study authors estimated, were driven by a history of smoking. Additionally, 3 percent of the cases were attributable to having a low CD4 count, and 3 percent were driven by HBV coinfection.  

For their analyses of the data regarding each of the four disease-category cohorts, the researchers adjusted the figures to account for differences among the cohort members in age, sex, race and history of injection drug use as well as all the risk factors they analyzed as potentially driving the specific disease diagnoses in question.

They found that the modifiable risk factors significantly and independently associated with an increased risk of NADC included: smoking (associated with a 1.61-fold increased risk of NADC), low CD4 count (1.69-fold), detectable viral load, meaning above 400 (1.32-fold), AIDS (1.36-fold) and HBV (1.46-fold). The researchers further parsed this data by excluding the 240 lung cancer cases and found that all these factors were still associated with an increased risk of NADCs, by 1.36-fold, 1.5-fold, 1.3-fold, 1.32-fold and 1.5-fold, respectively. HCV was factored into both analyses, but neither of them showed this virus was associated with NADC among people with HIV.

Heart attack:


Those who were diagnosed with a heart attack, compared with those who were not, had a higher prevalence of high cholesterol (at or above 240), high blood pressure, type 2 diabetes and Stage 4 chronic kidney disease (an estimated glomerular filtration rate of less than 30 milliliters per minute per 1.73 meters squared).

Of all heart attacks, smoking factored into 37 percent of diagnoses, high cholesterol into 44 percent, high blood pressure into 42 percent and low CD4 count into 6 percent.

The adjusted factors by which modifiable risk factors increased the risk of heart attack included: smoking (1.82-fold increased risk), high cholesterol (2.95-fold), high blood pressure (3.34-fold), Stage 4 chronic kidney disease (1.7-fold), low CD4 count (2.02-fold), detectable viral load (1.48-fold) and HCV (1.47-fold).

The study authors conducted a further analysis that looked only at data on the 57 percent of the cohort members for whom there were body mass index (BMI) data. This allowed them to control for obesity (a BMI of 30 or higher). In this case, the risk factors were associated with the following increased risks of heart attack: smoking (1.71-fold increased risk), high cholesterol (3.12-fold), high blood pressure (2.93-fold), type 2 diabetes (1.47-fold), low CD4 count (2.42-fold) and HCV (1.77-fold).

Having a detectable viral load was not statistically significantly associated with a higher risk of heart attack in the second analysis. Obesity was also not significantly associated with heart attack risk. AIDS was not linked to heart attack in either analysis.

End-stage liver disease:


Of all ESLD cases, 30 percent were driven by HCV, 19 percent by low CD4 count and 16 percent by HBV. 

The adjusted degrees to which risk factors increased the likelihood of ESLD included: low CD4 count (3.89-fold increased risk), detectable viral load (1.77-fold), AIDS (1.35-fold), HBV (3.11-fold) and HCV (3.13-fold).

Next, the investigators looked only at data on the 35 percent of cohort members for whom there were data on at-risk alcohol use—defined as at least three daily drinks or 7 weekly drinks for women and at least four daily or 14 weekly drinks for men. After controlling for at-risk drinking, the investigators found that the following risk factors were associated with an increase in ESLD risk by the following degrees: low CD4 count (4.3-fold increased risk), detectable viral load (1.62-fold), HBV (2.06-fold), HCV (2.76-fold) and at-risk alcohol use (1.78-fold).

AIDS was not associated with ESLD in the second analysis. A history of smoking was not linked to ESLD in either analysis.

End-stage renal disease:


Having a detectable viral load, high cholesterol, high blood pressure and type 2 diabetes were each more common among those diagnosed with ESRD compared with those who did not have the severe kidney disease.

The proportions of ESRD attributable to various modifiable risk factors included: high blood pressure (39 percent); high cholesterol (22 percent); detectable viral load (19 percent); low CD4 count (13 percent); and type 2 diabetes (6 percent).

After adjusting the data, the researchers found that the following risk factors were associated with an increased risk of ESRD by the following degrees: high cholesterol (2.54-fold increased risk), high blood pressure (5.18-fold), type 2 diabetes (1.67-fold), low CD4 count (3.03-fold), detectable viral load (1.87-fold) and AIDS (1.75-fold). Neither HCV nor a history of smoking were associated with ESRD.

What all this means for improving the health of people with HIV:


Researchers have proved that treating HIV earlier, when CD4 counts are higher (ideally above 500), provides numerous health benefits, including a reduced risk of cancer.

That said, the new study found that HIV-related risk factors turned out to have less of an impact on the risk of NADC, heart attack, ESLD and ESRD than numerous modifiable risk factors—in particular, smoking, high cholesterol, high blood pressure and HCV coinfection. This finding underlines the importance of addressing more than just treatment of the virus among people living with HIV.

“The evidence from our study is clear,” the study authors write, saying that “to avoid sizeable proportions of non-AIDS-defining cancers, myocardial infarction [heart attack], end-stage liver disease and end-stage renal disease, the continued focus on maintaining HIV viral suppression after [ARV treatment] initiation must be balanced with screening for traditional risk factors, effective interventions to reduce the burden of traditional risk factors and a sustainable model of care with the capacity to provide traditional risk factor interventions over the decades of life with HIV.”

Smoking poses a particularly grave danger to the health and well-being of people with HIV. Multiple previous studies have found that people with HIV lose more years of life to smoking than they do to living with the virus. One quarter of HIV-positive smokers are projected to die of lung cancer. All these findings notwithstanding, members of this demographic tend to underestimate the degree to which smoking threatens their health.

Smokers can quit with the help of various treatments; research has shown this reduces the risk of cancer and cardiovascular disease in particular. (A new study found that e-cigarettes are associated with a higher, albeit still modest, chance of quitting nicotine.) Otherwise, high cholesterol and high blood pressure are treatable through diet, exercise and medication. And HCV is curable with direct-acting antiviral (DAA) medications.

It is vital, the researchers stress, to modify such interventions, such as smoking cessation programs, to the specific needs, including financial considerations, of the HIV population.

One major factor that this study was not designed to consider is any degree to which specific ARVs themselves may raise the risk of the four major diseases considered. Nor did the study consider the health effects of variations in diet and physical activity. And importantly, because everyone in the study was living with HIV, the researchers could not determine how the virus itself, compared with its absence, might affect the risk of these diseases.

Benjamin Ryan is POZ’s editor at large, responsible for HIV science reporting. His work has also appeared in The New York TimesNew YorkThe NationThe Atlantic and The Marshall Project. Follow him on FacebookTwitter and on his website,