Abstract
Background:
The health benefits of antiretroviral treatment greatly depend on the extent to which people living with HIV/AIDS adhere to the prescribed daily dosing regimens. Peer support groups are expected to have a high impact on adherence to medication among people living with HIV/AIDS.
Objectives:
This study was designed to determine factors affecting adherence to medication among people living with HIV/AIDS in peer support and non-support groups attending antiretroviral clinics in Enugu State, Nigeria.
Method:
This was a cross-sectional, comparative study using qualitative and quantitative instruments among 840 people living with HIV/AIDS attending antiretroviral clinics in Enugu, Nigeria. Quantitative data was analysed using IBM SPSS version 22. A Chi-square test of statistical significance, a student t-test and multivariate analysis using binary logistic regression were used in the analysis. Manual content analysis was done for the qualitative data.
Results:
The mean age of respondents was comparable in both study groups: support group 38.5±9.6; and non-support group 38.5±10.1. A significantly higher proportion of respondents in the support group (91.9%) achieved good adherence compared with those in the non-support group (87.1%). A good medication adherence self-rating was a predictor of adherence in both study groups: support group (AOR 5.8, 95% CI: 2.7–12.2, p=<0.001) and non-support group (AOR 0.2, 95% CI: 0.1–0.4, p=<0.001).
Conclusion:
The peer-based intervention is a viable and effective tool for maintaining optimal adherence among people living with HIV/AIDS in resource-limited settings; hence, research should concentrate on interventions that do not focus on individuals alone, but rather on those that strengthen the capacity of groups to collectively participate in HIV programmes.
Introduction
In the context of the African HIV epidemic, support groups are not mere gatherings for discussions of social and health matters; neither do they function solely to make people living with HIV/AIDS (PLWHA) self-responsible and economically empowered. 1 The implementation of peer groups is, however, expected to improve their quality of life, have a high impact on morbidity and a moderate impact on mortality. 2 PLWHA have contributed immensely to HIV management in a wide range of peer-provided services since the onset of the HIV epidemic. 3 Clinic patients who are HIV positive and currently on highly active antiretroviral therapy (HAART) serve as ‘peers’ to other members, individually or in groups, by providing support during meetings and via telephone calls. 4
Antiretroviral treatment (ART) has been found to significantly improve the health, life expectancy and quality of life of PLWHA.5,6 HAART is the only known treatment for HIV/AIDS to date and is a combination of three or more drugs from at least two different classes of ART. A key determinant of successful HAART is drug adherence. More than 95% of the doses should be taken for optimal response, whereas lesser degrees of adherence are more often associated with virological failure. 7 How well these patients do depends largely on their ability to continuously adhere consistently and correctly to their respective treatment regimens.
Peer support in HIV care is consistent with the World Health Organization (WHO) guidelines on task shifting that encourage PLWHA to be part of the health workforce crisis solution, 8 and the promotion of the greater involvement of PLWHA in their own care has been a longstanding policy of WHO and the United Nations Programme on HIV/AIDS (UNAIDS). 9 Although the WHO promotes the role of support groups, 10 evidence of the impact peer support has on adherence has not been duly assessed.
In Nigeria, the earliest support groups for PLWHA were formed in 1997. 11 By 1998, the idea of a national network of people living with HIV/AIDS to bring together all existing support groups became a reality with the establishment of the Network of People Living with HIV/AIDS (NEPWHAN). 11 With increasing investment in HIV treatment programmes, the number of support groups across Nigeria has multiplied exponentially over the past decade. 1 In 2013, the Board of the Global Fund to Fights AIDS, Tuberculosis and Malaria approved $121 million funding for HIV in Nigeria based on recommendations from the Technical Review Panel (TRP) and the Grant Approvals Committee (GAC). 12 Eleven years into the free drug programme in Nigeria, findings from the Grant Approvals Committee revealed that although reducing barriers to access, increasing service utilization and improving adherence are important in the lives of PLWHA, there is no evidence available with regard to the effectiveness of the approaches currently being used and a lack of information about the potential impact of any proposed activities (scale-up of support groups, community dialogue and Mentor Mothers). 12
Although social services for the support of PLWHA exist, they are not directed by a clear set of directives or guidelines, resulting in haphazard provision of care and support services for PLWHA across the country. 13 With the rapid increase in patients’ access to ART in Nigeria, it has become vital to continuously monitor treatment adherence and identify interventions that can encourage its sustainability. 14 Establishing intervention efficacy is needed to build more confidence in the use of peer support interventions. It is hoped that the findings from this study will improve the management and lives of PLWHA. The findings will also be compared with those in other areas of Nigeria as well as other regions of the world. Furthermore, the results of this study will provide possible directions for future research and suggestions for intervention development to improve support services among PLWHA.
Methods
The study was conducted in Enugu State, one of the five states in the south-eastern geopolitical zone of Nigeria. Participants were PLWHA and a cross-sectional comparative study design was used. The study was facility-based and compared findings between PLWHA belonging to peer support groups and those not belonging to any support group using both qualitative and quantitative study instruments for a period of six weeks.
Specifically, the participants were PLWHA aged ≥18 years in Enugu State, who were enrolled in care in health facilities that provide comprehensive ART services, including peer support services, and who had been on HAART for more than a year prior to the study. In this study, a PLWHA peer support group is defined as a formal association of adult PLWHA who meet monthly at a health facility to address health-related issues and carry out activities related to individual and group management of HIV/AIDS at both group meetings, and through telephone calls and home visits. Discussions centre on the identification of barriers to HAART adherence, problem-solving strategies to overcome barriers and other life issues that impact adherence, including HIV status disclosure, dating, substance use and abuse, and struggles with other health issues.
A minimum of two-thirds attendance during the year at monthly support group meetings qualified PLWHA for inclusion into peer support groups. Participants who declined to give consent, those whose folders were incomplete or missing and those who were in stage three or four of the disease were excluded from the study.
The minimum sample size for the study was determined using the formula for comparing two independent proportions. 15 A study in Nigeria revealed an adherence rate of 95% among PLWHA in peer support groups and 87% among those not belonging to support groups.16,17 A sample size of 420 was estimated for each of the groups.
A multi-stage sampling technique was used. In the first stage, 14 health facilities in the state with functional peer support groups were first stratified according to their senatorial zones. A simple random sampling technique was used to select eight health facilities proportionately. Participants were then stratified into those belonging to peer support groups and those who did not. The sampling frame for each study group was determined using the last attendance register for the period of study (six weeks). A systematic sampling technique was used to select participants as they presented for their clinic visits on each day of data collection. By dividing the sampling frame by the sample size of 420, a sampling interval was obtained.
Participants were recruited purposively for the focus group discussions to reflect the heterogeneous characteristics inherent in the study population. Four sessions were held, each involving between 8–12 participants who were then excluded from responding to the questionnaire. The focus group discussions were held separately for males and females.
A trained interviewer administered the questionnaires as well as guiding the focus group discussions in both the local Igbo dialect and English. Data were collected from September to October 2016. The questionnaire assessed socio-demographic characteristics, an adherence self-report and an adherence self-rating. For this study, the ART medication adherence self-rating was assessed with questions adopted from a tool developed and validated by Chesney et al. 18 and was determined on a 4-point Likert scale (never, rarely, sometimes and always). Self-rating can identify individuals who may have difficulties with adherence and need assistance. A total score was calculated for self-rating with a range of 0 to 9, with scores 0–4 indicating a high self-rating and scores 5–9 indicating a low one.
Dosing frequency (as derived from the pharmacy dispensing record) and self-reported number of missed doses within the past 28 days were used to indirectly determine the rate of medication adherence calculated as a percentage using the formula
where
- total number of prescribed doses=number of doses per day x 28 days (a month)
- total number of doses of ART taken=total number of prescribed doses - total number of missed doses in the last 28 days
Based on WHO 20 recommendations, an adherence rate of ≥95% was classified as good adherence, whereas an adherence rate of <95% was classed as poor adherence.
Data entry and analysis were done using IBM SPSS version 22. A Chi-square test and multivariate analysis using binary logistic regression were used in the analysis and the level of statistical significance was set at a p-value of <0.05. Variables that were associated with medication adherence at p<0.2 in the bivariate analysis were entered into the logistic regression model to determine the predictors of medication adherence among people in the peer and non-peer support groups.
Manual content analysis was used to analyse the qualitative data. The recorded focus group discussions were immediately transcribed verbatim after each session after which four themes emerged.
Results
Quantitative data
A total of 840 PLWHAs, 420 each from the peer support and non-peer support groups in Enugu State participated in the study. The mean age of respondents in the peer support group and non-peer support group was 38.5±9.6 years and 38.5±10.2 years, respectively. Although a majority of the respondents in both groups were married, a higher proportion of these belonged to the non-peer support group and the difference was statistically significant (χ2=12.927, p=0.007). See Table 1. A higher proportion of respondents in the non-peer support group (13.1%) take alcohol when compared with those in the peer support group (8.1%) and the difference was statistically significant, (χ2=5.542, p=0.019). In addition, a higher proportion of respondents in the non-peer support group (2.6%) smoke cigarettes when compared with those in the peer support group (1.4%), but the difference was not statistically significant (χ2=1.501, p=0.221). See Table 2. A significantly higher proportion of respondents in the peer support group (91.9%) had good adherence when compared with those in the non-peer support group (87.1%) (χ2=5.077, p=0.024). The main reason participants gave for missing their medication was being away from home (peer support group=37.4%, non-peer support group=45.4%), although the difference in proportions was not statistically significant (χ2=10.668, p=0.058). See Table 3. When following the prescribed drug regimen, 81.2% of respondents in the peer support group and 65.5% in the non-peer support group took their drugs according to schedule (χ2=26.518, p<0.001). On an overall self-rating, 81.7% of respondents in the peer support group and 71% of those in the non-peer support group had a good self-rating with regard to adherence. This finding was statistically significant (χ2=13.335, p<0.001). See Table 4. Respondents who had a good self-rating with regard to adherence in peer support and non-peer support groups were about six and five times, respectively, more likely to achieve medication adherence when compared with those who had a poor self-rating with regard to adherence (95% CI: 2.7–12.2) (95% CI: 0.1–0.4), p<0.001. Among respondents in the non-peer support group, those who had been on HAART for less than four years were twice as likely to adhere to HAART than those who had been on HAART for four years or more (95% CI: 0.2–0.9), p=0.015. Likewise, respondents who did not take alcohol were about three times more likely to adhere when compared with those who take alcohol (95% CI: 2.1–9.4), p<0.001. See Table 5.
Socio-demographic characteristics.
student t-test.
Social attributes.
Fischer’s Exact Test.
Medication adherence.
student t-test, **likelihood ratio.
Adherence self-rating.
Factors affecting medication adherence among respondents in peer support and non-peer support groups.
Qualitative data
The themes that emerged during the analysis were: level of adherence to medication among the study groups; reasons for non-adherence; measures to improve adherence; and barriers to participation in support groups.
Level of adherence
Participants in both study groups were of the opinion that adherence is better with PLWHA who belong to peer support groups. A female participant from a non-support group was of the opinion that those in the peer support group do better because they discuss things and interact among themselves, giving them some measure of confidence to face life.
Reasons for non-adherence
Although the majority of the participants reported good adherence with regard to their medication, some provided reasons for their failure to fully adhere as required. One such reason was lack of money for travel. Some participants who had to travel from communities a long way away reported that occasionally they missed their appointments due to financial constraints. However, a female participant in the peer support group refuted the excuse and said, ‘It pains me that some people say no money for transport. If the drug is working for me as it is working for them, then there is no reason not to keep their appointments and take their medications.’
Another female peer support group participant recalled,
Nothing will make me not to take my drugs, these medicines have been of help. I was really sick and had lost a lot of weight before I started taking my medications. My life would have otherwise been short. Take a look at me now.
Another reason for missing medication is reticence. This was reported more among participants who had not disclosed their status. A female participant in the non-peer support group said she hides while taking her medication or can delay taking it if the location is not conducive to her doing so. She feared possible discrimination if her status became known. Contrary to her fears, another female participant in the non-peer support group always takes her drugs wherever she is or whoever she is with. She said, ‘sometimes I chew my drugs instead of taking it with water so that no one will know that I am taking drugs, I really pray for a cure’.
Measures to improve adherence
All participants were of the opinion that improving adherence requires patients to have a good understanding of the disease and accept the need for follow-up visits. Interestingly, some of them linked good understanding of the disease to belonging to support groups. A female participant from the non-peer support group put it this way, ‘Those that belong to support groups understand the importance of follow-up care. This is because they teach them all these things, I mean how to overcome barriers in taking medication.’
There was a call for everyone to join support groups and there was a suggestion that the support groups should be meeting every three months instead of every month in order to boost attendance.
Barriers to participation in support groups
Different views were expressed as to why people do not belong to support groups notwithstanding the advantages inherent in doing so. Most of the participants were of the view that the patients come from a long way away and this affects their ability to belong to support groups. A pregnant female participant recalls, ‘I come all the way from Owerri (212 kilometers away) to access care here. I was diagnosed here when I came to visit my mum and I have continued accessing care here because I love their services.’
Some participants had a keen interest in attending the meetings but had other competing demands on their time. A female participant regrettably reported thus, ‘I would love to attend support group meetings but I have tight work schedules. I also work on weekends. I’m a widow and also the bread winner of my family.’
Some were also of the opinion that other engagements in the community and efforts by the patients to make a living for themselves occupy the time of some PLWHA, making it impossible to belong to support groups and attend the meetings.
Discussion
PLWHA who belonged to peer support groups in this study significantly maintained higher levels of adherence to their medication when compared with those not in support groups. In a study that involved the six geopolitical zones in Nigeria, 16 a significantly higher proportion of respondents in a support group (95%) achieved self-reported adherence when compared with those who did not belong to a group (92%). The higher level of adherence in the above study 16 could be attributed to the fact that medication adherence was assessed using a three-day recall and involved patients who had been on HAART for less than four years prior to the study.
In another study in Port Harcourt, Nigeria 21 among non-adherent PLWHA, only 15.6% of the respondents in that study reportedly belonged to peer support groups. Therefore, it could be concluded that, in this study, poor adherence was observed more in PLWHA who were not involved in support group activities. 21 In addition, participants who attended support groups in a South African study were reported to be 1.6 times more likely to be taking HIV prophylaxis and ART medication than non-attenders. 22 This finding reaffirms the postulation that social support from PLWHA peers through a PLWHA support group mechanism plays an important part in ART adherence. Several studies in Mozambique, 23 Spain, 24 the USA 25 and Canada 26 have disclosed higher self-reported adherence among respondents in peer support groups when compared with respondents not in support groups. In other studies where pill count was used to assess adherence, higher rates of adherence were also observed among peer-supported respondents when compared with those not in peer support groups.27,28
Participants in the focus discussion groups were of the view that participation in peer support activities improves adherence to medication. They attributed this to what they were taught as well as the good interaction between members at group meetings. This is reaffirmed in a study by Wouters et al., 29 in which 89.9% of support group members reported that support group meetings helped create a forum for sharing knowledge and experiences, some of which related to taking medication.
Contrary to these findings, the results of a randomized controlled trial of a peer support intervention in Northern Nigeria reveal no significant difference in ART adherence level among HIV-infected adults who received peer education-based medication adherence interventions and their counterparts who received standard care services. 30 Similarly, there was no significant difference in the intervention and control groups during a 24-week follow-up study in the USA to assess the impact of monthly education sessions on adherence among HIV-positive patients. 31 This could be attributed to the fact that the intervention in that study was purely educational, with no psychosocial support element as could have been included with a formal peer support group. Although education is a necessary critical component in HIV care and management, it may not be enough on its own. The effects of an educational intervention may be short lived when compared with the interactive, ongoing and longer-lasting effects of peer support engagement. In another study, Simoni et al. demonstrated that receiving support from peers lasting three months might promote adherence, but that the effect did not persist when the support was discontinued. 32 It is, therefore, important to note that the duration of involvement and participation in support group activities could also influence adherence to medication among PLWHA belonging to such groups.
The proportion of respondents who consume alcohol was significantly higher among PLWHA not belonging to peer support groups. This reaffirms the postulation that peers may deliver behavioural modification interventions that reduce risky behaviours and may possibly improve treatment outcomes among PLWHA. 21
Being away from home and forgetfulness were reported as the main reasons for missing medication among respondents in both study groups in this study, and in other studies.18,21 Other reasons for missed doses among respondents in this study included: ran out of drugs; discomfort/side effects; ashamed of taking medication in front of others; and fasting. However, other reasons for missed doses reported in other studies included: sleeping through dose time; 21 non-availability of food18,19,33 (as most medications are required to be taken with food); and too many and too large tablets. 34
Adherence self-rating influenced medication adherence among respondents in peer support group in this study, whereas number of children, alcohol intake, number of years on HAART and adherence self-rating influenced adherence among respondents not in peer support groups. Multiple logistic regression analysis revealed that among respondents in peer support groups, those with a good adherence self-rating were about six times more likely to achieve adherence, whereas those not in peer support groups were about five times more likely to achieve adherence than those with a poor self-rating. This means that efforts towards improving patients’ adherence self-rating will invariably improve adherence to medication among PLWHA. Based on this observation, more emphasis should be placed on counselling, because this invariably improves self-rating among PLWHA.
Lack of money for travelling to health facilities was cited among the reasons for missed appointments and this was mentioned as a reason for poor adherence in the focus discussion groups. In a study to determine barriers to and facilitators of adherence to ART and consider the reasons why adult HIV-positive patients in Ethiopia remain in care, 34 lack of money for transport was one of the reasons that affected adherence among PLWHA.
Conclusion
A medication adherence of 95% or more is required for an optimal outcome from treatment. Given this, there is considerable need to increase the level of medication adherence considering the high incidence of HIV/AIDS in Nigeria. Our study revealed a significantly higher adherence level among respondents in the peer support group; hence, as a result of non-participation in peer support groups, optimal adherence may not be achieved. A poor adherence self-rating, alcohol intake, number of years on HAART, poverty and distance to health facilities all affected adherence in this study. These findings could be used as a benchmark to design programmes that will improve adherence to HAART in Nigeria. Programmes and research should concentrate on interventions that do not focus on individuals alone, but rather on those that strengthen the capacity of groups to participate collectively in HIV programmes.
Limitations
The measurement of adherence was based on participants’ self-reports over the past month. A self-report may be affected by recall bias. As the study was conducted on patients who had been on HAART for at least a year prior to the study, those who had only started on HAART during the past year were excluded. This may limit being able to generalize the study as it is not a clear representative of the population of PLWHA. Finally, using manual content analysis for qualitative analysis may have introduced bias during selection of codes. This would have been minimized had Nvivo or Atlas.ti software been used for the analysis.
Footnotes
Authors’ contributions
1. Onyinye Hope Chime designed, analysed and interpreted the work.
2. Sussan Uzoamaka Arinze-Onyia drafted the article.
3. Christopher Ndukife Obionu revised the article critically for intellectual content.
Declaration of conflicting interests
The authors declare that there is no conflict of interest.
Ethical approval
Ethical approval for the study was obtained from the Health Research Ethics Committee at the University of Nigeria Teaching Hospital in Enugu State with approval number NHREC/05/01/2008B – FWA00002458 – IRB00002323. Approval was also obtained from the Enugu State Ministry of Health and the management of the selected health facilities.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors for the research, authorship and/or publication of this article.
Informed consent
Written informed consents were obtained from the participants.
