Failure to Attain HIV Viral Suppression After Intensified Adherence Counselling—What Can We Learn About Its Factors?

Introduction

Since its discovery, antiretroviral therapy (ART) has saved the lives of many people infected with the human immunodeficiency virus.^1,2 Prior to the introduction of ART, many people living with HIV (PLHIV) died of HIV-related complications.³ ARTs have reduced the mortality of PLHIV from 1.95 to 0.95 million deaths.⁴ ART suppresses the viral load and boosts the immunity of the infected individual. The effects of ART are optimized by reasonable accessibility and availability of ART in health facilities.^5,6 The approach to ensure good coverage in most countries ART is either donor or government-funded and made available. Even though the availability and accessibility of ART have increased recently, some patients do not attain adequate virological suppression on receipt of their first-line ART globally.^7–9 Studies have reported several factors to be associated with non-suppressed viral loads. In South Africa, factors like age of less than 15 years during ART initiation, and being male were reported to be associated with virological failure.¹⁰ Male sex, alcoholism, and smoking were reported to be associated with virological failure in Morocco.¹¹ Low CD4 cell count, social isolation, and high stigma were shown to be significantly associated with virological failure in Vietnam.¹² In addition, adult patients were reported to have high risk of virological failure in Ethiopia and poor adherence to ART was reported to have significant association with virological failure in Kenya.^13,14

About 20% of PLHIV in Haiti switched to second-line regimens after failed first-line regimens.¹⁵ Studies in the Lake Zone of Tanzania revealed that approximately 12.18% of PLHIV had virological failure despite being in ART for at least six months.^15,16 Poor adherence to the treatment plan mainly causes virological failure, and adherence intervention is highly recommended.^17,18 Effort is made in different countries to reach the UNAIDS target which states that by 2030 95% of PLHIV will know their HIV status, 95% of people diagnosed with HIV will receive ART and 95% of PLHIV taking ART will achieve viral suppression.¹⁹

As per the national guidelines in Tanzania, adopted by the World Health Organization (WHO), patients with a viral load of more than or equal to 1000 HIV copies per mL of blood after being on ART for at least six months are suspected to have virological suppression failure, and therefore they are enrolled into intensified adherence counseling (IAC) sessions.²⁰ IAC includes a series of adherence counseling sessions offered once every month for three consecutive months, and this intervention maximizes the patients’ commitment to treatment schedules.²⁰ The guidelines affirm that viral load suppression could be achieved after three months of treatment. In addition, the intervention is expected to reduce the number of PLHIV to be switched to the second line, which is very expensive and with many side effects. Even though the ARTs are government or donor-funded, second-line ART is costly, and avoiding switching patients to the second line saves much money. In cases, this intervention demonstrates a failure in improving adherence and virological suppression, switching patients to the second line remains the only option to save the patient.²¹ The prevalence of switching patients from first-line to second-line ART is 3.1% by three years after the initiation of ART globally.²² In Tanzania, the incidence rate of switching patients from first-line to second-line is 1.7/100 person-years.²³ Switching of PLHIV from first line to second line was more higher before the introduction of IAC intervention not only in Uganda but also in Tanzania.²⁴ The prevalence of viral load suppression failure was 62.2%²⁵ in Tanzania and in Uganda.

Despite increased awareness of PLHIV with virological failure after IAC sessions, associated factors are not well studied in Tanzania. Therefore, this study investigated the factors associated with virological failure following three months of IAC to PLHIV.

Materials and Methods

Design and Settings

A multicenter retrospective cross-sectional study was conducted at two centers offering care and treatment clinics (CTCs) in Mwanza city. These centers were Bugando Medical Centre (BMC) and the Sekou-Toure Regional Referral Hospital. BMC is a teaching hospital and the main zonal and referral hospital in Tanzania serving eight regions with a population of over 13 million people.²⁶ Sekou Toure is a regional referral hospital of Mwanza city serving about 4 million people.²⁷ Both hospitals have high-load CTCs clients, serving as referral centers from other CTCs in the Lake Zone. Out of 1.6 million PLHIV on ART in Tanzania, 10,000 of them are registered and attend CTCs at BMC.¹⁶ More than 9600 PLHIV are attended at the Sekou-Toure Regional Referral Hospital in a year with an average of 40 clients attending a day.²⁸

Population

The study included PLHIV on first-line ART who were suspected to have virological failure between January, 2018 and December, 2019. According to Tanzania National guideline for management of HIV and AIDs (2019) patients is suspected to have virological failure whenever a patient has been on ART for at least six months and has viral load of more than 1000 copies/mL of blood. Viral suppression according to the WHO consolidated guidelines on the use of antiretroviral drug for treatment and preventing HIV infection recommendation for a public health approach is defined as a VL of less than 1000 copies/mL and undetectable viral load is defined as VL of less than 50 copies/mL of blood.²⁹ We excluded clients whose records missed IAC information, adherence assessment post counseling, and the three months HIV RNA viral load testing post counseling. Out of 267 PLHIVs suspected to have a virological failure, only 212 met the inclusion criteria of this study. Fifty-five (55) subjects were excluded due to missing key information.

Sample Size

The Estimated Sample Size Was Calculated Using Kish-Leslie Formula (1965)

The assumed prevalence (P) of virological failure as from previous study on HIV for adults aged 15 to 49 years 14.7% Z = 1.96, value of standard normal distribution at 95% confidence level and the accepted marginal of error of Ɛ = 5%. We assumed the minimum sample of 212.

Procedures and Measurements

Data of illegible participants in this study were extracted from CTC card number two (CTC-2) and the high viral load adherence assessment form using a checklist adopted and modified to suit the local setting.³⁰ The patient adherence profile was extracted from the high viral load assessment form and was assessed using Bugando Medical Centre adherence tool whereby adherence of 95% and above was considered as good adherence using pill count.³¹ According to this tool adherence of 95% and above means no missing or missing less than 2 pills during a prescribed time. The opposite of 95% is applicable for poor adherence. Participants’ PCR HIV RNA viral load after three months of IAC was also recorded from CTC-2. Additionally, age, gender, education, and drug regimen were also recorded from CTC-2. The researchers investigated to see how these independent variables interact with the dependent variables which were accepted for adherence and viral suppression.

Data were extracted from the patient medical records into an excel spreadsheet and then transferred to STATA version 15 for analysis. We used the Shapiro–Wilk normality test to assess the distribution of continuous variables. We summarized continuous variables as means () and standard deviation (SD). Categorical variables were summarized using frequencies and percentages. The ART duration among the participants was analyzed to see its effect on adherence and virological suppression. Median years of participants being on ART were calculated to divide them into short use and long use. We compared proportions using the chi-square test and a p-value of less than 0.05 was considered significant at a 95% confidence interval. We also used the Wilcoxon signed-rank test to compare the median viral load before and after IAC. Univariate logistic regression was used to determine the association between independent and dependent variables. Variables were included in the multivariate model if they had a p < 0.2 in univariate logistic regression analysis. Then, multivariate logistic regression was used to establish the factors associated with the viral load suppression failure after the IAC. According to the guidelines by the Tanzania National AIDS Control Program, we considered patients to have virological failure if they did not achieve a viral load of ≤1000 HIV copies/mL of blood following three months of IAC.²⁰

Ethical Considerations and Participant Consent

This study complies with the Declaration of Helsinki.

The ethical clearance was obtained from the joint Catholic University of Health and Allied Sciences (Bugando Medical Centre Ethics and Review Committee (CREC/410/2019)). Since it is a retrospective study, we received permission to access the patient’s records from Bugando Medical Centre (Ref: AB.317/440/Part. M/) and Sekou Toure regional referral hospital (Ref: FA. 137/264/01/G/32). We used the patient’s special codes instead of the patient’s name for confidentiality purposes. We received a waiver for the patient’s consent as we used secondary data.

Results

Participant’s Demographics and Clinical Characteristics

A total of 212 PLHIVs participated in this study. The mean age and standard deviation () of these participants were 43.2 ± 11.8 years. Out of 212 participants, 63.2%¹⁸ were female, 38.2% (81) were married, 3.8% (8) were separated, 43.4% (92) were in stage 3 as per WHO clinical staging of HIV, and 50% (106) were underweight (Table 1).

Table 1 Participants’ Demographic and Clinical Characteristics (N = 212)

Antiretral Viral Adherence and Median Viral Load Before and After IAC

Before IAC, only 66.5% (141) participants had good adherence of 95% and above using pill count. After IAC, we observed a significant increase in participants who achieved good adherence 85.9% (182) with a p-value of 0.001 using the same scale. Viral load data normality check was performed using swilk test (p-value=0.001) and therefore median was opted for further analysis. We also observed a significant decrease in median viral load after IAC using the Wilcoxon signed-rank test. However, with this significant reduction in the median viral load but some of the participants 24.5% (52), their viral load reduced but did not reach viral suppression and were confirmed to have virological failure after IAC (Table 2 and Table 3).

Table 2 ARTs Adherence and Virological Status Before and After IAC (N = 212)

Table 3 Prevalence of Virological Suppression After IAC (N = 212)

Factors Associated with ART Adherence Before and After IAC

In univariate analysis factors like being single, age group, ART duration, and use of local medicine qualified to be subjected to multivariate analysis. Being single shown to have significant contribution to ART adherence before the intervention. After IAC factors such as being divorced, being single, being on ART for more than 11 years, using local medicine, and living in Nyamagana had p-value of less than 0.2 in univariate analysis qualifying to be subjected to multivariate analysis. Being single remained the only factor that significantly associate with adherence before and after IAC in the univariate and multivariate analysis.(See Table 4).

Table 4 Factors Associated with Adherence Before and After Intensified Adherence Counselling (N = 212)

Factors Associated with Virological Failure Following Three Months of Intensified Adherence Counseling

During univariate logistic regression analysis, we discovered that being single (OR = 3.6 95% CI = 1.6–7.8), young (18–25 years) (OR = 7.9, 95% CI = 2.6–23.8), not disclosing the HIV status (OR = 3.3, 95% CI = 1.1–10.9) and being unstable (OR = 2.5, 95% CI = 2.3–4.7) during the start of ART were significantly associated with virological failure after IAC. All variables that had a p-value of less than 0.2 were subjected to multivariate logistic regression analysis.

In this regression analysis factors such as young age (AOR = 5.6, 95% CI = 1.1–29.6), poor adherence to ART after IAC (AOR = 4, 95% CI = 1.3–12.3), and unstable when starting ART (AOR = 0.3, 95% CI = 0.13–0.62) were associated with virological failure post-IAC (Table 5).

Table 5 Factors Associated with Virological Failure After Intensified Adherence Counselling (N = 212)

Discussion

This study has demonstrated promising results of improving ART adherence to PLHIV suspected to have virological failure while on first line ART. The findings from this study show that 85.9% and 75.5% PLHIV achieved more than 95% adherence and virological suppression respectively following IAC. Youngness and poor medication adherence increase the risk of virological failure after 3-months of intensification.

The improvement in adherence to medication observed in this study was probably because during the adherence counseling sessions, barriers like not disclosing information to partners, alcohol use, and herbal medicines were identified by the nurse counselors and discussed with the clients. The nurse counselors advised them to use mobile phone reminders, identifying a supporting person and peer clubs and these might have contributed to the improvement in medication adherence. Barriers for non-improved medication adherence reported being missing some counseling sessions, poor quality of counseling, and lack of remainder. In addition, another study by Lee et al³² reported chaotic home situations and busy work schedules as the main barriers for most clients not achieving medication adherence.³² Pill burden and forgetfulness among clients were the barriers for not achieving optimal medication adherence.³³ In China, clients report poor medication adherence due to forgetting, being away from home, being busy, and feeling worse after taking drugs.³⁴ Poor adherence to medication can cause treatment failure and even drug resistance³⁵ These findings were similar to what we observed in this study barriers like not disclosing information to partners, alcohol use, and herbal medicines as the main reasons for not adhering to medications.

In this study, we recorded the viral load before and after IAC. Unfortunately, 24.5% of the participants did not achieve viral suppression after the intervention. There was an increased risk of virological failure after IAC among single compared to those married, separated, or divorced. Single individuals have no support from partners who can remind them. This finding was different from studies done in Kombolcha town and South Wollo Zone Ethiopia, where they reported that being divorced or separated had increased risk of virological failure.^36,37 The possible reason for this discrepancy is that in their study, single was kept constant during analysis. Again, we found that a unit decrease in age increases the risk of virological failure after IAC among PLHIV, while a unit increase in age decreases the risk. Being young was significantly associated with virological failure after IAC sessions 7.9 times more odds. Similar findings from studies conducted in Ethiopia and Rwanda.^38,39 The risk is higher at a young age since young individuals have difficulty achieving accepted medication adherence.

Not achieving acceptable adherence of 95% and above, being unstable during admission to treatment had increased risk and was significantly associated with virological failure compared to their counterparts. This finding is consistent with findings reported by Lailulo et al.⁴⁰ In addition to peer-to-peer counselling, peer club or support groups, disclosure of HIV status, and ART mobile phone reminders by SMS reported in the previous studies, IAC could also be used as an alternative to improving ART adherence.

Furthermore, 66.5% of participants who had good adherence before IAC but suspected to have virological failure 68.8% suppressed their viral load and 59.6% retained their viral load of more than or equal to 1000 copies/mL even after the intervention. This could be contributed to the reason that IAC is a client centered intervention and aims to optimize client adherence to all guidelines including ART use probably those who suppressed virologic ally may be benefited from this. Those who failed virologic ally regardless of their good adherence may be due to drug resistance or may necessitate the use of other tools in assessing adherence apart from pill count. Study has reported several factors apart from adherence to be associated with virological failure. In the study conducted in Ethiopia revealed that participant who experience drug toxicity has high risk of virological failure.⁴¹ A study in children and adolescents discovered mismatch between pill count and viral load suppression.⁴²

In this study, we used pill count to assess medication adherence as these patients’ received ART during their CTC visit. The strength of this measure is that it is more objective than self-report adherence. However, due to retrospective nature of the study, small sample size, involving subjects from a single region, and exclusion of participants with missing key information limit generalizability of the findings.

Conclusion

Despite the fact that virological suppression is influenced by ART adherence, the findings in this study have shown existence of other factors including young age being unstable during ART initiation. Unstable during ART initiation is a new factor that has never being reported.