Coping Self-Efficacy as a Predictor of Adherence to Antiretroviral Therapy in Men and Women Living with HIV in Kenya

Abstract

The purpose of this study was to evaluate the relationship between the coping self-efficacy (CSE) scale and adherence to HIV medication in men and women enrolled in a large HIV treatment program in Kenya. Data were collected from a sample of 354 volunteers attending Nazareth Hospital's nine satellite clinics located in parts of Nairobi, and the central province of Kenya. A social demographic survey, Adult Clinical Trials Group adherence questionnaire, and CSE scale were used to obtain information. Descriptive statistics and logistic regressions were performed to analyze data and to test study hypotheses. Females were less likely to be nonadherent than males: the odds of adherence for females were 3.7 of the odds of adherence for males. Controlling for gender, CSE was significant. Adherence to antiretroviral therapy can be partially explained by CSE. Efforts aimed at building self-efficacy are likely to improve and maintain adherence to HIV and other medication. Implications, limitations, and future directions are discussed.

Introduction

T he proportion of men and women living with HIV who have been prescribed antiretroviral therapy (ART) in Kenya has dramatically increased from less than 10% in 2003 to over 75% in 2009.¹ This rise in ART access is projected to create beneficial health outcomes in the fight against the HIV pandemic in Kenya, including the reduction of HIV-related deaths, slowing diseases progression, and general improvement of health among people living with HIV.²

Despite widespread success of ART as a treatment for HIV/AIDS, adherence remains an issue.⁴ While recent ritonavir-boosted protease inhibitors and non-nucleoside reverse transcriptase inhibitors (NNRTIs) require moderate adherence of 70%,³ the literature also suggests that regardless of the regimen, higher levels of adherence (95–100%) may be required in the early phase of treatment before viral suppression is achieved.⁴ Furthermore, nonadherence is shown to be the main reason for virologic failure, which allows the virus to replicate rapidly, increasing the viral load, and consequently, allowing it to develop drug-resistant viral strains.⁵ Thus, a patient's ability to maintain high levels of adherence remains critical to optimizing treatment benefits.

Despite reports that adherence to ART among people in sub-Saharan Africa is as good as, if not better than, adherence to ART in developed countries, it is not clear whether patients' high motivation to adhere will continue over time as treatment is still novel in Africa.⁶ Moreover, recent studies indicate that an increasing proportion of participants (37.5%) in South Africa are not adherent to ART.⁴ It appears that adherence is still an issue, thus, it is important to identify adherences' best practices and the strategies that will promote and sustain high adherence consistently in this population.⁷

The literature suggests several emerging factors that influence adherence to ART including lack of support services and environmental constraints. Decreasing barriers such as lack of peer support in Haiti, introducing home-based counseling, home delivery of drugs and daily access to a cell phone in Kenya have demonstrated positive effect on adherence to ART.^8
–10 In this study, we explored coping self-efficacy (CSE) as a factor that may influence one's adherence to ART practices.

Within the literature, coping is defined as a person's constantly changing cognitive and behavioral efforts to manage specific external and internal demands that are appraised as exceeding his or her resources. Lazarus and Folkman's¹¹ theory of stress, appraisal, and coping postulates that there are two forms of coping: emotion focused and problem focused. While emotion-focused coping is directed toward regulating emotional distress to manage a problem, problem-focused coping is directed at defining the problem, enlisting various alternatives to solving it, weighing these alternatives in terms of their advantages and disadvantages, choosing among the available alternatives, and acting upon the choice. Notably, some problems may not have solutions. Folkman and Moskowitz¹² suggest that acquisition of meaning is another form of coping that can foster an understanding of why an event happened and promotes an adjustment to problems that may appear to have possible or known solutions. Thus, coping may target the regulation of emotional distress, strategies and efforts to eliminate the problem, and a search for the meaning of negative and unchangeable events in life.

Based upon the tenets of CSE, which is derived from stress, appraisal, and coping theory and the theory of self-efficacy, it is plausible that CSE can be associated with adherence to ART.^13,14 Caltz et al.¹⁴ suggest that people living with HIV are confronted with many stressors, such as fear of disclosure of HIV, stigma, discrimination, social exclusion, symptoms of AIDS, and HIV medication side effects, and it is these stressors that may interfere with patient ability to adhere to HIV medication.

Bandura¹³ defined self-efficacy as a person's belief about his or her ability and capacity to accomplish a task. A person's self-efficacy can be expressed in terms of the level of task demands and the belief that he or she can sustain efforts toward a goal despite impeding conditions. A person who perceives himself or herself as efficacious in coping with stress and in accomplishing difficult tasks with good strategies, more effort, and prolonged persistence can be said to have strong CSE.^11,15
–17 Coping self-efficacy can aid individuals in eliminating, buffering, or reducing their stress when faced with challenges or threats.¹⁷ Additionally, coping beliefs may relate to the execution of the courses of action required to manage prospective situations and to deal with barriers that arise in order to ensure sustained action.^11,15
–17

The purposes of this study were to investigate whether CSE is related to adherence to ART and whether each of the subscales of CSE (use of problem-focused coping [UPFC], stop unpleasant emotions and thoughts [SUET], and get support from friends and family [GSFF]) aide in predicting adherence to ART.

Methods

Participants

In accordance to the international guidelines for research conducted outside the United States, this research was submitted to the Institutional Research and Ethics Committee, Moi University, Eldoret, Kenya. The Institutional Research and Ethics committee approved the study protocol in August 2009. Subsequently, another application was submitted to IRB Seton Hall University, South Orange, New Jersey, and an approval was obtained in October 2009.

Convenience sampling was used to recruit 403 men and women volunteers, aged 18–64 years, living with HIV, and having ART prescribed, who were being cared for and treated for HIV in any of the nine comprehensive care and treatment centers under the umbrella of a health care provider serving parts of the Central Province and the metropolitan city of Nairobi, Kenya. Of the 403 who consented to participate and take the self-administered survey tools, 354 completed the questionnaires. Research assistants were recruited and trained on human subject research ethics, how to recruit patients, how to assist in the administration of the survey tools, and how to maintain confidentiality of patient information.

Recruitment

Participants were recruited through flyers displayed in physicians, counselors, and pharmacies' waiting areas at the nine approved sites of a major healthcare provider. The flyers contained information including (1) the purpose of the study, (2) eligibility (inclusion and exclusion criteria), and (3) the administration procedures of the survey tools. Interested participants either completed recruitment forms and placed them in an open box or presented themselves to the research assistants' desk. Participants who completed the forms were contacted and given an appointment. Written informed consent was obtained from all 403 participants during the appointment before the participants completed the survey tools. All participants received a token of appreciation worth US $7 for their time, regardless of their ability to fully complete the survey tools.

Inclusion and exclusion criteria

Men and women living with HIV who were: at least 18 years old (the legal age at which a person can give informed consent in Kenya); able to communicate in either Kiswahili, Kikuyu, or English; undergoing ART; and attending any of the HIV treatment clinics within Nairobi and the Central Province of Kenya. Individuals who were: under 18 years of age, prisoners, pregnant, and/or could not make competent or rational decisions were excluded from the study. To ensure the anonymity of the subjects, administration of the survey was carried out in private rooms at each of the nine sites, and all surveys were precoded to ensure that participants did not self-identify. The respondents self-administered three survey tools.

Measures

Demographic characteristics survey

The first was a demographic characteristics survey, gathering demographic information on participants' gender, age, education, employment status, marital status, and religion. Age was categorized as 18–29, 30–40, 41–50, and 51–64. Education was categorized as no formal education, primary, secondary, and college. Participants self-reported their employment status as employed, self-employed, or not employed. There were five categories for marital status: single, married, divorced, separated, and widowed. Religion was categorized as African Traditional religion, Roman Catholic, Protestant, or Muslim. Demographic characteristics were used to identify if any characteristic was related to adherence to ART as measured in the previous 4 days to be used a covariate(s) in the testing of the study hypotheses.

Coping self-efficacy

The CSE survey is a 26-item measure that measures an individual's belief that he or she can perform behaviors important to adaptive coping by sorting out what is controllable and what is uncontrollable.^16,17 On a scale of 0–10, participants were asked to indicate how confident they were that they can do certain tasks when things are not going well. For example, “When things are not going well with you, how confident are you that you can talk positively to yourself?” or “Sort out what can be changed and what cannot be changed” or “Get emotional support from friends and family.”¹⁶ The scale ranges from 0 (cannot do at all), through 5 (moderately certain I can do), to 10 (certain I can do). An overall CSE score is created by summing the item ratings (α=0.95, scale mean=137.4, standard deviation [SD]=45.6).

The third tool was the Adult Clinical Trial Group (ACTG) adherence questionnaire; this tool seeks to determine adherence to ART.¹⁷ Respondents self-reported adherence to HIV medication by completing the ACTG questionnaire. They were asked whether they had taken required doses and pills in each of the 4 days prior to the day they were responding to the questionnaire. They were also asked whether they took pills 2 h before or 2 h later than instructed. Participants were asked if they had missed a dose(s) or pills in the previous 28 days and whether they had ever skipped HIV medication. Additionally, participants responded to a list of 12 items that sought to determine some of the reasons for missed medication. Among reasons given for missed medication included “being away from home,” “being busy with other things,” “forgot,” “had too many pills to take,” and “did not want others to notice you taking medicine.” The reliability and validity of the ACTG questionnaire is well established.^17,18 For the purpose of this study, adherence was defined as taking all doses and pills, with compliance to all dosing instructions for the four days prior to completing the survey. A short period (4 days) was used to determine adherence because literature shows that use of fewer days reduces recall bias and has shown evidence of being closer to adherence as measured through electronic monitoring.¹⁹

Results

Analysis

Descriptive statistics were used to describe and summarize the sample characteristics and to summarize and present the data through tables and figures.²⁰ The majority of the participants were female (71.47%). Participants' ages were categorized into four classes: 18–30, 31–40, 41–50, and 51–64. Forty-two percent (42.2) of participants ranged from 31 to 40 years. The participants were classified into four religious groups: Catholic, Protestant, Muslim, and African religious traditional. The sample consisted of diverse religious affiliations, the majority being Protestants (49.2%) and Catholics (39%), while Muslims (0.6%) and Africans (11.3%) were the minority.

Marital status was classified into five groups: never married, married, separated, divorced, and widowed. The majority of the participants were married (56.5%), and 16.9% were widowed. More men were married (74%) than women (49%). A greater percentage of women were widowed (19%) than men (11%). A greater proportion of women were separated (13.6%) than men (5.7%). Only 4.7% of the participants were divorced.

Most participants had completed only a primary education (42.1%) or a secondary education (38.4%). A greater percentage of males (50.4%) had a secondary education than did females (33.6%). Most men and women (71.2%) were engaged in some form of economic activity (formal, informal, or farming sectors). Twenty-eight (28) percent were unemployed. A greater percentage of men were employed in the formal sector (40.6%) than women (32.2%). A greater percentage of women were employed in the farming sector (27.7%) or were not employed (30.4%) than men (18.6% and 24%, respectively).

The majority of respondents were medication adherent. More specifically, 94.5% of the participants indicated that they had not missed any of their medication in the previous 4 days.

Participants' own reasons for missing medications were as follows: “busy with other things” (11.3%); “simply forgot” (7.5%); “did not want others to notice you taking medications” (5.7%); “felt sick” (4.7%); “felt good” (4.8%); “had problem taking medication at specific times” (5.7%); and “slept through the dose” (3.8%). Only 1% indicated missing the dose because they ran out of pills.

All demographic variables were further examined to determine whether, as a group as well as individually, these variables predicted adherence to ART. A direct logistic regression analysis was performed on adherence to ART as an outcome with all six demographic predictors—gender, age, education, marital status, employment status, and religion—as predictor variables. A test of the full model with all six predictors against a constant-only model was not statistically reliable, χ² (19, n=354)=29.90, p<0.053. A further evaluation of the significance of the correlation between each of these predictors and outcomes indicated that gender was statistically significant as a predictor of adherence to ART, χ² (1, n=354), p<0.019, such that females were more adherent (less nonadherent) to ART. The odds of adherence among females were 3.7 (confidence interval [CI] 1.27–8.203) times greater than the odds for males.

Two direct logistic regression analyses were performed to assess (1) the relationship between CSE (composite) and adherence to ART and (2) the relationship between the subscales of CSE—UPFC, SUET, GSFF—and adherence to ART. These two regressions were run to examine CSE as a predictor of adherence and to find if each subscale of CSE was a significant predictor of adherence to ART.

The first model was fitted to assess the relationship between CSE and adherence to ART. The model was statistically reliable χ² (2, n=354), p=0.000, (p<0.05). Controlling for gender, the result was significant (p<0.05); indicating a positive significant relationship between CSE and adherence to ART. Thus, it was concluded that there is a significant relationship between CSE and adherence to HIV ART (Table 1).

Table 1.

Full Coping Self-Efficacy and Adherence to Antiretroviral Therapy (n=354)

								95% CI for EXP(B)
		B	SE	Wald	Df	Sig.	Exp(B)	Lower	Upper
Step 1	Gender(1)	1.145	0.470	5.929	1	0.015	3.141	1.250	7.892
	CSE	−0.022	0.006	13.895	1	0.000	0.978	0.967	0.990
	Constant	−0.014	0.845	0.000	1	0.987	0.986

Note: Variable(s) entered on step 1: Gender.

CSE, Coping self-efficacy; SE, standard error; CI, confidence interval.

Another model was fitted to assess the relationship between the subscales of CSE and adherence to ART. The model was statistically reliable: χ² (4, n=354), p=0.00, (p<0.05). However, only SUET relationship with adherence to ART was statistically significant (p=0.022). Another model run without SUET was statistically reliable, thus confirming that UPFC was the only statistically reliable predictor of adherence to ART among the three subscales of CSE (Table 2).

Table 2.

Subscales of Coping Self-Efficacy and Adherence to Antiretroviral Therapy (n=354)

								95% CI for EXP(B)
		B	SE	Wald	Df	Sig.	Exp(B)	Lower	Upper
Step 1	Gender(1)	1.172	0.476	6.068	1	0.014	3.228	1.271	8.203
	UPFC	−0.006	0.024	0.054	1	0.816	0.994	0.949	1.042
	SUET	−0.047	0.021	5.275	1	0.022	0.954	0.916	0.993
	GSFF	0.001	0.009	0.005	1	0.944	1.001	0.983	1.019
	Constant	0.183	0.862	0.045	1	0.832	1.201

Note: Variable(s) entered on step 1: Gender, use problem focused coping=UPFC, stop unpleasant emotions and thoughts=SUET, get Support from Friends and family=GSFF.

SE, standard error; CI, confidence interval.

Discussion

Based upon the data from this study, the majority of respondents were medication adherent. Of the 354 patients who participated in the study, 94.5% of the participants indicated that they had not missed any of their medication in the four days prior to the study. These findings support prior research that found that the majority of men and women enrolled in HIV programs in Africa are adherent. Specific to this population of interest, a previous meta-analysis study that had assessed adherence to ART across sub-Saharan Africa estimated that 77% were adherent to ART.⁶ Our findings of high levels of adherence to ART further support the findings of Sidle et al.²¹ in a longitudinal study in Kenya that showed that 77% of patients on ART reported perfect adherence at all visits and 89% reported perfect adherence at 90% or more of visits in the first year on treatment and Uganda, Byakika-Tusiime and colleagues'²² study which also found favorable levels of adherence to ART.

Most importantly, as hypothesized, this study found a significant relationship between CSE and adherence to ART. Like all people, individuals living with HIV face the hassles of daily living as well as HIV-related stressors (as there is little evidence suggesting that HIV-stigma is declining). In fact, recent research shows that individuals living with HIV continue to report experiences of social isolation and discrimination resulting from being HIV positive.²³ However, despite these barriers, our findings support that individuals living with HIV can appraise these barriers as surmountable with a positive belief that appears to have a significant relationship with adherence to ART.¹ Although this study's objective was not to identify sources of self-efficacy, it is probable that individuals with high CSE are likely to seek coping resources and learning strategies to solve problems.

Availability of social resources in one's network may enhance one's ability to cope with stressors. A case in point is a recent study that sought to determine the effects of short message service (SMS) reminders on adherence to ART among patients attending a rural clinic in Kenya; results suggest that SMS reminders may be an important tool in achieving optimal adherence to ART.²⁴ Individual perception that these reminders would be available provides answers to the questions in the secondary appraisal where people ask themselves, “What can I do, given that I have a problem?” Use of SMS reminders eliminates the tendency to forget to take medication at the right time or to take the wrong pills or dosage due to forgetfulness or because of being busy.

Knowledge of HIV medication, motivation, and behavioral skills are also likely to increase individuals' confidence in their ability to adhere to treatment. Rongtavilit et al.²⁵ noted that knowledge of HIV medication and the effects of nonadherence, motivation, and behavioral skills can enhance individuals' confidence in their ability to adhere, which in turn, influences behavioral efforts.

The finding that the odds of adherence among females were 3.7 times greater than the odds for males was surprising given that fewer women were unmarried than men and that a greater percentage of women were widowed, divorced, or separated than men. One would expect women to have more stressors than men, thus, be less adherent to ART than men. This unexpected finding leads us to further question the factors that may account for higher levels of adherence in women, or the barriers that prevent men from being as adherent as women.

As with all studies, this study has its limitations. The cross-sectional design of the study may not show how CSE varies with time. Furthermore, causal inferences can neither be inferred nor implied from data obtained in this study. The use of respondents' self-reports on adherence to ART may have resulted in underreporting of nonadherence or over-estimating of adherence to ART. Additionally, the use of a convenience sample may limit the degree to which the findings of the present study can be generalized to other populations in Kenya and Africa. Finally, the low number of nonadherent respondents in this study limits the interpretation of cases predicted correctly as nonadherent. However, the study findings do support the need to replicate this study in other settings in Kenya and in Africa and also to find other variables that could be significant contributors to adherence to HIV medication.

The findings in this study have many implications on possible adherence interventions. Interventions are needed not only to create knowledge of ART regimens and understanding of instructions on how to take medication, but also those that enhance patients' self-assurance of their ability to manage the demands of their daily life.¹³ Because self-efficacy can be constructed through vicarious experiences, use of social networks of people living with HIV may help develop confidence in those facing difficulties but who are unsure in their ability to adhere to ART. Since social comparison is a primary factor in the self-appraisal of capabilities, such individuals are likely to believe in their capabilities when they see others successfully confront and manage similar tasks.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

National AIDS and STD Programs. Kenya AIDS Indicator Survey. Nairobi: Ministry of Health, 2008.

Paterson

, Swindells

, Mohr

et al. Adherence to protease inhibitor therapy and outcome in patients with HIV infection. Ann Intern Med, 2000; 133:21–30.

Bhat

, Ramburuth

, Singh

et al. Factors associated with poor adherence to anti-retroviral therapy in patients attending a rural health centre in South Africa. Eur J Clin Microbiol Infect Dis, 2010; 29:947–953.

Bangsberg

, Kroetz

, Deeks

. Adherence-resistance relationships to combination HIV antiretroviral therapy. Curr HIV/AIDS Rep, 2007; 4:65–72.

Wainberg

, Martinez-Cajas

, Brenner

et al. Strategies for the optimal sequencing of antiretroviral drugs toward overcoming and preventing drug resistance. Future HIV Ther, 2007; 1:291–313.

Mills

, Nachega

, Buchan

et al. Adherence to ART in sub-Saharan Africa and North America: A meta-analysis. JAMA, 2006; 296:679–690.

Gill

, Hamer

, Simon

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

Mukherjee

, Ivers

, Leandre

et al. Antiretroviral therapy in resource-poor settings. Decreasing barriers to access and promoting adherence. J Acquir Immune Defic Syndr, 2006; 43,Suppl 1:S123–126.

Weidle

, Wamai

, Solberg

et al. Adherence to antiretroviral therapy in a home-based AIDS care programme in rural Uganda. Lancet, 2006; 1587–1594.

10.

Lester

, Ritvo

, Mills

et al. Effects of a mobile phone short message service on antiretroviral treatment adherence in Kenya (WelTel Kenya1): A randomised trial. Lancet, 2010; 376:1838–1845.

11.

Lazarus

, Folkman

. Stress, Appraisal, and Coping. New York, NY: Springer, 1984.

12.

Folkman

, Moskowitz

. Positive affect and the other side of coping. Am Psychol, 2000; 55:647–654.

13.

Bandura

. Self-Efficacy: The Exercise of Control. New York, NY: W.H. Freeman and Company, 1997.

14.

Catz

, Kelly

, Bogart

, Benotsch

, McAulife

. Patterns, correlates, and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychol, 2000; 19:124–133.

15.

Schwarzer

, Renner

. Social-cognitive predictors of health behavior: Action self-efficacy and coping self-efficacy. Health Psychol, 2000; 19:487–495.

16.

Chesney

, Chambers

, Taylor

, Johnson

, Folkman

. Coping effectiveness training for men living with HIV: Results from a randomized clinical trial testing a group-based intervention. Psychosom Med, 2003; 65:1038–1046.

17.

Chesney

, Ickovics

, Chambers

, Gifford

, Neidig

, Wu

. Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: AACTG Adherence Instruments. AIDS Care, 2000; 12:255–266.

18.

Reynolds

. Adherence to antiretroviral therapies: State of the science. Curr HIV Res, 2004; 2:207–214.

19.

Levine

, Hinkin

, Marion

et al. Adherence to antiretroviral medications in HIV: Differences in data collected via self-report and electronic monitoring. Health Psychol, 2006; 25:329–335.

20.

Portney

, Watkins

. Foundations of Clinical Research: Applications to Practice. Upper Saddle River, NJ: Prentice Hall, 2000.

21.

Sidle

, Kimaiyo

, Monahan

et al.

Patterns of antiretroviral adherence among Kenyan patients and factors related to non-adherence during the first year of treatment [Abstract no. WEPEB099]

Poster exhibition: 4th IAS Conference on HIV Pathogenesis, Treatment and Prevention.

22.

Byakika-Tusiime

, Oyugi

, Tumwikirize

, Katabira

, Mugyenyi

, Bangsberg

. Adherence to HIV ART in HIV+ Ugandan patients purchasing therapy [Abstract] Int J STD AIDS, 2005; 16:41.

23.

Cloete

, Simbayi

, Kalichman

, Strebel

, Henda

. Stigma and discrimination experiences of HIV-positive men who have sex with men in Cape Town, South Africa AIDS Care. Psychological and socio-medical aspects of AIDS/HIV. AIDS Care, 2008; 20:1105–1110.

24.

Pop-Eleches

, Thirumurthy

, Habyarimana

et al. Mobile phone technologies improve adherence to antiretroviral treatment in a resource-limited setting: A randomized controlled trial of text message reminders. AIDS, 2011; 25:825–834.

25.

Rongkavilit

, Naar-King

, Kaljee

et al. Applying the information-motivation-behavioral skills model in medication adherence among Thai youth living with HIV: A qualitative study. AIDS Patient Care STDs, 2010; 24:787–794.