Greater Depressive Symptoms and Higher Viral Load Are Associated with Poor Physical Function among Latino Men Living with HIV

Abstract

Physical function limitations have been associated with poor health outcomes, which have a negative impact on quality of life of older individuals. This study examined the association between depression, viral load, and acculturation with physical function among Latino men living with HIV. A secondary data analysis was performed using a cross-sectional data of 146 Latino immigrant men living with HIV in New York City and Washington, DC. Physical function was measured using the Short-Form Health Survey (SF-12). Uncontrolled HIV infection and depression were associated with worse physical function, thus implying the importance of adequate health care to address these conditions. Preserving physical function should start during middle adulthood, particularly among people living with HIV because of their greater risk of developing age-related challenges such as depression, diabetes, cardiovascular diseases among others. This study informs future interventions to preserve physical function and achieve the goal of successful aging.

Introduction

The use of the highly active antiretroviral treatment (HAART) has increased the life expectancy of people living with HIV (PLWH) and, consequently, age-related challenges such as limitations in physical function. Poor physical function, limitation in the capacity to perform independently various daily physical tasks, has been associated with high risk of mortality, hospitalization, and the need for long-term care.^1,2 Limitations in physical function will become more common as the HIV-infected population continues to age.^3,4 The current study examined self-reported physical function among Latino men living with HIV in the United States and investigated the role of viral load, depression, and acculturation relative to physical function in this population.

Depression and Physical Function

Depression is a risk factor associated with physical function limitations in adults. Previous studies found that depression increased the probability of functional limitations in adults without HIV^5–8 as well as among PLWH.⁹ Other studies, however, have noted that people with depression often reported limitations in physical function.^10,11 The association between depression and physical function is relevant for HIV-infected populations, because adults living with HIV have a greater risk of developing depression, which has been associated with worse health outcomes, including physical function.^11,12

HIV and Physical Function

Overall, the findings regarding the association between viral load and physical function are mixed. Higher viral load has been associated with reduced physical function and physical activity and capacity,^13–16 which are strong predictors of physical function. Moreover, a study comparing PLWH to an HIV-negative group with other common medical comorbidities (eg, diabetes, hypertension, and cardiovascular diseases) found that higher viral load was significantly related to worse physical function at the univariate level, but this association was no longer significant in the multivariable model with comorbidities.⁹ However, other studies have found no association between viral load and various functional outcomes among HIV-positive individuals.^17,18

Acculturation and Physical Function

Culture plays a role in shaping attitudes and behaviors related to physical activity and independent function. Latinos living in the United States could be affected by cultural norms from both the host-country (US) and their home-country (Latin America). Acculturation, which refers to the process of psychological and behavioral change experienced as a result of contact with another culture,¹⁹ has been associated with physical function in a study of Japanese immigrants to the United States: Japanese-speaking individuals reported fewer declines in function over time than English-speaking Japanese or white Americans.²⁰

Other studies on acculturation and health behaviors among Latinos have found varied effects of acculturation on health outcomes. For example, acculturation may affect physical functioning in Latinos through changes in health-related beha-viors.^21,22 Latino culture may emphasize physical activity to a lesser extent than US culture, which could have implications for physical function. Moreover, studies of older people have suggested greater limitation in physical function among individuals who are more embedded in Latino culture. For example, in a study on Latinos over 65 years, those who were interviewed in Spanish had a higher prevalence of limitations in activities of daily living than whites and Latinos interviewed in English.²³ Furthermore, they were more likely to develop disabilities over a 6-year follow-up.

Research Objectives

There has been limited research concerning physical function among Latino populations and particularly among those living with HIV. This study has 2 objectives: first, we will test the hypotheses that greater depressive symptoms and higher viral load will be associated with worse physical function, and second, we will explore the role of acculturation in physical function.

Methods

Design

We performed a secondary data analysis using data from a study on disclosure of positive HIV status among Latino men who have sex with men (MSM).²⁴ This cross-sectional study enrolled 155 HIV-positive Latino men from New York City and Washington, DC. Participants completed a survey using a touch-screen computer with audio computer–assisted self-interviewing software. Procedures for this study were approved by George Washington University Institutional Review Board [IRB].

Participants

The final sample consisted of 146 individuals. Nine participants were excluded from the regression analyses because they lacked data on physical function. Most of the participants were immigrants; fewer than 10% were born in the United States. They came from 16 different Latin American countries, with higher frequency from Colombia (14.9%), Mexico (12.3%), El Salvador (11.6%), and Puerto Rico (10.9%). Non-US-born participants had been living in the United States for an average of 12.4 years (standard deviation [SD] = 9.7). Participants’ age ranged from 18 to 67 years, with a mean age of 38.5 years (SD = 9.4). The majority of our sample can be described as having low income and being employed.

Measures

All measures were translated and back translated from Spanish to English to ensure language equivalence. In addition, native Spanish speakers from different countries examined all measures to ensure universal Spanish. Surveys covered sociodemo-graphic characteristics (age, country of birth, income, and education), HIV viral load, and HIV medication. Viral load was measured by a self-report question with 6 categories: (1) undetectable, (2) less than 10 000, (3) 10 000 to 49 999, (4) 50 000 to 99 999, (5) 100 000 or greater, and (6) unknown. Participants also reported whether the doctor had prescribed HIV medication; those who answered yes were asked a follow-up question to determine the medications prescribed (eg, zidovudine [ZDV], saquinavir [SQV]). Depression was measured with the 13-item short form of Beck Depression Inventory (BDI).²⁵ Response format was a 4-point Likert response scale (1–4), with higher mean scores indicating high depressive symptoms. Internal consistency reliability (Cronbach a) of the BDI was .89 in this sample. Acculturation was assessed using items from the Abbreviated Multidimensional Acculturation Scale.²⁶ The items evaluate acculturation to Latino culture (9 items) and US culture (11 items) and cover cultural identification, language competence, and cultural knowledge. Higher mean scores indicate greater acculturation to Latino and US culture. Cronbach α in this sample was .77 for Latino acculturation and .88 for US acculturation.

Physical function was measured using the physical component summary (PCS) from the Medical Outcomes Study 12-Item Short-Form Health Survey (SF-12).²⁷ The score was calculated using empirically derived weighted scoring algorithms that provide indications of physical functioning.²⁷ The scale scores range from 0 to 100, with higher scores indicating better physical function. The questions encompass 8 dimensions: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotions, and mental health. Cronbach α of the PCS was .84 in this sample.

Procedure

Detailed information about the procedures of this study can be found in the study by Zea et al. Participants were recruited using flyers distributed in bars, community centers, health clinics, and other establishments frequented by Latino MSM as well as through newspaper and radio advertisements and participant referrals. Inclusion criteria were being a man who self-identified as Latino and as gay or bisexual. The survey was administered in Spanish (87%) or English (13%), depending on the participant’s preference. The average time to complete the survey was 100 minutes. Reimbursement of USD $50 was given for participation.

Analysis

Descriptive statistics, t test, 1-way analysis of variance with Student-Newman-Keuls post hoc test, and bivariate correlations were calculated to assess differences and associations among physical function, demographic characteristics, viral load, depression, and acculturation to Latino and US culture. A hierarchical set regression was performed to test the hypothesis that viral load and depression would be related to physical function as well as to explore the association between acculturation and physical function. Age, income, type of antiretroviral treatment (ART), years living with HIV, and comorbidities (ie, unintentional weight loss and fatigue) were included as control variables. With this approach, we entered the predictors in sequential steps, and significance was tested with changes in adjusted R². In the first set, we entered the demographic characteristics, viral load, and depression. Because of the low number of participants in some categories, the variable viral load was recoded as (0) undetectable, (1) less than 10 000, (2) 10 000 to 49 000, (3) more than 50 000, and (4) unknown; and income was dichotomously recoded as (0) less than $400 and (1) more than $401 per month. Undetectable viral load was used as the reference category. Antiretroviral treatment was recoded as the use of nucleoside/nucleotide reverse transcriptase inhibitor (NRTI; 0 = no and 1 = yes), protease inhibitors (PIs; 0 = no and 1 = yes), and nonnucleosside reverse transcriptase inhibitor (NNRTI; 0 = no and 1 = yes). In the second set, we entered acculturation to Latino and US culture. Education level was not included in the final regression model because it was highly correlated with income (r = .35, P < .001). We did not control by alcohol and drug use in the linear regression model because they were measured as yes or no questions. To examine the effect of alcohol and drug use on physical function, we asked about alcohol and drug use, but because very few used substances, we did not breakdown by the type of drug or the amount of drinks per day. There were no differences in physical function between those who used and those who did not use alcohol and drugs (see Table 1). In order to obtain unbiased estimates in the regression model, we used full-rank reference parameterization (PROC GLMSELECT in SAS). All data were analyzed using SAS software (version 9.3).

Table 1.

Physical Function Scores by Demographic Characteristics, Viral Load Categories, Years Living with HIV, and Alcohol and Drug Use.

	N	Mean	Standard Deviation	F (df) and SNKa
Age (N = 146)				0.042 (3)
Less than 30	25	48.9	9.0	A
31–40	61	47.3	10.4	A
41–50	45	46.5	9.5	A
More than 50	15	39.9	11.3	Bb
Income (monthly, N = 146)				0.495 (2)
Less than $400	62	44.5	10.2	A
Between $401 and $800	47	47.1	10.3	A
More than $801	37	49.4	9.6	A
Education				2.04 (4)
Less than high school	27	45.3	10.4	A
High school diploma/GED	47	45.2	9.3	A
Some college	32	47.5	11.5	A
College studies	40	48.2	10.2	A
HIV viral load (N = 146)				0.002 (4)
Undetectable	68	49.1	8.6	A
Less than 10 000	27	47.9	8.9	A
10 000–49 999	16	44.1	13.4	Bb
More than 50 000	15	38.2	10.7	Bb
Unknown	20	44.4	10.5	Bb
Years living with HIV				4.02 (2)
Less than 6 years	41	49.8	8.5	A
6–8 years	31	43.2	10.7	Bb
More than 8 years	74	46.2	10.5	Bb
Alcohol usec
Yes	85	47.9	10.7	NS
No	61	44.7	9.3
Drug usec
Yes	51	46.7	11.8	NS
No	101	46.5	9.4

Abbreviations: GED, General Educational Development; NS, nonsignificant; SNK, Student-Newman-Keuls test.

^aOne-way analysis of variance was performed for comparison of mean. F (df) was used to test main effects in the model, and SNK was used for multiple comparisons indicated by uppercase letters.

^bP < .05.

^ct test was performed for comparison of mean.

Results

Descriptive Statistics

Table 1 shows the mean PCS scores by sociodemographic characteristics, viral load, years living with HIV as well as alcohol and drug use. Participants were more acculturated to Latino culture than to US culture and reported living with HIV for an average of 8.7 years (range: 5–12 years). The vast majority of participants (92%) reported receiving ART, but only 45% reported undetectable viral load. The mean of PCS score in this sample was 46.6 with an SD of 10.2. One-way analyses of variance with Student-Newman-Keuls post hoc test revealed significant mean differences in physical function score by age and viral load categories. Table 2 shows bivariate correlations between predictors and physical function. We found a significant negative correlation between physical function and age, viral load, and depressive symptoms, with lower physical function associated with older age, higher viral load, and more depressive symptoms.

Table 2.

Correlations among Predictors and Physical Function.

	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Physical function	1.00	−.20a	−.18b	−.25a	−.06	−.06	.01	−.08	−.47c	−.11	−.01	−.21a	−.32c
2. Age		1.00	−.06	−.06	.35a	.16	−.14	.11	.05	−.19a	.09	.03	.43
3. Incomed			1.00	−.09	.09	−.07	.06	−.19	−.20d	.02	.11	−.03	−.09
4. Viral loadd				1.00	−.05	−.22	−.13	−.02	.15	.05	−.03	.12	.16b
5. Years living with HIV					1.00	.15	.10	.04	−.07	−.16	.13	.04	−.12
6. Nucleoside/nucleotide reverse transcriptase inhibitor (NRTI)d						1.00	.39a	.32a	−.12	−.15	‒.01	.07	.08
7. Protease inhibitors (PIs)d							1.00	−.13	−.13	.04	−.06	−.06	−.02
8. Nonnucleoside reverse transcriptase inhibitors (NNRTIs)d								1.00	.12	−.10	−.01	−.02	−.02
9. Depression									1.00	.08	−.08	.05	.33a
10. Latino acculturation										1.00	−.07	.01	.03
11. US acculturation											1.00	−.08	.04
12. Unintentional weight lossd												1.00	.25a
13. Fatigued													1.00
N	146	155	155	155	155	155	155	155	149	155	155	155	155
Mean	46.57	38.54	1.58	2.32	8.07	0.79	0.51	0.37	1.48	3.44	2.66	0.27	0.43
Standard deviation	10.23	9.37	0.50	1.50	4.26	0.41	0.50	0.48	0.47	0.40	0.60	0.45	0.50

^aP < .01.

^bP < .05.

^cP < .001.

^dSpearman ρ coefficient was calculated for ordinal variables (income, viral load, antiretroviral treatments, unintentional weight loss, and fatigue).

In order to address the first aim, we performed a hierarchical logistic regression model. As shown in Table 3, the overall regression model containing demographic variables, viral load, ART, years living with HIV, depression, and comorbidities was significant, with an R² of .33. The addition of acculturation variables did not increase significantly the R², thus indicating that acculturation did not add to the explanation of physical function. The bottom of Table 3 shows the full model containing all predictors. Findings supported the hypothesis that greater depressive symptoms were associated with lower physical function.

Table 3.

Linear Regression Model Examining the Association between Viral Load, Depression, Acculturation, and Physical Function.

Model	N	F	AIC	BIC	Adjusted R2
Set 1	146	6.60a	780.85	637.81	.33
Set 2	146	5.90a	786.13	640.04	.34
Final Model			β		SE β
Intercept			81.29		9.32
Set 1: Demographics, viral load, and depression
Age			−0.19b		0.08
Income			1.39		1.48
Viral loadc
Less than 10 000			−0.04		1.91
10 000 to 49 999			−1.82		2.43
More than 50 000			−7.88d		2.44
Unknown			−4.41b		2.47
Years living with HIV			−0.18		0.18
Antiretroviral treatment
Nucleoside/nucleotide reverse transcriptase inhibitor (NRTI) use			−1.48		2.20
Protease inhibitor (PI) use			0.08		1.61
Nonnucleoside reverse transcriptase inhibitor (NNRTI) use			0.64		1.64
Depression			−8.37a		1.62
Comorbidities
Unintentional weight loss			−2.90		1.70
Fatigue			−2.81		1.59
Set 2: Acculturation
Latino acculturation			−2.97		1.91
US acculturation			−0.30		1.59

Abbreviation: SE, standard error; AIC, Akaike Information Criterion; BIC, Bayesian Information Criterion.

^aP < .001.

^bP < .05.

^cUndetectable viral load was used as reference category.

^dP < .01.

As hypothesized, viral load was also associated with physical function. Specifically, those with high viral loads (over 50 000) had significantly lower PCS scores than the reference group with undetectable viral load. In addition, the group that did not know their viral load had lower physical function scores than the reference group. Consistent with the small change in the R², neither US acculturation nor Latino acculturation was related to physical function.

Discussion

In this cross-sectional study, we examined the association of viral load, depression, and acculturation to Latino and US culture with physical function. Findings provided additional information about physical function in a sample of mainly immigrant Latino men living with HIV. Our findings are consistent with previous studies that have found that limitations in physical function are associated with both psychological (eg, depression, social support) and biological factors (eg, CD4 count, viral load) in PLWH.^9,13,28,29 However, the findings of functional limitations among PLWH have been mixed: 2 studies found greater limitations in physical function among HIV-infected individuals,^28,30 whereas 2 studies did not find significant differences between HIV-infected and uninfected individuals.^9,31

Depression diagnosed with the BDI was significantly associated with poor physical function. This finding is consistent with other studies on the general population^5–8,32 and among PLWH.^9,11 Findings support prior studies that relied on clinically diagnosed depression rather than prospective assessment of depression concomitant with physical function among PLWH.^9,33 We found a function score that was on average 10 points lower in patients with depression, adjusted for the effects of age, income, years living with HIV, ART, and comorbidities. As previously noted, however, causal relationships are complicated.³⁴ Although depression can lead to frailty and physical limitations, the reverse direction has also been found.^10,11 Therefore, additional longitudinal research is necessary to understand more fully the bidirectional relationship between depression and physical function. Regardless of causal direction, treating depression at early stages is important, because it can foster physical activity and prevent the development of new or further physical function limitations.

Findings indicated an impact of viral load on physical function only when the viral load level was particularly high (ie, over 50 000). Prior studies examining the association of viral load and physical function have shown that higher viral load was associated with poor physical function.^13,15,16 Another study found an association between viral load and poor physical function only among those transitioning from undetectable to detectable viral load.¹⁴ Changes in viral load could be an indication of disease progression, effectiveness of the antiretroviral (ARV) medications, and/or difficulties with treatment adherence leading to a stage in which individuals would be expected to experience indicators of functional limitations. On the other hand, in other studies, viral load was not independently associated with physical function but lower CD4 counts were.^9,30,35 In order to avoid disease progression, early detection and initiation of care as well as compliance (eg, medical appointments, medication adherence, and laboratory control) are crucial. This issue is of particular concern for Latino populations who often have inadequate access to health care³⁶ and receive HIV diagnosis within 1 year of transitioning to AIDS.³⁷

Although there is evidence that sociocultural factors can play a role in physical activity, we failed to find a relationship between acculturation and physical function. Our findings, however, do not imply that culture is irrelevant to health. We did not compare Latinos to other groups but rather examined differences within a group of Latinos who were overwhelmingly immigrant. Previous research shows lower physical activity among Latinos in the United States, which could contribute to worse physical function.³⁸

There were several limitations in this study. The sample size was small, and these results are not generalizable to non-Latino men or women. Measures were by self-report. A limitation is that this is not a performance-based measure of physical function, which tends to have greater validity and reliability than self-reported indicators. The SF-12, however, has been reliably used with HIV-infected individuals.^39,40 Although previous studies have found an association between body mass index (BMI), common medical comorbidities (eg, diabetes, hypertension), and physical function,^3,9 we could not include those variables in our model because this information was not available in the original data set used in these analyses. The cross-sectional design limited our understanding of the directional effects among variables and did not allow examination of changes in physical function over time.

Despite these limitations, the current study provides important information concerning a distinctive sample: Latino men living with HIV in the United States about which only limited information is available. Moreover, our study examined the relationship between acculturation and physical function in HIV-positive Latino men. Although researchers have posited several possible reasons for the worse physical function among PLWH,^17,28,30 and the current study does not include sufficient information to address these questions, it is important to recognize the possible underlying processes that might contribute to the discrepancy. Do HIV and/or HAART accelerate the aging process and thereby cause deterioration of physical function? Or does HIV increase the prevalence of age-related comorbidities, which then have an impact on physical function? More large-sample and longitudinal studies are needed to explain the ways in which HIV can affect physical function.

Future research on physical function among PLWH would be enhanced by the inclusion of objective measures of physical function, such as 6-minute walk and timed chair rise.³³ In addition, the complex role of HIV-related indicators of viral load and CD4 count should be explored as well as the impact of ART. Furthermore, other physical conditions, such as age-related comorbidities and body composition (eg, BMI), should be included as covariates in future studies. In immigrant populations, migration status could also be important because it directly affects the ability to access health care. Results from this study could stimulate future research on the role of psychological and behavioral factors, as well as disease state, on physical function.

In summary, our findings indicate that care of PLWH should aim to minimize both viral load and depressive symptoms as a means of preserving physical function. It is important to promote access to mental health screening and treatment in healthcare settings dealing with PLWH. Although the importance of controlling HIV through ARV medication is well established, this study demonstrates the potential long-term quality-of-life implications of uncontrolled virus. In addition, Latinos are a growing ethnic minority group in the United States that has been disproportionately affected by HIV. Cultural factors can shape health-related conditions and behaviors, and therefore, in order to address health disparities, it is important to understand the role of culture. Our findings suggest that HIV and mental health treatment are essential in clinical practice for preserving physical function among Latino men.