HIV Provider–Patient Communication Regarding Cardiovascular Risk

Abstract

Design:

Few global studies have assessed HIV clinician–patient communication regarding cardiovascular disease (CVD) risks.

Methods:

We conducted a multicountry, comparative, cross-sectional survey of HIV-infected individuals in 12 countries on 5 continents in 2010, with 100 to 200 enrollees per country. HIV-infected adults >17 years and on antiretroviral therapy were recruited in clinics and community organizations and surveyed via direct interview, telephone encounter, or online. Chi-square analyses were performed with an 80% power to detect a difference of >20%.

Results:

Of 2035 participants, 37% were women. Prevalence of self-reported CVD risk factors was 28% overall, and greater CVD risk was present in 55% of patients in North America, 12% in Africa, and 26% to 28% on other continents. Only 19% of patients ever discussed CVD with their physician, and 31% had ever discussed hypertension, hypercholesterolemia, family history of CVD, or smoking; these findings were true for HIV clinicians in all regions of the world. Forty-four percent of smokers reported never discussing smoking with their HIV clinician.

Conclusion:

We found that HIV clinicians worldwide are not sufficiently addressing CVD risk factors with their patients. Expanded training and education for HIV clinicians should include effective approaches to the mitigation of CV risk factors.

Introduction

HIV infection has recently been associated with an increased risk of cardiovascular disease (CVD). In the Data Collection on Adverse Events of Anti-HIV Drugs (DAD) Study, for example, CVD risk factors were highly prevalent; 25% of enrollees were aged 35 or older, 51.5% were cigarette smokers, 1.4% of enrollees had known ischemic heart disease, and hyperlipidemia was highly prevalent among treated individuals. ¹ Antiretroviral therapy (ART) can diminish the risk of CVD in people living with HIV. In the Strategies for Management of Antiretroviral Therapy (SMART) study, the study arm of patients in whom ART was either withdrawn or not started until the CD4 count fell below 250 cells/mm³ had a 70% increase in the risk of major cardiovascular (CV) complications compared to the study arm in whom ART was immediately provided. ² In another large cohort study of persons living with HIV, a greater risk of heart disease was observed in HIV-positive patients who had traditional CV risk factors, such as age more than 40 (3.3-fold increase), diabetes (3.2-fold increase), hyperlipidemia (1.9-fold increase), and hypertension (1.7-fold increase). ³

Accordingly, greater attention to the mitigation of CV risk factors has become an important responsibility for HIV clinicians. ⁴ Recent international guidelines for HIV care have stressed the importance of attention to CV risk assessment and reduction as part of the standard approach to people living with HIV and calls for expanded health worker training to cope with rising caseloads urgently address the need for attention to prevention and care of CV and metabolic diseases. ^{5 –10} However, no studies have been conducted on a global scale to assess the behavior of HIV clinicians in regard to attention to CV risks in their patients. Herein we report on the results of a global survey of people living with HIV regarding the risk of CVD and their clinician’s actions regarding CV risk mitigation.

Methods

The AIDS Treatment for Life International Survey (ATLIS) was a multicountry, comparative, cross-sectional survey of people living with HIV in 5 global regions: North America (the United States); Latin American (Brazil); Europe (France, Germany, Italy, Russia, Spain, and the United Kingdom); Asia (Australia and Korea); and Africa (Cote d’Ivoire and South Africa) in January to March 2010, with 100 to 200 enrollees per country. ¹¹ The project was initiated and coordinated by the International Association of Physicians in AIDS Care and governed by an ATLIS Task Force composed of HIV experts from 5 continents. The ATLIS survey fieldwork was conducted from January to March 2010 by Kantar Health, a global health research organization.

A standardized approach to patient recruitment, sampling frame, and recruitment setting was complemented by country-specific adaptations to account for local circumstances. Study eligibility included age more than 17 years, diagnosis of HIV or AIDS, and treatment with ART within the past 5 years. Patients were recruited in health care facilities and in community-based organizations and support groups and administered a 40-minute survey via direct interview or telephone encounter or an online self-administered questionnaire. Monetary incentives were offered to respondents for completion of interviews where customary and varied by country, depending on local practice. The ATLIS protocol, questionnaire, and patient recruitment materials were reviewed and approved by Essex Institutional Review Board, Lebanon, New Jersey, as well as by the research ethics review board in each of the countries involved.

To explore the differences in CVD risk, higher risk of CVD was defined as a diagnosis of heart disease or diabetes or both, or age more than 50, diagnosis of hypercholesterolemia or hypertension, smoking history, and obesity. A lower risk of CVD was defined by no history of heart disease or diabetes and not more than 1 CVD risk factor reported previously.

Data were entered into the study database via the Internet directly following the interviews. Data analyses were performed using STATA Release 10 (Stata Corporation, College Station, Texas). The main statistical analyses included proportions that were compared for statistical significance, where applicable, using chi-square test. Sample sizes varied by country (100-200) based on the overall national HIV/AIDS prevalence. For any independent variable, with at least 100 study participants evenly divided between 2 categories, there is 80% power to detect a 20% difference when comparing 2 prevalence rates, with a type 1 two-sided error of 5%.

Results

Of 2035 HIV-positive participants, 1,282 (63%) were men and 753 (37%) were women. More than one-quarter (28%) of patients reported a high risk of CVD defined as 2 or more CV risks and/or a diagnosis of diabetes or heart disease. Table 1 shows the prevalence of CVD risk reported by respondents. The reported prevalence of greater CVD risk factors was highest in patients in North America (55%), which was 2-fold higher than the risk in patients in Latin America, Asia, and Europe and 4-fold higher than that in patients in Africa.

Table 1.

Prevalence of CVD Risk among Respondents Overall and by Country.

ATLIS 2010 Key Regional Findings
Finding	Global Total	North America	Europe	Asia-Pacific	Latin America	Africa
Prevalence of reported CVD risk factors
High cholesterol, %	16	40	16	18	15	2
High blood pressure, %	15	32	14	15	15	8
Diabetes, %	6	12	6	7	7	2
Obesity, %	5	10	5	5	5	2
Heart disease, %	4	8	4	4	7	2
Degree of reported CVD risk
Lower CVD riska, %	72	45	74	69	73	88
Higher CVD riskb, %	28	55	26	32	27	12

Abbreviations: ATLIS, AIDS Treatment for Life International Survey; CVD, cardiovascular disease; DM, diabetes mellitus.

^a ≤1 CVD risks and no diagnosed DM or heart disease.

^b ≥2 CVD risk factors and/or DM or heart disease.

Fewer women (23%) than men (30%) reported an increased risk of CVD , but a greater proportion of women were diagnosed with obesity (8% versus 6%), and more women than men had a body mass index greater than 30 (17% versus 7%). Among other comorbidities that affect CVD risk, a history of kidney disease was prevalent in 4% of patients.

In spite of the high frequency of these risks, only 19% of all patients had ever discussed heart disease with their physician. Table 2 shows the number and frequency of patient–provider communication on various aspects of CVD risk overall and by country. There was little difference between countries, with the highest frequency of patient–physician communication on CVD occurring in North America (25%) and the lowest in Europe (17%). Fewer than one-third of patients had ever discussed hypertension, hypercholesterolemia, family history of CVD, or smoking with their HIV clinician. These findings were true in all regions of the world, as shown in Table 2, with no significant differences between regions. Forty-four percent of patients who smoked reported never having had a discussion regarding smoking with their HIV clinician. Two-thirds of respondents at higher risk of CVD were not engaged in frequent discussions of CVD with their HIV clinician.

Table 2.

Frequency of Patient–Provider Communication Regarding CVD Risk Overall and by Country.

Finding	Global Total	North America	Europe	Asia-Pacific	Latin America	Africa
Smoking, %	38	44	40	38	23	35
Family history of CVD, %	28	31	31	20	20	23
Cholesterol, %	34	58	32	39	38	18
High blood pressure, %	29	42	26	30	27	32
Heart disease, %	19	25	17	23	20	21

Abbreviations: CVD, cardiovascular disease.

Overall, 12% of patients reported side effects from ART related to CVD, such as hyperlipidemia or known heart disease, and the frequency was highest in North America (23%) and lowest in Africa (3%) but 8% to 16% in other regions. Only 31% of total patients reported having a conversation with their health care provider about CV effects of ART. Among patients who reported a side effect with potential impact on CV health, 77% reported having discussed the side effect with their HIV clinician.

Discussion

Our most important finding is that HIV clinicians worldwide are not sufficiently addressing CVD risk factors with their patients. Fewer than 1 in 5 patients reported that their doctor or caregiver had discussed heart disease, even in Europe and North America, where the risk of CVD is highest, and fewer than one-third had discussed a family history of heart disease. Of equal concern, only a minority of patients reported that their clinicians had addressed ART side effects that may affect CV health. Similarly, only a minority of patients reported discussing specific risk factors which can be corrected, such as smoking, blood pressure, and cholesterol, with their health care provider. Although there are trends in our data toward more attention to CVD risk in the region with the highest risk of CVD, that is, North America, the degree of attention that we observed was clearly substandard and inconsistent with current treatment guidelines in all regions of the world.

We found a high frequency of self-reported risk of CVD in the cohort overall, similar to the reports from HIV cohorts worldwide. Because self-report underestimates the true prevalence of CV risk, it is likely that the degree of risk among our patient population is even higher than reported. ¹²

There are many potential explanations for our observations. Several physician surveys have suggested that physicians are too busy to spend time on issues that are not perceived to be directly germane to HIV care. ¹³ Similarly, some may regard CV risk to be low in their young and otherwise healthy populations of persons living with HIV, and this appears to be more likely in less-affluent, resource-poor settings.

One likely reason for our findings is the lack of training and physician awareness of the growing importance of chronic inflammation and immune activation that is leading to accelerated CVD in people living with HIV. Similarly, some infectious disease physicians and other HIV clinicians may view the management of smoking, diabetes, and hyperlipidemia to be beyond their expertise and more within the province of the primary caregiver. ¹² However, in most settings in the world the HIV clinician is the primary caregiver and is responsible for the management of hypertension, diabetes, hyperlipidemia, smoking cessation, and other CVD risk factors.

Among the possible reasons for our finding is the methodology of relying on patient recall, which is one limitation of our study. Patient recall is highly imperfect, and in this setting it is likely to underreport a history of CVD risk. However, given the relative importance that patients place on heart disease and CVD risk, this is probably a minor issue, and it is not likely that our results would vary greatly if limits in patient recall were corrected.

Our findings require the urgent attention of HIV clinicians because there is ample evidence that interventions to reduce CVD risk led by HIV clinicians can be successful. In the DAD study, smoking cessation was associated with an 18% decline in the estimated risk of CVD per the Framingham calculation within the first year, and additional decrements of 25% in the risk of CVD occurred in subsequent years. ¹³ Recent guidelines have been developed to assist HIV clinicians with smoking cessation, weight reduction, and optimal lipid management in their unique patient populations. ^9,12,14,15

In sum, in this global survey of people living with HIV in which a high degree of CVD risk was found in patients in all regions of the world, fewer than one-fifth of patients reported having talked to their physician about heart disease, and fewer than one-third reported talking with their physician about common risk factors for CVD such as smoking, high blood pressure, and high cholesterol. Further studies of this issue are warranted, and expanded training and education for HIV clinicians worldwide should include effective approaches to the mitigation of CV risk factors.