Psychiatric disorders and adherence to antiretroviral therapy among a population of HIV-infected adults in Nigeria

Abstract

Summary

Psychiatric disorders are common among people living with HIV in Nigeria. Adherence is necessary to optimise the outcome of antiretroviral therapy. In this study, we aimed to identify associations between antiretroviral adherence, measured by one-week and one-month self-reported missed doses, and psychiatric illness in a cohort previously assessed for psychiatric disorders using the Composite International Diagnostic Interview. The study participants comprised 151 adults with major depression, anxiety or suicidal symptoms, and 302 matched-control participants. Two controls were randomly selected for each case within the same gender and education level. We compared participants with psychiatric disorders (WPDs) and no psychiatric disorders (NPDs) on selected demographic and clinical variables, in addition to adherence. Participants with one or more missed doses in the preceding month had twice the odds of having a major depressive episode as those with no missed doses during this period (OR 2.22, 95% CI 1.03, 4.79). This association remained significant after adjusting for selected risk factors. There was no statistically significant difference between WPD and NPD groups on either one-week or one-month adherence, or on age, marital status, occupational class, HIV viral load at enrolment or current CD4 cell count. Among Nigerian adults with HIV, suboptimal antiretroviral adherence is associated with, and could be a sign of, depression. Routine self-report adherence assessments may potentially be utilised in identifying individuals at risk among this population.

Keywords

HIV AIDS antiretroviral therapy adherence psychiatric disorder Nigeria

Introduction

Antiretroviral therapy (ART) is the mainstay of HIV/AIDS treatment and a high level of ART adherence is crucial to achieving viral control and preventing treatment resistance.^1,2 Recent research has reported higher rates of ART adherence in low-resource regions like Africa and Asia compared to the US, Europe and other well-resourced settings.^3–5 However, studies in Nigeria, a country with the second largest number of people with HIV in the world, report optimal adherence rates in less than 60% of treatment populations.^6–8

Among a variety of factors which influence ART adherence, the mental health of individuals with HIV has been recognised to play a crucial role.^9–11 Psychiatric disorders occur in up to 50% of individuals with HIV.^12–15 The most commonly reported of these are depression, anxiety disorders and substance use disorders.^14,16–18 These disorders act to reduce quality of life and accelerate the progression of HIV disease^19–21 through a number of mechanisms, including an increased likelihood of suboptimal treatment adherence.^22–24 Much research has demonstrated associations between psychiatric disorders and ART adherence.^10,25–27 Although this relationship is complex,²⁸ it has been theorized that depressive symptoms lead to hopelessness, cognitive impairment and isolation from social support, all of which could result in poor medication adherence.^29,30 On the other hand, non-adherence increases risk of further HIV-related complications, including chronically-activated inflammatory pathways, which may in turn confer additional risk for psychiatric and neurocognitive disorders.^31–33

Some studies have reported associations between poor ART adherence and psychiatric disorders among Nigerian adults with HIV. However, these either used broad screening tools to detect psychiatric morbidity rather than specific psychiatric diagnoses,^7,34 or focused solely on a diagnosis of depression,³⁵ without investigating associations with other psychiatric disorders. Furthermore, the studies on depression did not examine the influence of any mediating factors on this association with ART adherence.³⁵ In contrast, Oladeji et al. used the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI), and found a depression, anxiety and suicidality prevalence of 18.2% among 828 patients in the AIDS Prevention Initiative in Nigeria (APIN) clinic, Ibadan.³⁶ The CIDI is a comprehensive, fully-structured diagnostic tool designed for epidemiological and cross-cultural research,³⁷ and its validity and reliability have been reported in African countries including Ethiopia and Nigeria.^38–42 Therefore, we sought to identify possible associations between ART adherence and psychiatric disorders among an ART-using subset of persons with HIV attending the Ibadan APIN clinic, by comparing self-reported adherence in patients with mood disorders or a history of suicidality (i.e. with psychiatric diagnoses, WPD), versus those without these diagnoses (i.e. no psychiatric disorders [NPDs]). Our primary hypothesis was that rates of suboptimal ART adherence would be significantly higher among the WPD compared to the NPD group. In addition, we aimed to determine whether associations between psychiatric disorders and adherence were mediated by demographic factors such as age, marital status and occupational skill level, or clinical factors such as duration of HIV treatment and CD4 cell counts.

Methods

Study site

The study was conducted in the APIN clinic, University College Hospital, Ibadan, Nigeria. This outpatient clinic is staffed by specially-trained family physicians, medical officers and nurses, offers HIV testing and treatment services to referrals from primary, secondary and tertiary health care facilities in the region, and caters for attendees from all over South-West Nigeria. Over the past 13 years, the clinic has registered approximately 19,000 persons with HIV aged 18 years and older, and provides treatment to over 2000 persons every year. Treatment-eligible patients receive a monthly supply of ART at scheduled pharmacy appointments, free of charge. Dates of medication pick up and dates of expected return for a refill are recorded at each of these visits. The APIN clinic also maintains an electronic database of demographic information collected at patient enrolment, as well as routine clinical data including ART and non-ART medications prescribed during clinic visits, six monthly CD4 cell counts and HIV viral load.

Study participants

Participants were identified from the 828 adult APIN clinic patients previously assessed cross-sectionally for psychiatric disorders between May and August 2011.³⁶ Psychiatric diagnoses were made by trained and supervised research assistants using the WHO CIDI. Of the 828 patients, 744 had been receiving ART for durations ranging from one to 91 months, and were thus considered eligible for inclusion in the current analyses. In order to maximize the power to detect significant associations between psychiatric disorder and adherence, all participants with psychiatric diagnoses (the WPD group), numbering 151, were selected for inclusion. Thereafter the participants without psychiatric diagnoses were stratified by gender (man or woman) and educational status (no formal education, primary, secondary, or tertiary education). Two participants without psychiatric diagnoses (the NPD control group) were randomly selected for each case within the same gender and educational stratum, resulting in 302 controls. Thus the study consisted of a total of 453 participants.

Procedure

The current study was a retrospective comparative analysis of existing data from the study by Oladeji et al.³⁶ In the parent study, diagnoses of psychiatric disorder were made based on the International Statistical Classification of Diseases and Related Health Problems, tenth Revision (ICD-10) diagnostic criteria for major depressive disorder, generalized anxiety disorder, a history of suicidal ideation, plan or attempt since HIV diagnosis, or any combination of these diagnoses, as measured using corresponding modules of the CIDI. These psychiatric evaluations involved the use of the Yoruba version of the CIDI, earlier derived using standard iterative back-translation methods, for use in Nigeria.^42–44 Alongside other language versions, the Yoruba CIDI has been demonstrated to have acceptable reliability and validity by the WHO-CIDI Field Trials and other clinical calibration studies.³⁷ The CIDI is scored according to the number of diagnostic criteria present for each disorder, corresponding with definitions in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) and the ICD-10 Classification of Mental and Behavioral Disorders.^37,45,46

In addition, an interviewer-administered questionnaire was used to obtain information on demographic variables and on participant adherence. Participants were required to indicate the number of ART doses they had missed in the preceding one week and one month, from an option of none, one, two or greater than two missed doses. These interviews were conducted by two post-graduate students of Public Health at the University of Ibadan, who were selected, trained and supervised by the investigator of the parent study, a specialist psychiatrist, in the use of the CIDI. Additional available information was obtained from participants’ electronic medical records, including date of clinic enrolment and duration of ART, viral load before ART commencement and the most recent CD4 cell count before entry into the parent study. Ethical approval for the current analysis was given by the joint ethical committee of the University of Ibadan and University College Hospital. The Northwestern University IRB determined that the study met criteria for exemption from IRB review and approval.

Data management and statistical analysis

Analyses were conducted using SPSS version 20.0 (IBM Corp. Released 2011. IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY). Statistical testing was conducted at the 5% significance level. Participants with missing data on any variable were excluded from analyses involving that variable. Self-reported adherence was dichotomized as 100% adherent (no ART doses missed in the preceding week or month), or <100% non-adherent (one or more ART doses missed). Participants’ marital status was dichotomized as married or unmarried. Associations between these variables, with psychiatric disorder status (with psychiatric disorder [WPD] vs. without psychiatric disorder [NPD]), were tested using Pearson’s χ² test. Viral load was converted to log₁₀ values for ease of data handling and analysis. Participants’ ages, duration on ART, viral load at initial HIV diagnosis and their CD4 cell count at the time of psychiatric assessment in 2011 were analysed as continuous variables. Means of continuous variables were compared between the WPD and NPD groups, and mean differences were evaluated using independent-sample t tests.

In addition to comparing the WPD and NPD groups in this manner, eight separate analyses were conducted to similarly test associations between the same demographic and clinical variables, with each subgroup within the WPD group (depression, generalized anxiety disorder, suicidality or any depression or anxiety). Variables identified to have significant bivariate associations were then entered into a multivariable regression model to adjust for confounders and identify persisting associations of adherence WPD. Thus, one-month adherence was entered into a logistic regression model with dependent variable depression, along with other variables known from literature to be associated with depression in HIV, including age,⁴⁷ gender,⁴⁸ marital status,⁴⁹ educational status,⁵⁰ class of occupation,⁵¹ duration on ART,⁵² and current CD4 cell count.²⁴ For this purpose, age was grouped in 10-year intervals, and duration on ART was grouped in 12-month intervals.

Lastly, we evaluated the validity of the one-week and one-month reports of missed doses as accurate adherence measures in this study by testing associations between each of them and the closest viral load and CD4 cell counts to assessment, which have been used as surrogate measures of adherence in previous studies.^53–56 For participants who had data on the most recent viral load or CD4 cell counts (ranging from two weeks to six months before adherence assessment), we compared these variables between those identified as adherent or non-adherent by self-reported missed doses.

Results

Participant characteristics

From the parent study, 151 (20.2%) of the patients met criteria for at least one psychiatric disorder. One hundred and fourteen (15.3%) participants had a history of suicidal ideations, plans or attempts, 48 (6.5%) met criteria for a major depressive episode, and 57 (7.7%) met criteria for any depression and or anxiety disorder.

Of the 453 participants selected for this study, the majority (76.8%) were women, with mean age of 41.2 years (SD = 9.9 years), and mean length of time since HIV diagnosis of 3.6 years (SD = 1.8 years). Table 1 compares demographic and clinical characteristics of the WPD and NPD groups. There were no significant differences between the WPD and NPD groups in age (OR 0.84; 95% CI 0.57, 1.25), marital status (OR 0.80; 95% CI 0.54, 1.18), class of occupation (OR 2.19; 95% CI 0.99, 4.87), duration since HIV diagnosis (OR 1.03; 95% CI 0.69, 1.52), duration of ART (OR 0.84; 95% CI 0.47, 1.52), viral load before ART commencement (mean difference 0.22; 95% CI − 0.02, 0.45), or current CD4 cell count (mean difference 31.7; 95% CI − 10.18, 73.49) (Table 1).

Table 1.

Clinical and demographic characteristics by psychiatric disorder status.

	Mood disorder or suicidality present (WPD)	Mood disorder or suicidality absent (NPD)	p value	Odds ratio (95% CI)
	(n = 151) % (n)	(n = 302) % (n)	p value	Odds ratio (95% CI)
Age
≤40 years	54.3 (82)	50.0 (151)	0.39	0.84 (0.57–1.25)
>40 years	45.7 (69)	50.0 (151)
Missing	0.0 (0)	0.0 (0)
Gender^a
Men	23.2 (35)	23.2 (70)	1.00	Not applicable
Women	76.8 (116)	76.8 (232)
Missing	0.0 (0)	0.0 (0)
Level of education^a
No formal education	11.3 (17)	11.3 (34)	1.00	Not applicable
Primary	29.8 (45)	29.8 (90)
Secondary	41.7 (63)	41.7 (126)
Tertiary	17.2 (26)	17.2 (52)
Missing	0.0 (0)	0.0 (0)
Marital status
Married	53.0 (80)	58.6 (177)	0.25	0.80 (0.54–1.18)
Not married	47.0 (71)	41.4 (125)
Missing	0.0 (0)	0.0 (0)
Occupation^b
Skilled	5.3 (8)	10.9 (33)	0.05	2.19 (0.99–4.87)
Semi-skilled or unskilled	94.7 (143)	89.1 (269)
Missing	0.0 (0)	0.0 (0)
Duration since HIV diagnosis
£3 years	45.7 (69)	46.4 (140)	0.89	1.03 (0.69–1.52)
>3 years	54.3 (82)	53.6 (162)
Missing	0.0 (0)	0.0 (0)
Duration on ART
0–11 months	13.2 (20)	11.6 (35)	0.57	0.84 (0.47–1.52)
≥12 months	74.2 (112)	77.2 (233)
Missing	12.6 (19)	11.2 (34)
Adherence
Number of doses missed in preceding week
No doses missed	89.4 (135)	91.1 (275)	0.80	1.15 (0.57–2.34)
One or more doses missed	8.6 (13)	7.6 (23)
Missing	2.0 (3)	1.3 (4)
Number of doses missed in preceding month
No doses missed	84.0 (126)	88.4 (267)	0.42	1.44 (0.79–2.60)
One or more doses missed	14.0 (21)	10.3 (31)
Missing	2.0 (3)	1.3 (4)

Continuous variables
	n	Mean	SD	Range	25^th percentile	Median	75^th percentile	Mean difference (95% CI)	p
Duration of time on ART (months)
WPD	132	38.2	21.8	1–91	20.0	39.0	51.0	1.6 (−2.92–6.19)	0.48
NPD	268	39.8	21.8	1–88	23.0	41.0	57.0
Viral load before ART commencement (log₁₀ copies/ml)
WPD	134	4.8	1.1	2.3–7.7	4.1	4.9	5.7	0.2 (−0.2–0.45)	0.07
NPD	266	4.6	1.2	2.3–6.9	3.8	4.8	5.5
Current CD4 cell count (cells/mm³)
WPD	149	397.0	195.8	25.0–1175.0	241.0	391.0	520.0	0.1 (−0.02–0.24)	0.10
NPD	301	428.7	242.4	16.0–1903.0	251.0	379.0	571.0

WPD: participants with mood disorders or suicide-related symptoms; NPD: participants without mood disorders or suicide-related symptoms; ART: antiretroviral therapy; SD: standard deviation; CI: confidence interval.

Analyses on categorical variables were performed using Pearson’s χ² test of association, and differences between means of continuous variables were evaluated using the independent samples t test, with missing data excluded.

NPDs were matched on age and gender with WPDs; distribution on these characteristics is shown for descriptive purposes.

Occupational classification done according to International Labour Organisation’s ISCO-08, thus: skilled: managers, professionals, technicians & associate professionals, armed forces; semi-skilled or unskilled: clerical support workers, service & sales workers, skilled agricultural, forestry & fishery workers, craft & related trades workers, plant & machine operators, elementary occupations. Unemployed participants included in this group (International Labour Organisation, 2007).

ART adherence and psychiatric disorders

Reported adherence was high, with a total of 410 (90.3%) participants having no missed ART doses in the preceding week, while 36 (7.9%) of them had missed one or more doses. Similarly, 393 (86.8%) participants reported no missed doses in the preceding month, while 11.5% had missed at least one dose. Seven and eight participants, respectively, gave no reports on missed doses during these periods. Among WPDs, 8.6% had missed one or more doses in the previous week, compared to 7.6% of NPDs (OR 1.15; 95% CI 0.57, 2.34; p = 0.80). Also, within the previous month, 14.0% of WPDs, compared to 10.3% of NPDs, had missed at least one dose (OR 1.44; 95% CI 0.79, 2.60; p = 0.42), although none of these associations was statistically significant (Table 1).

Within the WPD group, there were no statistically significant associations between the participants’ number of mood diagnoses or suicidality symptoms, and their reported number of missed doses in the past week or month (OR 0.99; 95% CI 0.31, 3.21; p = 0.99).

One-month adherence and major depression

Among individual psychiatric disorders as shown in Table 2, a diagnosis of depression was significantly associated with non-adherence within the past month (OR 2.22; 95% CI 1.03, 4.79; p = 0.04) (Table 2). Adjusting for age, gender, marital status, education, class of occupation, duration of ART use and current CD4 cell count, the odds of having depression were three times higher in participants who were non-adherent in the preceding month, compared to those who were completely adherent (OR 3.23; 95% CI 1.41, 7.40; p = 0.01) (Table 3).

Table 2.

Odds of non-adherence by individual psychiatric diagnoses^a.

		Doses missed in past week				Doses missed in past month
		None	One or more	OR (95% CI)	p	None	One or more	OR (95% CI)	p
Major depressive episode		n = 446, % (n)				n = 445, % (n)
	Present	91.7 (44)	8.3 (4)	0.96 (0.33–2.85)	0.95	79.2 (38)	20.8 (10)	2.22 (1.03–4.79)	0.04
	Absent	92.0 (366)	8.0 (32)			89.4 (355)	10.6 (42)
Generalized anxiety disorder (GAD)		n = 446, % (n)				n = 445, % (n)
	Present	90.9 (10)	9.1 (1)	0.88 (0.11–7.04)	0.90	72.7 (8)	27.3 (3)	2.95 (0.76–11.48)	0.10
	Absent	92.0 (400)	8.0 (35)			88.7 (385)	11.3 (49)
Any GAD or depression		n = 446, % (n)				n = 445, % (n)
	Present	92.0 (358)	8.0 (31)	1.11 (0.41–2.98)	0.84	89.4 (347)	10.6 (41)	0.49 (0.24–1.03)	0.06
	Absent	91.2 (52)	8.8 (5)			80.7 (46)	19.3 (11)
Suicidality		n = 445, % (n)				n = 444, % (n)
	Present	92.2 (308)	7.8 (26)	1.17 (0.55–2.52)	0.68	88.9 (297)	11.1 (37)	0.85 (0.44–1.65)	0.64
	Absent	91.0 (101)	9.0 (10)				87.3 (96)	12.7 (14)

All analyses performed using χ² test of association.

Non-adherence is defined as one or more missed antiretroviral doses in the preceding week or month.

Table 3.

Associations with major depression.

	All % (n = 453)	With depression % (n = 48)	Without depression % (n = 405)	OR (95% CI)^a,b	p ^b,d	AOR (95% CI)^a,c	p ^c,d
Adherence (missed doses in preceding month)	98.2 (445)	100.0 (48)	98.0 (397)
One or more		20.8 (10)	10.6 (42)	2.22 (1.03–4.78)	0.04	3.23 (1.41–7.40)	0.01
None		79.2 (38)	89.4 (355)
Age (years)	100.0 (453)	100.0 (48)	100.0 (405)
11–20		0.0 (0)	1.2 (5)	0.99 (0.75–1.32)	0.95	1.07 (0.77–1.48)	0.68
21–30		10.4 (5)	13.3 (54)
31–40		37.5 (18)	37.3 (151)
41–50		37.5 (18)	30.6 (124)
51–60		14.6 (7)	12.1 (49)
61–70		0.0 (0)	5.4 (22)
Gender	100.0 (453)	100.0 (48)	100.0 (405)
Men		33.3 (16)	22.0 (89)	1.78 (0.93–3.38)	0.08	1.90 (0.88–4.10)	0.10
Women		66.7 (32)	78.0 (316)
Marital status	100.0 (453)	100.0 (48)	100.0 (405)
Married		60.4 (29)	46.3 (228)	1.19 (0.64–2.18)	0.59	1.09 (0.56–2.13)	0.79
Not married		39.6 (19)	43.7 (177)
Educational status	100.0 (453)	100.0 (48)	100.0 (405)
No formal education		16.7 (8)	10.6 (43)	1.20 (0.86–1.67)	0.29	1.42 (0.94–2.16)	0.10
Primary		31.2 (15)	29.6 (120)
Secondary		35.4 (17)	42.5 (172)
Tertiary		16.7 (8)	17.3 (70)
Class of occupation	100.0 (453)	100.0 (48)	100.0 (405)
Unskilled/semiskilled		95.8 (46)	90.4 (366)	2.45 (0.57-10.49)	0.23	1.56 (0.32-7.52)	0.58
Skilled		4.2 (2)	9.6 (39)
Duration on ART (months)	88.3 (400)	91.7 (44)	87.9 (356)
0–12		15.9 (7)	14.3 (51)	1.09 (0.92-1.30)	0.31	1.00 (0.83-1.22)	0.92
13–24		20.5 (9)	13.2 (47)
25–36		9.1 (4)	15.7 (56)
37–48		27.3 (12)	19.1 (68)
49–60		13.6 (6)	19.7 (70)
61–72		6.8 (3)	9.8 (35)
73–84		6.8 (3)	7.6 (27)
86–96		0.0 (0)	0.6 (2)
Current CD4 cell count (cells/mm³)	99.3 (450)	100.0 (48)	99.2 (402)
<200		25.0 (12)	2.9 (52)	0.60 (0.38–0.94)	0.03	1.71 (1.00–2.92)	0.05
200–499		52.1 (25)	53.7 (216)
≥500		22.9 (11)	33.3 (134)

OR: odds ratio; AOR: adjusted odds ratio.

Odds ratios compare the odds of major depression in participant groups by one-month adherence and selected risk factors.

These refer to bivariate analysis of relationships between major depression and each of the listed risk factors.

These refer to multivariate analysis of relationships between major depression and one-month self-reported adherence, adjusting for age, gender, marital status (married/unmarried), educational status (no formal education, primary, secondary or tertiary), class of occupation, duration of antiretroviral use and current CD4.

Only p values less than 0.05 are noted.

Validity of self-report adherence

There was no statistically significant difference between adherent and non-adherent participant groups, as measured by one-week recall, on clinical surrogate measures such as current log₁₀ viral load (mean difference 0.28; 95% CI − 0.21, 0.77) and current CD4 cell count (mean difference 15.31; 95% CI − 79.58, 110.20). Similarly, there was no association between adherent and non-adherent groups by one-month recall, on current log₁₀ viral load (mean difference 0.37; 95% CI − 0.24, 0.99) and current CD4 cell count (mean difference 1.53; 95% CI − 64.89, 67.95) (Table 4).

Table 4.

Differences between adherent and non-adherent participants on current viral load and CD4 count.

Current viral load^a (log₁₀ copies/ml)^b Mean (±SD)	Doses missed in preceding week N = 222
	None n = 205	One or more n = 17	Mean difference in current viral load (95% CI)	p
	2.75 (±0.96)	3.03 (±1.36)	0.28 (−0.21–0.77)	0.26
	Doses missed in preceding month N = 222
	None n = 201	One or more n = 21	Mean difference in current viral load (95% CI)	p
	2.74 (±0.95)	3.11 (±1.33)	0.37 (−0.24–0.99)	0.22
Current CD4 cell count (cells/mm³) Mean (±SD)	Doses missed in preceding week N = 444
	None n = 408	One or more n = 36	Mean difference in current CD4 (95% CI)	p
	417.47 (±224.98)	432.78 (±273.49)	15.31 (−79.58–110.20)	0.75
	Doses missed in preceding month N = 443
	None n = 391	One or more n = 52	Mean difference in current CD4 (95% CI)	p
	419.28 (+226.16)	420.81 (±249.23)	1.53 (−64.89–67.95)	0.97

Adherent participants were those with no missed doses, and non-adherent were those with one or more missed doses.

Analysis performed for comparison. Current viral load was not used in any other analysis due to unavailable values for 231 of the participants.

Log₁₀ values were used because viral load is highly skewed.

Discussion

Neuropsychiatric disorders often go undetected among individuals with HIV, especially when unassociated with prominent changes in judgment and behaviour. However, psychiatric disorders are increasingly recognized to present a potential threat to patients’ functioning, quality of life and medication management.^21,25 Although mood disorders, especially depression, have been associated with suboptimal ART adherence,¹⁰ this relationship is complex as poor adherence may in turn increase the risk of HIV disease progression and associated neuropsychiatric complications.^57,58

Our analysis was aimed at identifying associations between self-reported ART adherence and psychiatric disorders among adults with HIV in Nigeria. We also sought to compare WPDs and NPDs on selected demographic and clinical factors. We found that participants who missed at least one ART dose in the preceding month had twice the odds of major depression compared to those who had no missed doses. This increased to three times the odds after adjusting for demographic variables such as age, gender, marital status, educational status and class of occupation, and clinical factors such as duration of ART use and current CD4 cell count.

Among participants with anxiety disorders or suicidality, we failed to rule out the absence of any significant effects in either one-week or one-month adherence. WPDs grouped together, we also failed to rule out the absence of an effect in one-week or one-monthly adherence between WPD and NPD groups. Similarly, we failed to rule out the absence of effects in age, marital status, class of occupation, duration since HIV diagnosis, duration on ART, pre-ART viral load and CD4 cell counts at the time of the study, between the WPD and NPD groups.

Our findings suggest that suboptimal ART adherence in Nigerian adults with HIV is not associated with increased rates of broad psychopathology. This contrasts with the findings of Adewuya et al., who found an association between general psychopathology and low ART adherence among Nigerian adults with HIV.⁷ Adewuya et al. measured general psychopathology using the General Health Questionnaire-12 (GHQ-12), an extensively-used self-administered screener for general psychiatric wellbeing, which has been proposed to have three factors: anxiety and depression; social dysfunction; and loss of confidence.^59,60 It is possible the GHQ identified psychological distress among several participants in that study who had experienced any of these three factors. Thus, a relatively larger proportion (65.4%) was identified to have symptoms of psychopathology than was found in our parent study, which identified participants with specific diagnoses. Correspondingly, our findings may be related to the relatively low rates (18.2%) of any mood disorder or suicidality in our sample, which although higher than the 4.7% reported rate of any mental health disorder in the Nigerian general population,⁶¹ are lower than the reported rates among other HIV-infected populations in Nigeria.^62,63

Elevated rates of depression have been widely-reported among individuals with HIV, compared to other psychiatric disorders, and multiple investigations of adherence and mental health in sub-Saharan Africa have demonstrated similar associations with depression.^26,64,65 While it is recognized that reduced motivation, social withdrawal and cognitive impairment associated with depression could result in suboptimal ART adherence, poor viral control due to impaired adherence increases the risk for HIV-associated central nervous system injury resulting in inflammation and changes in emotional and hormonal regulation, which could manifest in depression.^66–69 This bi-directional relationship could explain the associations between ART adherence and depression consistently reported in research, and supported by our findings among Nigerian adults.

Our analysis identified an association between depression and one-month self-report adherence, but not one-week self-report. Although more recall bias would be expected for longer periods of recollection, the use of one-month self-recollection periods has shown validity for adherence measurement, being less associated with over-reporting than three- or seven-day recall durations,⁷⁰ and has also been shown to correlate with adherence measured by electronic monitoring systems, and viral load.⁷¹ It may be likely that patients are more willing to acknowledge non-adherence over a longer than shorter time period. However, it is also important to note that both participants with and without psychiatric disorders reported high rates of adherence. While it is one of the most widely-used methods for assessing antiretroviral adherence, self-report may be unreliable as patients typically overestimate their ART use.^72,73 Possible evidence for this in this study may have been our findings of no association between self-reported adherence and either HIV viral load or CD4 cell count close to the time of assessment, both of which have been used as surrogate measures of adherence in previous research.^53–55,74 It should be noted, however, that self-reported adherence over a preceding one-month period may not correspond with recent virologic indices obtained up to six months preceding, as was the case with some participants in this study. Thus, durations of such length between the assessments, rather than non-validity of self-reports, may have been responsible for the lack of association we found between self-reported adherence and virologic indices.

In the Nigerian context, the tendency to overestimate self-reported adherence may be particularly strong given cultural dynamics of the provider–patient relationship in treatment settings. The often paternalistic nature of Nigerian doctor–patient relationships⁷⁵ may contribute to a tendency to overestimate adherence for social desirability reasons. Persons receiving ART in the study location may also have felt indebted to their providers because ART is received free of charge, and may tend to portray themselves as taking full advantage of this benefit to prevent its being withdrawn. In addition, such inaccurate estimation may have resulted from recall bias, another recognized limitation of self-report measures.^76,77 In keeping with existing data indicating better adherence rates in Africa compared to Europe and North America,^3,78 such positive rates in Nigeria could also be reflective of successful widespread efforts (such as community-level HIV testing and ART provision) by government and international organizations, targeted at improving access to care in response to the generalized HIV/AIDS epidemic in this region. In contrast, the concentration of the epidemic among relatively focused, vulnerable and often hard-to-reach core risk subpopulations in well-resourced regions, may contribute to a relative lack of access and potentially poorer adherence rates among such groups, despite disproportionately better resources in these regions.

We acknowledge some limitations to our investigation, including the cross-sectional design of the parent study. The one-time assessment of psychiatric disorders did not permit assessment of symptom onset, duration and temporal relationships, and made it impossible to hypothesize causative relationships between non-adherence and psychiatric disorders. Also, although the use of the CIDI for diagnostic assessments of psychiatric disorders was an improvement over prior studies among similar Nigerian populations, it did not provide for a graded assessment of these psychiatric symptoms. Thus, the psychiatric assessments yielded only categorical classifications of disorder presence or absence, making it difficult to study associations between adherence and severity of psychiatric disorder. Finally, the missing data on a large numbers of participants’ nadir CD4 cell counts and their viral load at point of assessment, as well as absence of data on stressful life events, resulted in gaps in our ability to measure important clinical and psychosocial factors known to be significant predictors of HIV outcome, and possibly of depression. It is also important to highlight the possibility that our observed association between one-month adherence and depression may have occurred due to chance, given the number of independent tests of association we conducted at various stages of analysis.

The strengths of our study include the adoption of a matching strategy aimed at reducing the potential confounding effects of gender and educational status, since the parent study included a large number of participants from a population rarely studied for psychiatric symptoms in the past. Our findings are therefore an important contribution to the literature on adherence and mental health among adults living with HIV in sub-Saharan Africa, and specifically in Nigeria. Self-reports of missed doses within the preceding month may be a potentially useful tool for assessing adherence in this setting, and suboptimal adherence identified through this method could be a signal of undiagnosed depression in adults with HIV. There is, however, need for further research into the utility of other adherence measures such as pharmacy refill records, as well as a need for longitudinal studies to not only explore causative relationships between psychiatric disorder and adherence in this population, but also inform prevention and intervention efforts.

Footnotes

Acknowledgements

The authors acknowledge Kevin Robertson, PhD (Department of Neurology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA), Professors Isaac F Adewole the Principal Investigator, and David Olaleye the co-Principal Investigator of the APIN clinic, University College Hospital, Ibadan.

Authors’ note

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: research training for this publication was supported by the Fogarty International Center and National Institute of Mental Health of the National Institutes of Health under Award Number D43TW009608.

References

Arnsten

Demas

Farzadegan

. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: comparison of self-report and electronic monitoring. Clin Infect Dis 2001; 33: 1417–1423.

Bangsberg

Hecht

Charlebois

. Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS 2000; 14: 357–366.

Kim

S-H

Gerver

Fidler

. Adherence to antiretroviral therapy in adolescents living with HIV: systematic review and meta-analysis. AIDS 2014; 28: 1945–1956.

Vreeman

Wiehe

Pearce

. A systematic review of pediatric adherence to antiretroviral therapy in low- and middle-income countries. Pediatr Infect Dis J 2008; 27: 686–691.

Mills

Nachega

Bangsberg

. Adherence to HAART: a systematic review of developed and developing nation patient-reported barriers and facilitators. PLoS Med 2006; 3: e438–e438.

Oku

Owoaje

Ige

. Prevalence and determinants of adherence to HAART amongst PLHIV in a tertiary health facility in south-south Nigeria. BMC Infect Dis 2013; 13: 401–401.

Adewuya

Afolabi

Ola

. The effect of psychological distress on medication adherence in persons with HIV infection in Nigeria. Psychosomatics 2010; 51: 68–73.

Oku

Owoaje

Oku

. Prevalence and determinants of adherence to highly active antiretroviral therapy amongst people living with HIV/AIDS in a rural setting in south-south Nigeria. Afr J Reprod Health 2014; 18: 133–143.

Wagner

Goggin

Remien

. A closer look at depression and its relationship to HIV antiretroviral adherence. Ann Behav Med Publ Soc Behav Med 2011; 42: 352–360.

10.

Nel

Kagee

. Common mental health problems and antiretroviral therapy adherence. AIDS Care 2011; 23: 1360–1365.

11.

Kagee

. Addressing psychosocial problems among persons living with HIV. Afr J Psychiatry 2012; 15: 424–426.

12.

Atkinson

Heaton

Patterson

. Two-year prospective study of major depressive disorder in HIV-infected men. J Affect Disord 2008; 108: 225–234.

13.

Chander

Himelhoch

Moore

. Substance abuse and psychiatric disorders in HIV-positive patients: epidemiology and impact on antiretroviral therapy. Drugs 2006; 66: 769–789.

14.

Adewuya

Afolabi

Ola

. Psychiatric disorders among the HIV-positive population in Nigeria: a control study. J Psychosom Res 2007; 63: 203–206.

15.

Lopes

Olfson

Rabkin

. Gender, HIV status, and psychiatric disorders: results from the National Epidemiologic Survey on alcohol and related conditions. J Clin Psychiatry 2012; 73: 384–391.

16.

Jallow

Ljunggren

Wändell

. Prevalence, incidence, mortality and co-morbidities amongst human immunodeficiency virus (HIV) patients in Stockholm County, Sweden – The Greater Stockholm HIV Cohort Study. AIDS Care 2015; 27: 142–149.

17.

Garey

Bakhshaie

Sharp

. Anxiety, depression, and HIV symptoms among persons living with HIV/AIDS: the role of hazardous drinking. AIDS Care 2015; 27: 80–85.

18.

Olagunju

Adeyemi

Erinfolami

. Factors associated with anxiety disorders among HIV-positive attendees of an HIV clinic in Lagos, Nigeria. Int J STD AIDS 2012; 23: 389–393.

19.

Farinpour

Miller

Satz

. Psychosocial risk factors of HIV morbidity and mortality: findings from the Multicenter AIDS Cohort Study (MACS). J Clin Exp Neuropsychol 2003; 25: 654–670.

20.

Venturini

Giannini

Montefiori

. Quality of life of people living with HIV, preliminary results from IANUA (Investigation on Antiretroviral Therapy) study. J Int AIDS Soc 2014; 17: 19581–19581.

21.

Hughes

Jelsma

Maclean

. The health-related quality of life of people living with HIV/AIDS. Disabil Rehabil 2004; 26: 371–376.

22.

Phiri

Bussmann

. Psychosocial factors affecting medication adherence among HIV-1 infected adults receiving combination antiretroviral therapy (cART) in Botswana. AIDS Res Hum Retroviruses 2010; 26: 685–691.

23.

Vranceanu

Safren

. The relationship of post-traumatic stress disorder and depression to antiretroviral medication adherence in persons with HIV. AIDS Patient Care STDs 2008; 22: 313–321.

24.

Sledjeski

Delahanty

Bogart

. Incidence and impact of posttraumatic stress disorder and comorbid depression on adherence to HAART and CD4+ counts in people living with HIV. AIDS Patient Care STDs 2005; 19: 728–736.

25.

Springer

Dushaj

Azar

. The impact of DSM-IV mental disorders on adherence to combination antiretroviral therapy among adult persons living with HIV/AIDS: a systematic review. AIDS Behav 2012; 16: 2119–2143.

26.

Nakimuli-Mpungu

Bass

Alexandre

. Depression, alcohol use and adherence to antiretroviral therapy in sub-Saharan Africa: a systematic review. AIDS Behav 2012; 16: 2101–2118.

27.

Starace

Ammassari

Trotta

. Depression is a risk factor for suboptimal adherence to highly active antiretroviral therapy. J Acquir Immune Defic Syndr 1999 2002; 31: S136–S139.

28.

Gross

Bellamy

Chapman

. The effects of a problem solving-based intervention on depressive symptoms and HIV medication adherence are independent. PloS One 2014; 9: e84952–e84952.

29.

DiMatteo

. Enhancing patient adherence to medical recommendations. JAMA 1994; 271: 79, 83–79, 83.

30.

DiMatteo

Lepper

Croghan

. Depression is a risk factor for noncompliance with medical treatment: meta-analysis of the effects of anxiety and depression on patient adherence. Arch Intern Med 2000; 160: 2101–2107.

31.

Dantzer

O’Connor

Freund

. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci 2008; 9: 46–56.

32.

Lawson

Kelley

Dantzer

. Intracerebroventricular administration of HIV-1 Tat induces brain cytokine and indoleamine 2,3-dioxygenase expression: a possible mechanism for AIDS comorbid depression. Brain Behav Immun 2011; 25: 1569–1575.

33.

Del Guerra

Fonseca

JLI

Figueiredo

. Human immunodeficiency virus-associated depression: contributions of immuno-inflammatory, monoaminergic, neurodegenerative, and neurotrophic pathways. J Neurovirol 2013; 19: 314–327.

34.

Yunusa

Njoku

Obembe

. Self-reported adherence to treatment: a study of socioeconomic factors and psychiatric morbidity among male and female patients with HIV infection in Sokoto, Nigeria. Niger J Med 2014; 23: 33–39.

35.

Olisah

Baiyewu

Sheikh

. Adherence to highly active antiretroviral therapy in depressed patients with HIV/AIDS attending a Nigerian university teaching hospital clinic. Afr J Psychiatry 2010; 13: 275–279.

36.

Oladeji B, Fayemiwo S, Mosuro O, et al. Suicidal behaviour and quality of life in patients with HIV/AIDS in Nigeria. In: 19th international AIDS society world AIDS conference, Washington, DC, 2012.

37.

Kessler

Ustün

. The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). Int J Methods Psychiatr Res 2004; 13: 93–121.

38.

Gureje

Uwakwe

Oladeji

. Depression in adult Nigerians: results from the Nigerian Survey of Mental Health and Well-being. J Affect Disord 2010; 120: 158–164.

39.

Bella

Omigbodun

. Social phobia in Nigerian university students: prevalence, correlates and co-morbidity. Soc Psychiatry Psychiatr Epidemiol 2009; 44: 458–463.

40.

Oladeji

Makanjuola

Esan

. Chronic pain conditions and depression in the Ibadan Study of Ageing. Int Psychogeriatr IPA 2011; 23: 923–929.

41.

Gelaye

Williams

Lemma

. Diagnostic Validity of the Composite International Diagnostic Interview (CIDI) depression module in an East African population. Int J Psychiatry Med 2013; 46: 387–405.

42.

Gureje

Obikoya

. The GHQ-12 as a screening tool in a primary care setting. Soc Psychiatry Psychiatr Epidemiol 1990; 25: 276–280.

43.

Gureje

Lasebikan

Kola

. Lifetime and 12-month prevalence of mental disorders in the Nigerian Survey of Mental Health and Well-Being. Br J Psychiatry 2006; 188: 465–471.

44.

Gureje

Ogunniyi

Kola

. Functional disability among elderly Nigerians: results from the Ibadan Study of Ageing. J Am Geriatr Soc 2006; 54: 1784–1789.

45.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders : DSM-IV-TR, Washington, DC: American Psychiatric Association, 2000.

46.

World Health Organisation. The ICD-10 classification of mental and behavioural disorders: clinical descriptions and diagnostic guidelines. Geneva: World Health Organization, 1992.

47.

Olley

Seedat

Nei

. Predictors of major depression in recently diagnosed patients with HIV/AIDS in South Africa. AIDS Patient Care STDs 2004; 18: 481–487.

48.

Jagannath

Unnikrishnan

Hegde

. Association of depression with social support and self-esteem among HIV positives. Asian J Psychiatry 2011; 4: 288–292.

49.

Qiu

Luo

Cheng

. [Emotional problems and related factors in patients with HIV/AIDS]. Zhong Nan Da Xue Xue Bao Yi Xue Ban 2014; 39: 835–841.

50.

Ngo

Wagner

Nakasujja

. Effectiveness of antidepressants and predictors of treatment response for depressed HIV patients in Uganda. Int J STD AIDS 2015; 26: 998–1006.

51.

Wang

Zhang

Ruan

. Depressive symptoms and social support among people living with HIV in Hunan, China. J Assoc Nurses AIDS Care JANAC 2014; 25: 568–576.

52.

Erdbeer

Sabranski

Sonntag

. Everything fine so far? Physical and mental health in HIV-infected patients with virological success and long-term exposure to antiretroviral therapy. J Int AIDS Soc 2014; 17: 19673–19673.

53.

Gill

Hamer

Simon

. No room for complacency about adherence to antiretroviral therapy in sub-Saharan Africa. AIDS Lond Engl 2005; 19: 1243–1249.

54.

Ferguson

Donnelly

Hooper

. Adherence to antiretroviral therapy and its impact on clinical outcome in HIV-infected patients. J R Soc Interface R Soc 2005; 2: 349–363.

55.

Piacenti

. An update and review of antiretroviral therapy. Pharmacotherapy 2006; 26: 1111–1133.

56.

Martin

Elliott-DeSorbo

Wolters

. Patient, caregiver and regimen characteristics associated with adherence to highly active antiretroviral therapy among HIV-infected children and adolescents. Pediatr Infect Dis J 2007; 26: 61–67.

57.

Ances

Ellis

. Dementia and neurocognitive disorders due to HIV-1 infection. Semin Neurol 2007; 27: 86–92.

58.

Watkins

Treisman

. Cognitive impairment in patients with AIDS – prevalence and severity. HIVAIDS 2015; 7: 35–47.

59.

Graetz

. Multidimensional properties of the General Health Questionnaire. Soc Psychiatry Psychiatr Epidemiol 1991; 26: 132–138.

60.

Gao

Luo

Thumboo

. Does the 12-item General Health Questionnaire contain multiple factors and do we need them? Health Qual Life Outcomes 2004; 2: 63–63.

61.

Demyttenaere

Bruffaerts

Posada-Villa

. Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys. JAMA 2004; 291: 2581–2590.

62.

Olagunju

Ogundipe

Erinfolami

. Toward the integration of comprehensive mental health services in HIV care: an assessment of psychiatric morbidity among HIV-positive individuals in sub-Saharan Africa. AIDS Care 2013; 25: 1193–1198.

63.

Ofovwe

. Psychological disorders among human immunodeficiency virus-infected adults in southern Nigeria. Afr J Reprod Health 2013; 17: 177–182.

64.

Nel

Kagee

. The relationship between depression, anxiety and medication adherence among patients receiving antiretroviral treatment in South Africa. AIDS Care 2013; 25: 948–955.

65.

Memiah

Shumba

Etienne-Mesubi

. The effect of depressive symptoms and CD4 count on adherence to highly active antiretroviral therapy in sub-Saharan Africa. J Int Assoc Provid AIDS Care 2014; 13: 346–352.

66.

Smith

Stebbins

Bartt

. White matter anisotropy and depression symptoms in patients with HIV. J Neuropsychiatry Clin Neurosci 2008; 20: 458–465.

67.

Langford

Baron

Joy

. Contributions of HIV infection in the hypothalamus and substance abuse/use to HPT dysregulation. Psychoneuroendocrinology 2011; 36: 710–719.

68.

Paul

Brickman

Navia

. Apathy is associated with volume of the nucleus accumbens in patients infected with HIV. J Neuropsychiatry Clin Neurosci 2005; 17: 167–171.

69.

Stubbe-Drger

Deppe

Mohammadi

. Early microstructural white matter changes in patients with HIV: a diffusion tensor imaging study. BMC Neurol 2012; 12: 23–32.

70.

Safren

Skolnik

. Optimal recall period and response task for self-reported HIV medication adherence. AIDS Behav 2008; 12: 86–94.

71.

Walsh

Mandalia

Gazzard

. Responses to a 1 month self-report on adherence to antiretroviral therapy are consistent with electronic data and virological treatment outcome. AIDS 2002; 16: 269–277.

72.

Liu

Golin

Miller

. A comparison study of multiple measures of adherence to HIV protease inhibitors. Ann Intern Med 2001; 134: 968–977.

73.

Ammassari

Trotta

Murri

. Correlates and predictors of adherence to highly active antiretroviral therapy: overview of published literature. J Acquir Immune Defic Syndr 2002; 31: S123–S127.

74.

Martin

Elliott-DeSorbo

Wolters

. Patient, caregiver and regimen characteristics associated with adherence to highly active antiretroviral therapy among HIV-infected children and adolescents. Pediatr Infect Dis J 2007; 26: 61–67.

75.

Taiwo

Kass

. Post-consent assessment of dental subjects’ understanding of informed consent in oral health research in Nigeria. BMC Med Ethics 2009; 10: 11–11.

76.

Wilson

Carter

Berg

. Improving the self-report of HIV antiretroviral medication adherence: is the glass half full or half empty? Curr HIV/AIDS Rep 2009; 6: 177–186.

77.

Chkhartishvili

Rukhadze

Svanidze

. Evaluation of multiple measures of antiretroviral adherence in the Eastern European country of Georgia. J Int AIDS Soc 2014; 17: 18885–18885.

78.

Mills

Nachega

Buchan

. Adherence to antiretroviral therapy in sub-Saharan Africa and North America: a meta-analysis. JAMA 2006; 296: 679–690.