Abstract
Past studies of barriers to HIV care have not comprehensively assessed psychiatric symptoms, and few have assessed barriers to care among people living with HIV (PLWH) who are lost to care (LTC). We examined psychiatric symptoms, barriers to HIV care, and immune functioning in PLWH who were retained in care (RIC; n = 21) or LTC (n = 21). Participants completed diagnostic interviews for posttraumatic stress disorder (PTSD) and other psychiatric disorders, self-report measures of HIV risk behaviors and psychiatric symptoms, and a blood draw to assess viral load. Compared to RIC participants, LTC participants met criteria for a greater number of psychiatric disorders and reported greater depressive symptoms and more barriers to HIV care. There were no group differences in PTSD severity, risk behaviors, or viral load, suggesting that LTC individuals experience greater psychiatric problems and perceive more barriers to care than RIC participants, but are not less likely to have achieved viral suppression.
Introduction
Full HIV care engagement requires being diagnosed with HIV, being linked with and retained in care, adhering to antiretroviral therapy (ART), and achieving viral suppression. 1 -3 Unfortunately, many people living with HIV (PLWH) stop attending HIV care appointments. 4 A study in the greater Philadelphia area found that only 37% of individuals newly diagnosed with HIV and linked with care were retained in care 3 to 4 years later. 5
Given that poor HIV care retention is linked with inconsistent use of antiretroviral medication, virologic failure, 6 and poor health outcomes, 3 it is critical to understand why some PLWH drop out of care. Unfortunately, PLWH who are lost to care (LTC) are, by definition, difficult to engage. As a result, many studies examining barriers to care have not recruited LTC participants but those who are retained in care (RIC) 7 -14 or HIV care providers. 15,16 Only a handful of studies in the United States have successfully recruited LTC PLWH, 3 of which included LTC participants only, using focus groups (n = 29) 17 or survey methods (n = 179 and n = 247), 18,19 and 2 of which were interview studies that included both LTC and RIC participants (n = 334 and n = 51). 20,21
These studies found that in addition to structural barriers, psychiatric problems such as depression and substance use and HIV stigma are often reported as barriers to HIV care retention among PLWH who are LTC. Indeed, the presence of psychiatric symptoms has been found to be associated with adherence to HIV care appointments 13,22,23 and ART 24 -27 among PLWH who are RIC. Although psychiatric symptoms appear to be an important consideration when understanding HIV care retention, little is currently known about psychiatric symptoms among PLWH who are LTC. Previous studies with this population have not used standardized measures of psychiatric symptoms, and most have focused only on depression or alcohol/substance abuse, even though other disorders such as posttraumatic stress disorder (PTSD) is known to be elevated among PLWH compared to the general population. 28
The current study examines the relationships between psychiatric symptoms, immune functioning, and barriers to care in PLWH who were either RIC or LTC. Based on previous research, we hypothesized that LTC participants would report greater psychiatric difficulties, more barriers to care (in terms of both HIV care appointments and ART adherence), and have lower viral suppression. We aimed to extend prior work by interviewing PLWH who are LTC and RIC and inquiring specifically about whether psychiatric disorders functioned as barriers to care. In addition, we aimed to explore the relationships between psychiatric difficulties and barriers to care.
Methods
Participants
Participants were PLWH who were RIC (n = 21) and LTC (n = 21). Retained-in-care participants had attended 2 or more HIV medical visits separated by at least 90 days in the past calendar year. 29 Lost-to-care participants had attended at least 1 HIV care appointment in their lifetime (ie, had connected with care at some point), not attended any HIV care appointments at least the past 6 months, and had missed at least 2 appointments in the past 2 years. Exclusion criteria were: (1) a lifetime history of a psychotic disorder or bipolar disorder; (2) current suicidal ideation and intent; (3) current substance dependence; and (4) physical disability or medical condition (other than HIV) that would interfere with attending HIV care appointments.
Procedure
Potential participants were recruited from metropolitan hospitals, local HIV care clinics, community mental health clinics, and a homeless shelter in the [Philadelphia, PA] area using fliers, reviewing clinic registries and attendance data, and word of mouth. For those who responded, a brief phone screen was used to assess initial study eligibility. Figure 1 shows the number of participants contacted, screened, and enrolled.
Participant flowchart.
Eligible participants provided informed consent and then completed a 60- to 90-minute audio-recorded assessment consisting of a diagnostic interview with a trained independent evaluator who was blind to HIV care status, self-report questionnaires, and a 40-mL blood draw. Assessment fidelity was monitored by reviewing a random subsample of audio-recordings. Lost-to-care participants were encouraged to schedule an HIV care appointment before they left. Participants were compensated $50 for their time. This study was approved by the University of Pennsylvania’s institutional review board.
Measures
Demographic Information
Participants reported their gender, age, age when diagnosed with HIV, ethnicity, race, relationship status, number of children, education history, employment status and occupation, income, and religion.
Managed Problem-Solving
A modified version of the Managed Problem-Solving (MAPS) 30 was used to assess the barriers to care engagement (both medication and treatment adherence) for PLWH. Symptoms of mental health, schedule, and finances were some of the potential barriers to adherence assessed.
The Mini International Neuropsychiatric Interview
The Mini International Neuropsychiatric Interview (MINI) 31 is a brief, structured diagnostic interview for 17 Diagnostic and Statistical Manual of Mental Disorders (4th Edition; [DSM-IV]) axis 1 psychiatric disorders, including mood disorders, anxiety disorders, and substance use disorders. The MINI demonstrates good test–retest reliability, internal consistency, and convergent validity.
The PTSD Symptom Scale Interview for DSM-5
The PTSD Symptom Scale Interview for DSM-5 (PSSI-5) 32 is a brief, 24-item semistructured interview that assesses PTSD symptom severity and diagnosis based on the criteria from DSM-5. The measure demonstrates good test–retest reliability, internal consistency, interrater reliability, convergent validity with the Clinician Administered PTSD Scale (CAPS), and discriminant validity with the Beck Depression Inventory-II (BDI-II) (Pearson). The Cronbach’s α for the PSSI-5 in the current sample was .93.
Risk Assessment Battery
Risk Assessment Battery (RAB) 33 is a 45-item self-report measure of drug risk and sexual risk behaviors associated with HIV transmission. The Cronbach’s α for the RAB in the current sample was .95, .65, and .80 for the drug risk subscale, sexual-risk subscale, and total scale, respectively.
Quick Inventory of Depressive Symptomatology
The Quick Inventory of Depressive Symptomatology (QIDS) 34 is a 16-item measure of the severity of DSM-IV depressive symptoms. The measure has demonstrated good construct validity, content validity, and criterion validity with major depressive disorder diagnosis. 34 The Cronbach’s α for the QIDS in the current sample was .89.
Viral Load
Serum was collected to examine markers of immune functioning and viral load. Plasma HIV-1 RNA quantitative was performed at Quest Diagnostics using the COBAS AmpliPrep/COBAS Taqman HIV-1 test kit version 2.0 (Roche Molecular Systems). Individuals with a viral load of ≤50 copies/mL were considered virally suppressed.
Data Analyses
Means and standard deviations for psychiatric variables were computed. Group differences were analyzed using chi-square tests for categorical variables and analyses of variance for continuous variables. Correlation analyses examined relationships between psychiatric variables and barriers to care. Psychiatric diagnostic status was coded as any mood disorders (major depressive disorder, bipolar disorder, or dysthymic disorder), any anxiety disorders (social anxiety disorder, panic disorder, agoraphobia, generalized anxiety disorder, obsessive compulsive disorder, or PTSD), and any suicidality (response of ≥1 on the MINI suicide section). Statistical analyses were performed using SPSS, version 23.
Results
Demographic characteristics are reported in Table 1. There were no significant group differences in the demographic variables (all Ps > .16)
Participant Demographics and Means and Standard Deviations for All Psychiatric Variables.a
Abbreviations: DSM-5, Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; LTC, lost to care; MINI, Mini International Neuropsychiatric Interview; PSSI, PTSD symptom Scale Interview for DSM-5; PTSD, posttraumatic stress disorder; QIDS, Quick Inventory of Depressive Symptomatology; RAB, Risk Assessment Battery; RIC, retained in care; SD, standard deviation.
aTotal psychiatric disorders include number of axis I disorders diagnosed using MINI. Mood disorders are number of mood disorder diagnoses including major depressive disorder, bipolar disorder, and/or dysthymic disorder on the MINI. Suicidality indicates response of greater ≥1/3 on the MINI suicide screen section. Anxiety disorders include number of anxiety disorder diagnoses including social anxiety disorder, panic disorder, agoraphobia, generalized anxiety disorder, obsessive compulsive disorder, and PTSD on the MINI.
Psychiatric Symptoms
Means and standard deviations for all psychiatric variables are also reported in Table 1. Compared to RIC participants, LTC participants were more likely to have a psychiatric disorder, F 1,40 = 6.81, P = .013, including a mood disorder, F 1,40 = 5.070, P = .030, and an anxiety disorder, F 1,40 = 4.44, P = .041, and to report suicidality, F 1,40 = 4.71, P = .036. There were no group differences in PTSD symptom severity (P =.503) or diagnosis (P =.367).
Lost-to-care participants reported greater depressive symptoms than RIC participants, F 1,40 = 6.52, P = .015. There were no significant group differences in total, sexual, or drug risk scores on the RAB (P > .238).
Barriers to Care
Lost-to-care participants report more barriers to attending HIV care appointments, F 1,40 = 16.01, P < .001, and adhering to their HIV medication, F 1,40 = 15.06, P < .001, than RIC participants.
Viral Load
There were no differences between LTC and RIC participants in their viral load count (P = .819) or the proportion with an undetectable (≤50 copies/mL) viral load (80% in LTC versus 81.25% in RIC; P = .641).
Psychiatric Symptoms and Barriers to Care
The number of appointment barriers was significantly correlated with all psychiatric variables examined (see Table 2). The number of medication barriers was correlated with the number of psychiatric disorders, suicidality, anxiety disorder status, and depressive symptoms.
Correlations between Barriers and Psychiatric Variables.a
Abbreviations: DSM-5, Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; LTC, lost to care; MINI, Mini International Neuropsychiatric Interview; PSSI, PTSD symptom Scale Interview for DSM-5; PTSD, posttraumatic stress disorder; QIDS, Quick Inventory of Depressive Symptomatology; RAB, Risk Assessment Battery; RIC, retained in care.
aTotal psychiatric disorders include number of axis I disorders diagnosed using MINI. Mood disorders are number of mood disorder diagnoses including major depressive disorder, bipolar disorder, and/or dysthymic disorder on the MINI. Suicidality indicates response of greater ≥1/3 on the MINI suicide screen section. Anxiety disorders include number of anxiety disorder diagnoses including social anxiety disorder, panic disorder, agoraphobia, generalized anxiety disorder, obsessive compulsive disorder, and PTSD on the MINI.
bCorrelation is significant at the .01 level (2-tailed).
cCorrelation is significant at the .05 level (2-tailed).
dCannot be computed because no RIC participants reported any suicidality.
Among LTC participants, the number of appointment barriers was correlated with the number of psychiatric disorders, presence of an anxiety disorder, depressive symptom severity, and PTSD symptom severity, whereas none of the correlations between medication barriers and psychiatric variables remained significant (all Ps > .07). Among RIC participants, there were no significant correlations between barriers and psychiatric variables (all Ps > .09).
Discussion
We examined the relationships between psychiatric symptoms, immune functioning, and barriers to care in PLWH who were either RIC or LTC. Consistent with hypothesis, LTC participants met diagnostic criteria for more psychiatric disorders as a whole as well as for mood or anxiety disorder specifically. The use of psychiatric diagnostic interviews extends previous research by demonstrating the presence of clinically significant distress and dysfunction that warrants mental health intervention. Lost-to-care participants also reported greater depressive symptoms and suicidal ideation, pointing to the need for psychiatric screening as part of HIV care relinkage programs.
Consistent with prior research, the rate of PTSD in the sample (33% in LTC and 24% in RIC) was elevated relative to the general population rate of 7%. 35 However, contrary to hypothesis, LTC participants did not report greater PTSD severity or more drug risk behaviors. Possibly, these null findings may have been due to insufficient power or to the fact that participants with psychosis or substance dependence were excluded (although many participants in both group reported substance use).
As expected, LTC participants reported more barriers to attending HIV care appointments and to adhering to their HIV medication than RIC participants. All psychiatric were correlated with the number of barriers to attending HIV appointments; only the number of psychiatric disorders, suicidality, and anxiety disorder status was correlated with barriers to medication adherence. Most of these correlations remained significant among LTC but not RIC participants. This suggests that LTC PLWH experience more psychiatric problems and that their psychiatric problems are more strongly associated with barriers to HIV care appointments and medication adherence compared to RIC PLWH.
Conclusion
The finding that LTC participants showed equal levels of viral suppression (80%) as the RIC participants (81.5%) is heartening and suggests that some LTC PLWH may not be out of care but simply receiving less care. While conclusions must be tempered, given the small sample size of this study, it may be that estimates of viral suppression across the HIV care continuum are underestimating the proportion of PLWH who are virally suppressed.
Several study limitations should be noted. First, this was a small pilot study and resources allowed only for 31 (74%) participants to complete the blood draw to assess viral load. Second, we assessed the number of barriers identified but did not examine whether the barriers were perceived as important as did other researchers. 19 Third, although few participants (5 of 53 screened) were excluded due to psychosis and substance dependence, these exclusion criteria may have biased the results by impacting recruitment (eg, we did not recruit from a local needle exchange program). Finally, the definition of LTC we used was more liberal than others have used; our sample may better represent PLWH who recently dropped from care or who are intermittently in care than PLWH who have not engaged in care for a longer period of time. Clinically valid and standardized measures of HIV care retention are needed.
In conclusion, the results showed that LTC participants carry a greater psychiatric burden and perceive more barriers to care than RIC but have equal rates of viral suppression. The study also points to the feasibility of engaging LTC PLWH in research. Encouragingly, 14 of the 20 LTC participants in the study agreed to reengage with care and made an HIV clinic appointment during their study visit. Future studies should examine whether efforts to identify and treat psychiatric problems at the point of HIV care engagement can reduce subsequent dropout rates from care.
Footnotes
Authors’ Note
Preliminary data from this study were presented at the annual convention for the International Society for Traumatic Stress Studies in Dallas, Texas, in November 2016.
Acknowledgments
The authors would like to acknowledge and thank the individuals who volunteered to participate in this study: Dr Patria Alvelo and Ms Samantha Silverberg for serving as independent evaluators, and Dr Sandy Capaldi for serving as the assessment trainer and fidelity monitor.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Penn Mental Health AIDS Research Center (PMHARC) through a grant (P30-MH097488-03) from the National Institute of Mental Health (NIMH).