Factors Associated with Antiretroviral Therapy Adherence among Adolescents Living with HIV in Ethiopia: Structural Equation Modeling

doi:10.21203/rs.3.rs-8212039/v1

Download PDF

Research Article

Factors Associated with Antiretroviral Therapy Adherence among Adolescents Living with HIV in Ethiopia: Structural Equation Modeling

https://doi.org/10.21203/rs.3.rs-8212039/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

This study aims to develop and validate a model of antiretroviral therapy (ART) adherence determinants among Ethiopian adolescents living with HIV by extending the information-motivation-behavioral skills (IMB) model through the integration of HIV-related stigma as an additional construct, given its significant influence on adherence in this population.

Methods

A cross-sectional study was conducted from March to October 2024 among 219 adolescents living with HIV in Addis Ababa. The data were analyzed via SPSS and the lavaan 0.6–18 package in R. The pathways through which the study constructs influenced ART adherence were estimated via the diagonally weighted least squares (DWLS) method. Model fit was evaluated via the chi-square to degrees of freedom ratio (χ²/df), comparative fit index (CFI), Tucker–Lewis index (TLI), and root mean square error of approximation (RMSEA).

Results

The findings of the study revealed that ART knowledge and ART attitudes had significant indirect effects on ART adherence (β = .12, p = .021 and β = .10, p = .021, respectively). Social support had direct (β = .18, p = .039) and indirect (β = .08, p = .012) effects on ART adherence. HIV stigma had no significant direct (β=-.21, p = .192) or indirect (β = .10, p = .082) effects on ART adherence. Treatment self-efficacy mediated the relationships among ART knowledge, ART attitudes, social support, and ART adherence.

Conclusion

Treatment self-efficacy mediated ART adherence among adolescents, with social support showing direct and indirect effects, whereas ART knowledge, attitudes, and stigma had no significant direct effects, highlighting the importance of treatment self-efficacy in ART adherence.

antiretroviral therapy

adherence

HIV

adolescents

Ethiopia

structural equation modeling

The introduction of antiretroviral therapy (ART) and advancements in pediatric care have significantly prolonged the lifespan of perinatal HIV-infected children into adolescence and adulthood (1, 2). ART effectively suppresses viral replication, improves immune function, and reduces HIV transmission; however, its success is contingent on sustained adherence to prescribed regimens (3). Adolescents often exhibit lower adherence rates than children and adults do (4, 5) due to unique developmental, emotional and cognitive changes, increasing autonomy, and the transition from the pediatric to the adult healthcare system (6–8). Moreover, this age group faces structural and contextual barriers such as HIV-related stigma (9), financial constraints (10), and limited healthcare access (11). Poor adherence to ART can lead to insufficient plasma drug concentrations, increasing the risk of resistance to one or more medications in the regimen (4, 12). Consequently, this could lead to a greater chance of viral replication, reduce the number of potential future treatment options, and result in disease progression (4, 13, 14).

In Ethiopia, adolescents living with HIV face additional adherence challenges due to sociocultural and religious practices. The fasting tradition of the Ethiopian Orthodox Church and Ramadan among Muslims frequently leads to disruptions in medication schedules, resulting in missed or delayed doses (15). Moreover, religious practices such as visiting holy sites for water-based rituals, which are often perceived as a form of spiritual healing, can interfere with consistent ART adherence (15). Despite these barriers, national HIV policies focus primarily on children and adults, neglecting the specific developmental and cultural needs of adolescents.

Although studies have examined factors influencing ART adherence, their findings often lack consistency and fail to utilize a comprehensive theoretical framework (5). The information-motivation-behavioral skills (IMB) model provides a robust framework for understanding health behaviors by integrating information, motivation, and behavioral skills (16). In HIV care, the IMB model is widely used to provide a comprehensive approach to elucidate the multifactorial influences on adherence (17). This study extended the IMB model by incorporating HIV stigma as an additional construct on the basis of its significant role in adolescent adherence behavior (9, 18, 19).

Using structural equation modeling (SEM), this study aims to construct and validate a model of determinants of ART adherence among Ethiopian adolescents living with HIV. The findings will enhance our understanding of adherence in this group and inform the development of effective interventions tailored to the unique needs of this age group.

Participants

This study employed a cross-sectional design to obtain data concerning factors associated with ART adherence among adolescents receiving ART in a public hospital in Addis Ababa, Ethiopia. Participants were included in the study if they 1) were aged 13–18 years, 2) disclosed or were aware of their disease, indicating the presence of the HIV virus in their body, 3) received ART in the healthcare facility, 4) had family or legal caregivers who could provide consent, and 5) were willing to participate. The participants who were 19 years or older and younger than 13 years old, who had not disclosed their status, who had no family or legal caregivers, except mature minors, and who had not received ART in the healthcare facility were excluded from the study.

The standard for determining the appropriate sample size in structural equation modeling has long been a topic of debate among scholars. While there is no universally accepted standard, the literature suggests that a sample size of 200 or higher is generally recommended to ensure reliable and valid results (20, 21). Therefore, with a 10% nonresponse consideration, the study's total sample size was determined to be 220. However, the study ultimately recruited a sample of 219 adolescents living with HIV who met the inclusion criteria. After thorough data cleaning, the final model included 210 study participants. With this sample size, a power of .94 was achieved, which was calculated on the basis of the root mean square error of approximation (RMSEA-based power) method (22).

Measures

All instruments were used with written permission obtained via email from the original authors.

ART adherence

ART adherence was assessed via a self-reported 6-item simplified medication adherence questionnaire. The questionnaire included “yes” or “no” response questions. Additionally, participants were asked about the frequency of missed doses over the past week and weekend, as well as the number of days without taking any medicine over the past three months. After each response was coded, a total adherence score ranging from 0 to 6 was then calculated. Participants with a total score of 4 or higher indicated adherence, whereas those with scores below 4 indicated nonadherence. For the SEM, all six items were included in the model. This instrument had a Cronbach’s alpha of .72 in the previous study (23) and .83 in the current study.

ART Knowledge

ART knowledge was assessed via a 14-item scale consisting of core domains related to ART knowledge. The participants responded to each item with "right," "wrong," or "I don’t know." One point was assigned for each correct answer, with no points awarded for incorrect or "I don’t know" responses, resulting in a total score ranging from 0–14 points. The scale demonstrated reliability, with a Kuder–Richardson Formula 20 (KR-20) coefficient of .90 in a previous study (24) and .71 in this study.

ART Attitude

Attitude toward the ART was assessed via the modified motivation subscale of the Life Windows Information-Motivation-Behavioral Skills Antiretroviral Therapy Adherence Questionnaire. This subscale comprises 6 items scored on a scale from 1 (strongly disagree) to 5 (strongly agree). The subscale demonstrated reliability, with Cronbach’s α values of 0.81 in a previous study (25) and 0.84 in this study.

Social support

Social support was assessed via the 12-item Multidimensional Scale of Perceived Social Support. The scale has a 7-point Likert-type response format ranging from 1 (very strongly disagree) to 7 (very strongly agree). The scale had reliability, with Cronbach’s α values of .93 in a previous study (26) and .90 in the current study.

HIV stigma

HIV stigma was measured via the 12-item HIV Stigma Scale, which uses a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree). The reliability of the scale was assessed, yielding Cronbach’s α values of .70 in previous studies (27) and .73 in this study.

Treatment self-efficacy

Treatment self-efficacy was assessed via the 12-item HIV Adherence Self-Efficacy Scale, which measures an individual’s confidence in executing essential treatment-related behaviors concerning adherence plans. Responses on the scale range from 1 (cannot do it at all) to 10 (certainly can do it). The scale demonstrated reliability, with a Cronbach’s α of .90 in a previous study (28) and .94 in the current study.

Data collection

The study participants were recruited when they visited the hospital for ART services. For the data collection, two bachelor’s degree nurses were hired. Then, one day of training was given by the researcher for the data collectors. Data collection was conducted from March to October 2024 via a structured, self-administered questionnaire in which adolescents completed the questionnaire in a designated confidential area within the hospital; assistance was provided by data collectors to study participants who encountered difficulties in reading and comprehending the questionnaire. The data collection process lasted approximately 40 minutes per participant, with each study participant receiving 100 Ethiopian Birr to cover their transportation expenses. The principal investigator reviewed the collected data daily to ensure completeness and overall data quality, with incomplete questionnaires exceeding 5% being excluded from the final analysis.

Data Analysis

The data analysis plan involved employing various software programs for specific purposes, including SPSS version 26.0 for descriptive analysis and lavaan 0.6–18 for structural equation modeling (SEM). Descriptive statistics such as frequencies, percentages, means, and standard deviations were used to summarize participant characteristics, and chi-square tests, t tests and Fisher’s exact tests were performed to evaluate the differences between the adherent and nonadherent groups. A significance threshold of a p value < .05 was used to declare statistically significant results.

Two-step structural equation modeling (SEM) was executed, comprising a measurement model that established the validity and reliability of the instruments and a structural model that elucidated the relationships between the constructs and their path relationships in the model. Since the outcome variable was categorical, the diagonally weighted least squares (DWLS) estimation method was employed. To evaluate the model fit, a χ²/df ≤ 5 is considered acceptable, CFI and TLI values ≥ .95 indicate good fit, and an RMSEA value ≤ 0.05 denotes good fit (24).

Demographic and ART-related characteristics

This section presents the test for the general characteristics of the study participants by categorizing them into adherent (n = 144) and nonadherent (n = 66) groups. In this study, 68.6% of the study participants adhered to their ART medication. The mean age of the adherent participants was 15.41 ± 1.51 years, whereas the nonadherent participants were significantly older, with a mean age of 16.24 ± 1.52 years (t = 3.80, p < .001). A significantly greater proportion of adherent participants than nonadherent participants were enrolled in primary school (51.4% and 27.3%, respectively) (χ² = 11.26, p = .002). Moreover, nonadherent participants were significantly more likely to miss clinic appointments due to a lack of transportation fees than their adherent counterparts were (57.6% vs. 40.3%; χ² = 5.46, p = .019). While social support, HIV stigma, and ART knowledge did not significantly differ between the two groups, treatment self-efficacy (t = 2.02, p = .045) and ART attitudes (t = 1.98, p = .049) significantly differed (Table 1).

Table 1

Demographic and ART-related characteristics of the study participants (N = 210)
Variables	Categories	Total	Adherent groups (n = 144)	Nonadherent groups (n = 66)	t or χ²	p
Variables	Categories	n(%) or M\(\:\pm\:\)SD)	n(%) or M\(\:\pm\:\)SD	n(%) or M\(\:\pm\:\)SD	t or χ²	p
Age (Years)		15.7\(\:\pm\:\)1.56	15.41\(\:\pm\:\)1.51	16.24\(\:\pm\:\)1.52	3.80	< .001
Gender	Female	118(56.2)	78(54.2)	40(60.6)	0.76	.383
Gender	Male	92(43.8)	66(45.8)	26(39.4)	0.76	.383
Educational level	Secondary school (9–12)	114(54.3)	68(47.2)	46(69.7)	11.26*	.002
	Primary school (1–8)	92(43.8)	74(51.4)	18(27.3)
	No formal education	4(1.9)	2(1.4)	2(3.0)
Having close person	Yes	191(91.0)	134(93.1)	57(86.4)	2.46	.117
Having close person	No	19(9.0)	10(6.9)	9(13.6)	2.46	.117
Currently living with	Parent (s)	163(77.6)	112(77.7)	51(77.3)	4.26*	.366
	Relatives	16(7.6)	11(7.6)	5(7.6)
	Caregiver	14(6.7)	7(4.9)	7(10.6)
	Sibling (s)	10(4.8)	9(6.3)	1(1.5)
	Alone	7(3.3)	5(3.5)	2(3.0)
Waiting time at clinic	> 1 hour	140(66.7)	94(65.3)	46(69.7)	0.40	.528
Waiting time at clinic	≤ 1 hour	70(33.3)	50(34.7)	20(30.3)	0.40	.528
Distance from clinic	> 10 km	104(49.5)	67(46.6)	37(56.1)	2.02	.365
	5–10 km	68(32.4)	48(33.3)	20(30.3)
	< 5 km	38(18.1)	29(20.1)	9(13.6)
Lack of transportation fee	No	114(54.3)	86(59.7)	28(42.4)	5.46	.019
Lack of transportation fee	Yes	96(45.7)	58(40.3)	38(57.6)	5.46	.019
(Table continued)
Table 1. (Continued)
Variables	Categories	Total	Adherent groups (n = 144)	Nonadherent groups (n = 66)	t or χ²	p
Variables	Categories	n(%) or M\(\:\pm\:\)SD)	n(%) or M\(\:\pm\:\)SD	n(%) or M\(\:\pm\:\)SD	t or χ²	p
ART knowledge		7.32\(\:\pm\:\)2.65	7.35\(\:\pm\:\)2.59	7.26\(\:\pm\:\)2.79	-0.23	.820
ART attitude		3.34\(\:\pm\:\)1.05	3.43\(\:\pm\:\)1.01	3.12\(\:\pm\:\)1.13	1.98	.049
Social support		4.66\(\:\pm\:\)1.34	4.67\(\:\pm\:\)1.34	4.63\(\:\pm\:\)1.53	0.17	.864
HIV stigma		2.61\(\:\pm\:0\).54	2.58\(\:\pm\:0\).57	2.67\(\:\pm\:0\).49	-1.19	.235
Treatment self-efficacy		7.12\(\:\pm\:\)2.29	7.33\(\:\pm\:\)2.14	6.65\(\:\pm\:\)2.52	2.02	.045
* Fisher’s Exact Test M = Mean; SD = Standard deviation; HIV = Human immunodeficiency virus; ART = Antiretroviral therapy; km = Kilometer; N = Total sample size; n = Subsample; χ²= Chi-square; t = t test; p = p value.

Factors associated with ART adherence

Measurement model

The reliability and validity of the instruments were evaluated through confirmatory factor analysis (CFA). The latent variables included in the CFA were treatment self-efficacy, social support, HIV stigma, ART attitudes, ART knowledge, and ART adherence. The results showed that the constructs were generally well measured, as indicated by good reliability and convergent validity (Table 2).

Table 2

Validity and reliability of the measurements(N = 210)
Latent variables	Items	Scale ranges	Standardized estimates (ꞵ)	S.E	z(p)	CR	AVE
Treatment self-efficacy	Total	1–10				.90	.60
	selfe1		.77
	selfe2		.70	.15	5.59(< .001)
	selfe6		.76	.16	5.46(< .001)
	selfe7		.79	.22	4.73(< .001)
	selfe8		.87	.17	6.39(< .001)
	selfe11		.73	.18	5.16(< .001)
Social support	Total	1–7				.92	.74
	sup1		.85
	sup2		.80	.12	7.44(< .001)
	sup3		.91	.15	6.46(< .001)
	sup4		.88	.18	5.43(< .001)
HIV stigma	Total	1–4				.81	.51
	stigm4		.66
	stigm5		.85	.36	3.09(.002)
	stigm6		.62	.27	2.35(.019)
	stigm9		.71	.30	2.60(.009)
ART attitude	Total	1–5				.83	.51
	attit1		.92
	attit2		.74	.15	6.00(< .001)
	attit3		.58	.15	4.67(< .001)
	attit4		.56	.14	4.65(< .001)
	attit5		.71	.18	4.68(< .001)
ART knowledge	Total	1–3				.86	.51
	know2		.75
	know4		.63	.19	5.28(< .001)
	know6		.67	.22	4.43(< .001)
	know8		.65	.21	4.12(< .001)
	know10		.73	.20	4.89(< .001)
	know14		.84	.20	5.64(< .001)
ART adherence	Total	Yes/No				.95	.76
	adher1		.71
	adher2		.84	.11	10.44(< .001)
	adher3		.80	.11	10.95(< .001)
	adher4		.96	.12	13.73(< .001)
	adher5		.91	.12	11.58(< .001)
	adher6		.98	.12	12.65(< .001)
HIV = Human immunodeficiency virus; ART = Antiretroviral therapy; S.E = Standardized error; z(p) = z value and p value; CR = Composite reliability; AVE = Average variance extracted; N = Total sample; selfe = Self-efficacy; sup = Social support; attit = Attitude; adher = Adherence; know = Knowledge; stigm = Stigma.

Structural model

The results of the structural model presented in Table 3 illustrate the direct, indirect, and total effects of various exogenous variables (ART knowledge, ART attitudes, social support, and HIV stigma) on two endogenous variables, namely, treatment self-efficacy and ART adherence. The results highlight that ART knowledge (β = .33, p = .005), ART attitudes (β = .27, p = .003), social support (β = .21, p = .003) and HIV stigma (β = .26, p = .039) have direct significant effects on treatment self-efficacy. These findings suggest that better ART knowledge, a positive attitude toward ART, stronger social support, and HIV stigma are associated with higher levels of treatment self-efficacy. The z values, all above 2, further support the significance of these direct effects (Table 3).

On the other hand, the effects on ART adherence exhibit a more complex pattern. ART knowledge has a nonsignificant direct effect (β=-.22, p = .105) but a significant indirect effect (β = .11, p = .021), resulting in a nonsignificant total effect (β=-.12, p = .395). Similarly, ART attitude has a nonsignificant direct effect (β = .12, p = .234) but a significant indirect effect (β = .09, p = .026) and a nonsignificant total effect (β = .21, p = .056). Social support has a significant direct effect (β = .20, p = .029) and a significant indirect effect (β = .07, p = .014), with a significant total effect of β = .26, p = .004. HIV stigma has a nonsignificant direct effect (β=-.23, p = .143) and an indirect effect (β = .08, p = .088), resulting in a nonsignificant total effect (β=-.14, p = .315). Furthermore, there was a significant direct effect of treatment self-efficacy on adherence to ART (β = .32, p = .002). These results demonstrate the critical roles that social support and treatment self-efficacy play in ART adherence, whereas HIV stigma, ART knowledge and attitudes toward ART have more mediating effects (Table 3).

Moreover, the model fit indices of χ²/df = 1.23, CFI = .99, TLI = .98, and RMSEA = 0.033 confirm that the structural model provides a robust representation of the data, supporting the validity of the proposed relationships between the variables. The good fit of the model underscores the relevance of the identified factors and the pathways through which they influence ART adherence (Table 3).

Table 3

Estimation of the structural model: direct, indirect and total effects(N = 210)
Endogenous variables	Exogenous variables	Direct effect ꞵ(p)	Indirect effect ꞵ(p)	Total effect ꞵ(p)	z value	SMC
Treatment self-efficacy	ART knowledge	.33(.005)	-	.33(.005)	2.80	.38
	ART attitude	.27(.003)	-	.27(.003)	2.98
	Social support	.21(.003)	-	.21(.003)	2.97
	HIV stigma	.26(.039)	-	.26(.039)	2.06
ART adherence	ART knowledge	− .22(.105)	.11(.021)	− .12(.395)	2.31	.23
	ART attitude	.12(.234)	.09(.026)	.21(.056)	2.23
	Social support	.20(.029)	.07(.014)	.26(.004)	2.85
	HIV stigma	− .23(.143)	.08(.088)	− .14(.315)	-1.01
	Treatment self-efficacy	.32(.002)		.32(.002)	3.17
Model fit indices: χ²=514.96, df = 419, p < .001, χ²/df = 1.23, CFI = .99, TLI = .98, RMSEA = 0.033
HIV = Human immunodeficiency virus; ART = Antiretroviral therapy; ꞵ(p) = Standardized estimates (p value); SMC = Squared multiple correlation; χ²=Chi-square; df = Degree of freedom; χ²/df = Chi-square to degree of freedom ratio; CFI = Comparative fit index; TLI = Tucker‒Lewis index; RMSEA = Root mean square error of approximation.

Path diagram

The structural model illustrates the complex interplay of factors influencing ART adherence. The key constructs in the model include ART knowledge, ART attitudes, social support, HIV stigma, treatment self-efficacy, and ART adherence. Notably, the model shows that treatment self-efficacy is a critical mediator in the relationship between these factors and ART adherence.

This study aimed to determine the factors that influence ART adherence among adolescents living with HIV by using the IMB model as a guiding framework and structural equation modeling. The study revealed significant demographic influences and verified the direct and indirect effects of several factors on ART adherence.

The overall ART adherence level in this study was 68.6%, which falls far behind the recommended adherence rate of 95% required to maximize the benefits of ART (29). This adherence level is lower than the 79.1% four-day adherence rate reported in a previous study conducted in Addis Ababa (30). However, this rate exceeds the 65% pooled adherence rate reported in a meta-analysis of 66 studies involving adolescents living with HIV in sub-Saharan Africa (5). Additionally, this rate is much higher than the 34.8% adherence rate reported in Ethiopian children under 15 years of age when adherence was assessed via unannounced home-based pill counts (31). These differences may be attributed to differences in adherence measurement methods, study settings, and populations. While previous studies have examined self-reported four-day adherence (30) or objective measures through unannounced home-based pill counts (31), this study employed a more comprehensive approach by assessing adherence over the last week, weekend and past three months.

The results from SEM analysis indicate that ART knowledge and ART attitudes had significant indirect effects mediated by treatment self-efficacy on ART adherence, whereas social support had direct and indirect effects on ART adherence. HIV stigma had a nonsignificant effect on ART adherence. The model fit indices (χ²/df = 1.23, CFI = .99, TLI = .98, and RMSEA = 0.033) suggest a well-fitting model, supporting the hypothesized relationships.

In this study, ART knowledge was found to have a significant indirect effect on ART adherence via treatment self-efficacy; however, it did not have a significant direct effect on ART adherence. This implies that while extensive knowledge of ART is important, it is insufficient on its own to ensure ART adherence without treatment self-efficacy to apply that ART knowledge effectively. Moreover, traditional health education methods that focus primarily on increasing knowledge might fail to achieve the desired health behaviors unless they also enhance treatment self-efficacy (32, 33). Thus, treatment self-efficacy is a critical mediating construct that converts ART knowledge into ART adherence. This finding was similar to the findings of previous studies indicating that greater knowledge has a positive effect on adolescents’ treatment self-efficacy (33, 34) and the mediating role of treatment self-efficacy (32, 33). Therefore, targeted interventions that enhance both ART knowledge and treatment self-efficacy could significantly promote ART adherence among adolescents living with HIV.

As hypothesized, ART attitude had an indirect significant effect mediated by treatment self-efficacy on ART adherence, although it did not have a direct effect on ART adherence. This finding indicated that without sufficient treatment self-efficacy, even a positive ART attitude might not lead to the desired ART adherence. Hence, treatment self-efficacy acts as a vital mediator that converts a positive ART attitude into ART adherence. This finding contributes to our understanding of the influence of ART attitudes on ART adherence. While previous studies have demonstrated a direct effect of ART attitudes on ART adherence (35–37), this study revealed an alternative pathway in which treatment self-efficacy mediates the relationship between these two factors. This study also revealed a positive relationship between ART attitudes and treatment self-efficacy. This finding supports social cognitive theory, which posits that a positive attitude enhances treatment self-efficacy (38). Therefore, fostering both a positive ART attitude and treatment self-efficacy is essential for improving ART adherence among adolescents living with HIV.

Social support for adolescents living with HIV was found to be a critical factor that had significant positive direct and indirect effects on ART adherence, with the indirect effects being mediated through treatment self-efficacy. This finding highlights the multifaceted ways in which social support can promote adherence to ART. Similar to previous studies (9, 36), this finding confirmed the vital role of social support in enhancing ART adherence among adolescents. Social support can have a direct effect on ART adherence by providing tangible assistance, such as medication reminders or financial support (39). However, the effect of social support on ART adherence is amplified when it enhances treatment self-efficacy, as adolescents living with HIV who feel supported and confident in managing their ART are more likely to sustain optimal ART adherence over time (40). This implies that social support promotes ART adherence not only by providing direct support to adolescents but also by increasing adolescents’ confidence in their ability to manage their ART regimen (10, 41). Therefore, strengthening social support and interventions that enhance treatment self-efficacy can more effectively improve ART adherence among adolescents living with HIV.

In this study, HIV stigma did not have significant direct or indirect effects on ART adherence, despite participants reporting slightly higher-than-average stigma scores. One possible explanation could be the presence of a supportive environment, including healthcare provider support, family support, and spiritual copying, helping adolescents mitigate the effects of stigma (42, 43). However, this interpretation is speculative, as specific supportive environments or religious influences were not directly measured in this study. Additionally, this study is the first to model and empirically test the impact of HIV stigma on ART adherence within the IMB model framework, highlighting the need for further research to determine this relationship. Interestingly, unlike previous studies that reported a negative correlation between HIV stigma and treatment self-efficacy (44), this study revealed a positive correlation. This unexpected finding may suggest that some adolescents develop resilience in response to stigma, potentially bolstered by social support (45), a positive attitude (46) and spiritual copying (42). However, resilience was not directly assessed, and this relationship warrants further investigation. Future studies should incorporate measures of resilience and explore contextual factors to better understand this dynamic.

Treatment self-efficacy is a vital construct in this study, functioning as a mediator between exogenous factors such as ART knowledge, ART attitudes, and social support and the endogenous factor, ART adherence. Consistent with this finding, prior studies have reported that treatment self-efficacy is a significant mediator of ART adherence (47) and the self-management of chronic diseases (32, 33). Strong treatment self-efficacy enhances ART adherence, as individuals with greater confidence are more likely to engage actively and persistently with their healthcare providers' instructions (48). This process is further supported by adolescents’ adequate ART knowledge, positive ART attitudes, and strong social support, which collectively bolster their confidence in managing treatment regimens (49). Therefore, interventions targeting ART adherence in adolescents living with HIV should prioritize enhancing treatment self-efficacy combined with addressing ART knowledge, ART attitudes, and social support to create a synergistic effect that maximizes ART adherence.

Overall, the results from the SEM analysis highlighted the effects of the factors modeled in the IMB model on ART adherence, with treatment self-efficacy serving as a critical mediating factor. While ART knowledge and ART attitudes were significantly associated with treatment self-efficacy, neither demonstrated a direct effect on ART adherence. These findings suggest that knowledge and attitudes toward ART promote ART adherence only through their influence on adolescents’ treatment self-efficacy in managing their treatment effectively. In contrast, social support had both direct and indirect effects on ART adherence, with the indirect effects mediated through treatment self-efficacy. This dual pathway highlights the multifaceted role of social support in promoting ART adherence. These results highlight the importance of designing comprehensive interventions that simultaneously target ART knowledge, promote positive attitudes and strengthen social support to improve treatment self-efficacy and, in turn, ART adherence. Notably, HIV stigma, which was incorporated into the IMB model on the basis of its impact on ART adherence, did not exhibit significant direct or indirect effects on ART adherence in this study, unlike previous findings (19, 50). This difference may be attributable to unexamined factors that mitigate the negative impact of stigma or to the resilience developed by adolescents in response to stigma. Future research is necessary to address these matters.

ART knowledge and ART attitudes were found to have no significant direct effect on ART adherence. According to the IMB model, in the absence of novel behavioral skills, information and motivation can directly influence health behaviors (16). However, the presence of such skills may attenuate their direct effects. In the context of the current study, the mediating role of treatment self-efficacy can be interpreted as a novel behavioral skill that diminishes the direct influence of ART knowledge and ART attitudes on ART adherence. This finding suggests that possessing ART knowledge and a positive attitude toward ART alone may be insufficient to achieve the desired ART adherence. Instead, these factors contribute to enhancing adolescents’ treatment self-efficacy, which then translates ART knowledge and ART attitudes into improved ART adherence. Therefore, interventions aimed at improving ART adherence among adolescents living with HIV should prioritize strengthening treatment self-efficacy, recognizing its essential role in linking ART knowledge and ART attitudes to ART adherence.

In this study, the significant demographic differences observed between the adherent and nonadherent groups highlight the influence of age and educational background on adherence to ART. Younger adolescents and those in primary school were more likely to adhere to ART, whereas older adolescents and those in secondary school were less adherent. These findings are in line with the findings of previous studies (12, 51, 52). These findings suggest that as adolescents grow older and have higher education levels, they may encounter challenges that negatively affect their adherence, potentially due to increased responsibilities such as academic pressure (9, 53), being responsible for their health, as most of them are transitioning to adult care (8), and developing personal identities during this stage of life (6, 7). Thus, interventions that specifically address the unique challenges faced by older adolescents and those in secondary school might be critical in enhancing their ART adherence.

In this study, a lack of money for transportation significantly affected the clinic attendance of the study participants, as nonadherent participants were more likely to miss their clinic appointments than adherent participants were. Similarly, Audi et al. (2021) reported that a significant factor negatively affecting ART adherence among adolescents living with HIV was a lack of transportation fees (10). Therefore, it is important to consider the burden of financial problems that adolescents face when designing an intervention that enhances ART adherence.

Strengths and limitations of the study

To the best of our knowledge, this study represents the first comprehensive investigation into the various factors influencing ART adherence among adolescents living with HIV in Ethiopia. It also utilized the IMB model as a theoretical framework and advanced statistical modeling techniques such as structural equation modeling to construct and empirically test the relationships among the constructs that influence ART adherence. However, this study has several limitations. First, it was conducted at a single center, which might limit the generalizability of its findings to broader settings. It also employs a cross-sectional design, where data are collected from the population at a single point in time, which limits its ability to clearly establish causal relationships among the studied variables because the chronological sequence of events is not captured within this design (54). Furthermore, self-report questionnaires are prone to social desirability bias, where participants may over report positive behaviors (e.g., ART adherence) or underreport negative experiences (e.g., HIV stigma), and recall bias, which can affect the accuracy of responses, particularly for questions requiring memory of past experiences. Therefore, even though the results of this study provide insightful knowledge about factors affecting ART adherence among adolescents living with HIV within a particular study setting, consideration should be taken when extending these findings to other settings.

This study aimed to determine the factors influencing ART adherence among adolescents living with HIV by using structural equation modeling and the IMB model as a guiding framework. The study has identified and shown the specific pathways through which the modeled constructs influence ART adherence. Notably, treatment self-efficacy emerged as a vital mediator in the relationships among ART knowledge, ART attitudes, social support, and ART adherence. While ART knowledge and ART attitudes did not have significant direct effects on ART adherence, social support was found to have both significant direct and indirect effects on ART adherence. Conversely, HIV-related stigma did not demonstrate any significant direct or indirect impact on ART adherence among the study participants. Stigma is still a major obstacle for many, even if it did not have significant effects in this study. Thus, to guarantee thorough support for ART adherence, integrating stigma-reduction strategies into adolescent HIV care systems is essential.

AIDS

acquired immunodeficiency syndrome

ART

antiretroviral therapy

CFA

confirmatory factor analysis

CFI

comparative fit index

DWLS

diagonally weighted least squares

HIV

human immunodeficiency virus

IMB

information–motivation–behavioral skill

RMSEA

root mean square error of approximation

SEM

structural equation modeling

TLI

Tucker–Lewis index

Ethical consideration

Ethical approval was obtained from Keimyung University (IRB No. 40525-202310-HR-049-03) and ALERT/AHRI (PO-04-24). Informed consent and assent were obtained from the caregivers and adolescents.

Consent for publication: Not applicable

Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no known competing interests that could have influenced the work reported in this study.

Funding sources

This work was supported by the authors’ personal contribution.

Authors’ contributions

GGG: conceptualization, methodology, statistical analysis, writing-original draft, writing-review & editing, and visualization.

SJ: conceptualization, methodology, validation, supervision, and review of the manuscript.

Acknowledgments

We extend our sincere gratitude to all the participants for generously volunteering their time and contributing to this study.

Nyakato P, Schomaker M, Sipambo N, Technau KG, Fatti G, Rabie H, et al. Virologic response of adolescents living with perinatally acquired HIV receiving antiretroviral therapy in the period of early adolescence (10–14 years) in South Africa. AIDS. 2021;35(6):971–8.
Slogrove AL, Mahy M, Armstrong A, Davies MA. Living and dying to be counted: What we know about the epidemiology of the global adolescent HIV epidemic. J Int AIDS Soc. 2017;20(3):21520.
Arafat Y, Mohamed Ibrahim MI. The Use of Measurements and Health Behavioral Models to Improve Medication Adherence. Social and Administrative Aspects of Pharmacy in Low- and Middle-Income Countries2018. pp. 53–69.
Kim SH, Gerver SM, Fidler S, Ward H. Adherence to antiretroviral therapy in adolescents living with HIV: Systematic review and meta-analysis. AIDS. 2014;28(13):1945–56.
Hlophe LD, Tamuzi JL, Shumba CS, Nyasulu PS. Barriers and facilitators to anti-retroviral therapy adherence among adolescents aged 10 to 19 years living with HIV in sub-Saharan Africa: A mixed-methods systematic review and meta-analysis. PLoS ONE. 2023;18(5):e0276411.
Tanyi WNG, Odero O, Farquhar T, Kimosop C, Mayi D. Factors affecting adherence to antiretroviral therapy among children and adolescents living with HIV in the Mbita Sub-County Hospital, Homa Bay- Kenya. Afr Health Sci. 2021;21:18–24.
Hegde A, Chandran S, Pattnaik JI. Understanding adolescent sexuality: A developmental perspective. J Psychosexual Health. 2022;4(4):237–42.
Tanner AE, Philbin MM, DuVal A, Ellen J, Kapogiannis B, Fortenberry JD, et al. Transitioning HIV-positive adolescents to adult care: Lessons learned from twelve adolescent medicine clinics. J Pediatr Nurs. 2016;31(5):537–43.
Madiba S, Josiah U. Perceived Stigma and Fear of Unintended Disclosure are Barriers in Medication Adherence in Adolescents with Perinatal HIV in Botswana: A Qualitative Study. BioMed Research International. 2019;2019.
Audi C, Jahanpour O, Antelman G, Guay L, Rutaihwa M, van de Ven R, et al. Facilitators and barriers to antiretroviral therapy adherence among HIV-positive adolescents living in Tanzania. BMC Public Health. 2021;21(1):2274.
Cluver L, Meinck F, Toska E, Orkin FM, Hodes R, Sherr L. Multitype violence exposures and adolescent antiretroviral nonadherence in South Africa. AIDS. 2018;32(8):975–83.
Adejumo OA, Malee KM, Ryscavage P, Hunter SJ, Taiwo BO. Contemporary issues on the epidemiology and antiretroviral adherence of HIV-infected adolescents in sub-Saharan Africa: A narrative review. J Int AIDS Soc. 2015;18(1):20049.
Natukunda J, Kirabira P, Ong KIC, Shibanuma A, Jimba M. Virologic failure in HIV-positive adolescents with perfect adherence in Uganda: A cross-sectional study 11 Medical and Health Sciences 1117 Public Health and Health Services. Trop Med Health. 2019;47(1).
Okonji EF, van Wyk B, Mukumbang FC, Hughes GD. Determinants of viral suppression among adolescents on antiretroviral treatment in Ehlanzeni district, South Africa: a cross-sectional analysis. AIDS Res Therapy. 2021;18(1).
Bezabhe WM, Chalmers L, Bereznicki LR, Peterson GM, Bimirew MA, Kassie DM. Barriers and facilitators of adherence to antiretroviral drug therapy and retention in care among adult HIV-positive patients: A qualitative study from Ethiopia. PLoS ONE. 2014;9(5):e97353.
Fisher WA, Fisher JD, Harman J. The Information-Motivation-Behavioral Skills Model: A General Social Psychological Approach to Understanding and Promoting Health Behavior. Social Psychological Foundations of Health and Illness2003. pp. 82–106.
Fisher JD, Fisher WA, Amico KR, Harman JJ. An information-motivation-behavioral skills model of adherence to antiretroviral therapy. Health Psychol. 2006;25(4):462–73.
Apondi E, Wachira J, Ayikukwei R, Kafu C, Onyango J, Omollo M, et al. Barriers to ART adherence among school students living with HIV in Kenya. Afr J AIDS Res. 2021;20(3):232–7.
Kerkhoff AD, Villiera JB, Katsabola H, Bvumbwe M, Mhango J, Khosa J et al. Factors associated with antiretroviral therapy adherence among adolescents living with HIV in the era of isoniazid preventive therapy as part of HIV care. PLOS Global Public Health. 2022;2(6).
Wolf EJ, Harrington KM, Clark SL, Miller MW. Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety. Educ Psychol Meas. 2013;76(6):913–34.
Wang J, Wang X. Sample size for structural equation modeling. In: David. J. Balding NACC, Garrett M. Fitzmaurice, Harvey Goldstein, lain M. Johnstone, Geert Molenberghs, David W. Scott, Adrian F. M. Smith, Ruey S. Tsay, Sanford Weisberg, editor. Structural equation modeling: Applications using Mplus. 2 ed: John Wiley & Sons Ltd; 2019. pp. 443–81.
Jak S, Jorgensen TD, Verdam MGE, Oort FJ, Elffers L. Analytical power calculations for structural equation modeling: A tutorial and Shiny app. Behav Res Methods. 2021;53(4):1385–406.
Agala CB, Fried BJ, Thomas JC, Reynolds HW, Lich KH, Whetten K, et al. Reliability, validity and measurement invariance of the Simplified Medication Adherence Questionnaire (SMAQ) among HIV-positive women in Ethiopia: a quasiexperimental study. BMC Public Health. 2020;20(1):567.
Xu D, Shi Y, Pan L, Duan Q, Huang N, Liu P, et al. Assessment of an instrument scale measuring the knowledge of antiretroviral therapy among people living with HIV. BMC Public Health. 2023;23(1):278.
Peng Z, Yu Y, Wei W, Hou Y, Sun Z, Wang Y, et al. Reliability and Validity of the LifeWindows Information–Motivation–Behavioral Skills Antiretroviral Therapy Adherence Questionnaire Among HIV + Patients in Shanghai. Patient Prefer Adherence. 2020;14507–15.
Wang X, Cheng Z Cross-sectional studies: Strengths, weaknesses, and recommendations.Chest. 2020; 158((1S):):S65-S71.
Zimet GD, Canty-Mitchell J. Psychometric properties of the multidimensional scale of perceived social support in urban adolescents. Am J Community Psychol. 2000;28(3):391–400.
Reinius M, Wettergren L, Wiklander M, Svedhem V, Ekstrom AM, Eriksson LE. Development of a 12-item short version of the HIV stigma scale. Health Qual Life Outcomes. 2017;15(1):115.
Johnson MO, Neilands TB, Dilworth SE, Morin SF, Remien RH, Chesney MA. The role of self-efficacy in HIV treatment adherence: validation of the HIV Treatment Adherence Self-Efficacy Scale (HIV-ASES). J Behav Med. 2007;30(5):359–70.
Paterson DL, Swindells S, Mohr J, Brester M, Vergis EN, Squier C, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med. 2000;133(1):21–30.
Firdu N, Enquselassie F, Jerene D. HIV-infected adolescents have low adherence to antiretroviral therapy: a cross-sectional study in Addis Ababa, Ethiopia. Pan Afr Med J. 2017;27:80.
Biressaw S, Abegaz WE, Abebe M, Taye WA, Belay M. Adherence to Antiretroviral Therapy and associated factors among HIV infected children in Ethiopia: unannounced home-based pill count versus caregivers' report. BMC Pediatr. 2013;13:132.
Chuang LM, Wu SV, Lee MC, Lin LJ, Liang SY, Lai PC, et al. The effects of knowledge and self-management of patients with early-stage chronic kidney disease: Self-efficacy is a mediator. Japan J Nurs Sci. 2021;18(2):e12388.
Wu SFV, Hsieh NC, Lin LJ, Tsai JM. Prediction of self-care behavior on the basis of knowledge about chronic kidney disease using self-efficacy as a mediator. J Clin Nurs. 2016;25(17–18):2609–18.
Hornschuh S, Dietrich JJ, Tshabalala C, Laher F. Antiretroviral treatment adherence: Knowledge and experiences among adolescents and young adults in Soweto, South Africa. AIDS Res Treat. 2017;2017:5192516.
Mussa FM, Massawe HP, Bhalloo H, Moledina S, Assenga E. Magnitude and associated factors of antiretroviral therapy adherence among children attending HIV care and treatment clinics in Dar es Salaam, Tanzania. PLoS ONE. 2022;17(9 September).
Jjumba I, Kanyesigye M, Ndagijimana G, Wattira J, Olong C, Olok RA, et al. Perceived barriers and facilitators to antiretroviral therapy adherence among youth aged 15–24 years at a regional HIV clinic in SouthWestern Uganda: a qualitative study. Afr Health Sci. 2022;22(2):54–62.
Langebeek N, Elizabeth HG, Reiss P, Sigrid C, Vervoort TBH, Clemens Richter, Mirjam AG, Sprangers, Pythia T, Nieuwkerk EH, Gisolf P, Reiss SC, Vervoort. Thóra B Hafsteinsdóttir, Clemens Richter, Mirjam AG Sprangers and Pythia T Nieuwkerk. Predictors and correlates of adherence to combination antiretroviral therapy (ART) for chronic HIV infection: A meta analysis. BMC Med. 2014;12:142.
Bandura A. Self-efficacy. In V. S. Ramachaudran,Encyclopedia of human behavior (Vol. 4, pp. 71–81). New York: Academic Press. (, Reprinted. in H. Friedman, editor, Encyclopedia of mental health. San Diego: Academic Press, 1998). 1994.
Steinert JI, Shenderovich Y, Smith M, Zhou S, Toska E, Cluver L. Economic Well-being and Associated Mediating Pathways to Improved Antiretroviral Therapy Adherence Among Adolescents Living With HIV: A Prospective Cohort Study in South Africa. J Acquir Immune Defic Syndr. 2022;91(4):343–52.
Kekwaletswe CT, Jordaan E, Nkosi S, Morojele NK. Social support and the mediating roles of alcohol use and adherence self-efficacy on antiretroviral therapy (ART) adherence among ART recipients in Gauteng, South Africa. AIDS Behav. 2017;21(7):1846–56.
Huang H, Tu L, Zhang X, Zhang L, Zhang J, Liu Q, et al. Social support, self-efficacy, self-esteem, and self-management behaviors among people living with HIV/AIDS in China: A structural equation modeling analysis. BMC Public Health. 2024;24(1):3070.
Martinez J, Harper G, Carleton RA, Hosek S, Bojan K, Clum G, et al. The impact of stigma on medication adherence among HIV-positive adolescent and young adult females and the moderating effects of coping and satisfaction with health care. AIDS Patient Care STDs. 2012;26(2):108–15.
Zou J, Yamanaka Y, John M, Watt M, Ostermann J, Thielman N. Religion and HIV in Tanzania: Influence of religious beliefs on HIV stigma, disclosure, and treatment attitudes. BMC Public Health. 2009;9:75.
Umar E, Levy JA, Donenberg G, Mackesy-Amiti ME, Pujasari H, Bailey RC. The influence of self-efficacy on the relationship between depression and HIV-related stigma with ART adherence among the youth in Malawi. Jurnal Keperawatan Indonesia. 2019;22(2):147–60.
Masten AS. Resilience from a developmental systems perspective. World Psychiatry. 2019;18(1):100–1.
Pennock SF, Latif S. Resilience in Positive Psychology: How to Bounce Back 2017 [updated 2024 November 22. Available from: https://positivepsychology.com/resilience-in-positive-psychology/.
Zhang T, Wilson IB, Youn B, Lee Y, Shireman TI. Use of Antiretroviral Therapy for a US Medicaid Enrolled Pediatric Cohort with HIV. AIDS Behav. 2021;25(8):2455–62.
Zhang L, Li X, Lin Z, Jacques-Tiura AJ, Xu J, Zhou Y, et al. Side effects, adherence self-efficacy, and adherence to antiretroviral treatment: a mediation analysis in a Chinese sample. AIDS Care. 2016;28(7):919–26.
Chen YY, Weng LC, Li YT, Huang HL. Mediating effect of self-efficacy on the relationship between social support and self-management behaviors among patients with knee osteoarthritis: a cross-sectional study. BMC Geriatr. 2022;22(1):635.
Madiba S, Mokgatle M. Fear of stigma, beliefs, and knowledge about HIV are barriers to early access to HIV testing and disclosure for perinatally infected children and adolescents in rural communities in South Africa. South Afr Family Pract. 2017;59(5):175–81.
Hudelson C, Cluver L. Factors associated with adherence to antiretroviral therapy among adolescents living with HIV/AIDS in low- and middle-income countries: A systematic review. AIDS Care - Psychol Socio-Medical Aspects AIDS/HIV. 2015;27(7):805–16.
Wiggins L, Gabrielle OM, Anjuli DW, Mutisya. I, Wilson. KS, Lawrence S, et al. They can stigmatize you’: A qualitative assessment of the influence of school factors on engagement in care and medication adherence among adolescents with HIV in western Kenya. Health Educ Res. 2022;37(5):355–63.
MacCarthy S, Saya U, Samba C, Birungi J, Okoboi S, Linnemayr S. How am I going to live? exploring barriers to ART adherence among adolescents and young adults living with HIV in Uganda. BMC Public Health. 2018;18(1):1158.

No competing interests reported.

Download PDF

Reviews received at journal
27 Dec, 2025
Reviews received at journal
23 Dec, 2025
Reviewers agreed at journal
23 Dec, 2025
Reviewers agreed at journal
21 Dec, 2025
Reviewers agreed at journal
19 Dec, 2025
Reviewers agreed at journal
19 Dec, 2025
Reviewers invited by journal
19 Dec, 2025
Editor assigned by journal
26 Nov, 2025
Submission checks completed at journal
26 Nov, 2025
First submitted to journal
26 Nov, 2025

You are reading this latest preprint version

Factors Associated with Antiretroviral Therapy Adherence among Adolescents Living with HIV in Ethiopia: Structural Equation Modeling

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Background

Methods and Materials

Participants

Measures

ART adherence

ART Knowledge

ART Attitude

Social support

HIV stigma

Treatment self-efficacy

Data collection

Data Analysis

Results

Demographic and ART-related characteristics

Factors associated with ART adherence

Measurement model

Structural model

Path diagram

Discussion

Strengths and limitations of the study

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1