Skip to main content
NCBI home page
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences logoLink to The Journals of Gerontology Series B: Psychological Sciences and Social Sciences
. 2021 Feb 13;77(1):50–60. doi: 10.1093/geronb/gbab027

Links Between Inflammation, Mood, and Physical Function Among Older Adults With HIV

Heather M Derry 1,✉, Carrie D Johnston 2, Chelsie O Burchett 1,3, Mark Brennan-Ing 4, Stephen Karpiak 5,6, Yuan-Shan Zhu 7, Eugenia L Siegler 1,#, Marshall J Glesby 2,#
Editor: Lynn Martire
PMCID: PMC8755907  PMID: 33580236

Abstract

Objectives

People living with human immunodeficiency virus (PLWH) treated with antiretrovirals have life spans similar to their HIV-negative peers. Yet, they experience elevated inflammation-related multimorbidity. Drawing on biopsychosocial determinants of health may inform interventions, but these links are understudied in older PLWH. We investigated cross-sectional relationships between psychosocial factors (mood, loneliness, and stigma), inflammatory markers, and age-related health outcomes among 143 PLWH aged 54–78 years.

Method

Participants provided blood samples for serum cytokine and C-reactive protein (CRP) analyses, completed surveys assessing psychosocial factors and health, and completed frailty assessments. Regression models tested relationships between key psychosocial-, inflammation, and age-related health variables, adjusting for relevant sociodemographic and clinical factors.

Results

Participants with more depressive symptoms had higher composite cytokine levels than those with fewer depressive symptoms (β = 0.22, t(126) = 2.71, p = .008). Those with higher cytokine levels were more likely to be prefrail or frail (adjusted odds ratio = 1.72, 95% confidence interval = 1.01–2.93) and reported worse physical function (β = −0.23, t(129) = −2.64, p = .009) and more cognitive complaints (β = −0.20, t(129) = −2.16, p = .03) than those with lower cytokine levels. CRP was not significantly related to these outcomes; 6-month fall history was not significantly related to inflammatory markers.

Discussion

Novel approaches are needed to manage comorbidities and maximize quality of life among older PLWH. Illustrating key expected biopsychosocial links, our findings highlight several factors (e.g., depressive symptoms, poorer physical function) that may share bidirectional relationships with chronic inflammation, a key factor driving morbidity. These links may be leveraged to modify factors that drive excessive health risk among older PLWH.

Keywords: Cytokines, Depression, Frailty, HIV, Stigma

Older adults are the fastest-growing segment of those living with human immunodeficiency virus (HIV). In 2018, the Center for Disease Control and Prevention estimated that 51% of people living with HIV (PLWH) were older than 50 years (Centers for Disease Control and Prevention, 2020), an age often used to mark older adulthood in HIV clinical guidelines (Department of Health and Human Services, 2019). Although those accessing HIV care frequently achieve viral suppression using effective antiretroviral regimens, PLWH face substantial health disparities as they age compared to their counterparts without HIV (Ogletree et al., 2019). For example, PLWH had approximately 50% higher risk for acute myocardial infarction over 6 years than demographically matched individuals without HIV (Freiberg et al., 2013). Although adequate comparison groups can be difficult to establish, similar patterns continue to emerge when comparing PLWH with other groups who are vulnerable to health disparities. For example, among older gay and bisexual men, those with HIV reported higher rates of health problems compared to those without HIV (Emlet et al., 2020). Furthermore, PLWH experience advanced aging processes in comparison to adults without HIV infection, such as earlier onset of geriatric syndromes (Pathai et al., 2014). Accordingly, current research priorities in the HIV and aging field include identifying and improving the complex, intersecting factors that drive adverse aging outcomes (Shiau et al., 2020).

Chronic inflammation is a likely contributor to excess health burden and frailty among PLWH (Fukui et al., 2018). Proinflammatory biomarkers such as C-reactive protein (CRP) and interleukin-6 (IL-6) are robust predictors of mortality and age-related health conditions (Pearson et al., 2003; Walker et al., 2019). PLWH have higher levels of chronic inflammation than those without HIV despite viral suppression, at least partially due to persistent HIV infection and immune activation (Peterson & Baker, 2019). Inflammation in turn elevates risk for geriatric syndromes among PLWH, though studies have largely focused on middle-aged adults (Margolick et al., 2017). Given age-related rises in inflammation and frailty risk, these links are most applicable in older adults, an understudied group in HIV research. In addition, the absence of clear interventions for blunting inflammation in PLWH (Peterson & Baker, 2019) suggests a need for identifying modifiable and behavioral factors that may influence inflammation in this group.

The biopsychosocial model, which suggests that biological, psychological, and social factors interact to influence health (Engel, 1980), is a particularly useful approach for considering modifiable pathways affecting health outcomes in those aging with HIV (Heckman & Halkitis, 2014). Indeed, psychosocial factors contribute to immune dysregulation, which is associated with poor health outcomes (Kiecolt-Glaser et al., 2015). For example, there is a strong evidence base to suggest that depressive symptoms and perceived social disconnection (e.g., feeling lonely, rejected, or excluded) promote inflammation (Kiecolt-Glaser et al., 2015; Leschak & Eisenberger, 2019). These links are complex, such that they are stronger in some subgroups than in others and are often bidirectional (Majd et al., 2020). Of note, older PLWH often experience relevant psychosocial risks, including depressive symptoms, loneliness, and HIV-related stigma, at greater rates than people without HIV and younger PLWH (Grov et al., 2010), suggesting it is particularly important to investigate the physiological correlates of these factors among older PLWH.

Initial work in middle-aged PLWH suggests that depressive symptoms are associated with inflammation. In a study of middle-aged adults, PLWH with clinically significant depressive symptoms had higher CRP, an inflammatory biomarker linked to cardiovascular risk, than those with fewer depressive symptoms; a similar pattern was observed in linear analyses, suggesting this relationship extends to subclinical symptoms (Poudel-Tandukar et al., 2014). Higher levels of inflammatory markers (indexed by empirically derived factors of multiple immune markers) also increased the risk of both clinically significant and mild depressive symptoms over time in a prospective study of PLWH who were primarily middle-aged (Lu et al., 2019). However, the literature on these links in the context of aging with HIV is relatively nascent. Furthermore, associations between inflammation and loneliness or stigma have not been demonstrated in PLWH to our knowledge. These are key gaps, because depressive symptoms, loneliness, and HIV-related stigma, which elevate health risks, are common, notable, often interrelated problems that are more prevalent with age in PLWH (Grov et al., 2010; Nanni et al., 2015).

In summary, PLWH have greater systemic inflammation and adverse age-related health outcomes, as well as psychosocial risk factors, compared to those without HIV. Determining relationships between key physiological and psychosocial risks may help to identify strategies to target modifiable factors that contribute to the disproportionate medical burden for those aging with HIV. In doing so, our approach was informed by a conceptual model in which psychological (e.g., depression) and social (e.g., loneliness, HIV-related stigma) factors may impact key physiological mechanisms (e.g., inflammation) that drive relevant age-related health outcomes among PLWH. In this study, we examined these biopsychosocial links cross-sectionally, a needed initial step to investigating these pathways among older PLWH. This study’s goals were to (a) determine whether key psychosocial factors were related to inflammatory markers and (b) examine age-related health correlates of inflammation among older adults with HIV. We expected that participants with greater depressive symptoms, loneliness, and HIV-related stigma would have higher levels of inflammation than those with lower depressive symptoms, loneliness, and stigma. We also hypothesized that higher levels of inflammation would be associated with greater age-related health problems, including frailty, falls, self-reported physical function, and cognitive complaints.

Method

Participants

Older adults with HIV who were English- or Spanish-speaking were recruited from the HIV clinics at New York-Presbyterian Hospital/Weill Cornell Medicine using an age-stratified random selection strategy as part of the larger Research on Older Adults with HIV 2.0 multisite survey study (Erenrich et al., 2018). Of these participants, a subset was invited to complete an additional study visit in order to obtain more detailed biomedical assessments of health and aging. Participants were invited to the biomedical substudy if they were aged 55 or older (selected to increase the representation of aging-related phenotypes of interest, including frailty), had documented HIV infection, and reported comfort completing the study procedures in English (due to capability of available staff). Exceptions were made to enroll two individuals who were aged 54 and met the other inclusion criteria. The study procedures were approved by the Weill Cornell Medical College Institutional Review Board (IRB# 1603017050). Participants provided informed consent, and received compensation for attending the study visits.

Planned enrollment for the biomedical substudy was 200 participants. A preliminary power calculation suggested that 193 participants would be sufficient for 80% power to detect small-to-medium bivariate correlations (ρ = .20) between psychosocial factors and inflammation. Accrual for the biomedical substudy was stopped early due to funding constraints, and there were 161 participants who completed both the psychosocial survey and the biomedical study visits and had blood samples available for inflammation assays. Given the current study’s intended focus on chronic low-grade inflammation, we excluded 18 individuals from the analytic sample with CRP levels above 10 mg/L, a threshold indicating likely acute infection (Boylan & Ryff, 2013; Pearson et al., 2003; Wilson et al., 2019). Accordingly, 143 participants are included in the current analysis. Those with CRP values above 10 mg/L had higher disease burden as measured by Veterans Aging Cohort Study (VACS) index scores (Tate et al., 2013) than those with lower CRP levels (t(159) = 2.88, p = .005); there were no statistically significant differences in age, gender, race, ethnicity, smoking status, or body mass index (BMI) between those included versus excluded based on CRP values (p-values for comparisons > .19). In order to illustrate how this analytic decision impacted the results, we completed ancillary analyses that included participants with high CRP values; this approach yielded largely similar results and is available in the Supplementary Material (Tables S1–S4).

Procedures

Participants completed an initial visit during which they completed surveys that included measures of sociodemographic information, depressive symptoms, loneliness, stigma, and smoking status. At a separate biomedical research visit at the Weill Cornell Clinical and Translational Science Center, participants provided fasting blood samples, which were processed, then stored at −80 °C until assayed for inflammatory markers. After eating breakfast, participants completed a frailty assessment that was administered by trained research staff, as well as a survey of self-reported health and function.

Measures

Inflammation biomarkers

Serum samples were assayed for CRP and proinflammatory cytokines IL-6, interferon-gamma (IFN-γ), tumor necrosis factor-alpha (TNF-α), and IL-1β using an electrochemiluminescence method on the MESO QuickPlex SQ120 Analyzer according to Meso Scale Discovery (Rockville, MD) kit instructions (K151STG and K15052G, respectively). Assays were run in duplicate with quality controls; 10% were repeated for confirmation. The intra-assay coefficient of variation (CV) for CRP, IL-6, IFN-γ, TNF-α, and IL-1β ranged from 2.4% to 6.4%, and inter-assay CVs ranged from 5.0% to 10.2%. The detection limits of CRP, IL-1β, IL-6, TNF-α, and IFN-γ are 0.01 ng/mL, 0.10 pg/mL, 0.10 pg/mL, 0.20 pg/mL, and 0.40 pg/mL, respectively. IL-1β was below the limit of detection in the majority of samples (98%), and thus IL-1β data were not analyzed.

Psychosocial factors

Depressive symptoms.

—Participants reported depressive symptoms on the well-validated 10-item Center for Epidemiological Studies - Depression (CES-D-10) scale (Andresen et al., 1994). The CES-D has been widely used in studies of PLWH, including older adults (Grov et al., 2010), and the shorter 10-item version has high concordance with the full scale among PLWH (Zhang et al., 2012). Representative items include “I was bothered by things that usually don’t bother me” and “I felt hopeful about the future” (reverse-scored), with items scored on a 0–3 scale. Higher total scores indicate greater depressive symptoms, and scores above 10 suggest clinically significant depressive symptoms (Andresen et al., 1994). When the response for one item was missing, the mean of the other items was substituted; total scores were not computed for those with more than one missing item (n = 5; Andresen et al., 1994). In this sample, total scores ranged from 0 to 24, and Cronbach’s alpha was .83.

Loneliness.

—Participants completed a modified version of the UCLA Loneliness Scale (Russell et al., 1978). An abbreviated version consisting of 13 items (items 1–7, 10–12, 14, 18, and 19 from the original scale) was used to reduce participant burden. Participants rated how often they experienced the items (such as “I lack companionship,” “I feel left out,” and “I feel as if nobody really understands me”) on a scale of 1 (never) to 4 (often). Higher total scores indicated greater levels of loneliness. Consistent with the approach to CES-D data above, we substituted the mean of the other items when one was missing, and total scores were not computed for those with more than one missing item (n = 6). In this sample, total scores ranged from 13 to 52, and Cronbach’s alpha was .94.

HIV-related stigma.

—Participants completed the well-validated 13-item HIV Stigma Scale (Emlet, 2005; Sowell et al., 1997). According to the scale instructions, participants rated the items (such as “I feared I would lose my friends if they learned about my illness,” “I felt blamed by others for my illness,” and “I thought other people were uncomfortable being with me”) in regard to their HIV status, using Likert-type options ranging from 1 (not at all) to 4 (often). Consistent with the approach described above, we substituted the mean of the other items when one was missing, and total scores were not computed for those with more than one missing item (n = 10). In this sample, total scores ranged from 13 to 44, with higher scores indicating greater stigma, and Cronbach’s alpha was .87.

Age-related health correlates

Self-rated function and symptoms.

—Participants completed the Medical Outcomes Study-HIV Health Survey (MOS-HIV), a well-validated quality-of-life instrument (Wu et al., 1997). Subscale scores were computed for physical function and cognitive function. Higher scores indicate better self-reported well-being on physical function and cognitive function subscales.

Recent falls.

—To assess history of recent falls, participants were asked to respond “yes” or “no” to the question “In the past six months, have you had a fall?”

Frailty.

—Frailty was assessed according to the well-established Fried phenotype, which includes weak grip, slow walk, unintentional weight loss, low physical activity, and exhaustion components (Fried et al., 2001). To classify each frailty criterion, we used the following methods and cutoffs from a prior large cohort study (HIV Infection, Aging, and Immune Function Long-Term Observational [HAILO] Study) of PLWH, as described by Erlandson and colleagues (2017): Grip strength was assessed using the average of three dynamometer trials with the participant’s dominant hand; sex- and BMI-specific cutoffs indicating weak grip were applied (values listed in Fried et al., 2001). To assess slowness, participants completed two 4-m walk trials at their usual speed, with assistive devices as needed. The average walk time in seconds was computed, and slow walk was defined by sex- and height-specific cutoffs (average walk time of ≥6.22 s for men ≤173 cm and women ≤159 cm in height; average walk time of ≥5.33 s for men >173 cm and women >159 cm). Unintentional weight loss of 10 pounds or more in the past year was assessed via self-report; for six participants who did not provide this information, medical chart data revealed lack of unintentional weight loss and was used to code their status. Activity level was assessed with the question “Does your health limit you in vigorous activities such as running, lifting heavy objects, or participating in strenuous sports?,” and low activity was defined by responding that their health limited vigorous activity “a lot” (rather than “not at all” or “a little”). Exhaustion was indicated by endorsing “everything I do is an effort” or “sometimes I just cannot get going” at least three to four times per week.

Participants who met frailty criteria for none of the components were categorized as robust, those who met criteria for one or two components were categorized as prefrail, and those who met criteria for three or more components were categorized as frail. Participants with missing data on one or two components were considered evaluable (Fried et al., 2001) if their final frailty category would remain unchanged by the addition of the remaining scores (n = 3).

Additional health-related variables/covariates

BMI was calculated using height and weight measured at the biomedical visit. Smoking status was assessed via self-report on the psychosocial survey. Medical chart data were used to calculate VACS index scores, which were used to measure the physiological burden of HIV and multimorbidity. Research nursing staff reconciled participants’ medication lists at the study visit. We coded a dichotomous variable to summarize any versus no use of regular, systemic anti-inflammatory medications (immunosuppressants, systemic glucocorticoids, antigout medications, and statins; WHO ATC categories L04A, H02A, M04A, and C10AA, respectively). In order to maximize the sample for adjusted analyses, we used relevant medical chart data from the nearest clinical encounter for those with missing covariate data.

Statistical Analysis

Analyses were conducted in SPSS, version 24.0 (IBM Corp., Armonk, NY). Inflammation data were examined for outliers, and each biomarker was winsorized to 3 SDs from the mean (Boylan & Ryff, 2013; Gouin et al., 2020). Due to positive skew, inflammation variables were then log-transformed for normality. Given moderate-to-strong correlations between IL-6, TNF-α, and IFN-γ cytokine levels (r = .32–.57), a composite score was computed (i.e., mean of standardized variables) in order to facilitate interpretation and limit the number of statistical comparisons. Correlations with CRP (a downstream marker stimulated by IL-6) were weaker and thus this variable was examined separately.

Pearson correlations were used to evaluate bivariate associations between inflammatory markers, psychosocial factors, and age-related health outcomes. For those with significant bivariate associations using an alpha level of .05, we conducted follow-up adjusted analyses using regression models. Separate linear regression models were used to examine links between psychosocial predictors and inflammatory markers, and to test whether inflammatory markers were key predictors of age-related health outcomes. Separate logistic regression models were used to test relationships between inflammation (key predictors) and frailty. Due to the relatively low number of frail participants and prior work suggesting negative health consequences of prefrailty (Paolillo et al., 2020), we combined prefrail and frail categories (key dichotomous outcome), with robust as the reference category for this analysis. Adjusted analyses controlled for age, sex, race, BMI, VACS score, smoking status, and medication use based on their relationships with inflammation and health outcomes in the broader literature (Lu et al., 2019; Wilson et al., 2019). Cases with missing data were excluded listwise from adjusted analyses. Residual plots and collinearity statistics were examined to verify that regression assumptions were met; variables were entered simultaneously.Given literature suggesting possible age and gender differences in the relationships between depression, inflammation, and health symptoms (Majd et al., 2020; Walker et al., 2020), we conducted exploratory analyses to examine whether these factors moderated the relationships of interest. In separate unadjusted linear and logistic regression models, we examined the interactions between sex and psychosocial factors (i.e., depressive symptoms, loneliness, and stigma) as predictors of inflammation levels; similarly, we examined interactions between sex and inflammation as predictors of self-rated physical function, cognitive complaints, falls, and frailty status. The same approach was used to examine age as a moderator.

Results

Of the 143 participants in the analytic sample, individuals were primarily Black (50%), male (68%), and completed education beyond high school (71%). Participant age ranged from 54 to 78 years, with a mean of 61 years (SD = 5.88). On average, participants had been diagnosed with HIV for 23 years (SD = 5.96). The majority (93%) were virally suppressed. Seventy-two participants (50%) were taking substantial systemic anti-inflammatory medications, with four (3%) taking immunosuppressants, six (4%) taking systemic glucocorticoids, two (1%) taking antigout medications, and 68 (48%) taking statins; some had multiple prescriptions. Additional sample characteristics are presented in Table 1.

Table 1.

Characteristics of Participants in Sample

Characteristic Participants with data Mean (SD) or n (%)
Age 143 61.10 (5.88)
Sex 143
 Female 45 (32%)
 Male 98 (68%)
Race 139
 Black 69 (50%)
 White 44 (32%)
 Asian or Pacific Islander 2 (1%)
 Bi- or multiracial 24 (17%)
Ethnicity 124
 Hispanic/Latino 39 (31%)
 Non-Hispanic/Latino 85 (69%)
Education level 139
 ≤12 years 40 (29%)
 >12 years 99 (71%)
Viral load 143
 <200 copies/mL 133 (93%)
 ≥200 copies/mL 10 (7%)
Time since HIV diagnosis (years) 143 23.16 (5.95)
VACS score 143 31.45 (16.45)
BMI (kg/m2) 143 28.19 (6.99)
Current tobacco use 139 23 (16%)
Frailty category 140
 Robust 44 (31%)
 Prefrail 81 (58%)
 Frail 15 (11%)
Depressive symptoms (CES-D score) 138 9.98 (6.27)
HIV-related stigma (HIV Stigma Scale score) 133 20.81 (7.36)
Loneliness (13-item UCLA Loneliness Scale score) 137 24.71 (9.60)
Open in a new tab

Note: BMI = body mass index; CES-D = Center for Epidemiological Studies Depression scale; HIV = human immunodeficiency virus; VACS = Veterans Aging Cohort Study. % indicates percentage of those with available data for each variable.

Approximately half (49%) of the sample scored 10 or higher on the CES-D-10, suggesting significant levels of depressive symptoms (Andresen et al., 1994). Using Fried frailty phenotype criteria, 11% were frail, 58% were prefrail, and 31% were robust. The most commonly displayed frailty characteristics were exhaustion (37%) and low activity (30%) followed by weak grip (19%), slow walk (18%), and unintentional weight loss (6%). Twenty-two percent reported that they had fallen in the past 6 months. The median VACS index score was 28 (interquartile range = 18–39), corresponding to a 10.8% 5-year mortality risk.

Psychosocial Factors and Inflammation

Bivariate analyses (Table 2) showed a significant correlation between depressive symptoms and composite cytokine levels (r = .26, p = .002), but not CRP levels (r = −.02, p = .82). In adjusted analyses (Table 3), those with greater depressive symptoms had higher cytokine levels than those with fewer depressive symptoms (standardized beta coefficient, β = 0.22, t(126) = 2.71, p = .008), above and beyond the effects of age, sex, race, BMI, VACS scores, smoking status, and anti-inflammatory medication use.

Table 2.

Bivariate Pearson Correlations Between Inflammation Markers and Key Outcomes

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
[1] Log IL-6 —
[2] Log IFN-γ .32** —
[3] Log TNF-α .43** .57** —
[4] Log CRP .38** .26* .24* —
[5] Cytokine composite .74** .80** .84** .37** —
[6] Depressive symptoms .20* .25* .19* −.02 .26* —
[7] Loneliness .08 .04 −.04 −.09 .03 .58** —
[8] Stigma .05 −.09 .01 −.18* −.03 .36** .44** —
[9] Physical function −.28** −.20* −.16 −.12 −.26* −.37** −.13 −.17* —
[10] Cognitive function −.17* −.18* −.13 −.04 −.20* −.46** −.21* −.27* .38** —
[11] Fall in past 6 months .04 .13 −.03 −.01 .06 .27* .22* .07 −.35** −.15 —
[12] Prefrail/frail status .19* .18* .13 .06 .21* .40** .21* .11 −.52** −.39** .21*
Open in a new tab

Notes: CRP = C-reactive protein; IFN = interferon; IL = interleukin; TNF = tumor necrosis factor.

*p < .05. **p < .001.

Table 3.

Adjusted Linear Regression Model With Depressive Symptoms as a Predictor of Cytokine Levels

Outcome: composite cytokine levels
Predictor β t p
Age (years) −0.004 −0.04 .97
Sex (male) 0.04 0.45 .65
Race (White) 0.02 0.17 .87
BMI 0.17 2.07 .04
VACS score 0.27 2.72 .008
Smoking (yes) 0.12 1.39 .17
Anti-inflammatory use 0.04 0.52 .61
Depressive symptoms 0.22 2.70 .008
Open in a new tab

Note: BMI = body mass index; VACS = Veterans Aging Cohort Study. Depressive symptoms are measured by continuous 10-item Center for the Epidemiological Studies - Depression scale total scores; composite cytokine levels summarize log-transformed interleukin-6, tumor necrosis factor-α, and interferon-γ.

Loneliness was not significantly associated with composite cytokine levels or CRP levels (p-values > .29). Unexpectedly, those who reported greater stigma had lower CRP levels than those who reported lower stigma (r = −.18, p = .04), and this association remained in adjusted analyses (β = −0.24, t(120) = −2.64, p = .01). Stigma was not associated with composite cytokine levels (r = −.03, p = .74). In exploratory analyses, sex did not significantly moderate the effect of depression, loneliness, or perceived stigma on composite cytokine levels or CRP levels in separate unadjusted models (p-values for all interaction effects > .37). Similarly, age was not a statistically significant moderator of these relationships (p-values for all interaction effects > .14).

Inflammation and Age-Related Health Outcomes

In bivariate analyses, composite cytokine levels were associated with self-reported physical function and cognitive well-being (Table 2). These relationships remained in adjusted linear regression models (Table 4). Participants with higher composite cytokine levels had poorer physical function (β = −0.23, t(129) = −2.64, p = .009) and greater cognitive complaints (β = −0.20, t(129) = −2.16, p = .03) than those with lower cytokine levels. Furthermore, those with higher composite cytokine levels were more likely to be prefrail or frail than those with lower cytokine levels in adjusted analyses (adjusted odds ratio = 1.72, 95% confidence interval = 1.01–2.93, p = .047). CRP was not significantly related to these examined age-related outcomes, and self-reported 6-month fall history was not associated with inflammation levels. In exploratory moderation analyses, the relationships between inflammation and these health outcomes did not vary significantly by age (p-values for all interaction effects > .12) or sex (p-values for all interaction effects > .24) in separate models.

Table 4.

Adjusted Linear and Logistic Regression Models With Cytokine Composite Levels as Predictors of Age-Related Health Outcomes

Outcome: MOS-HIV
physical function		 Outcome: MOS-HIV
cognitive function		 Outcome: prefrail/frail status
Predictor β t p β t p OR 95% CI p
Age (years) 0.10 1.00 .32 0.05 0.51 .61 1.07 0.98–1.17 .11
Sex (male) 0.11 1.25 .21 −0.002 −0.02 .98 0.96 0.39–2.36 .93
Race (White) 0.16 1.80 .08 −0.001 −0.009 .99 0.66 0.28–1.59 .36
BMI −0.15 −1.83 .07 0.03 0.33 .74 0.99 0.94–1.04 .66
VACS index score −0.06 −0.60 .55 −0.09 −0.81 .42 1.01 0.98–1.04 .65
Smoking (yes) −0.01 −0.12 .91 0.11 1.27 .21 1.02 0.36–2.87 .97
Anti-inflammatory use −0.05 −0.53 .60 −0.01 −0.07 .94 1.10 0.50–2.40 .82
Composite cytokine levels −0.23 −2.64 .009 −0.20 −2.16 .03 1.72 1.01– 2.93 .05
Open in a new tab

Note: BMI = body mass index; CI = confidence interval; MOS-HIV = Medical Outcomes Study-HIV Health Survey; OR = odds ratio; VACS = Veterans Aging Cohort Study.

Discussion

In this single-site observational study, we examined cross-sectional links between psychosocial factors, inflammatory markers, and age-related health outcomes among older adults (ages 54–78 years) with HIV who were receiving care at an outpatient clinic. Those who reported greater depressive symptoms on the CES-D-10 had higher cytokine levels than those with fewer depressive symptoms. In addition, higher cytokine levels were associated with subjective health symptoms on the MOS-HIV survey, including poorer self-reported physical function and more cognitive complaints. Finally, participants with higher cytokine levels were more likely to exhibit prefrailty or frailty phenotypes than their counterparts with lower cytokine levels. These relationships were present above and beyond other relevant sociodemographic and health-related factors. This study suggests key, expected links between depression, inflammation, and health among older PLWH, an understudied group with excessive health risk during aging.

The observed links between depressive symptoms and inflammatory cytokines in this study builds upon prior well-established research in adults without HIV. Depressed individuals have heightened inflammation over time, and elevated inflammation enhances risk for depression (Kiecolt-Glaser et al., 2015; Sonsin-Diaz et al., 2020). In experimental studies, inflammatory responses can elevate depressive symptoms, cognitive dysfunction, and social withdrawal, termed “sickness behaviors” (Dantzer et al., 2008); those with a history of depression have higher inflammatory responses to immune challenges (Christian et al., 2010). In addition, this study adds to the emerging literature on mood and inflammation among PLWH. In recent cross-sectional studies of PLWH, somatic depressive symptoms were associated with higher levels of IL-6 and D-dimer (Norcini Pala et al., 2016; Stewart et al., 2020). We observed similar inflammation–depression links in this sample, suggesting a pattern that is consistent across emerging research studies of PLWH.

There were not consistent relationships between inflammation levels and loneliness or stigma. Unexpectedly, greater perceived stigma was associated with lower CRP levels. Although there is ample evidence that experiences of social disconnection including loneliness can elevate inflammation (Leschak & Eisenberger, 2019; Smith et al., 2020), these links have not been previously examined among PLWH to our knowledge. It is possible that ties between loneliness and inflammation are not as strong among older PLWH as they are in younger PLWH or those without HIV. On the other hand, a potential reason we did not observe the expected correlations between loneliness and HIV-related stigma with inflammatory markers could be that most participants were men, and some studies (but not all [Vingeliene et al., 2019]) suggest that social factors and inflammation may be more strongly linked among women (Elliot et al., 2018). Future studies of perceived social disconnection and inflammation among PLWH will be informative to further understand these effects or lack thereof.

Consistent with prior studies documenting that inflammation contributes to adverse aging outcomes among PLWH, we observed that participants with higher cytokine levels were more likely to be prefrail or frail than those with lower cytokine levels. These effects are especially meaningful in PLWH given that health risks may converge, accumulating to increase vulnerability to physical health challenges. For example, in a large clinical database, PLWH with higher CRP levels had fourfold higher rates of acute myocardial infarction compared with individuals without HIV who had lower CRP levels (Triant et al., 2009). Our findings also support the view that these patterns extend to key patient-reported outcomes that affect quality of life during aging, such that depressive symptoms, self-reported physical function, and cognitive complaints are poorer among PLWH with higher cytokine levels. While the current study focused on establishing within-group links between psychosocial factors, inflammation, and health outcomes among older PLWH, future studies that include participants without HIV (i.e., demographically matched healthy controls or those with other chronic conditions) would be helpful for establishing whether these links are stronger, weaker, or similar among those with HIV and those without HIV. For example, determining how the strength of these patterns in PLWH compares to those without HIV could have implications for understanding health disparities, strategies for identifying health risks, and intervention approaches.

These findings contribute needed data on a broad sample of older participants living with HIV, a strength of the current study. Our sample appears to be older (median age = 60) with higher rates of prefrailty and frailty (58% and 11%, respectively) than a well-characterized large cohort of 1,016 PLWH (median age = 51) in which 38% were prefrail and 6% were frail (HAILO Study, e.g., Erlandson, Wu, et al., 2017). By recruiting from an outpatient HIV clinic using an age-stratified random selection strategy as an entry point to the study, we studied a group of participants that reflects a broader spectrum of PLWH than in prior studies that selected specialized subgroups, such as long-term cohort studies focusing on PLWH who inject drugs (Piggott et al., 2020) or men who have sex with men (Erlandson et al., 2017). This approach strengthened the generalizability of the current findings. Yet, there are also limits of generalizability; all participants were recruited from a single outpatient program where they were engaged in HIV care. Thus, the extent to which these findings represent patterns for individuals in other locations or those with limited health care access or engagement is unknown. Additional strengths of the study include the use of validated assessments of frailty, psychosocial factors, and self-reported health symptoms, as well as inclusion of relevant covariates in analyses (omission of which has limited interpretation of several prior studies).

The primary limitation of this study is that the cross-sectional nature of the analysis prevents conclusions about the directionality of these effects. The broader literature suggests that inflammation–depression links are bidirectional. Accordingly, it is possible that higher levels of inflammation drive both depressive symptoms and age-related health outcomes over time, or that depressive symptoms lead to higher inflammation levels that carry health risks. Future prospective studies should evaluate these factors longitudinally in order to inform intervention strategies, such as those that target psychosocial factors to reduce inflammation. An additional limitation is that depressive symptoms were measured at the separate survey visit before the biomedical assessment (median = 2 months). While it is possible that we would have observed different patterns if both assessments occurred on the same day, this also suggests that earlier and/or stable depressive symptoms could have lasting relationships with inflammation. In addition, the substantial number of samples with undetectable IL-1β was unexpected, although prior work with other populations has also noted a high frequency of undetectable values (Kiecolt-Glaser et al., 2015; Pfitzenmaier et al., 2003). In future studies, researchers may wish to consider more sensitive methods for assessing this marker in older PLWH, or alternative approaches such as assessing lipopolysaccharide-stimulated production of IL-1β. Finally, the accrued sample was smaller than originally anticipated, which may have limited our ability to detect several effects of interest (including links with CRP and loneliness); a larger study is needed in order to make stronger conclusions about the null effects observed in this study.

In summary, as mortality in PLWH has shifted toward non-acquired immunodeficiency syndrome-related chronic diseases (Smith et al., 2014), novel approaches are needed to manage inflammation-related comorbidities, thereby maximizing quality of life among older PLWH (Erlandson & Karris, 2019; Siegler & Brennan-Ing, 2017). Our results highlight several potential risk factors (e.g., depressive symptoms, poorer physical function) that may share bidirectional relationships with low-grade inflammation, a key physiological driver of morbidity and mortality. Potential research and practice implications include leveraging these links in order to address factors that contribute to multiple biopsychosocial outcomes. This would be consistent with utilizing key geriatric principles to guide comprehensive, multidomain assessments and holistic care plans (Singh et al., 2017). For example, improving depressive symptoms could be a promising strategy to mitigate inflammation in this at-risk group (or vice versa). Further longitudinal work is needed to elucidate how these psychosocial and physical risks influence each other over time among older PLWH, to promote innovative strategies toward achieving health equity for those aging with HIV.

Supplementary Material

gbab027_suppl_Supplementary_Material
Click here for additional data file. (41.6KB, docx)

Acknowledgments

The authors express appreciation to the participants in this study and to the staff at the Weill Cornell Clinical and Translational Center. This study was not preregistered. De-identified data will be shared on reasonable request to the corresponding author.

Funding

This work was supported by funding from the American Psychological Foundation (Visionary Grant), the MAC AIDS Fund, the New York Community Trust, National Institute on Aging (T32 AG049666), the National Cancer Institute (K99 CA245488), the National Institute of Allergy and Infectious Diseases (T32 AI007613), and the National Center For Advancing Translational Sciences (UL1TR000457).

Conflict of Interest

H. M. Derry has a financial relationship (spouse employment) with Merck. M. J. Glesby’s institution has received research support from Gilead Sciences and Regeneron to support his work.

References

  1. Andresen, E. M., Malmgren, J. A., Carter, W. B., & Patrick, D. L. (1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale). American Journal of Preventive Medicine, 10(2), 77–84. doi: 10.1016/s0749-3797(18)30622-6 [DOI] [PubMed] [Google Scholar]
  2. Boylan, J. M., & Ryff, C. D. (2013). Varieties of anger and the inverse link between education and inflammation: Toward an integrative framework. Psychosomatic Medicine, 75(6), 566–574. doi: 10.1097/PSY.0b013e31829683bd [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Centers for Disease Control and Prevention. (2020). HIV surveillance report, 2018 (updated): vol. 31.http://www.cdc.gov/hiv/library/reports/hiv-surveillance.html
  4. Christian, L. M., Franco, A., Iams, J. D., Sheridan, J., & Glaser, R. (2010). Depressive symptoms predict exaggerated inflammatory responses to an in vivo immune challenge among pregnant women. Brain, Behavior, and Immunity, 24(1), 49–53. doi: 10.1016/j.bbi.2009.05.055 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dantzer, R., O’Connor, J. C., Freund, G. G., Johnson, R. W., & Kelley, K. W. (2008). From inflammation to sickness and depression: When the immune system subjugates the brain. Nature Reviews. Neuroscience, 9(1), 46–56. doi: 10.1038/nrn2297 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Department of Health and Human Services. (2019). Panel on antiretroviral guidelines for adults and adolescents. Guidelines for the use of antiretroviral agents in adults and adolescents living with HIV.https://aidsinfo.nih.gov/contentfiles/lvguidelines/adultandadolescentgl.pdf [DOI] [PubMed]
  7. Elliot, A. J., Heffner, K. L., Mooney, C. J., Moynihan, J. A., & Chapman, B. P. (2018). Social relationships and inflammatory markers in the MIDUS cohort: The role of age and gender differences. Journal of Aging and Health, 30(6), 904–923. doi: 10.1177/0898264317698551 [DOI] [PubMed] [Google Scholar]
  8. Emlet, C. A. (2005). Measuring stigma in older and younger adults with HIV/AIDS: An analysis of an HIV Stigma Scale and initial exploration of subscales. Research on Social Work Practice, 15(4), 291–300. doi: 10.1177/1049731504273250 [DOI] [Google Scholar]
  9. Emlet, C. A., Fredriksen-Goldsen, K. I., Kim, H. J., & Jung, H. (2020). Accounting for HIV health disparities: Risk and protective factors among older gay and bisexual men. Journal of Aging and Health, 32(7–8), 677–687. doi: 10.1177/0898264319848570 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Engel, G. L. (1980). The clinical application of the biopsychosocial model. American Journal of Psychiatry, 137(5), 535–544. PMID:7369396. doi: 10.1176/ajp.137.5.535 [DOI] [PubMed] [Google Scholar]
  11. Erenrich, R., Seidel, L., Brennan-Ing, M., & Karpiak, S. (2018). HIV and aging in San Francisco: Findings from the research on older adults with HIV (ROAH) 2.0 San Francisco Study.https://bit.ly/2AhRvJV
  12. Erlandson, K. M., Wu, K., Koletar, S. L., Kalayjian, R. C., Ellis, R. J., Taiwo, B., Palella, F. J., Jr, Tassiopoulos, K. (2017). Association between frailty and components of the frailty phenotype with modifiable risk factors and antiretroviral therapy. Journal of Infectious Diseases, 215(6), 933–937. doi: 10.1093/infdis/jix063 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Erlandson, K. M., & Karris, M. Y. (2019). HIV and aging: Reconsidering the approach to management of comorbidities. Infectious Disease Clinics of North America, 33(3), 769–786. doi: 10.1016/j.idc.2019.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Erlandson, K. M., Ng, D. K., Jacobson, L. P., Margolick, J. B., Dobs, A. S., Palella, F. J., Lake, J. E., Bui, H., Kingsley, L., & Brown, T. T. (2017). Inflammation, immune activation, immunosenescence, and hormonal biomarkers in the frailty-related phenotype of men with or at risk for HIV infection. The Journal of Infectious Diseases, 215(2), 228–237. doi: 10.1093/infdis/jiw523 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Freiberg, M. S., Chang, C. C., Kuller, L. H., Skanderson, M., Lowy, E., Kraemer, K. L., Butt, A. A., Bidwell Goetz, M., Leaf, D., Oursler, K. A., Rimland, D., Rodriguez Barradas, M., Brown, S., Gibert, C., McGinnis, K., Crothers, K., Sico, J., Crane, H., Warner, A., … Justice, A. C. (2013). HIV infection and the risk of acute myocardial infarction. JAMA Internal Medicine, 173(8), 614–622. doi: 10.1001/jamainternmed.2013.3728 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fried, L. P., Tangen, C. M., Walston, J., Newman, A. B., Hirsch, C., Gottdiener, J., Seeman, T., Tracy, R., Kop, W. J., Burke, G., & McBurnie, M. A. (2001). Frailty in older adults evidence for a phenotype. The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 56(3), M146–M157. doi: 10.1093/gerona/56.3.M146 [DOI] [PubMed] [Google Scholar]
  17. Fukui, S. M., Piggott, D. A., & Erlandson, K. M. (2018). Inflammation strikes again: Frailty and HIV. Current HIV/AIDS Reports, 15(1), 20–29. doi: 10.1007/s11904-018-0372-5 [DOI] [PubMed] [Google Scholar]
  18. Gouin, J. P., Wrosch, C., McGrath, J., & Booij, L. (2020). Interpersonal capitalization moderates the associations of chronic caregiving stress and depression with inflammation. Psychoneuroendocrinology, 112, 104509. doi: 10.1016/j.psyneuen.2019.104509 [DOI] [PubMed] [Google Scholar]
  19. Grov, C., Golub, S. A., Parsons, J. T., Brennan, M., & Karpiak, S. E. (2010). Loneliness and HIV-related stigma explain depression among older HIV-positive adults. AIDS Care, 22(5), 630–639. doi: 10.1080/09540120903280901 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Heckman, T. G., & Halkitis, P. N. (2014). Biopsychosocial aspects of HIV and aging. Behavioral Medicine (Washington, D.C.), 40(3), 81–84. doi: 10.1080/08964289.2014.937630 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kiecolt-Glaser, J. K., Derry, H. M., & Fagundes, C. P. (2015). Inflammation: Depression fans the flames and feasts on the heat. The American Journal of Psychiatry, 172(11), 1075–1091. doi: 10.1176/appi.ajp.2015.15020152 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Leschak, C. J., & Eisenberger, N. I. (2019). Two distinct immune pathways linking social relationships with health: Inflammatory and antiviral processes. Psychosomatic Medicine, 81(8), 711–719. doi: 10.1097/PSY.0000000000000685 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lu, H., Surkan, P. J., Irwin, M. R., Treisman, G. J., Breen, E. C., Sacktor, N., Stall, R., Wolinsky, S. M., Jacobson, L. P., & Abraham, A. G. (2019). Inflammation and risk of depression in HIV: Prospective findings from the multicenter AIDS cohort study. American Journal of Epidemiology, 188(11), 1994–2003. doi: 10.1093/aje/kwz190 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Majd, M., Saunders, E. F. H., & Engeland, C. G. (2020). Inflammation and the dimensions of depression: A review. Frontiers in Neuroendocrinology, 56, 100800. doi: 10.1016/j.yfrne.2019.100800 [DOI] [PubMed] [Google Scholar]
  25. Margolick, J. B., Bream, J. H., Martínez-Maza, O., Lopez, J., Li, X., Phair, J. P., Koletar, S. L., & Jacobson, L. P. (2017). Frailty and circulating markers of inflammation in HIV+ and HIV− men in the Multicenter AIDS Cohort Study. JAIDS Journal of Acquired Immune Deficiency Syndromes, 74(4), 407. doi: 10.1097/QAI.0000000000001261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nanni, M. G., Caruso, R., Mitchell, A. J., Meggiolaro, E., & Grassi, L. (2015). Depression in HIV infected patients: A review. Current Psychiatry Reports, 17(1), 530. doi: 10.1007/s11920-014-0530-4 [DOI] [PubMed] [Google Scholar]
  27. Norcini Pala, A., Steca, P., Bagrodia, R., Helpman, L., Colangeli, V., Viale, P., & Wainberg, M. L. (2016). Subtypes of depressive symptoms and inflammatory biomarkers: An exploratory study on a sample of HIV-positive patients. Brain, Behavior, and Immunity, 56, 105–113. doi: 10.1016/j.bbi.2016.02.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ogletree, A. M., Brennan-Ing, M., Blieszner, R., Karpiak, S. E., & Sands, L. P. (2019). Health burden, support adequacy, and depressive symptoms in older men with HIV. The Gerontologist, 59(6), 1131–1140. doi: 10.1093/geront/gny169 [DOI] [PubMed] [Google Scholar]
  29. Paolillo, E. W., Sun-Suslow, N., Pasipanodya, E. C., Morgan, E. E., Ellis, R. J., Jeste, D. V., & Moore, D. J. (2020). Pre-frailty predicts cognitive decline at 2-year follow-up in persons living with HIV. Journal of Neurovirology, 26(2), 168–180. doi: 10.1007/s13365-019-00814-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Pathai, S., Bajillan, H., Landay, A. L., & High, K. P. (2014). Is HIV a model of accelerated or accentuated aging? The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 69(7), 833–842. doi: 10.1093/gerona/glt168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Pearson, T. A., Mensah, G. A., Alexander, R. W., Anderson, J. L., Cannon, R. O., Criqui, M., Fadl, Y. Y., Fortmann, S. P., Hong, Y., Myers, G. L., Rifai, N., Smith, S. C., Taubert, K., Tracy, R. P., & Vinicor, F. (2003). Markers of inflammation and cardiovascular disease application to clinical and public health practice: A statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation, 107(3), 499–511. doi: 10.1161/01.CIR.0000052939.59093.45 [DOI] [PubMed] [Google Scholar]
  32. Peterson, T. E., & Baker, J. V. (2019). Assessing inflammation and its role in comorbidities among persons living with HIV. Current Opinion in Infectious Diseases, 32(1), 8–15. doi: 10.1097/QCO.0000000000000510 [DOI] [PubMed] [Google Scholar]
  33. Pfitzenmaier, J., Vessella, R., Higano, C. S., Noteboom, J. L., Wallace, D.Jr, & Corey, E. (2003). Elevation of cytokine levels in cachectic patients with prostate carcinoma. Cancer, 97(5), 1211–1216. doi: 10.1002/cncr.11178 [DOI] [PubMed] [Google Scholar]
  34. Piggott, D. A., Bandeen-Roche, K., Mehta, S. H., Brown, T. T., Yang, H., Walston, J. D., Leng, S. X., & Kirk, G. D. (2020). Frailty transitions, inflammation, and mortality among persons aging with HIV infection and injection drug use. AIDS (London, England), 34(8), 1217–1225. doi: 10.1097/QAD.0000000000002527 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Poudel-Tandukar, K., Bertone-Johnson, E. R., Palmer, P. H., & Poudel, K. C. (2014). C-reactive protein and depression in persons with human immunodeficiency virus infection: The Positive Living with HIV (POLH) Study. Brain, Behavior, and Immunity, 42(suppl. C), 89–95. doi: 10.1016/j.bbi.2014.06.004 [DOI] [PubMed] [Google Scholar]
  36. Russell, D., Peplau, L. A., & Ferguson, M. L. (1978). Developing a measure of loneliness. Journal of Personality Assessment, 42(3), 290–294. doi: 10.1207/s15327752jpa4203_11 [DOI] [PubMed] [Google Scholar]
  37. Shiau, S., Bender, A. A., O’Halloran, J. A., Sundermann, E., Aggarwal, J., Althoff, K. N., Baker, J. V., Deeks, S., Fried, L. P., Karpiak, S., Karris, M. Y., Marcotte, T. D., Nachega, J. B., Margolick, J. B., Erlandson, K. M., & Moore, D. J. (2020). The current state of HIV and aging: Findings presented at the 10th International Workshop on HIV and Aging. AIDS Research and Human Retroviruses, 36(12), 973–981. doi: 10.1089/AID.2020.0128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Siegler, E. L., & Brennan-Ing, M. (2017). Adapting systems of care for people aging with HIV. The Journal of the Association of Nurses in AIDS Care, 28(5), 698–707. doi: 10.1016/j.jana.2017.05.006 [DOI] [PubMed] [Google Scholar]
  39. Singh, H. K., Del Carmen, T., Freeman, R., Glesby, M. J., & Siegler, E. L. (2017). From one syndrome to many: Incorporating geriatric consultation into HIV care. Clinical Infectious Diseases, 65(3), 501–506. doi: 10.1093/cid/cix311 [DOI] [PubMed] [Google Scholar]
  40. Smith, K. J., Gavey, S., RIddell, N. E., Kontari, P., & Victor, C. (2020). The association between loneliness, social isolation and inflammation: A systematic review and meta-analysis. Neuroscience and Biobehavioral Reviews, 112, 519–541. doi: 10.1016/j.neubiorev.2020.02.002 [DOI] [PubMed] [Google Scholar]
  41. Smith, C. J., Ryom, L., Weber, R., Morlat, P., Pradier, C., Reiss, P., Kowalska, J. D., de Wit, S., Law, M., el Sadr, W., Kirk, O., Friis-Moller, N., Monforte, A. d., Phillips, A. N., Sabin, C. A., & Lundgren, J. D. (2014). Trends in underlying causes of death in people with HIV from 1999 to 2011 (D:A:D): A multicohort collaboration. Lancet (London, England), 384(9939), 241–248. doi: 10.1016/S0140-6736(14)60604-8 [DOI] [PubMed] [Google Scholar]
  42. Sonsin-Diaz, N., Gottesman, R. F., Fracica, E., Walston, J., Windham, B. G., Knopman, D. S., & Walker, K. A. (2020). Chronic systemic inflammation is associated with symptoms of late-life depression: The ARIC Study. The American Journal of Geriatric Psychiatry, 28(1), 87–98. doi: 10.1016/j.jagp.2019.05.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sowell, R. L., Lowenstein, A., Moneyham, L., Demi, A., Mizuno, Y., & Seals, B. F. (1997). Resources, stigma, and patterns of disclosure in rural women with HIV infection. Public Health Nursing (Boston, Mass.), 14(5), 302–312. doi: 10.1111/j.1525-1446.1997.tb00379.x [DOI] [PubMed] [Google Scholar]
  44. Stewart, J. C., Polanka, B. M., So-Armah, K. A., White, J. R., Gupta, S. K., Kundu, S., Chang, C. H., & Freiberg, M. S. (2020). Associations of total, cognitive/affective, and somatic depressive symptoms and antidepressant use with cardiovascular disease-relevant biomarkers in HIV: Veterans Aging Cohort Study. Psychosomatic Medicine, 82(5), 461–470. doi: 10.1097/PSY.0000000000000808 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Tate, J. P., Justice, A. C., Hughes, M. D., Bonnet, F., Reiss, P., Mocroft, A., Nattermann, J., Lampe, F. C., Bucher, H. C., Sterling, T. R., Crane, H. M., Kitahata, M. M., May, M., & Sterne, J. A. (2013). An internationally generalizable risk index for mortality after one year of antiretroviral therapy. AIDS (London, England), 27(4), 563–572. doi: 10.1097/QAD.0b013e32835b8c7f [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Triant, V. A., Meigs, J. B., & Grinspoon, S. K. (2009). Association of C-reactive protein and HIV infection with acute myocardial infarction. Journal of Acquired Immune Deficiency Syndromes (1999), 51(3), 268–273. doi: 10.1097/QAI.0b013e3181a9992c [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Vingeliene, S., Hiyoshi, A., Lentjes, M., Fall, K., & Montgomery, S. (2019). Longitudinal analysis of loneliness and inflammation at older ages: English Longitudinal Study of Ageing. Psychoneuroendocrinology, 110, 104421. doi: 10.1016/j.psyneuen.2019.104421 [DOI] [PubMed] [Google Scholar]
  48. Walker, J. L., Slavish, D. C., Dolan, M., Dietch, J. R., Wardle-Pinkston, S., Messman, B., Ruggero, C. J., Kohut, M., Borwick, J., Kelly, K., & Taylor, D. J. (2020). Age-dependent associations among insomnia, depression, and inflammation in nurses. Psychology & Health, 0(0), 1–18. doi: 10.1080/08870446.2020.1805450 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Walker, K. A., Walston, J., Gottesman, R. F., Kucharska-Newton, A., Palta, P., & Windham, B. G. (2019). Midlife systemic inflammation is associated with frailty in later life: The ARIC Study. The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 74(3), 343–349. doi: 10.1093/gerona/gly045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wilson, S. J., Woody, A., & Kiecolt-Glaser, J. K. (2019). Inflammation as a biomarker method in lifespan developmental methodology. In B. Knight (Ed.), The oxford research encyclopedia of psychology and aging. New York, NY: Oxford University Press. doi: 10.1093/acrefore/9780190236557.013.355 [DOI] [Google Scholar]
  51. Wu, A. W., Revicki, D. A., Jacobson, D., & Malitz, F. E. (1997). Evidence for reliability, validity and usefulness of the Medical Outcomes Study HIV Health Survey (MOS-HIV). Quality of Life Research, 6(6), 481–493. doi: 10.1023/a:1018451930750 [DOI] [PubMed] [Google Scholar]
  52. Zhang, W., O’Brien, N., Forrest, J. I., Salters, K. A., Patterson, T. L., Montaner, J. S. G., Hogg, R. S., & Lima, V. D. (2012). Validating a shortened depression scale (10 item CES-D) among HIV-positive people in British Columbia, Canada. PLoS ONE, 7(7). doi: 10.1371/journal.pone.0040793 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gbab027_suppl_Supplementary_Material
Click here for additional data file. (41.6KB, docx)

Articles from The Journals of Gerontology Series B: Psychological Sciences and Social Sciences are provided here courtesy of Oxford University Press

RESOURCES