Abstract
In order to achieve the programmatic goals established in the National HIV/AIDS Strategy, virologic suppression remains the most important outcome within the HIV care continuum for individuals receiving antiretroviral therapy (ART). Therefore, clinicians have dedicated substantial resources to improve adherence and clinic retention for individuals on ART; however, these efforts should be focused first on those most at risk of morbidity and mortality related to AIDS. Our study aimed to characterize the factors that are associated with AIDS-defining illnesses (ADIs) amongst people living with HIV (PLHIV) who are poorly adherent or retained in care in order to identify those at highest risk for poor clinical outcomes. We recruited 99 adult PLHIV with a history of poor adherence to ART, poor clinic attendance, or unsuppressed viral load from the Infectious Disease Program (IDP) of the Grady Health System in Atlanta, GA between January and May of 2011 to participate in a survey investigating the acceptability of a financial incentive for improving adherence. Clinical outcomes including the number of ADI episodes in the last five years, viral loads, and CD4 counts were abstracted from medical records. Associations between survey items and number of ADIs were performed using Chi-square analysis. In our study, 36.4% of participants had ≥1 ADI in the last five years. The most common ADIs were Pneumocystis jirovecii pneumonia, recurrent bacterial pneumonia, and esophageal candidiasis. Age <42.5 years (OR 2.52, 95%CI 1.08-5.86), Male gender (OR 3.51, 95%CI 1.08-11.34), CD4 nadir <200 cells/µL (OR 11.92, 95%CI 1.51-94.15), unemployment (OR 3.54, 95%CI 1.20-10.40), and travel time to clinic <30 minutes (OR 2.80, 95%CI 1.20-6.52) were all significantly associated with a history of ≥1 ADI in the last five years. Awareness of factors associated with ADIs may help clinicians identify which poorly adherent PLHIV are at highest risk for HIV-related morbidity.
Keywords: AIDS-defining illness, opportunistic infection, HIV adherence
Introduction
Despite the availability of antiretroviral therapy (ART) for HIV infection, achieving virologic suppression remains a significant challenge for many people living with HIV (PLHIV). Virologic suppression depends on two components: clinic attendance and ART adherence. Several studies have identified barriers to ART adherence including alcohol consumption, employment outside the home(Chesney et al., 2000), medication availability, forgetfulness, and being busy with other things(Amico et al., 2007; Chesney et al., 2000). On the other hand, barriers to clinic retention have been more difficult to define because they tend to be more diverse and regionally dependent. However, retention in care is just as vital as medication adherence, as missed appointments have been shown to predict HIV progression(Berg et al., 2005; Park et al., 2007; Rastegar, Fingerhood, & Jasinski, 2003) and increased mortality(Giordano et al., 2007; Mugavero et al., 2009). Despite many reasons for poor adherence to ART or clinical care, little is known about which of these patients will have poor clinical outcomes. In this study, we aimed to identify risk factors that are associated with AIDS-defining illnesses (ADIs) in poorly adherent PLHIV.
Methods
Patients
Ninety-nine adult (≥18 years old) PLHIV were recruited from the Infectious Disease Program (IDP) of the Grady Health System in Atlanta, GA between January and May of 2011. The IDP is an outpatient clinic that served approximately 5,200 PLHIV (46.72% uninsured, 25.78% Medicaid, and 22.84% Medicare) in 2011 who had a diagnosis of AIDS. Participants were identified by chart review and referred from clinic providers in two phases. The first phase enrolled patients with a clinic visit within the last 1.5 years but not within the prior six months, and who had a history of poor ART adherence and/or a most recent HIV-1 viral load (VL) > 200 copies/ml. In the second phase, recent clinic attendance was not an exclusion criteria, but the most recent VL had to be > 200 copies/mL after ≥6 months of ART. All participants could not be actively abusing illicit substances since the survey was designed to determine the feasibility of a financial incentive.
Survey methods
Data was obtained from a survey investigating the acceptability of a financial incentive for improving adherence among poorly adherent PLHIV (Clinical Trials # NCT01305590). The survey was conducted using a semi-structured interview by trained interviewers in a private setting and included questions regarding medication adherence, attitudes toward proposed interventions for improving adherence, demographics, transportation, and employment. Participants provided written informed consent and were compensated for their time. All study procedures and instruments were approved by the Institutional Review Board at Emory University.
Definitions and statistical analysis
Study variables included age, gender, actively taking prescribed ART at the time of survey (as documented by providers in medical chart), employment, travel time to clinic, and mode of transportation. Participants were divided into two groups based on the mean age (42.5 years). The most recent CD4 counts (≤9) were abstracted up to five years prior to enrollment and CD4 nadir was defined as the lowest abstracted CD4 count. Travel time to clinic was stratified into < and ≥ 30 minutes. Participants were grouped into two categories based on mode of transportation: public transport and walking; and car, motorcycle, and other. Car and motorcycle were combined because these participants were thought to have greater control over their transportation, compared with those who took public transportation or walk to clinic.
Medical charts were abstracted for each unique ADI event as defined by the CDC(Schneider et al., 2008) from the preceding five years. ADIs were validated by reviewing diagnostic studies and cultures if available, but diagnoses were otherwise assumed if listed in the medical record. Participants were stratified based on the presence or absence of ADIs.
Chi-square analyses were performed to determine the associations between study variables and history of ADI using bivariate analyses. All data analyses were performed using SPSS version 22.
Results
Characteristics of study population
The study participants were predominantly male and African-American; mean age was 42.5 years (Table 1). The median CD4 count at enrollment was 154 cells/µL, the median CD4 nadir was 79 cells/µL, 56% were taking ART according to provider documentation, and 30.7% had virologic suppression at study enrollment. Participants had HIV for a median of 11.1 years. Sixty-nine percent were unemployed. 42.4% traveled <30 minutes to clinic, and 53.5% walked or took public transport to clinic.
Table 1. Characteristics of poorly adherent PLHIV, Infectious Diseases Program, Atlanta, GA, 2011, N=99.
| Overall | Phase 1 n=52 |
Phase 2 n=47 |
|
|---|---|---|---|
| Mean age at enrollment (years) | 42.5 | 43.4 | 41.5 |
| Gender: | |||
| Male | 76% | 65.4% | 88.6% |
| Female | 24% | 34.6% | 11.4% |
| Race: | |||
| African American | 85.9% | 84.6% | 87.2% |
| Other | 14.1% | 15.4% | 12.8% |
| Median CD4 at enrollment, cells/μL (Q1,Q3)* | 154 (94,277) | 171.5 (90.3,334.8) | 146 (97,247) |
| Median CD4 nadir in last 5 years, cells/μL(Ql,Q3)* | 79 (19,154) | 98.5 (19.3,192.8) | 56 (17,105) |
| Taking antiretroviral therapy at enrollment** | 56.6% | 50% | 63.8% |
| Virologic suppression (VL>200 copies/mL) at enrollment | 30.7% | 38.5% | 23.4% |
| Median years since diagnosis (Q1,Q3)* | 11.1 (6.1,15.8) | 10.9 (6.7,15.8) | 12.5 (4.3,15.8) |
| Employment: | |||
| Employed | 31% | 26% | 31.9% |
| Unemployed | 69% | 74% | 68.1% |
| Travel time to clinic: | |||
| <30 minutes | 42.4% | 32.7% | 53.2% |
| ≥30 minutes | 57.6% | 67.3% | 46.8% |
| Mode of transport to clinic: | |||
| Walking or public transport | 53.5% | 65.4% | 40.4% |
| Car, motorcycle, other | 46.5% | 34.6% | 59.6% |
| Number of AIDS-defining illnesses in last 5 years | |||
| 0 | 63.6% | 69.2% | 57.4% |
| ≥1 | 36.4% | 30.8% | 42.6% |
Q1, Q3: First and third quartile
as documented by HIV provider in medical chart
AIDS-defining illnesses
In the last five years, 25.3% had 1-2 ADIs, and 11.1% had 3 or more. The most common ADIs were Pneumocystis jirovecii pneumonia (24 events), recurrent bacterial pneumonia (23 events), and esophageal candidiasis (14 events) (Table 2). Age <42.5 years (OR 2.52, 95%CI=1.08-5.86)), Male gender (OR 3.51, 95%CI=1.08-11.34)), CD4 nadir <200 cells/µL (OR 11.92, 95%CI=1.51-94.15), unemployment (OR 3.54, 95%CI=1.20-10.40), and travel time to clinic <30 minutes (OR 2.8, 95%CI=1.20-6.52) were all significantly associated with a history of ≥1 ADI in the last five years (Table 3). No association was found between current ART usage and mode of transport with the presence of ADIs.
Table 2. Frequency of AIDS-defining illnesses among study participants.
| AIDS-defining illness | # of Unique Events |
|---|---|
| Pneumocystis jirovecii pneumonia | 24 |
| Pneumonia, recurrent | 23 |
| Candidiasis of esophagus | 14 |
| Herpes simplex, chronic ulcers (>1 month's duration) or bronchitis, pneumonitis, or esophagitis | 11 |
| Encephalopathy, HIV-related | 4 |
| Cytomegalovirus disease | 3 |
| HIV wasting syndrome | 3 |
| Ciyptococcosis, extrapulmonary | 2 |
| Cryptosporidiosis | 1 |
| Kaposi sarcoma | 1 |
| Mycobacterium tuberculosis | 1 |
| Progressive multifocal encephalopathy | 1 |
| Toxoplasmosis of brain | 1 |
Table 3. Factors associated with a history of AIDS-defining illnesses (ADIs) in the last live years, N=99.
| 0 ADI Participants, n (%) |
≥ 1 ADI Participants, n (%) |
Odds Ratio (95% CI) |
|
|---|---|---|---|
| Age at enrollment: | |||
| <42.5 years | 26 (53.1%) | 23 (46.9%) | 2.52*** |
| ≥42.5 years | 37 (74.0%) | 13 (26.0%) | (1.08-5.86) |
| Gender*: | |||
| Male | 42 (57.5%) | 31 (42.5%) | 3.51*** |
| Female | 19 (82.6%) | 4 (17.4%) | (1.08-11.34) |
| CD4 nadir | |||
| <200 cells/μL | 47 (57.3%) | 35 (42.7%) | 11.92*** |
| ≥200 cells/μL | 16 (94.1%) | 1 (5.9%) | (1.51-94.15) |
| Taking antiretroviral therapy at enrollment: | |||
| No | 29 (67.4%) | 14 (32.6%) | 0.75 |
| Yes | 34 (60.7%) | 22 (39.3%) | (0.32-1.72) |
| Employment**: | |||
| Unemployed | 39 (56.5%) | 30 (43.5%) | 354*** |
| Employed | 23 (82.1%) | 5 (17.9%) | (1.20-10.40) |
| Travel time to clinic | |||
| <30 minutes | 21 (50.0%) | 21 (50.0%) | 2.80*** |
| ≥30 minutes | 42 (73.7%) | 15 (26.3%) | (1.20-6.52) |
| Mode of transport to clinic: | |||
| Public transport or walking | 37 (69.8%) | 16 (30.2%) | 0.56 |
| Car, motorcycle other | 26 (56.5%) | 20 (43.5%) | (0.25-1.29) |
3 transgender participants excluded from analysis
2 participants who omitted question excluded from analysis
Chi-square p<0.05
Discussion
Overall, the participants were mostly male, African-American, middle aged, unemployed, and had been diagnosed with HIV for more than a decade. The participants traveled variable amounts of time to clinic, which is at least partly explained by the clinic's wide catchment area, which includes 20 counties in the eligible Atlanta metropolitan area. The low mean CD4 count and nadir, low percentage who achieved virologic suppression, and low percentage taking ART, in conjunction with the long duration of time since diagnosis confirm that this is a group with poor adherence.
In our study, age <42.5 years was significantly associated with a history of ≥1 ADI in the last five years. Prior studies have suggested that older age (>50 years old) reduced the risk of non-adherence(Ghidei et al., 2013; Weintrob et al., 2008), and that younger patients had a higher risk of poor retention in care(Hall et al., 2013). By extension, older individuals would likely have lower risk of ADIs despite studies showing poorer immune reconstitution and immunosenescence(Pirrone et al., 2013).
We also found that male gender was significantly associated with a history of ADI. Previous large US studies have shown that the incidence of opportunistic infections in women has been higher than men (Buchacz et al., 2010). However, our study participants were predominantly African American, while the aforementioned study included mostly non-Hispanic Whites. In our study, the average CD4 count at enrollment was 198 cells/µL for males and 270 cells/µL for females (p=0.151). Despite the non-significant p-value, the mean enrollment CD4 count in males was below 200 cells/µL, the clinical cutoff for AIDS, which may also explain the increased risk for ADI among males in our study.
Moreover, a CD4 nadir <200 cells/µL was associated with a history of ≥1 ADIs and had the highest odds ratio (11.92) of any of the factors, confirming the known biological relationship between low CD4 count and ADIs. Of note, few participants had a CD4 nadir ≥200 cells/µL, reflecting the enrollment criteria of the IDP clinic, which requires a history of a diagnosis of AIDS.
Unemployment was also found to be linked to a history of ADI in the last five years. While this association may seem intuitive, different studies have shown mixed effects of employment on HIV outcomes. One study identified employment as a predictor of CD4 count improvement after ART initiation(Simoni, Yard, & Huh, 2013). Conversely, another showed an association between employment outside the home and poorer adherence to ART, though its effect on ADI or CD4 count was not described(Chesney et al., 2000).
Transportation has also been described as a barrier to receiving HIV care by patients(Kempf et al., 2010) and providers in the Southeast US(Reif, Golin, & Smith, 2005). In our study, a travel time to clinic <30 minutes but not the mode of transportation was associated with a history of ADI. This apparent paradox may be due to the clinic's urban setting and may be confounded by secondary factors associated with urban living such as drug abuse(Ohl et al., 2010), homelessness(Milloy, Marshall, Montaner, & Wood, 2012), and poverty(Shacham, Lian, Onen, Donovan, & Overton, 2013).
One of the limitations of the study is the cross-sectional design, which prevents the determination of causality. In addition, there may be an element of prevalence-incidence bias since patients who may have died of ADIs could not be included in the study. The study also only includes PLHIV in Atlanta, and the results may not be generalizable to all regions. Finally, the study excluded participants with active drug abuse, so conclusions cannot be generalized to this group.
In conclusion, our data show that there is still substantial HIV-related morbidity among PLHIV in the Southeast US, and many challenges remain in achieving widespread virologic suppression. Our study has identified certain factors—young age, male gender, CD4 nadir <200 cells/µL, unemployment, and travel time <30 minutes—that are associated with ADI in poorly adherent PLHIV. Awareness of these factors may help direct further studies to decrease the risk for ADI in this region where a significant burden of HIV remains.
Acknowledgments
The authors would like to acknowledge the efforts by Elyse LaFond and Chris Foster in recruiting patients, chart abstraction, and administering the study. We would also like to thank the patients of the Infectious Disease Program of the Grady Health System for their participation in this study. Their involvement has enabled us to enhance the care of other people living with HIV.
Funding/Support: This work was supported by the National Institutes of Health under Grant P30AG034532 and was facilitated by the Center for AIDS Research at Emory University under Grant P30AI050409.
Footnotes
Disclosures: The authors have no financial disclosures or conflicts of interest.
References
- Amico KR, Konkle-Parker DJ, Cornman DH, Barta WD, Ferrer R, Norton WE, Fisher WA. Reasons for ART non-adherence in the Deep South: adherence needs of a sample of HIV-positive patients in Mississippi. AIDS Care. 2007;19(10):1210–1218. doi: 10.1080/09540120701426516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Berg MB, Safren SA, Mimiaga MJ, Grasso C, Boswell S, Mayer KH. Nonadherence to medical appointments is associated with increased plasma HIV RNA and decreased CD4 cell counts in a community-based HIV primary care clinic. AIDS Care. 2005;17(7):902–907. doi: 10.1080/09540120500101658. [DOI] [PubMed] [Google Scholar]
- Buchacz K, Baker RK, Palella FJ, Jr, Chmiel JS, Lichtenstein KA, Novak RM, Brooks JT. AIDS-defining opportunistic illnesses in US patients, 1994-2007: a cohort study. AIDS. 2010;24(10):1549–1559. doi: 10.1097/QAD.0b013e32833a3967. [DOI] [PubMed] [Google Scholar]
- Chesney MA, Ickovics JR, Chambers DB, Gifford AL, Neidig J, Zwickl B, Wu AW. Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: the AACTG adherence instruments. Patient Care Committee & Adherence Working Group of the Outcomes Committee of the Adult AIDS Clinical Trials Group (AACTG) AIDS Care. 2000;12(3):255–266. doi: 10.1080/09540120050042891. [DOI] [PubMed] [Google Scholar]
- Ghidei L, Simone MJ, Salow MJ, Zimmerman KM, Paquin AM, Skarf LM, Rudolph JL. Aging, Antiretrovirals, and Adherence: A Meta Analysis of Adherence among Older HIV-Infected Individuals. Drugs Aging. 2013 doi: 10.1007/s40266-013-0107-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Giordano TP, Gifford AL, White AC, Jr, Suarez-Almazor ME, Rabeneck L, Hartman C, Morgan RO. Retention in care: a challenge to survival with HIV infection. Clin Infect Dis. 2007;44(11):1493–1499. doi: 10.1086/516778. [DOI] [PubMed] [Google Scholar]
- Hall HI, Frazier EL, Rhodes P, Holtgrave DR, Furlow-Parmley C, Tang T, Skarbinski J. Differences in human immunodeficiency virus care and treatment among subpopulations in the United States. JAMA Intern Med. 2013;173(14):1337–1344. doi: 10.1001/jamainternmed.2013.6841. [DOI] [PubMed] [Google Scholar]
- Kempf MC, McLeod J, Boehme AK, Walcott MW, Wright L, Seal P, Moneyham L. A qualitative study of the barriers and facilitators to retention-in-care among HIV-positive women in the rural southeastern United States: implications for targeted interventions. AIDS Patient Care STDS. 2010;24(8):515–520. doi: 10.1089/apc.2010.0065. [DOI] [PubMed] [Google Scholar]
- Milloy MJ, Marshall BD, Montaner J, Wood E. Housing status and the health of people living with HIV/AIDS. Curr HIV/AIDS Rep. 2012;9(4):364–374. doi: 10.1007/s11904-012-0137-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mugavero MJ, Lin HY, Willig JH, Westfall AO, Ulett KB, Routman JS, Allison JJ. Missed visits and mortality among patients establishing initial outpatient HIV treatment. Clin Infect Dis. 2009;48(2):248–256. doi: 10.1086/595705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ohl M, Tate J, Duggal M, Skanderson M, Scotch M, Kaboli P, Justice A. Rural residence is associated with delayed care entry and increased mortality among veterans with human immunodeficiency virus infection. Med Care. 2010;48(12):1064–1070. doi: 10.1097/MLR.0b013e3181ef60c2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Park WB, Choe PG, Kim SH, Jo JH, Bang JH, Kim HB, Choe KW. One-year adherence to clinic visits after highly active antiretroviral therapy: a predictor of clinical progress in HIV patients. J Intern Med. 2007;261(3):268–275. doi: 10.1111/j.1365-2796.2006.01762.x. [DOI] [PubMed] [Google Scholar]
- Pirrone V, Libon DJ, Sell C, Lerner CA, Nonnemacher MR, Wigdahl B. Impact of age on markers of HIV-1 disease. Future Virol. 2013;8(1):81–101. doi: 10.2217/fvl.12.127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rastegar DA, Fingerhood MI, Jasinski DR. Highly active antiretroviral therapy outcomes in a primary care clinic. AIDS Care. 2003;15(2):231–237. doi: 10.1080/0954012031000068371. [DOI] [PubMed] [Google Scholar]
- Reif S, Golin CE, Smith SR. Barriers to accessing HIV/AIDS care in North Carolina: rural and urban differences. AIDS Care. 2005;17(5):558–565. doi: 10.1080/09540120412331319750. [DOI] [PubMed] [Google Scholar]
- Schneider E, Whitmore S, Glynn KM, Dominguez K, Mitsch A, McKenna MT. Revised surveillance case definitions for HIV infection among adults, adolescents, and children aged <18 months and for HIV infection and AIDS among children aged 18 months to <13 years--United States, 2008. MMWR Recomm Rep. 2008;57(RR-10):1–12. [PubMed] [Google Scholar]
- Shacham E, Lian M, Onen NF, Donovan M, Overton ET. Are neighborhood conditions associated with HIV management? HIV Med. 2013;14(10):624–632. doi: 10.1111/hiv.12067. [DOI] [PubMed] [Google Scholar]
- Simoni JM, Yard SS, Huh D. Prospective prediction of viral suppression and immune response nine months after ART initiation in Seattle, WA. AIDS Care. 2013;25(2):181–185. doi: 10.1080/09540121.2012.687821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weintrob AC, Fieberg AM, Agan BK, Ganesan A, Crum-Cianflone NF, Marconi VC, Wortmann GW. Increasing age at HIV seroconversion from 18 to 40 years is associated with favorable virologic and immunologic responses to HAART. J Acquir Immune Defic Syndr. 2008;49(1):40–47. doi: 10.1097/QAI.0b013e31817bec05. [DOI] [PubMed] [Google Scholar]