Abstract
Aim
The aim of this study was to assess the relationship between quality of life (QOL), age and other demographic and clinical factors in persons with HIV (PWH).
Background
Patient-centered outcomes such as QOL are increasingly relevant for PWH with their advancing age. QOL by age group and other predictors have not been fully explored.
Design
This study was conducted using a cross-sectional study design.
Methods
Overall, 614 Kaiser Permanente Northern California members with HIV completed baseline questionnaires from May 2013 - May 2015 at enrollment in an alcohol treatment study (n=310 for PWH <50 years of age and n=304 for PWH ≥50 years of age). QOL was assessed by the 12-item Short Form Survey (SF-12), which includes physical and mental QOL domains. Linear regression models were used to evaluate the association of age and other predictors of mental and physical QOL.
Results
PWH <50 years of age reported poorer mental QOL but better physical QOL compared with PWH ≥50 years of age. Other factors associated with worse mental QOL included substance use, depression, and anxiety. Other factors associated with worse physical QOL included injection drug use and depression.
Conclusions
In this primary care-based sample of PWH, we identified age group differences in QOL. Addressing substance use and depression in PWH can also improve mental and physical QOL. These findings can be used by healthcare providers to guide patient-centered care.
Keywords: alcohol, depression, HIV, mental health, quality of life, HIV/AIDS
Persons living with HIV (PLWH) are living longer due to effective combination antiretroviral therapy (ART). As a result, PLWH and their providers have been faced with long-term management of co-morbidities associated with aging, and ongoing care for multiple chronic conditions (High et al., 2012; Valcour, Paul, Neuhaus, & Shikuma, 2010). It has, therefore, become increasingly important to examine the burden of HIV and co-morbid conditions on quality of life (QoL), to inform the delivery of high-quality, integrated, patient-centered care (Balderson et al., 2013). The Institute of Medicine and the National Institutes of Health have recognized QoL as a global priority for tracking health trends and measuring health outcomes (High et al., 2012). In HIV care settings, poor QoL has been associated with increased HIV transmission risk due to unprotected sex, poor medication adherence, and detectable HIV RNA levels (de Oliveira e Silva, Reis, Nogueira, & Gir, 2014). In addition, poor QOL is recognized as an independent predictor of increases in HIV-related morbidity and mortality and associated higher health system costs (High et al., 2012).
It is unclear to what extent HIV clinical factors (CD4+ T cell count, HIV RNA levels), substance use characteristics (hazardous alcohol use, marijuana use, other substance use), and mental health symptoms (depression, anxiety) are independently associated with QoL in PLWH. Furthermore, few studies have separately examined physical QoL, which corresponds to functional ability or the ability to perform activities of daily living, and mental QoL, which corresponds to the degree of patient satisfaction with his or her psychological health and cognitive function (de Oliveira e Silva et al., 2014; Vance, McGuinness, Musgrove, Orel, & Fazeli, 2011). Differences by age are important to consider, because older PLWH experience different physical and mental co-morbidities than younger PLWH (Moore et al., 2014; Vance et al., 2011).
Our aims in this study were to: (a) evaluate the independent effect of age on QoL; and (b) identify other factors associated with QoL (demographics, HIV clinical characteristics, substance use, and mental health symptoms) in PLWH. We hypothesized that QoL would differ by age and that older PLWH would report poorer mental and physical QoL compared to younger PLWH. We further hypothesized that the association between mental and physical QoL and other factors would differ by age. Understanding the association of age and other factors with mental and physical QoL has the potential to enhance patient-centered care for PLWH.
Methods
Study Setting, Design, and Participants
Our cross-sectional analysis was conducted at Kaiser Permanente Northern California (KPNC), a large, private, non-profit, integrated health system of 4 million members, where more than 8,500 members with HIV receive care (Satre et al., 2016). Study subjects were recruited with purposive sampling and included 614 PLWH from KPNC recruited for a randomized clinical trial evaluating an alcohol use behavioral intervention called the Health and Motivation study (Silverberg et al., 2018). At enrollment, participants completed a self-administered questionnaire that included baseline measures of QoL, demographic characteristics, HIV clinical factors, substance use, mental health symptoms, and QoL, the outcome variable for this study, from May 2013 to May 2015. We used baseline assessment data for our study. Participants in the trial met the following inclusion criteria: (a) inclusion in the KPNC HIV registry, which has maintained a list of all PLWH in KPNC since the 1980s; (b) ages 18 years and older receiving HIV services at KPNC San Francisco medical center; and (c) exceeded low-risk alcohol use limits (≥ 3 drinks in a day for women, ≥ 4 drinks in a day for men) in any 24-hour period during the prior year, a time frame consistent with other hazardous drinking studies (Satre, Chi, Eisendrath, & Weisner, 2011). Additional enrollment information and details of inclusion criteria for the study can be found in the baseline paper (Silverberg et al., 2018). The study received IRB approval from KPNC and the University of California, San Francisco, and informed consent was obtained from each participant.
Measures
Quality of Life (QoL)
QoL, the primary outcome variable, was assessed using the 12-item Short Form Survey (SF-12). The SF-12 includes two domains: physical QoL and mental QoL, both scored from 0 to100, with higher scores representing better QoL (Ware, Kosinscki, & Keller, 1996). High validity and reliability have been established with this tool in PLWH, including PLWH who use substances and across ages older than 18 years (Aden et al., 2015; Wu, Revicki, Jacobson, & Malitz, 1997).
Demographic characteristics and HIV clinical factors
Age, gender, race/ethnicity, marital status, education, and employment status were obtained from surveys, and HIV exposure risk groups (i.e., men who have sex with men [MSM] or bisexual, injection drug use, heterosexual, and unknown/other), CD4+ T cell count, and HIV RNA levels (viral load) were obtained from the KPNC HIV registry. The median age at baseline for our sample was 49 years (interquartile range: 43, 57). Age groups were categorized as (a) 50 years and older and (b) younger than 50 years. This age group categorization was consistent with other studies (Monteiro, Canavarro, & Pereira, 2016; Moore et al., 2014).
Substance use characteristics
Substance use was measured using self-report questions about hazardous alcohol use in the previous 30 days, and any tobacco, marijuana, or other substance use, in the previous 30 days. Other substance use (which was grouped together for analysis) included prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy in the previous 30 days. For the analyses, hazardous alcohol use was defined as four or more drinks per day more than four times in the past 30 days for women and five or more drinks per day more than four times in the past 30 days for men.
Mental health symptoms
Depression symptoms were measured using the Patient Health Questionnaire (PHQ-9), a 9-item instrument based on DSM-IV depression criteria (Kroenke, Spitzer, & Williams, 2001). The PHQ-9 is a valid and reliable measure of depression severity and has been found to make criteria-based diagnoses of depressive disorders; the conciseness of the PHQ-9 makes it a useful clinical and research tool (Kroenke et al., 2001). Scores range from 0 to 27, with higher scores indicating more symptoms (Dum, Pickren, Sobell, & Sobell, 2008). For our analyses, PHQ-9 scores were dichotomized, with scores 0–9 indicating no/mild depression and 10–27 indicating moderate/severe depression. Anxiety was measured using the Generalized Anxiety Disorder-7 (GAD-7), a valid and reliable measure of symptoms of anxiety (Shacham, Morgan, Önen, Taniguchi, & Overton, 2012). This test has good internal and test-retest reliability, as well as construct, convergent, and factorial validity to diagnose generalized anxiety disorder (Spitzer, Kroenke, Williams, & Löwe, 2006). Scores range from 0 to 21, with higher scores indicating more symptoms. GAD-7 was dichotomized with scores 0–9 rated as no/mild anxiety and 10–21 as moderate/severe anxiety.
Statistical Analyses
Summary statistics, including percentages, for patient characteristics were reported for the total sample (N = 614) and by age group (< 50 years, ≥ 50 years). Chi-square tests compared categorical variables and Student’s t-tests compared continuous variables between age groups, given that normality assumptions were not violated.
For our first objective, linear regression models were constructed to evaluate the association of age on the two primary outcomes of mental and physical QoL. To assess the influence of confounding factors, models were adjusted sequentially for the following variables: (a) age group; (b) additional demographic characteristics (gender, race/ethnicity with White as the control group; marital status, education, employment, and HIV exposure risk factor with MSM as the control group); (c) HIV clinical characteristics (CD4+ T cell count, HIV RNA levels); and (d) alcohol and substance use (tobacco, other substances) and mental health symptoms (depression, anxiety).
For our second objective, we again used linear regression to identify a final model for other factors associated with mental and physical QoL. Other factors possibly associated with mental and physical QoL included demographic variables, HIV clinical characteristics, alcohol and substance use, and mental health symptoms. The final model was determined by backward selection, which involved iteratively removing the least significant variable that did not meet p < .10 levels. Once the variable was removed, it remained excluded. This process was repeated until all remaining variables were below the p < .10 levels. Final models in strata defined by age were also constructed using backward selection. Beta coefficients with 95% confidence intervals (CIs) are presented, with statistical significance defined as p < .05. All analyses were conducted using SAS Version 9.3 (SAS Inc., Cary, NC).
Results
A total of 614 PLWH met study qualifications and completed study questionnaires to be included in the sample. The mean age of participants was 48.9 years (SD = 11.0). Participants were predominantly male (97.1%), non-Hispanic White (62.7%), and included PLWH 50 years of age and older (n = 304) and PLWH younger than 50 years of age (n = 310). Table 1 shows group differences based on age. PLWH 50 years of age and older were more likely to be employed (p < .001); were less likely to report recent hazardous alcohol use (p < .001), tobacco use (p < .001), and other substance use (p < .05); and reported less anxiety (p < .05) than PLWH younger than 50 years of age (see Table 1). The mean mental QOL score was lower in PLWH younger than 50 years of age compared to PLWH 50 years of age and older (M = 42.7, SD = 11.7 and M = 46.1, SD = 11.6, respectively; p < .001). The mean physical QoL score was higher in PLWH younger than 50 years of age compared to PLWH 50 years of age and older (M = 51.6, SD = 8.4 and M = 47.7, SD 10.1, respectively; p < .001).
Table 1.
Baseline Characteristics of Study Sample Living With HIV, Total and by Age
| Characteristic | Total | < 50 years of age | ≥ 50 years of age | p-valuea |
|---|---|---|---|---|
| N = 614 | n = 310 | n = 304 | ||
| Gender (Male), % | 96.7 | 95.8 | 97.7 | 0.19 |
| Race/ Ethnicity, % | < .001 | |||
| White | 62.7 | 53.2 | 72.4 | |
| Hispanic | 14.0 | 7.7 | 8.9 | |
| Black | 9.1 | 20 | 10.5 | |
| Unknown/Otherb | 14.2 | 19.0 | 8.2 | |
| Married, % | 43.8 | 41.3 | 46.4 | 0.20 |
| Education, % | 0.69 | |||
| < HS/GED | 25.4 | 26.1 | 24.7 | |
| AA/AS | 17.1 | 17.7 | 16.5 | |
| College | 35.2 | 35.8 | 34.5 | |
| ≥ Graduate School | 22.3 | 20.3 | 24.3 | |
| Employed, % | 72.5 | 82.3 | 62.5 | < .001 |
| HIV exposure risk factor, % | 0.26 | |||
| MSM or bisexual | 73.1 | 70.3 | 76.0 | |
| Injection Drug Use | 6.4 | 8.1 | 4.6 | |
| Heterosexual | 4.2 | 4.2 | 4.3 | |
| Unknown/Otherc | 16.3 | 17.4 | 15.1 | |
| CD4+ T cell count (≥ 500 cells/mm3), % | 71.8 | 76.1 | 67.4 | 0.05 |
| HIV RNA levels < 75 copies/mL, % | 93.5 | 91.9 | 95.1 | 0.12 |
| Hazardous alcohol used | 23.1 | 31.9 | 14.1 | < .001 |
| Tobacco usee | 24.4 | 30.7 | 18.1 | < .001 |
| Marijuana usee | 49.8 | 51.0 | 48.7 | 0.57 |
| Other substance usee,f | 27.4 | 31.3 | 23.4 | 0.03 |
| Moderate/Severe Depressiong, % | 16.1 | 18.7 | 13.5 | 0.079 |
| Moderate/Severe Anxietyh, % | 13.8 | 17.1 | 10.5 | 0.018 |
Note. HS/GED = High school/General Education Development; AA/AS = Associates in Arts/Associates in Science; MSM = men who have sex with men.
Based on Pearson Chi-square test comparing < 50 years old to ≥ 50 years old
Includes Asian, American Indian, Other, and Unknown
Includes Transfusion/Other/Unknown
Defined as 4 drinks for women and 5 drinks for men in a day for 4+ days in the past 30 days
Use in the past 30 days
Other substances include prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy
Patient Health Questionnaire (PHQ)-9 score
Generalized Anxiety Disorder (GAD)-7 score
In bivariate models, PLWH younger than 50 years of age had a better mental QoL score compared to PLWH 50 years of age and older (difference = 3.4, 95% CI: 1.6, 5.2; p < .001). PLWH younger than 50 years of age had a poorer physical QoL score compared to PLWH 50 years of age and older (difference = −4.0, 95% CI: −5.4, −2.5; p < .001; see Table 2). Age group remained significant after sequentially including all other factors. In the multivariable model, adjusting for other demographic variables, HIV clinical characteristics, substance and alcohol use, and mental health symptoms, age remained significant for both mental QoL (difference = 2.4, 95% CI: 0.8, 4.0; p = .005) and physical QoL (difference = −3.1, 95% CI: −4.6, −1.6; p < .001; see Table 2).
Table 2.
Associations of Mental and Physical Quality of Life by Age
|
Mental Quality of Life | |||
| Factors in Model | Age differencea | 95% CI | p-value |
| Age onlyb | |||
| Age ≥ 50 years (< 50 years reference) | 3.4 | (1.6, 5.2) | < .001 |
| Age, Demographicsc | |||
| Age ≥ 50 years (< 50 years reference) | 4.1 | (2.2, 6.0) | < .001 |
| Age, Demographic, HIV Clinicald | |||
| Age ≥ 50 years (< 50 years reference) | 4.0 | (2.0, 5.9) | < .001 |
| Age, Demographic, HIV Clinical, Substance use, Mental Health Characteristicse | |||
| Age ≥ 50 years (< 50 years reference) | 2.4 | (0.8, 4.0) | .0047 |
|
Physical Quality of Life | |||
| Factors in Model | Age differencea | 95% CI | p-value |
| Age onlyb | |||
| Age ≥ 50 years (< 50 years reference) | −4.0 | (−5.4, −2.5) | < .001 |
| Age, Demographicsc | |||
| Age ≥ 50 years (< 50 years reference) | −3.0 | (−4.5, −1.5) | < .001 |
| Age, Demographic, HIV Clinicald | |||
| Age ≥ 50 years (< 50 years reference) | −3.0 | (−4.4, −1.4) | < .001 |
| Age, Demographic, HIV Clinical, Substance use, Mental Health Characteristicse | |||
| Age ≥ 50 years (< 50 years reference) | −3.1 | (−4.6, −1.6) | < .001 |
Note. CI = confidence interval.
Differences by age from linear regression models and interpreted as mean quality of life summary scores for persons living with HIV ≥ 50 years of age minus scores for persons living with HIV < 50 years of age.
Bi-variate model including only age group.
Demographic variables included in model: gender, race/ethnicity, marital status, education, employment status, and HIV risk group.
HIV clinical characteristics included in model: CD4+ T cell counts and HIV RNA levels.
Substance use and psychosocial variables included in the model: Alcohol use defined as 4 drinks for women and 5 drinks for men in a day for 4+ days in the past 30 days, Marijuana and Tobacco use in past 30 days, and other substance use (prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy); Mental Health Characteristics include depression as assessed by the Patient Health Questionnaire (PHQ)-9 score and anxiety as assessed by the Generalized Anxiety Disorder (GAD)-7 score
Mental and Physical QoL for Total Sample
We identified the final model with strongest independent factors associated with mental QoL in PLWH. PLWH 50 years of age and older had a 2.6-unit greater mental QoL compared to PLWH younger than 50 years of age (difference = 2.6, 95% CI: 1.0, 4.2; p < .01). Hispanic race/ethnicity compared to White race/ethnicity was significantly associated with better mental QoL (difference = 3.8, 95% CI: 1.5, 6.1; p < .01). Other substance use (difference = −3.6, 95% CI: −5.3, −1.8; p < .001), having moderate to severe depression symptoms (difference = −8.4, 95% CI: −11.0, −5.8; p < .001), and having moderate to severe anxiety symptoms (difference = −10.3, 95% CI: −13.0, −7.5; p < .001) were significantly associated with poorer mental QoL in our sample.
Next, we identified the final model for physical QoL in PLWH (Table 3). Having a college education compared to lacking a high school diploma or equivalent (difference = 2.4, 95% CI: 0.6, 4.2; p < .01) and employment (difference = 5.4, 95% CI: 3.8, 7.0; p < .001) were associated with better physical QoL. PLWH 50 years of age and older compared to ages younger than 50 years (difference = −3.2, 95% CI: −4.6, −1.7; p < .001), HIV exposure risk group of using injection drugs compared to bisexual or MSM (difference = −3.8, 95% CI: −6.7, −0.9; p < .05), and moderate to severe depression symptoms (difference = −4.6, 95% CI: −6.6, −2.7; p < .001) were associated with poorer physical QoL.
Table 3.
Demographics, HIV Clinical Characteristics, Alcohol/Substance Dependence, and Psychosocial Factors Associated With Mental and Physical Quality of Life (n = 614)
| Variablesa | Mental Quality of Life | Physical Quality of Life | ||||
|---|---|---|---|---|---|---|
| Difference | 95% CI | p-value | Difference | 95% CI | p-value | |
| Age ≥ 50 years (< 50 years reference) | 2.6 | (1.0, 4.2) | 0.0013 | −3.2 | (−4.6, −1.7) | <.001 |
| Race/Ethnicity | ||||||
| Hispanic | 3.8 | (1.5, 6.1) | 0.0011 | |||
| Black | 2.1 | (−0.6, 4.8) | 0.10 | |||
| Unknown/Otherb | 0.3 | (−2.0, 2.6) | 0.79 | |||
| White (reference) | 0.0 | |||||
| Married | 1.5 | (−0.004, 3.1) | 0.05 | |||
| Education | ||||||
| < HS/GED (reference) | 0.0 | |||||
| AA/AS | 0.9 | (−1.3, 3.1) | 0.42 | |||
| College | 2.4 | (0.6, 4.2) | 0.0094 | |||
| ≥ Graduate School | 1.7 | (−0.3, 3.7) | 0.10 | |||
| Employed | 1.7 | (−0.03, 3.5) | 0.05 | 5.4 | (3.8, 7.0) | < .001 |
| HIV exposure risk factor, % | ||||||
| MSM or bisexual | 0.0 | |||||
| Injection Drug Use | −3.8 | (−6.7, −0.9) | 0.01 | |||
| Heterosexual | −0.3 | (−3.8, 3.1) | 0.85 | |||
| Unknown/Otherc | −0.6 | (−2.4, 1.3) | 0.56 | |||
| Other substance used | −3.6 | (−5.3, −1.8) | < .001 | 1.5 | (−0.1, 3.1) | 0.06 |
| Moderate to Severe Depressione | −8.4 | (−11.0, −5.8) | < .001 | −4.6 | (−6.6, −2.7) | < .001 |
| Moderate to Severe Anxietyf | −10.3 | (−13.0, −7.5) | < .001 | |||
Note. CI = confidence interval; HS/GED = High school/General Education Development; AA/AS = Associates in Arts/Associates in Science; MSM = men who have sex with men.
Linear regression models using backward selection for final model selection. Variables without results were excluded based on p > .10 criterion.
Includes Asian, American Indian, Other, and Unknown
Includes Transfusion/Other/Unknown
Use in the past 30 days and includes prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy
None to mild depression is the reference group, from Patient Health Questionnaire (PHQ)-9 score
None to mild anxiety is the reference group, from Generalized Anxiety Disorder (GAD)-7 score
Factors Associated with Mental Quality of Life, by Age Group
In PLWH younger than 50 years (n = 310), Hispanic race/ethnicity compared to White race/ethnicity (difference = 4.2, 95% CI: 1.3, 7.2; p < .01), being married (difference = 2.5, 95% CI: 0.3, 4.8; p < .05), and being employed (difference = 3.6, 95% CI: 0.7, 6.5; p < .05) were associated with better mental QoL (Table 4). Moderate to severe depression symptoms (difference = −9.2, 95% CI: −12.7, −5.7; p < .001) and moderate to severe anxiety symptoms (difference = −8.0, 95% CI: −11.6, −4.3; p < .001) were associated with poorer mental QoL for PLWH younger than 50 years of age (Table 4).
Table 4.
Demographics, HIV Clinical Characteristics, Alcohol/Substance Dependence, and Psychosocial Factors Associated With Mental Quality of Life by Age Groups
| Variablesa | < 50 years of age | ≥ 50 years of age | ||||
|---|---|---|---|---|---|---|
| Difference | 95% CI | p-value | Difference | 95% CI | p-value | |
| Race/Ethnicity | ||||||
| Hispanic | 4.2 | (1.3, 7.2) | < .01 | 3.9 | (0.1, 7.8) | 0.043 |
| Black | 3.1 | (−1.1, 7.2) | 0.15 | 2.5 | (−1.0, 6.1) | 0.16 |
| Unknown/Otherb | 1.2 | (−1.7, 4.1) | 0.403 | −1.2 | (−5.1, 2.8) | 0.0036 |
| White (reference) | 0.0 | |||||
| Married | 2.5 | (0.3, 4.8) | 0.027 | |||
| Education | ||||||
| < HS/GED (reference) | 0.0 | |||||
| AA/AS | 2.2 | (−1.1, 5.6) | 0.1857 | |||
| College | −0.4 | (−3.2, 2.4) | 0.78 | |||
| ≥ Graduate School | −2.8 | (−6.1, 0.5) | 0.092 | |||
| Employed | 3.6 | (0.7, 6.5) | 0.015 | |||
| Tobacco usec | −2.3 | (−4.7, 0.04) | 0.054 | |||
| Other substance used | −4.0 | (−6.6, −1.3) | 0.0036 | |||
| Moderate to Severe Depressione | −9.2 | (−12.7, −5.7) | < .001 | −8.5 | (−11.6, −3.9) | < .001 |
| Moderate to Severe Anxietyf | −8.0 | (−11.6, −4.3) | < .001 | −13.7 | (−18.0, −9.4) | <. 001 |
Note. CI = confidence interval; HS/GED = High school/General Education Development; AA/AS = Associates in Arts/Associates in Science.
Linear regression models using backward selection for final model selection. Variables without results were excluded based on p > .10 criterion.
Includes Asian, American Indian, Other, and Unknown
Use in the past 30 days
Use in the past 30 days and other substances include prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy
None to mild depression is the reference group, from Patient Health Questionnaire (PHQ)-9 score
None to mild anxiety is the reference group, from Generalized Anxiety Disorder (GAD)-7 score
For PLWH 50 years of age and older (n = 304), Hispanic race/ethnicity (difference = 3.9, 95% CI: 0.1, 7.8; p < .05) was associated with better mental QoL compared to White participants. Other substance use in the prior 30 days (difference = −4.0, 95% CI: −6.6, −1.3; p < .01), moderate to severe depression symptoms (difference = −8.5, 95% CI: −11.6, −3.9; p < .001), and moderate to severe anxiety symptoms (difference = −13.7, 95% CI: −18.0, −9.4; p < .001) were associated with poorer mental QoL.
Factors Associated with Physical Quality of Life by Age Group
In PLWH younger than 50 years of age (n = 310), college (difference = 2.5, 95% CI: 0.1, 4.8; p < .05) or graduate school education (difference = 2.9, 95% CI: 0.2, 5.6; p < .05) compared to less than high school or equivalent education, employment (difference = 3.8, 95% CI: 1.4, 6.2; p < .01), and other substance use (difference = 2.0, 95% CI: 0.04, 3.9; p < .05) were associated with better physical QoL (Table 5). Black race/ethnicity compared to White (difference = −4.5, 95% CI: −8.1, −1.0; p < .05), HIV exposure risk factor of injection drug use compared to MSM or bisexual exposure (difference = −4.4, 95% CI: −7.7, −1.1; p < .05), and having moderate to severe depression (difference = −3.5, 95% CI: −5.8, −1.2; p < .01) were associated with poorer physical QoL.
Table 5.
Demographics, HIV Clinical Characteristics, Alcohol/Substance Dependence, and Psychosocial Factors Associated With Physical Quality of Life by Age Groups
| Variablesa | < 50 years of age | ≥ 50 years of age | ||||
|---|---|---|---|---|---|---|
| Difference | 95% CI | p-value | Difference | 95% CI | p-value | |
| Race/Ethnicity | ||||||
| Hispanic | −2.0 | (−4.4, −0.5) | 0.11 | |||
| Black | −4.5 | (−8.1, −1.0) | 0.013 | |||
| Unknown/Otherb | 0.9 | (−2.0, 3.8) | 0.54 | |||
| White (reference) | 0.0 | |||||
| Education | ||||||
| < HS/GED (reference) | ||||||
| AA/AS | 2.5 | (−0.2, 5.3) | 0.07 | |||
| College | 2.5 | (0.1, 4.8) | 0.04 | |||
| ≥ Graduate School | 2.9 | (0.2, 5.6) | 0.033 | |||
| Employed | 3.8 | (1.4, 6.2) | < .01 | 6.3 | (4.1, 8.5) | < .001 |
| HIV risk exposure group | ||||||
| Unknown/Otherc | −1.5 | (−4.6, 1.5) | 0.32 | |||
| Injection Drug Use | −4.4 | (−7.7, −1.1) | 0.01 | |||
| Heterosexual | 1.2 | (−3.4, 5.8) | 0.62 | |||
| MSM or bisexual (reference) | 0.0 | |||||
| Hazardous alcohol used | −3.6 | (−6.6, −0.6) | 0.019 | |||
| Other substance usee | 2.0 | (0.04, 3.9) | 0.046 | |||
| Moderate to Severe Depressionf | −3.5 | (−5.8, −1.2) | 0.0035 | −5.8 | (−8.9, −2.7) | < .001 |
Note. CI = confidence interval, HS/GED = High school/General Education Development; AA/AS = Associates in Arts/Associates in Science; MSM = men who have sex with men.
Linear regression models using backward selection for final model selection. Variables without results were excluded based on p > 0.10 criterion.
Includes Asian, American Indian, Other, and Unknown
Includes Transfusion/Other/Unknown
Defined as 4 drinks for women and 5 drinks for men in a day for 4+ days in the past 30 days
Use in the past 30 days and other substances include prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy
None to mild depression is the reference group, from Patient Health Questionnaire (PHQ)-9 score
In PLWH 50 years of age and older (n = 304), employment (difference = 6.3, 95% CI: 4.1, 8.5; p < .001) compared to unemployment was associated with better physical QoL. Hazardous alcohol use (difference = −3.6, 95% CI: −6.6, −0.6; p < .05) and moderate to severe depression (difference = −5.8, 95% CI: −8.9, −2.7; p < .001) were associated with poorer physical QoL.
Discussion
In a large, primary-care based sample of PLWH from an integrated health system, we found that PLWH younger than 50 years of age had significantly poorer mental QoL and better physical QoL than older PLWH, even after controlling for demographic, HIV clinical characteristics, substance use, depression, and anxiety factors. In addition to age, key factors associated with poorer mental QoL included other substance use, moderate to severe depression, and moderate to severe anxiety. Finally, in addition to age, key factors associated with poorer physical QoL included injection drug use and moderate to severe depression. Of note, HIV clinical characteristics, including CD4+ T cell levels and HIV RNA levels, were not associated with QoL.
Our finding of poorer mental QoL but better physical QoL for younger PLWH was consistent with some, but not all, prior research. Previous studies suggested that PLWH report poorer QoL based on age (Chen & Barbour, 2017). Although one study in China found that older PLWH reported worse mental and physical QoL and more depression compared to younger PLWH (Liu et al., 2014), our results were consistent with findings from other studies indicating that older PLWH reported poorer physical QoL compared to their younger counterparts (Liu et al., 2014; Monteiro et al., 2016). Older PLWH reported more physical symptoms compared to younger PLWH, such as more co-morbidities and declines in activities of daily living, which related to poor physical QoL (Monteiro et al., 2016). As people age, more symptoms associated with co-morbid conditions negatively affect physical health, which can relate to poor physical QoL (Liu et al., 2014; Monteiro et al., 2016).
Other demographic variables, including race/ethnicity, education, and employment were significantly associated with mental or physical QoL in our study. Contrary to the literature, Hispanic race/ethnicity was associated with better mental QoL compared to White PLWH in our sample. One explanation could be that in KPNC, given equal access to care, there are fewer racial/ethnic differences in response to ART and clinical outcomes (Silverberg, Leyden, Quesenberry, & Horberg, 2009). One group showed that higher education was associated with better physical QoL (Degroote, Vogelaers, & Vandijck, 2014), a trend also seen in our sample. Employment was also significantly related to better mental QoL, which was consistent with another study (Degroote et al., 2014). Employment constitutes a large part of daily life, even as people age, and can provide structure, role identity, meaning, and a social support network. It is important to highlight the roles of education, and employment in poor physical QoL, as declines in physical QoL have been associated with adverse physical symptoms (Degroote et al., 2014).
Contrary to another study (Armon & Lichtenstein, 2012), HIV clinical characteristics in our sample were not associated with QoL. One study found that lower CD4+ T cell count and more advanced stage of HIV disease were associated with poorer QOL (Degroote et al., 2014). Despite low HIV RNA levels (≤ 50 copies/mL) and fairly high CD4+ T cell count (≥ 350/mm3), older PLWH reported poor physical QoL due to increased problems of mobility, daily function, and self-care, but reported improved mental QoL (McGowan et al., 2014). It is possible that we were not able to also observe this association because our sample was well treated and overall had high CD4+ T cell counts; in addition, a high proportion had undetectable HIV RNA levels.
HIV exposure risk group (injection drug use compared to MSM or bisexual) was associated with poorer physical QoL and other substance use was associated with poorer mental QoL, a finding consistent with the literature (Millar, Starks, Gurung, & Parsons, 2017). In our sample, other substance use (including prescription drug use other than as prescribed, tranquilizers, stimulants, cocaine, painkillers, heroin, hallucinogens, and ecstasy) was associated with poorer mental QoL. The relationship of recent substance use and history of injection drug use with QoL could be attributed to indirect causes, including exacerbation of mental health symptoms. Millar et al. (2017) found no relationship between substance use and QoL after adjusting for mental health issues (depression and anxiety), but we found that a history of injection drug use was strongly associated with poorer physical QoL and recent substance use was strongly associated with poorer mental QoL.
In our study, depression was a significant factor associated with both mental and physical QoL; anxiety was a significant factor associated with mental QoL. Depression can cause physical problems, including lack of sleep, poor appetite, and fatigue, and negatively affect physical QoL in PLWH (Degroote et al., 2014). Furthermore, depression has a direct effect on adherence to medications, which can directly impact a person’s physical health (Degroote et al., 2014). Anxiety was strongly associated with poor mental QoL in our study, a similar finding in PLWH in Uganda (Ezeamama et al., 2016). The study by Zimpel and Fleck (2014) emphasized the need to assess and treat anxiety and depression in PLWH, because mental health symptoms directly affect HIV risk behavior and QoL. Overall, our findings supported findings that having moderate to severe depression and anxiety contributed to poor QoL in PLWH (Mays, Jones, Delany-Brumsey, Coles, & Cochran, 2017). Our results reinforce the importance of treating mental health symptoms to enhance overall QoL in PLWH.
Limitations
Several study limitations are noted. First, the sample was primarily male, and included patients who were insured and had well-controlled HIV disease. Thus, findings are not generalizable to women living with HIV or populations with publicly-funded health care. Nevertheless, the sample was highly generalizable to PLWH with insurance, as KPNC provides care to approximately 30% of all insured Californians (Gordon, 2011). In addition, due to the cross-sectional nature of the data, we cannot draw conclusions about the directionality of relationships between many clinical factors and QoL. Finally, we did not analyze other key factors that may impact QoL, such as common medical co-morbidities, and psychosocial variables such as social support.
Conclusion
In this large primary care-based sample, we found that QoL in PLWH with well-controlled HIV disease varied by age, with older patients reporting poorer physical but better mental QoL compared to younger patients. Thus, it is important to develop effective treatment planning and age-appropriate interventions to improve QoL across the lifespan. Other factors, including substance use and mental health symptoms, were also strongly associated with QoL, highlighting the potential impact of effective screening and treatment of these conditions in PLWH. Together, results suggest that measurement of broader functional outcomes such as QoL in clinical practice improves understanding of patient perspectives, which health care providers can use as a foundation for directing comprehensive, patient-centered care.
Key Considerations.
Clinicians should consider tailoring interventions to improve physical QoL in older PLWH, because older PLWH report lower physical QoL scores compared to their younger counterparts.
Addressing mental health and substance use in PLWH is important, as both are associated with mental and physical QOL.
The nursing profession can contribute to improved QoL for PLWH across the lifespan by addressing mental and physical QoL by age groups.
It is important to treat mental health symptoms, specifically depression and anxiety, in all ages, because they affect QoL and health in PLWH.
Acknowledgement Statement
This work was supported by the Kaiser Permanente Delivery Sciences Fellowship at the Division of Research and from the National Institute on Alcohol Abuse and Alcoholism (grant number U01 AA021997). The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of Kaiser Permanente or any other funders. We are grateful to the study participants and staff for their dedication to HIV research. We would also like to acknowledge Julie Schmittdiel, PhD, Richard Grant, MD and the Delivery Science fellows.
Footnotes
Declaration of interest statement
The authors report no real or perceived vested interests that relate to this article that could be construed as a conflict of interest.
References
- Aden B, Dunning A, Nosyk B, Wittenberg E, Bray JW, & Schackman BR (2015). Impact of illicit drug use on health-related quality of life in opioid-dependent patients undergoing HIV treatment. Journal of Acquired Immune Deficiency Syndromes, 70(3), 304–310. doi: 10.1097/qai.0000000000000768 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Armon C, & Lichtenstein K (2012). The associations among coping, nadir CD4+ T-cell count, and non-HIV-related variables with health-related quality of life among an ambulatory HIV-positive patient population. Quality of Life Research, 21, 993–1003. doi: 10.1007/s11136-011-0017-2 [DOI] [PubMed] [Google Scholar]
- Balderson BH, Grothaus L, Harrison RG, McCoy K, Mahoney C, & Catz S (2013). Chronic illness burden and quality of life in an aging HIV population. AIDS Care, 25(4), 451–458. doi: 10.1080/09540121.2012.712669 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chen W-T, & Barbour R (2017). Life priorities in the HIV-positive Asians: A text-mining analysis in young vs. old generation. AIDS Care, 29(4), 507–510. doi: 10.1080/09540121.2016.1221029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Degroote S, Vogelaers D, & Vandijck DM (2014). What determines health-related quality of life among people living with HIV: An updated review of the literature. Archives of Public Health, 72(1), 1–10. doi: 10.1186/2049-3258-72-40 [DOI] [PMC free article] [PubMed] [Google Scholar]
- de Oliveira e Silva AC, Reis RK, Nogueira JA, & Gir E (2014). Quality of life, clinical characteristics and treatment adherence of people living with HIV/AIDS. Revista Latino-Americana de Enfermagem, 22(6), 994–1000. doi: 10.1590/0104-1169.3534.2508 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dum M, Pickren J, Sobell LC, & Sobell MB (2008). Comparing the BDI-II and the PHQ-9 with outpatient substance abusers. Addictive Behaviors, 33(2), 381–387. doi: 10.1016/j.addbeh.2007.09.017 [DOI] [PubMed] [Google Scholar]
- Ezeamama AE, Woolfork MN, Guwatudde D, Bagenda D, Manabe YC, Fawzi WW, & Fawzi MCS (2016). Depressive and anxiety symptoms predict sustained quality of life deficits in HIV-positive Ugandan adults despite antiretroviral therapy: A prospective cohort study. Medicine, 95(9), e2525. doi: 10.1097/MD.0000000000002525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gordon NP (2011). How does the adult Kaiser Permanente membership in Northern California compare with the larger community? Retrieved from https://divisionofresearch.kaiserpermanente.org/projects/memberhealthsurvey/SiteCollectionDocuments/chis_non_kp_2011.pdf
- High KP, Brennan-Ing M, Clifford DB, Cohen MH, Currier J, Deeks SG, … Volberding P (2012). HIV and aging: State of knowledge and areas of critical need for research: A report to the NIH Office of AIDS Research by the HIV and Aging Working Group. Journal of Acquired Immune Deficiency Syndromes, 60(Suppl. 1), S1–S18. doi: 10.1097/QAI.0b013e31825a3668 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kroenke K, Spitzer RL, & Williams JB (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606–613. doi: 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu H, He X, Levy JA, Xu Y, Zang C, & Lin X (2014). Psychological impacts among older and younger people living with HIV/AIDS in Nanning, China. Journal of Aging Research, 2014, 6, 1–6. doi: 10.1155/2014/576592 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mays VM, Jones AL, Delany-Brumsey A, Coles C, & Cochran SD (2017). Perceived discrimination in health care and mental health/substance abuse treatment among Blacks, Latinos, and Whites. Medical Care, 55(2), 173–181. doi: 10.1097/MLR.0000000000000638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McGowan J, Sherr L, Rodger A, Fisher M, Miners A, Johnson M, . . . Lampe F (2014). Effects of age on symptom burden, mental health and quality of life amongst people with HIV in the UK. Journal of the International AIDS Society, 17(4 Suppl. 3), 19511. doi: 10.7448/IAS.17.4.19511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Millar BM, Starks TJ, Gurung S, & Parsons JT (2017). The impact of comorbidities, depression, and substance use problems on quality of life among older adults living with HIV. AIDS and Behavior, 21(6), 1684–1690. doi: 10.1007/s10461-016-1613-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Monteiro F, Canavarro MC, & Pereira M (2016). Factors associated with quality of life in middle-aged and older patients living with HIV. AIDS Care, 28(Supp1.), 92–98. doi: 10.1080/09540121.2016.1146209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Moore RC, Fazeli PL, Jeste DV, Moore DJ, Grant I, & Woods SP (2014). Successful cognitive aging and health-related quality of life in younger and older adults infected with HIV. AIDS and Behavior, 18(6), 1186–1197. doi: 10.1007/s10461-014-0743-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Satre DD, Altschuler A, Parthasarathy S, Silverberg M, Volberding P, & Campbell C (2016). Affordable Care Act implementation in a California health care system leads to growth in HIV-positive patient enrollment and changes in patient characteristics. Journal of Acquired Immune Deficiency Syndromes, 73, e76–e82. doi: 10.1097/QAI.0000000000001188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Satre DD, Chi FW, Eisendrath S, & Weisner C (2011). Subdiagnostic alcohol use by depressed men and women seeking outpatient psychiatric services: Consumption patterns and motivation to reduce drinking. Alcoholism: Clinical and Experimental Research, 35(4), 695–702. doi: 10.1111/j.1530-0277.2010.01387.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shacham E, Morgan J, Önen NF, Taniguchi T, & Overton ET (2012). Screening anxiety in the HIV clinic. AIDS and Behavior, 16(8), 2407–2413. doi: 10.1007/s10461-012-0238-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverberg MJ, Leyden W, Quesenberry CPJ, & Horberg MA (2009). Race/ethnicity and risk of AIDS and death among HIV-infected patients with access to care. Journal of General Internal Medicine, 24(9), 1065–1072. doi: 10.1007/s11606-009-1049-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverberg MJ, Leyden WA, Leibowitz A, Hare BC, Jang HJ, Sterling S, . . . Satre DD (2018). Factors associated with hazardous alcohol use and motivation to reduce drinking among HIV primary care patients: Baseline findings from the Health & Motivation Study. Addictive Behaviors, 84, 110–117. doi: 10.1016/j.addbeh.2018.03.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Spitzer RL, Kroenke K, Williams JW, & Löwe B (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092–1097. doi: 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
- Valcour V, Paul R, Neuhaus J, & Shikuma C (2010). The effects of age and HIV on neuropsychological performance. Journal of the International Neuropsychological Society, 17(1), 190–195. doi: 10.1017/S1355617710001438 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vance DE, McGuinness T, Musgrove K, Orel NA, & Fazeli PL (2011). Successful aging and the epidemiology of HIV. Clinical Interventions in Aging, 6, 181–192. doi: 10.2147/CIA.S14726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ware JEJ, Kosinscki M, & Keller SD (1996). A 12-item Short-Form Health Survey: Construction of scales and preliminary tests of reliability and validity. Medical Care, 34(3), 220–233. doi: 10.1097/00005650-199603000-00003 [DOI] [PubMed] [Google Scholar]
- Wu AW, Revicki DA, Jacobson D, & Malitz FE (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]
- Zimpel RR, & Fleck MP (2014). Depression as a major impact on the quality of life of HIV-positive Brazilians. Psychology, Health & Medicine, 19(1), 47–58. doi: 10.1080/13548506.2013.772302 [DOI] [PubMed] [Google Scholar]