Symptom distress, or the degree of discomfort from symptoms experienced by the patient, remains one of the most challenging complications of living with HIV, despite effective HIV antiretroviral therapy. Because HIV has become a chronic disease, HIV-associated symptoms have evolved from being related to illness progression and medication side effects to those associated with inflammation, accelerated aging, and age-related comorbidities (Webel et al., 2019; Willig et al., 2015). Although symptom distress in people living with HIV (PLWH) varies by gender and race (Lee, Dziadkowiec, & Meek, 2014; Webel et al., 2015), it significantly affects daily function and quality of life for all PLWH.
PLWH often simultaneously experience multiple distressing symptoms, yet symptom management strategies tend only to target one symptom at a time (Schnall et al., 2017). In particular, pharmacologic interventions individually treat pain, insomnia, and depression with varying degrees of success, but their use by PLWH is limited by pill burden and side effects (Gimeno-Gracia, Crusells-Canales, Armesto-Gomez, Compaired-Turlan, & Rabanaque-Hernandez, 2016; Siefried et al., 2018). By contrast, nonpharmacologic strategies such as physical activity and diet may provide symptom relief with a lower side effect profile than most prescription drugs.
Indeed, physical activity and a healthy dietary intake can reduce symptoms for individuals with a wide variety of chronic diseases. In individuals without HIV infection, physical activity has been shown to reduce chronic pain (Geneen et al., 2017), depression (Harvey et al., 2018), and fatigue (Salmon, Hewlett, Walsh, Kirwan, & Cramp, 2017) and to improve cognitive function (Barha, Davis, Falck, Nagamatsu, & Liu-Ambrose, 2017), whereas a diet comprising high fiber, moderate fat, and low in carbohydrates has been associated with reduced pain, depression, and fatigue (Jacka et al., 2017). Similar responses to physical activity have been observed in PLWH in smaller studies (Fazeli et al., 2015; O'Brien, Tynan, Nixon, & Glazier, 2016), but little is known about the relationship between dietary intake and symptoms in PLWH. Furthermore, PLWH in the United States consume diets low in fiber and high in fat and carbohydrates (Webel et al., 2017; Willig, Wright, & Galvin, 2018), which may exacerbate symptoms. Definitive evidence of the impact of physical activity and dietary intake on symptoms in PLWH is needed to provide specific physical activity and dietary recommendations to improve their health and well-being.
To address these gaps in the literature, we are currently conducting the “Impact of Physical Activity and Diet on Symptom Experience in People Living with HIV (PROSPER-HIV)” study. The purpose of the PROSPER-HIV study is to understand not just whether but which specific aspects of physical activity and dietary intake affect the symptom experience in PLWH. The PROSPER-HIV study will be one of the first investigations to longitudinally assess physical activity and dietary intake in PLWH using gold standard assessment techniques. Specifically, we aim to
- characterize longitudinal, objectively measured physical activity and diet patterns in PLWH;
- determine which aspects of physical activity patterns (e.g., intensity, frequency, duration) and diet quality are associated with decreased symptom burden and intensity in PLWH and whether this relationship is moderated by age and gender; and
- explore the potential mediating effects of anthropomorphic and physical fitness (waist–hip ratio, body mass index, handgrip strength, Short Performance Physical Battery) variables on the relationships between physical activity, diet patterns, and symptom burden in PLWH.
The PROSPER-HIV study is a 4-year, prospective, observational study of 850 participants from the Centers for AIDS Research (CFAR) Network of Integrated Clinical Systems (CNICS). Enrolled participants will have to complete assessments of physical activity, dietary intake, and physical fitness once a year for 3 years (Figure 1). As part of their routine clinical care, all participants will also have to complete a standard CNICS patient-reported outcome assessment (which includes the HIV Symptom Index) and clinical assessment procedures (Kitahata et al., 2008).
Centers for AIDS Research Network of Integrated Clinical Systems
Any single HIV clinical site typically will have insufficient numbers of patients to provide adequate physical activity and dietary intake data to understand associations with reported symptoms. As such, the PROSPER-HIV study is being conducted as a CNICS (R24AI067039) research protocol. The CNICS has incorporated high-quality clinical data from PLWH in care for the past 25 years. The CNICS was established to better define relationships between patient and treatment factors and long-term outcomes in PLWH (Kitahata et al., 2008). The CNICS incorporates new PLWH as they enter care, ensuring relevance to modern HIV care. It has eight HIV clinical care sites with more than 32,000 PLWH enrolled since 1995 with more than 35 million clinical observations and more than 200,000 person-years of follow-up. The CNICS is diverse in gender, race/ethnicity geography, and age with 18% women, 44% White, 38% Black, 12% Hispanic race/ethnicity, and more than 42% are between 50 years and older. In addition to working with a large, diverse, generalizable cohort, nesting the PROSPER-HIV study within the CNICS cohort allows us to leverage two other important resources:
- Robust yet brief patient-reported outcomes are routinely collected on all participants. These data are collected at least annually in self-administered surveys on encrypted touch screen computers and include symptom burden, HIV medication adherence, smoking and substance use, quality of life, interpersonal violence, depression and anxiety, and sexual risk behaviors (Fredericksen et al., 2012). The CNICS has also recently included measures of housing stability and social support.
- Established standards for terminology, format, data verification, and quality assurance for a variety of laboratory, medication, demographic, and health care utilization data that will be used as descriptive variables or covariates in the PROSPER-HIV study. The use of these resources has resulted in dozens of important studies that have improved HIV clinical care (e.g., Hartzler et al., 2017; Nance et al., 2017); the CNICS is an ideal platform to incorporate measures of physical activity and dietary intake into the existing data collection framework.
Before proposing the PROSPER-HIV study to the National Institutes of Health, a concept sheet had to be approved by the CNICS Research Coordinating Committee. This carefully delineated process is best described on the CNICS website (https://www.uab.edu/cnics), but, in brief, the process involved the following: (a) conducting preliminary work to demonstrate the significance of the unmitigated symptoms in PLWH and data supporting the hypothesis that physical activity and a healthy dietary intake are likely to reduce symptom burden; (b) checking study feasibility, which included ensuring that the PROSPER-HIV study was unique from other CNICS projects and contained the data elements needed to complete the study; (c) working with a CNICS collaborator to develop a concept proposal application; (d) participating in a review call with the CNICS Research Coordinating Committee to discuss the study purpose, overview, procedures, and impact on the individual participating sites (goals and content of this call were tailored to the type of study proposed); and (e) receiving notification of the Research Coordinating Committee's decision. For the PROSPER-HIV study, steps b through e of the review process took approximately 4 months, with most of that time devoted to writing the concept proposal.
The PROSPER-HIV study is being conducted at four academic medical centers (Case Western Reserve University, University of Alabama at Birmingham, University of Washington at Seattle, and Fenway Health Institute) that provide HIV care for diverse PLWH who are broadly representative of the U.S. People living with HIV population (Table 1). All sites are longstanding members of the CNICS and have dedicated space and technological infrastructure for research visits. All PROSPER-HIV procedures take place in existing CNICS research space.
The PROSPER-HIV study enrollment goal is 850 PLWH. This sample size was determined using a power analysis based on our preliminary work describing symptom rates and SDs of self-reported physical activity (Webel, Willig et al., 2019). We determined that with 850 PLWH, we would be able to detect the odds of symptom reduction being 93% higher with an increase of 30 minutes of moderate to vigorous physical activity (MVPA) with 85% power. To be eligible, participants must (a) be an active CNICS participant (i.e., they must have a CNICS consent and have completed the current patient-reported outcome assessment); (b) be at least 18 years of age; (c) be prescribed antiretroviral therapy as part of CNICS care; and (d) have an HIV viral load less than 200 copies/ml at the time of enrollment (>90% of PLWH in CNICS care are undetectable; Nance et al., 2017). Participants will be excluded if they (a) did not complete the HIV Symptom Index in a recent assessment; (b) are pregnant, breastfeeding, or planning a pregnancy during the study period; (c) do not have telephone or Internet access to complete 24-hour diet recalls; or (d) plan to move out of the area in the next 36 months. In addition, we plan to have a minimum of 50% (n = 425) of African American, Hispanic, and Asian participants and 30% of female (n ≥ 255) participants, resulting in representative findings that will allow us to examine race/ethnicity and gender as potential moderators on our outcomes.
As is usual, participants will complete the standard CNICS patient-reported outcome assessment battery at routine patient clinic visits. After the participants complete the assessment, the research assistant will explain PROSPER-HIV study procedures and potential risks and benefits, and ask the individual whether she or he would like to enroll. On agreement, the research assistant will obtain informed consent and then help the participant complete the PROSPER-HIV assessment, which includes (a) waist and hip circumference measures, (b) handgrip strength using the Jamar Hand Grip Dynamometer, (c) Short Physical Performance Battery (SPPB), and (d) Food Security Questionnaire. The research assistant will explain and schedule three telephone dietary interviews with the University Hospitals Cleveland Medical Center's Clinical Research Unit Bionutrition Core during the following 30 days. The research assistant will then initialize the participant's accelerometer, instruct the participant in the proper use of the ActiGraph accelerometer, and provide instructions for returning it approximately 10 days later (either in person or by mail). In the first months of study recruitment, this assessment has been taking approximately 20–25 minutes to complete and has typically been performed at the end of the participant's regular scheduled clinic visit. These procedures are repeated once every 12 months for 3 years. The study was approved by the University Hospitals Cleveland Medical Center's Institutional Review Board, which is the Institutional Review Board of record for the PROSPER-HIV study. Institutional Review Boards at the other three sites have sanctioned the study protocol. The PROSPER-HIV study is registered at ClinicalTrials.gov #NCT03790501.
The primary outcome in the PROSPER-HIV study is symptom distress, which is assessed with the 20-item HIV Symptom Index (Justice et al., 2001) that evaluates the presence and intensity of 20 common symptoms reported by PLWH (e.g., pain, anxiety, fatigue) during the past 4 weeks. All symptoms are measured on a one to five ordinal scale, where 1 is I do not have the symptom and 5 is I have this symptom and it bothers me a lot. The reliability of the HIV Symptom Index in the CNICS cohort is 0.92 (Webel, Willig, et al., 2019). The PROSPER-HIV endpoints are (a) total symptom count (sum all symptoms that are reported as having the symptom and it bothers the participant at least a little) and (b) a total count of symptoms that bother the participant a lot (high symptom distress).
Physical activity is measured with the ActiGraph accelerometer (ActiGraph, LLC, Fort Walton Beach, FL; Strath et al., 2013). The intraclass correlation for actigraphy ranges from 0.94 to 0.99 (Sasaki, John, & Freedson, 2011). Participants wear the monitor for 7–10 consecutive days on the nondominant hip. A valid wear cycle has data recorded for a minimum of 10 hours per day for at least 4 days (Haskell et al., 2012). Nonwear time is defined as 0 counts per minute for at least 60 minutes. Participants not meeting wear-time standards are asked to re-wear the ActiGraph. Data are sampled at 30 Hz, using 60-second epochs and the low-frequency filter (Migueles et al., 2017). Activity of at least 2,690 counts per minute and at least 10 minutes is defined as exercise. The primary exercise endpoints are (a) time spent in MVPA and (b) sedentary time. These endpoints are set with the Sasaki et al. (2011) adult cut points.
Food security status is determined using the two-item food security questionnaire, which classifies respondents as (a) food secure, (b) food secure with hunger (low food security), or (c) food insecure with hunger (very low food security; Muhammad et al., 2019; Young, Jeganathan, Houtzager, Di Guilmi, & Purnomo, 2009). Diet intake is measured using a standardized triple-pass 24-hour recall (partial correlation = 0.27) obtained by a registered dietitian trained in the process. Recalls capture dietary intake for two weekdays and one weekend, within a 30-day window (Moshfegh et al., 2008; Satija, Yu, Willett, & Hu, 2015). The registered dietitian calls each participant to conduct a nutrition interview. During the interview, the participant recalls what and how much was consumed in the previous 24 hour. Responses are simultaneously entered into the Nutrition Data System for Research Nutritional Analysis Software (University of Minnesota). The Healthy Eating Index-2015, a composite measure of diet, is the PROSPER-HIV endpoint (Freedman, Guenther, Krebs-Smith, & Kott, 2008; Guenther, Reedy, Krebs-Smith, & Reeve, 2008). We will also assess the intake of dietary fiber, protein, and simple carbohydrates.
Physical fitness is assessed with two measures: handgrip strength and the Short Performance Physical Battery. Handgrip strength is commonly used and correlates with overall muscle strength. It is measured with the Jamar handheld dynamometer. Each participant completes two trials with the self-reported dominant hand (Webel, Jenkins, et al., 2019). The maximum strength of dominant hand (all attempts) is recorded as the endpoint.
The SPPB is a brief measure of physical performance or functional status that includes a timed walk, repeated chair stands, and several balance tests. The SPPB is a well-regarded, valid (test–rest reliability = 0.87), objective assessment of physical function, particularly lower extremity function, and is associated with short-term mortality, disability, hospitalizations, and nursing home admission (Guralnik et al., 1994). Each measure on the SPPB is assigned a score from 0 to 4, with 0 indicating inability to complete the test, yielding a summary score ranging from 0 (frail) to 12 (not frail; Guralnik et al., 1994). The summery score will be analyzed as our endpoint (Crane et al., 2019; Willig et al., 2018).
Outcomes for the PROSPER-HIV study are to (a) characterize longitudinal physical activity and dietary intake patterns in PLWH, (b) determine which aspects of physical activity patterns and diet quality are associated with symptom burden, and (c) explore mediating effects of other factors that may affect the relationships of activity and diet with symptom burden. To characterize physical activity (MVPA) and dietary intake (HEI-2015), linear models maximized with the method of generalized estimating equations (GEEs) will be used. In addition, individuals will be grouped based on similar trajectories for activity and nutrition using a semiparametric group-based trajectory estimation (Jones, Nagin, & Roeder, 2001). These trajectory groups will be investigated as “exposure” groups in the analyses.
To determine which physical activity and diet patterns are associated with symptom burden, we will implement multiple analyses to determine the relationship between activity, nutrition, and HIV symptoms. We will use GEE, as described above, with autoregressive working correlation and robust standard errors. HIV symptoms in two forms (symptom count and serious symptom count) will be examined with separate Poisson regressions within GEE, resulting in estimated risk ratios to assess associations of physical activity, nutrition, time, and other covariates to HIV symptoms.
Finally, to assess mediating factors on the relationship between physical activity, dietary intake, and symptoms, a counterfactual approach to mediation analysis will be used, allowing for the estimation of direct and indirect effects (VanderWeele, 2009; 2015). Given the assumption of normality for each of the potential mediators (waist–hip ratio, body mass index, handgrip strength, and Short Performance Physical Battery score), we anticipate using the SAS macro published by Valeri and VanderWeele (2015) or the newly created PROC CAUSALMED in SAS and examine potential mediator–mediator interactions.
As in any longitudinal study, it is anticipated that some participants may die, withdraw consent, or be lost to follow up before the end of the study, which will result in missing data. To deal with missing data, investigation of the mechanisms for missing data will be conducted. Apart from changing the analysis from GEE to mixed models, sensitivity analysis will include multiple imputation, as described by Schafer (1997). Once missing values have been imputed, each multiply imputed data set can be analyzed. Final parameter estimates and their standard errors will be calculated using Rubin's (2004) formula for combining results from multiply imputed data sets. We will report the final study results with and without using the multiple imputation strategy and examine and describe any discrepancies found.
The PROSPER-HIV study will prospectively examine symptom burden over time and have 4 years of research procedures. Data collection started in January 2019 at two sites and was scaled up to all four sites in April 2019. Enrollment and baseline procedures are expected to continue through March 2020, and all study procedures should be completed by March 2023.
At the conclusion of the PROSPER-HIV study, we will have (a) characterized longitudinal physical activity and dietary patterns in PLWH; (b) determined which aspects of physical activity and diet quality are associated with decreased symptom burden; and (c) described the effect of potential mediating factors on the relationships between physical activity, dietary patterns, and symptom burden in PLWH. Furthermore, although we anticipate a limited number of participants who identify as transgender, we plan to disaggregate data by gender to better understand these relationships and the potential influence of gender-affirming hormones. This new evidence will fill important gaps in the literature, enabling future research on the biological underpinnings of symptoms in PLWH. These insights will set the stage for feasible, targeted, nonpharmacologic interventions for use by clinicians and PLWH beyond the CNICS to reduce symptom burden and improve quality of life.
Nurses care for PLWH who have high symptom burdens every day but have few tools to help reduce those symptoms. PROSPER-HIV data have tremendous potential to affect care for these patients. We will provide data on symptom patterns that can be used to help clinicians develop a clinic-based physical activity and diet intervention to improve symptom management, physical function, and quality of life for PLWH.
The authors report no real or perceived vested interests related to this article that could be construed as a conflict of interest.
This study was funded by the National Institutes of Nursing Research #R01 NR018391 (PIs: A. R. Webel and A. L. Willig) and by the National Institute of Allergy and Infectious Diseases #R24 AI067039 (PI: M. S. Saag). Registration number: NCT03790501.
Barha C. K., Davis J. C., Falck R. S., Nagamatsu L. S., Liu-Ambrose T. (2017). Sex differences in exercise efficacy to improve cognition: A systematic review and meta-analysis of randomized controlled trials in older humans. Frontiers in Neuroendocrinology, 46, 71–85. doi:10.1016/j.yfrne.2017.04.002
Crane H. M., Miller M. E., Pierce J., Willig A. L., Case M. L., Wilkin A. M., High K. P. (2019). Physical functioning among patients aging with human immunodeficiency virus (HIV
) versus HIV
uninfected: Feasibility of using the short physical performance battery in clinical care of people living with HIV
aged 50 or older. Open Forum Infectious Diseases, 6(3), ofz038. doi:10.1093/ofid/ofz038
Fazeli P. L., Marquine M. J., Dufour C., Henry B. L., Montoya J., Gouaux B., Moore D. J. (2015). Physical activity
is associated with better neurocognitive and everyday functioning among older adults with HIV
disease. AIDS and Behavior, 19, 1470–1477. doi:10.1007/s10461-015-1024-z
Fredericksen R. J., Crane P. K., Tufano J., Ralston J., Schmidt S., Brown T., Crane H. M. (2012). Integrating a web-based, patient-administered assessment into primary care for HIV
-infected adults. Journal of AIDS and HIV
Research, 4(2), 47–55. doi:10.5897/jahr11.046
Freedman L. S., Guenther P. M., Krebs-Smith S. M., Kott P. S. (2008). A population's mean Healthy Eating Index-2005 scores are best estimated by the score of the population ratio when one 24-hour recall is available. Journal of Nutrition, 138, 1725–1729. doi:10.1093/jn/138.9.1725
Geneen L. J., Moore R. A., Clarke C., Martin D., Colvin L. A., Smith B. H. (2017). Physical activity
and exercise for chronic pain in adults: An overview of cochrane reviews. The Cochrane Database of Systematic Reveiws, 4, CD011279. doi:10.1002/14651858.CD011279.pub3
Gimeno-Gracia M., Crusells-Canales M. J., Armesto-Gomez F. J., Compaired-Turlan V., Rabanaque-Hernandez M. J. (2016). Polypharmacy in older adults with human immunodeficiency virus infection compared with the general population. Clinical Interventions in Aging, 11, 1149–1157. doi:10.2147/CIA.S108072.
Guenther P. M., Reedy J., Krebs-Smith S. M., Reeve B. B. (2008). Evaluation of the Healthy Eating Index-2005. Journal of the American Dietetic Association, 108, 1854–1864. doi:10.1016/j.jada.2008.08.011.
Guralnik J. M., Simonsick E. M., Ferrucci L., Glynn R. J., Berkman L. F., Blazer D. G., Wallace R. B. (1994). A short physical performance battery assessing lower extremity function: Association with self-reported disability and prediction of mortality and nursing home admission. Journal of Gerontology, 49, M85–M94. doi:10.1093/geronj/49.2.m85
Hartzler B., Dombrowski J. C., Crane H. M., Eron J. J., Geng E. H., Mathews C. W., Donovan D. M. (2017). Prevalence and predictors of substance use disorders among HIV
care enrollees in the United States. AIDS and Behavior, 21, 1138–1148. doi:10.1007/s10461-016-1584-6
Harvey S. B., Overland S., Hatch S. L., Wessely S., Mykletun A., Hotopf M. (2018). Exercise and the prevention of depression: Results of the HUNT cohort study. The American Journal of Psychiatry, 175, 28–36. doi:10.1176/appi.ajp.2017.16111223
Haskell W. L., Troiano R. P., Hammond J. A., Phillips M. J., Strader L. C., Marquez D. X., Ramos E. (2012). Physical activity
and physical fitness: Standardizing assessment with the PhenX Toolkit. American Journal of Preventive Medicine, 42, 486–492. doi:10.1016/j.amepre.2011.11.017
Jacka F. N., O'Neil A., Opie R., Itsiopoulos C., Cotton S., Mohebbi M., Berk M. (2017). A randomised controlled trial of dietary improvement for adults with major depression (the “SMILES” trial). BMC Medicine, 15, 23. doi:10.1186/s12916-017-0791-y
Jones B. L., Nagin D. S., Roeder K. (2001). A SAS procedure based on mixture models for estimating developmental trajectories. Sociological Methods and Research, 29, 374–393. doi:10.1177/0049124101029003005
Justice A. C., Holmes W., Gifford A. L., Rabeneck L., Zackin R., Sinclair G., Wu A. W. (2001). Development and validation of a self-completed HIV
symptom index. Journal of Clinical Epidemiology, 54, S77–S90. doi:10.1016/s0895-4356(01)00449-8
Kitahata M. M., Rodriguez B., Haubrich R., Boswell S., Matthews W. C., Lederman M. M., Saag M. (2008). Cohort profile: The centers for AIDS research network of integrated clinical systems. International Journal of Epidemiology, 37, 948–955. doi:10.1093/ije/dym231
Lee K. A., Dziadkowiec O., Meek P. (2014). A systems science approach to fatigue management in research and health care. Nursing Outlook, 62, 313–321. doi:10.1016/j.outlook.2014.07.002
Migueles J. H., Cadenas-Sanchez C., Ekelund U., Delisle Nystrom C., Mora-Gonzalez J., Lof M., Ortega F. B. (2017). Accelerometer data collection and processing criteria to assess physical activity
and other outcomes: A systematic review and practical considerations. Sports Medicine, 47(9), 1821–1845. doi:10.1007/s40279-017-0716-0
Moshfegh A. J., Rhodes D. G., Baer D. J., Murayi T., Clemens J. C., Rumpler W. V., Cleveland L. E. (2008). The US department of agriculture automated multiple-pass method reduces bias in the collection of energy intakes. The American Journal of Clinical Nutrition, 88(2), 324–332. doi:10.1093/ajcn/88.2.324
Muhammad J. N., Fernandez J. R., Clay O. J., Saag M. S., Overton E. T., Willig A. L. (2019). Associations of food insecurity and psychosocial measures with diet
quality in adults aging with HIV
. AIDS Care, 31(5), 554–562. doi:10.1080/09540121.2018.1554239
Nance R. M., Delaney J. C., Golin C. E., Wechsberg W. M., Cunningham C., Altice F., Crane H. M. (2017). Co-calibration of two self-reported measures of adherence to antiretroviral therapy. AIDS Care, 29, 464–468. doi:10.1080/09540121.2016.1263721
O'Brien K. K., Tynan A. M., Nixon S. A., Glazier R. H. (2016). Effectiveness of aerobic exercise for adults living with HIV
: Systematic review and meta-analysis using the cochrane collaboration protocol. BMC Infectious Diseases, 16, 182. doi:10.1186/s12879-016-1478-2
Rubin D. B. (2004). Multiple imputation for nonresponse in surveys (Vol. 81). Hoboken, NJ: John Wiley & Sons.
Salmon V. E., Hewlett S., Walsh N. E., Kirwan J. R., Cramp F. (2017). Physical activity
interventions for fatigue in rheumatoid arthritis: A systematic review. Physical Therapy Reviews, 22, 12–22. doi:10.1186/s12891-019-2558-4
Sasaki J. E., John D., Freedson P. S. (2011). Validation and comparison of ActiGraph activity monitors. Journal of Science and Medicine in Sport, 14, 411–416. doi:10.1016/j.jsams.2011.04.003
Satija A., Yu E., Willett W. C., Hu F. B. (2015). Understanding nutritional epidemiology and its role in policy. Advances in Nutrition, 6(1), 5–18. doi:10.3945/an.114.007492
Schafer J. L. (1997). Analysis of incomplete multivariate data. London, UK: Chapman & Hall.
Schnall R., Liu J., Cho H., Hirshfield S., Siegel K., Olender S. (2017). A health-related quality-of-life measure for use in patients with HIV
: A validation study. AIDS Patient Care and STDS, 31, 43–48. doi:10.1089/apc.2016.0252
Siefried K. J., Mao L., Cysique L. A., Rule J., Giles M. L., Smith D. E., Carr A. (2018). Concomitant medication polypharmacy, interactions and imperfect adherence are common in Australian adults on suppressive antiretroviral therapy. AIDS, 32, 35–48. doi:10.1097/QAD.0000000000001685
Strath S. J., Kaminsky L. A., Ainsworth B. E., Ekelund U., Freedson P. S., Gary R. A., Swartz A. M. (2013). Guide to the assessment of physical activity
: Clinical and research applications: A scientific statement from the American Heart Association. Circulation, 128, 2259–2279. doi:10.1161/01.cir.0000435708.67487.da
Valeri L., VanderWeele T. J. (2015). SAS macro for causal mediation analysis with survival data. Epidemiology, 26(2), e23–e24. doi:10.1097/EDE.0000000000000253
VanderWeele T. J. (2009). Conceptual issues concerning mediation, interventions and composition. Statistics and Its Interface, 2(4), 457–468. doi:10.4310/sii.2009.v2.n4.a7
VanderWeele T. J. (2015). Explanation in causal inference: Methods for mediation and interaction. New York, NY: Oxford University Press.
Webel A., Jenkins T., Longenecker C., Vest M., Davey C. H., Currie J., Josephson R. (2019). Relationship of HIV
status and fatigue, cardiorespiratory fitness, myokines, and phyiscal activity. The Journal of the Association of Nurses in AIDS Care, 30, 392–404. doi:10.1097/JNC.0000000000000022
Webel A. R., Sattar A., Funderburg N. T., Kinley B., Longenecker C. T., Labbato D., McComsey G. A. (2017). Alcohol and dietary factors associated with gut integrity and inflammation in HIV
-infected adults. HIV
Medicine, 18, 402–411. doi:10.1111/hiv
Webel A., Wantland D., Rose C. D., Kemppainen J., Holzemer W. L., Chen W. T., Portillo C. (2015). A cross-sectional relationship between social capital, self-compassion and perceived HIV symptoms
. Journal of Pain and Symptom Management, 50(1), 59–68. doi:10.1016/j.jpainsymman.2014.12.013
Webel A., Willig A., Liu W., Sattar A., Boswell S., Crane H. M., Rodriguez B. (2019). Physical activity
intensity is associated with symptom distress in the CNICS cohort. AIDS and Behavior, 23(3), 627–635. doi:10.1007/s10461-018-2319-7
Willig A. L., Westfall A. O., Overton E. T., Mugavero M. J., Burkholder G. A., Kim D., Willig J. H. (2015). Obesity is associated with race/sex disparities in diabetes and hypertension prevalence, but not cardiovascular disease, among HIV
-infected adults. AIDS Research and Human Retroviruses, 31(9), 898–904. doi:10.1089/AID.2015.0062
Willig A., Wright L., Galvin T. A. (2018). Practice paper of the academy of nutrition and dietetics: Nutrition intervention and human immunodeficiency virus infection. Journal of the Academy of Nutrition and Dietetics, 118(3), 486–498. doi:10.1089/AID.2015.0062
Young J., Jeganathan S., Houtzager L., Di Guilmi A., Purnomo J. (2009). A valid two-item food security questionnaire for screening HIV
-1 infected patients in a clinical setting. Public Health Nutrition, 12(11), 2129–2132. doi:10.1017/S1368980009005795