A comparative study of the predictors of treatment adherence among patients on antiretroviral therapy at rural and urban centers in Cross River State, Nigeria

INTRODUCTION

Human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/AIDS) has remained a burden to most families in sub-Saharan Africa which harbors about 70% of the world’s HIV infected population.^[1] The prevalence of HIV/AIDS in Nigeria based on the sentinel surveillance of 2019 was 1.4% and Cross River State had an estimated prevalence of 1.7% which was the 5^th medium prevalence in Nigeria with a range of 8.2% (in the urban areas) to 1.8% (in the rural areas).^[1]

In the absence of a cure for HIV/AIDS, antiretroviral therapy (ART) has remained the only available option that offers the possibility of reducing HIV/AIDS-related morbidity and mortality while improving the quality of life of the people living with HIV and AIDS (PLWHA).^[2,3] However, ART has to be taken as a lifelong therapy and its success depends on continual adherence to the medication regimen.^[4,5]

ART has improved the health of many HIV positive individuals who would have died due to this infection.^[6,7] This evidence was shown in the annual number of HIV/ AIDS deaths which declined from 2.5 million in 2005 to 0.67 million in 2017.^[8] In part, this decline in the death of PLWHA was likely a result of the substantial increase in access to HIV/AIDS treatment and adherence to ART.^[8,9]

The World Health Organization defined adherence as the extent to which a person’s behavior: taking medication, following a diet, and/or executing lifestyle changes, corresponds with agreed recommendations from a healthcare provider.^[10] and the Nigerian National Guidelines for HIV and AIDS Treatment and Care in Adolescents and Adults agree with this ≥ 95% cutoff for ART adherence, asserting that for a patient to be tagged as adherent to ART, he must not miss more than one dose in 10 days if on a twice daily regimen or one dose in 20 days if on a once daily regime.^[11]

The identified factors that can influence medication adherence in PLWHA can be categorized into three groups: Patient-related, health system-related, and clinical factors.^[12] Predictors of adherence to ART vary among patients, environments, and communities.

The health system-related factors includes: Drugs being out of stock, insufficient amount of time a clinician spends with the patient due to overtaxing of the healthcare system, lack of utilization of health information, lack of feedback from patient, and meeting different doctors at each consultation follow-up visits.

The patient-related factors that may lead to non-adherence are lack of patient’s understanding of the relationship between the disease, the required percentage of adherence and quality of life, lack of involvement in the treatment decision-making process, feelings of stigmatization, sharing drugs with spouse and other family members, feelings of wellness, undetectable level of virus in peripheral blood following long-term ART and faith in miraculous healing.^[13,14]

The above listed factors can vary based on the environment such as urban and rural communities. This study aims to assess, identify, and compare the factors that predict treatment adherence among patients on ART in an urban and a rural clinic, to improve patients’ management outcome.

MATERIAL AND METHODS

This was a cross-sectional analytical study conducted within 4 months from March to June 2015. Using interviewer’s administered questionnaire of adapted Morisky 8-item medication adherence questionnaire,^[15] Duke functional social questionnaire,^[16] modified family support scale,^[17] data for this study were collected at the Family Medicine Clinic of the University of Calabar Teaching Hospital (UCTH), Calabar and General Hospital (GH), Ikot Ene in Akpabuyo LGA, Cross River State, Nigeria. These intervention sites were established and funded by President’s Emergency Plan for AIDS Relief (PEPFAR) and had been operational since 2005.

A total of 322 participants (161 from each study sites) were recruited into the study after calculating sample size using formula for demonstrating significant difference in two independent groups.^[18] and systematic sampling technique was utilized for selection criteria into the study. The inclusion criteria were PLWHA 18 years of age and above that were on ART for at least 6 months, who had consented to participate in the study. All critically ill patients were excluded from the study.

Data collected for this study was analyzed using the Statistical Package for the Social Sciences version 20 software, manufactured by International Business Machines Corporations, United State of Ameriaca. Descriptive and inferential statistics were done.

Ethical approved for this study was given by the Health Research and Ethics Committee of the UCTH with protocol number NHREC/07/2012/UCTH/HREC/33/250. There was no conflict of interest in any of the authors.

RESULTS

Comparing the socio-demographic characteristics of the study participants in the urban and rural centers [Table 1]. The trend in the tribe distribution had a statistically significant difference ranging from 1.9% to 10.5% as seen among the Ejagham and Ibibio ethic groups respectively. There was a significant difference of 12.4% in the distribution of religion between the urban and rural respondents. The ratio of females to males in both centers was 1.1:1.0, however, this difference was not statistically significant. The difference in the trend of age group and marital status in the respondents in the urban and rural centers ranged from 1.3–5.5% and 0–9.9%, respectively. These differences were not statistically significant.

Table 2 compares the socio-economic characteristics of the study participants in the urban and rural centers. It was noticed that the trend in educational level had a differences ranging from 0.6% to 15.5%. The differences in the trend of the distribution of the employment of the respondents in both centers ranged from 0 to 8.6% with self-employment having the highest difference. Comparing the average monthly income and place of residence in both study centers, differences were 6.8% and 23.6%, respectively

Table 3 presents the comparison of the patient-related factors that predicted adherence to ART at the two study centers.

The result showed more participants (4.4%) in the rural center were aware of the need to take ART for life and this difference was statistically significant (P = 0.012).

The adherence to ART on pharmacy records had a difference 16.4% between the urban and rural centers. The trend in the distribution of clinical stages of the respondents in both study centers ranged from 4.9 to 11.2%. The difference in the self-reported adherence and awareness of the need for optimal adherence to ART was 7.5% and 3.1%, respectively. The trend in the distributions of CD4 and BMI had differences that ranged from 0.6–5% and 0.7–9.1%, respectively.

Table 4 presents the predictors of adherence to ART among the study participants in the urban and rural centers. The two centers had the following predictors in common: awareness that ≥95% adherence was a requirement, adherence on pharmacy records, clinical stage, family support, and stigmatization experience. In addition to, these predictors the rural center had family support.

The urban center had 7% more respondents that were aware of the need to take ≥ 95% of their ART. The pharmacy record in the rural center was 0.3% more relationship than the urban center. The differences between the trends in clinical stages of the respondents in the urban and rural centers range from 2.2–86.6% and 11.2–66.6%, respectively.

The trend in social support in the rural respondents ranged between 11.2 and 32.8%. The differences in the trend in family support between the urban and rural centers were in a range of 4.8–68%. About 8.6% of the participants in the urban center experienced high stigmatization than those in the rural center.

Table 5 shows binary logistic regression of predictors of antiretroviral treatment adherence among study participants in the urban and rural centers. The result revealed that the two centers had in common adherence on pharmacy records and awareness that ≥95% adherence was a requirement as the predictors of adherence to ART. In addition to these, sex was a predictor in the urban center while family support and stigmatization were noticed as predictors in the rural center.

There were three times more respondents in the urban center that were aware that ≥95% adherence was required than in the rural center. The ratio of adherence on pharmacy records between the urban and rural centers was 2.2:1. In the rural center participants with no stigmatization experience were more than 4 times more likely to adhere to ART. In the urban center, males were < ½ times likely to adhere to ART.

DISCUSSION

In this study, 66.5% of urban respondents were adherent to ART based on self-reported method while 65.8% of them were adherent as shown on pharmacy record. The difference between the two methods used was not significant. The adherence to ART was lower in the rural respondents both in the self-reported method (59.0%) and pharmacy record (49.4%). This difference between the methods could be explained by the suggestion that the use of pharmacy records in assessing adherence to ART is superior to self-reporting method, though not up to the 10–20% proposed difference.^[19]

The proportion of adherence among urban respondents in this study was similar to a study in Ibadan (2015) with 63% adherence level.^[20] However, it was different from a study carried out in Ethiopia (2015) where 83.1% of the respondents were adherent to ART.^[21]

In this study, the predictors of ART adherence were: Sex, body mass index, family/community support, awareness of the need for ≥ 95% adherence, stigma experience, and adherence as shown on pharmacy records.

The difference in the ART adherence proportion between the rural and urban respondents in this study may be due to some environmental factors such as community/social supports. A study carried out in a Teaching Hospital in Wolaita Soddo, Ethiopia by Alagaw et al. (2013), revealed that the predictors of ART adherence were: Sources of food for consumption, food scarcity, the person or people the client lives with and presence of depression in the patient.^[20] This was different from a study in Northern Ethiopia by Demeke and Chanie (2014), which showed that duration of treatment, family disclosure, living condition, and taking other medications along with ART were the predictors.^[22] Another study in South-East Ethiopia by Lencha et al. (2015), reported that history of drug abuse, relationship with clinician and keeping to regular follow-up were the determinants of ART adherence.^[23]

This study is in agreement with the studies cited above that predictors of ART vary with geographic areas and it further portrays that even in the same districts there could be difference between the urban and rural area.

CONCLUSION

There was the possibility of response bias by participants who may have falsely reported to be adherent to medications in order to impress the interviewer. However, the study showed that there are similarities and differences in the predictors of ART adherence between the urban and rural centers. The identified predictors above may serve as a guide for developing interventions aimed at improving the proportion of patient on ART and sustain the adherence among these clients.