Depression and patterns of self-reported adherence to antiretroviral therapy in Rwanda

Abstract

We determined the prevalence of depression in HIV-infected adults on antiretroviral therapy in rural Rwanda and measured the association of depression with non-adherence. In all, 292 patients on antiretroviral therapy for ≥6 months were included. Adherence was self-reported by four-day recall, two- and seven-day treatment interruptions, and the CASE Index, which is a composite score accounting for difficulty taking medications on time, the average number of days per week a dose is missed, and the most recent missed dose. A total of 84% and 87% of participants reported good adherence by the four-day recall and CASE Index, respectively; 13% of participants reported two-day treatment interruptions; and 11% were depressed. Depression was significantly associated with two-day treatment interruptions but not with other measures of non-adherence. Self-reported adherence to antiretroviral therapy in rural Rwanda is high. Adherence assessments that do not consider treatment interruptions may miss important patterns of non-adherence, which may be especially prevalent among depressed individuals. Mental health interventions incorporated into routine HIV care may lead to improvements in mental health and adherence.

Keywords

HIV AIDS Africa antiretroviral therapy screening viral disease treatment adherence depression

Introduction

Since 2000, Rwanda has made large strides in healthcare, with the highest immunisation coverage in Africa, declining infant and under-five mortality, and rapidly increasing life expectancy.^1–4 The Rwandan government has prioritised improving the accessibility and affordability of health services and the quality of care, increasing government spending on health since the 1994 genocide and decentralising services to the district level.² Similar progress has been seen in HIV treatment in Rwanda. With an estimated HIV prevalence of 3%,¹ approximately 99,000 patients started antiretroviral therapy (ART) between 2005 and 2010, and of these, 84% were retained on ART as of June 2010.⁵ Every year, there is evidence that patients are initiating treatment earlier, with the proportion of patients presenting with stages 3 and 4 dramatically declining from 64% in 2005 to 24% in 2010.⁵

Previous studies in sub-Saharan Africa have addressed ART adherence rates but very few have looked at concurrent challenges to adhere to the medications. Adherence rates range from 77% overall adherence with some studies greater than 90%.^6–8 However, there exist significant social, economic, and structural barriers to accessing HIV treatment, adhering to treatment, and remaining in care,^9,10 and more research is needed to understand risk factors for poor adherence in order to better support patients over the course of lifelong therapy.⁸ Multiple factors must be considered, including mental health, social capital, and variables related to poverty.^11,12

HIV-positive individuals have high rates of depression,^13–15 and depressive symptoms have been shown to predict HIV disease progression and mortality in the United States. ^16,17 There is evidence that depression is a risk factor for non-adherence to ART in developing contexts,^10,15,18 and management of depressive symptoms is a priority for adherence.¹⁹ One multi-centre study conducted in the US found that intermittent use of ART, defined as stopping and starting again at least 1 time, explained the observed association between depression and higher mean viral load.¹⁷

In this paper, we summarise results from a cross-sectional study of patients on ART in the rural Burera District of Rwanda, where health services are provided through the Ministry of Health in partnership with Partners In Health. These data were collected prior to the initiation of an integrated model using community health workers through which patients on HIV treatment receive increased social, economic, and psychological support. The objectives of this study were to (1) estimate the levels of adherence in HIV-infected patients on ART; (2) report the prevalence of depression, and (3) examine the association between depression and ART adherence.

Methods

Study setting

The study was conducted in three health centres in Burera district, a mountainous rural region in northern Rwanda where the majority of the population relies on subsistence agriculture. Prior to 2008, this district of approximately 350,000 people did not have a district hospital and had one practicing physician, and there were no non-governmental implementation partners supporting the Ministry of Health (MOH) in ART delivery in the district. In 2008, the Rwandan Ministry of Health partnered with Partners In Health and the Clinton Health Access Initiative to strengthen healthcare services in Burera. At that time, approximately 800 patients in the district were on ART according to national guidelines. A key part of the healthcare scale-up was to expand quality HIV treatment services to all health centres with additional services for patients on ART, including nutritional support, community healthcare workers, and minimisation of out-of-pocket expenses such as user fees and transportation.

Study design and enrollment

This is a cross-sectional study that recruited participants from June to September 2008 at the three rural health centres. Individuals visiting the clinic for routine visits and who had been on ART for 6 months or more were eligible for inclusion in the study. Exclusion criteria were age less than 18 years and receipt of ART for less than 6 months. On randomly chosen clinic days, study staff explained the study to all eligible HIV-infected patients waiting for their clinic appointment and invited them to participate. Two hundred ninety-two patients who provided informed consent were enrolled in the study. At the study visit a trained study nurse administered a questionnaire to assess self-reported adherence and depression.

Outcome assessment and definitions

Adherence was measured using multiple self-report methods including four-day recall, two- and seven-day treatment interruptions, and the CASE Index. Self-reported adherence measures have been shown to correlate with viral load results,^20,21 and in rural Uganda, treatment interruptions greater than 48 hours have been shown to predict drug–resistance.⁶ The measures of adherence were designed to capture different aspects of adherence. For example, the four-day recall was used to assess recent adherence, where non-adherence was defined as missing one or more doses in the four days prior to interview. Two-day and seven-day treatment interruptions assess non-adherence on sequential days, which can have important implications for persistent viraemia and viral resistance.^6,17 Treatment interruptions suggest intermittent difficulties in adherence caused by missed clinic visits or stock outs, for example. Finally, the three-item CASE Index is a composite score that asks individuals to report three items: (1) the frequency of difficulty taking HIV medications on time, (2) the average number of days per week at least one dose of HIV medications was missed, and (3) how long ago s/he missed a dose of HIV medications.²² Composite scores from the CASE Index range from 3 to 16, with scores >10 indicating good adherence; however, pilot data suggested that participants had difficulty distinguishing between two response categories: missing a dose an average of “zero times per week” and “less than once a week.” We therefore combined these response categories for a total maximum score of 15 and used a CASE index score of 9 or less indicating suboptimal adherence.²²

Exposure assessment and definitions

Depression was measured using the Hopkins Symptoms Checklist-15 (HSCL-15), a commonly used tool in primary care settings that has been shown to correlate with other measures of depression.^23,24 Participants classify their degree of each of 15 symptoms such as low energy, poor appetite, sleep disturbances, and suicidal thoughts during the past week, scoring each response as 1–4 with higher numbers correlating with more severe depressive symptoms. The scale has been shown to have good reliability (Cronbach’s alpha 0.87) in an HIV-infected population in Rwanda.²⁵ Individuals with an average score of 1.75 or higher across the depressive symptoms are classified as depressed.

Statistical analysis

We report the following descriptive statistics: proportion of participants who were adherent (by each self-report method) and proportion depressed. To examine the association between depression with each self-reported measure of adherence, we used Fisher's exact test. Multivariable analysis was performed using logistic regression and adjusting for age, gender, and baseline CD4 count. Significance was determined by a two-sided p value of <0.05. Individuals with missing data were excluded from analysis. All data were analysed using the Stata version 12.

Ethics

All participants gave informed consent in Kinyarwanda. For the cross-sectional component of the study, patients were given 2500 Rwandan francs (approximately 5 USD) for transport reimbursement. All study procedures received ethical approval from the Rwandan National Research Ethics Committee and the Institutional Review Board of Partners Healthcare in Boston, MA.

Results

Of the 292 participants, 63% were women with an average age of 38.2 years. The median time on ART at the time of the survey was 19 months, and almost 50% of patients were WHO Stage III or IV at initiation of treatment (Table 1). Response rates for all of the adherence measures were 96–97%. Results were recorded for 97% of participants for depression.

Table 1.

Participant characteristics and self-reported adherence.

Characteristic	N	n (%) unless otherwise indicated
Women	292	184 (63.0)
Age in years^a	291	38.2 (11.8)
Time since starting ART in months^b	292	18.7 (14–29)
CD4 at ART initiation^b	284	187.5 (95–256.5)
WHO stage at ART initiation	288
Stage 1		31 (10.8)
Stage 2		118 (41.0)
Stage 3		128 (44.4)
Stage 4		11 (3.8)
Good CASE Index score^c	285	247 (86.7)
No missed doses on 4-day recall^d	288	241 (83.7)
Never had two day interruption^e	286	250 (87.4)
Never had seven day interruption^f	287	283 (98.6)

ART: antiretroviral therapy.

Mean (standard deviation).

Median (interquartile range).

95% confidence interval: 82.2–90.4.

95% confidence interval: 78.9–87.8.

95% confidence interval: 83.0–91.0.

95% confidence interval: 96.5–99.6.

By four-day recall, 84% of patients were adherent (95% CI: 78.9–87.8%), and 87% (95% CI: 82.2–90.4%) of patients met the cutoff for good adherence by the modified CASE Index (Table 1). Treatment interruptions were uncommon, with 12.6% (95% CI: 9–17%) reporting two-day interruptions and 1.4% (95% CI: 0.4–3.5%) reporting seven-day interruptions in the last year. Eleven percent of participants were depressed (Table 2).

Table 2.

Univariable associations between adherence and depression.

		Adherent by CASE score			Adherent by 4-day recall			Adherent by 2-day treatment interruption
		N = 275^b			N = 278^b			N = 276^b
Depressed	n (%)	Yes	No	p ^a	Yes	No	p ^a	Yes	No	p ^a
Yes	32 (11.2)	10.1%	15.8%	0.27	10.4%	17.0%	0.15	10.0%	22.2%	0.05
No	254 (88.8)	89.9%	84.2%		89.6%	83.0%		90.0%	77.8%

By Fisher's exact test.

Sample size once individuals with missing outcome or exposure data were excluded.

Due to the small number of individuals who reported a seven-day treatment interruption, we examined risk factors for non-adherence as reported by four-day recall, CASE Index, and two-day treatment interruption. Patients reporting two-day treatment interruptions were more likely to be depressed compared to those who did not report any two-day interruptions (p = 0.05). On four-day recall, 17% of non-adherent patients were depressed compared to 10% of adherent patients (p = 0.15). Depression was also not significantly related to CASE scores (Table 2). In a multivariable analysis adjusting for age, gender, and baseline CD4 cell count, depression remained associated with two-day treatment interruptions (OR: 2.7; 95%CI [1.1, 6.9]; p = 0.03) but was not significantly associated with adherence as reported using the CASE index (OR: 1.7; 95% CI [0.6, 4.8]; p = 0.17) or four-day recall (OR: 1.7; 95%CI [0.66, 4.3]; p = 0.28). Among individuals who were not depressed, concordancy was high between two-day treatment interruptions and both the CASE index and four-day recall: only 10% of those who were adherent by the CASE Index or four-day recall reported a two-day treatment interruption. In contrast, among individuals who were depressed, 29% and 25% of those who met the definition for good adherence using the CASE Index and 4-day recall, respectively, reported a treatment interruption of at least two days.

Discussion

Despite high rates of self-reported adherence to ART in this cohort, depression was a risk factor for two-day treatment interruptions. This suggests that findings from previous studies linking depression to non-adherence, including treatment interruptions, may be generalisable to this rural East African setting.^{10,15,17–19} Our findings build on prior work to show that while depression was significantly associated with two-day treatment interruptions, it was not significantly associated with other types of self-reported adherence. This indicates that individuals who are depressed may be at higher risk for adherence patterns associated with drug resistance and that interruptions may be present even in the absence of non-adherence according to other measures. Simple assessments with tools such as four-day recall or the CASE Index may fail to detect meaningful patterns of suboptimal adherence if treatment interruptions are not also considered.

Screening for depression is also important as it is common and a major burden of disease worldwide.²⁶ Noting high rates of depression in HIV-infected populations^14,27,28 and evidence that depression is linked to barriers at every stage of care,^27–30 prevention, early detection, and treatment of depression should be an integrated part of HIV treatment programmes. Further, general community-level support for patients on treatment could contribute to reductions in depression, as a recent study demonstrated a 44% reduction in prevalence of depression among individuals receiving a community-based accompaniment model.³¹

The high levels of self-reported adherence may have limited power to detect relationships between depression and the other measures of self-reported adherence. Pharmacy refill data or other more objective measures were not available for this study. If self-reported adherence tended to be over-estimated due to social desirability or other factors, this would be expected to reduce power providing that over-reporting was not associated with depression. Hence over-reporting would be unlikely explain the observed association between two-day treatment interruptions and depression. A second limitation is that this study was conducted among patients that survived at least the first six months of ART. If individuals with severe depression were more likely to die or exit care before six months, this may have also limited our ability to detect an association between depression and the other measures of self-reported adherence at six months.

Conclusion

These results indicate that adherence in rural Rwanda is high and that depression is linked to two-day treatment interruptions. Given the low cost and ease of assessing adherence through self-report, it is a commonly used method to screen for non-adherence. Because treatment interruptions have been shown to be related to viral resistance, careful monitoring of treatment interruptions, especially among patients with depression, may be critical to detecting important patterns of non-adherence.

Footnotes

Acknowledgements

The authors express their gratitude to and admiration for all of the patients and healthcare professionals who contributed to this study. We give specific appreciation to the work done by Dr Valens Hafashimana, Dr Michael Rich, Dr Patrick Almazor, Anne Sosin, and all of our colleagues in Burera District.

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: The Doris Duke Clinical Research Fellowship Program provided funding for EBW to travel to Rwanda that made this work possible. BHG and MF received support from the Department of Global Health and Social Medicine Research Core (at Harvard Medical School).

References

World Health Organization, Country Health Profile, Rwanda. http://www.who.int/countries/rwa/en/ (accessed 23 September 2012).

Logie

Rowson

Ndagije

. Innovations in Rwanda’s health system: looking to the future. Lancet 2008; 372: 256–261.

Farmer

Nutt

Wagner

. Reduced premature mortality in Rwanda: lessons from success. BMJ 2013; 346: f65–f65.

National Institute of Statistics of Rwanda. The third integrated household living conditions survey (EICV3). 2012. www.statistics.gov.rw/publications/third-integrated-household-living-conditions-survey-eicv-3-main-indicators-report (accessed 17 March 2013).

Nsanzimana

Ruton

Lowrance

. Cell phone-based and internet-based monitoring and evaluation of the national antiretroviral treatment program during rapid scale-up in Rwanda: TRACnet, 2004–2010. J Acquir Immune Defic Syndr 2012; 59: e17–e23.

Oyugi

Byakika-Tusiime

Ragland

. Treatment interruptions predict resistance in HIV-positive individuals purchasing fixed-dose combination antiretroviral therapy in Kampala, Uganda. AIDS 2007; 21: 965–971.

Mills

Nachega

Buchan

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

Nachega

Mills

Schechter

. Antiretroviral therapy adherence and retention in care in middle-income and low–income countries: current status of knowledge and research priorities. Curr Opin HIV AIDS 2010; 5: 70–77.

Hardon

Akurut

Comoro

. Hunger, waiting time and transport costs: time to confront challenges to ART adherence in Africa. AIDS Care 2007; 19: 658–665.

10.

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.

11.

Castro

. Adherence to antiretroviral therapy: merging the clinical and social course of AIDS. PLoS Med 2005; 2: e338–e338.

12.

Reda

Biadgilign

. Determinants of adherence to antiretroviral therapy among HIV-infected patients in Africa. AIDS Res Treat 2012; 2012: 574656–574656.

13.

Franke

Kaigamba

Socci

. Improved retention associated with community-based accompaniment for antiretroviral therapy delivery in rural Rwanda. Clin Infect Dis 2013; 56: 1319–1326.

14.

Ciesla

Roberts

. Meta-analysis of the relationship between HIV infection and risk for depressive disorders. Am J Psychiatry 2001; 158: 725–730.

15.

Wagner

Goggin

Remien

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

16.

Ickovics

Hamburger

Vlahov

. Mortality, CD4 cell count decline, and depressive symptoms among HIV-seropositive women: longitudinal analysis from the HIV epidemiology research study. JAMA 2001; 285: 1466–1474.

17.

Carrico

Riley

Johnson

. Psychiatric risk factors for HIV disease progression: the role of inconsistent patterns of antiretroviral therapy utilization. J Acquir Immune Defic Syndr 2011; 56: 146–150.

18.

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.

19.

Gonzalez

Batchelder

Psaros

. Depression and HIV/AIDS treatment nonadherence: a review and meta-analysis. J Acquir Immune Defic Syndr 2011; 58: 181–187.

20.

Sethi

Celentano

Gange

. Association between adherence to antiretroviral therapy and human immunodeficiency virus drug resistance. Clin Infect Dis 2003; 37: 1112–1118.

21.

Chesney

Morin

Sherr

. Adherence to HIV combination therapy. Soc Sci Med 2000; 50: 1599–1605.

22.

Mannheimer

Mukherjee

Hirschhorn

. The CASE adherence index: A novel method for measuring adherence to antiretroviral therapy. AIDS Care 2006; 18: 853–861.

23.

Derogatis

Lipman

Covi

. SCL-90: an outpatient psychiatric rating scale—preliminary report. Psychopharmacol Bull 1973; 9: 13–18.

24.

Lee

Schulberg

Raue

. Concordance between the PHQ-9 and the HSCL-20 in depressed primary care patients. J Affect Disord 2007; 99: 139–145.

25.

Epino HM, Rich ML, Kaigamba F, et al. Reliability and construct validity of three health-related self-report scales in HIV-positive adults in rural Rwanda. AIDS Care Epub ahead of print 19 Mar 2012.

26.

Murray

Lopez

. Global mortality, disability, and the contribution of risk factors: global burden of disease study. Lancet 1997; 349: 1436–1442.

27.

Tegger

Crane

Tapia

. The effect of mental illness, substance use, and treatment for depression on the initiation of highly active antiretroviral therapy among HIV-infected individuals. AIDS Patient Care STDS 2008; 22: 233–243.

28.

Bhatia

Hartman

Kallen

. Persons newly diagnosed with HIV infection are at high risk for depression and poor linkage to care: results from the steps study. AIDS Behav 2011; 15: 1161–1170.

29.

Holzemer

Corless

Nokes

. Predictors of self-reported adherence in persons living with HIV disease. AIDS Patient Care STDS 1999; 13: 185–197.

30.

Fairfield

Libman

Davis

. Delays in protease inhibitor use in clinical practice. J Gen Intern Med 1999; 14: 395–401.

31.

Thomson DR, Rich ML, Kaigamba F, et al. Community-based accompaniment and psychosocial health outcomes in HIV-infected adults in Rwanda: a prospective study. AIDS Behav Epub ahead of print 27 Feb 2013.