Abstract
Background:
Nonadherence to antiretroviral therapy (ART) remains a critical barrier to achieving optimal treatment outcomes among people living with HIV (PLHIV). Understanding multidimensional factors influencing adherence is essential for targeted interventions.
Aim and Objective:
This study aimed to assess determinants of nonadherence to ART among PLHIV attending a tertiary care hospital in Jabalpur, Madhya Pradesh, India.
Materials and Methods:
A cross-sectional study was conducted among 1,067 PLHIV aged ≥18 years receiving ART for ≥6 months at a tertiary care hospital in Jabalpur, Madhya Pradesh. Adherence was assessed using the pill count method, with <95% defined as nonadherence. Data on sociodemographic, patient-related, condition-related, therapy-related, and health system-related factors were collected through structured interviews and medical records review. Chi-square tests and odds ratios were calculated to identify significant associations.
Results:
Nonadherence prevalence was 17.5%. Significant predictors included dissatisfaction with family support (aOR = 38.61), NGO linkage (aOR = 3.82), non-heterosexual transmission routes (aOR = 3.09), current CD4 count <200 cells/mm3 (aOR = 2.90), and age ≤35 years (aOR = 1.51).
Conclusion:
This study found a 17.5% nonadherence rate among PLHIV on ART in Jabalpur. Younger age, non-heterosexual transmission, poor family support, and low CD4 counts were key predictors. Strengthening social support, addressing stigma, and focusing on vulnerable groups may help improve ART adherence and contribute to progress toward the 95-95-95 targets.
Keywords: Antiretroviral therapy, determinants, family support, HIV, nonadherence, stigma
INTRODUCTION
Human immunodeficiency virus (HIV) remains a major global health concern despite significant advances in prevention and treatment. According to UNAIDS 2024 estimates, approximately 39 million people are living with HIV globally. People Living with HIV (PLHIV) require sustained adherence to antiretroviral therapy (ART) to achieve optimal treatment outcomes.[1] India accounts for 6.3% of global HIV cases, with an estimated 25.44 lakh people living with HIV and an adult prevalence of 0.20%.[2] In Madhya Pradesh, the adult prevalence is 0.10%, with approximately 70,170 PLHIV.[2] Antiretroviral therapy (ART) has transformed HIV from a fatal illness into a manageable chronic condition, improving survival and quality of life for PLHIV. Free ART centers have been crucial in reducing HIV-related mortality.[3] However, maintaining high adherence—at least 95% as recommended by the World Health Organization (WHO)—is essential for effective viral suppression, preventing disease progression, and avoiding drug resistance.[4]
The Joint United Nations Programme on HIV/AIDS (UNAIDS) has set ambitious targets of 95-95-95: 95% of people living with HIV knowing their status, 95% of those diagnosed receiving sustained ART, and 95% on ART achieving viral suppression.[5] Achieving these targets requires a comprehensive understanding of adherence barriers at the local level to design effective retention strategies and patient-centered care models.[5]
Nonadherence to ART represents a multidimensional challenge influenced by various interconnected factors. The World Health Organization identifies five key dimensions affecting adherence: socioeconomic factors, patient-related factors, condition-related factors, therapy-related factors, and healthcare system-related factors. Poor adherence results in inadequate viral suppression, immunological failure, drug resistance, disease progression, and treatment failure.[4]
Despite free ART availability and comprehensive support services, adherence challenges persist across diverse Indian settings. Previous studies have documented nonadherence rates ranging from 10.9% to 41% in different regions, with considerable heterogeneity in adherence patterns, barriers, and facilitators.[6,7,8] In the context of central India, particularly Madhya Pradesh, limited region-specific data exists on ART adherence patterns and associated factors. Geographic, cultural, socioeconomic variations, and local healthcare delivery systems necessitate localized evidence to inform context-appropriate interventions. Jabalpur, a major urban center in Madhya Pradesh, serves a diverse population with varying degrees of healthcare access, educational attainment, and socioeconomic status—all potentially impacting ART adherence behaviors. Therefore, this study was conducted to assess the prevalence of nonadherence to ART and factors associated with nonadherence among PLHIV registered at an ART center in Jabalpur.
MATERIALS AND METHODS
Study Design and Setting: A cross-sectional study was conducted at the Antiretroviral Therapy (ART) center of Netaji Subhash Chandra Bose Medical College, Jabalpur, Madhya Pradesh, between November 2023 and August 2024.
Study Population and Eligibility: The study population comprised adult PLHIV registered at the ART center. Inclusion criteria were: (1) age ≥18 years, (2) receiving ART for at least six months, and (3) willing to provide written informed consent. Exclusion criteria included: (1) diagnosed mental illness that would impair understanding or consent, and (2) severely ill patients unable to participate in interviews at the time of data collection.
Sample Size Calculation: Assuming a 50% prevalence of nonadherence (to maximize sample size in absence of precise local estimates), 95% confidence level, and 3% absolute precision, the minimum required sample size was calculated using the formula for single proportion: n = z²pq/e², Where z = 1.96 (for 95% confidence), P = 0.5, q = 0.5, and e = 0.03, This yielded a minimum sample size of 1,067 participants. A 3% absolute precision was adopted to maximize statistical accuracy and narrow confidence intervals, accounting for variability in adherence behavior within a large tertiary care population. Recruitment feasibility was ensured by the high patient load at the ART center.
Sampling Technique: Systematic random sampling was employed. From approximately 10,000 registered ART patients, a sampling interval of 9 was calculated (9,895/1,067 ≈ 9). Data collection was conducted three days per week (to capture different days and time slots). The first participant was randomly selected from the first nine eligible patients, followed by every 9th eligible patient thereafter. If a selected patient was unavailable or declined participation, the next eligible patient in sequence was approached. All eligible patients provided written informed consent, and 1,067 participants were enrolled consecutively and included in the final analysis, with no exclusions due to incomplete outcome data.
Data Collection: Data were collected through structured face-to-face interviews using a pre-designed, pre-tested questionnaire developed based on the WHO adherence framework and adapted from previously published ART adherence questionnaires used in Indian settings.[7,8] The questionnaire was pilot-tested among 30 participants before final administration. Reliability analysis for key domains, including family support and perceived stigma, demonstrated acceptable internal consistency (Cronbach’s alpha >0.7). Necessary modifications were made based on pilot findings. Data collected included:
Sociodemographic variables: Age, gender, education, occupation, socioeconomic status (Modified BG Prasad classification, 2023), marital status, family type, and locality.
Patient-related factors: Substance use (alcohol, smoking, tobacco), risk group classification, route of HIV transmission, family support, stigma experiences, and disclosure status
Condition-related factors: Comorbidities, WHO clinical staging, CD4 count (baseline and current), viral load, opportunistic infections, and HIV-TB coinfection. The CD4 count threshold of 200 cells/mm3 was used as it is a well-established clinical cut-off to define advanced immunosuppression and increased risk of opportunistic infections, as recommended in WHO guidelines for HIV management.[4]
Therapy-related factors: ART regimen, treatment duration, medication side effects. Medication side effects assessed included nausea, vomiting, dizziness, fatigue, and other patient-reported adverse effects.
Health system-related factors: NGO linkages, counseling received.
High-risk groups included injection drug users (IDU), men who have sex with men (MSM), transgender individuals (TG), and sex workers, as per NACO classification. Comorbidity was defined as the presence of one or more chronic conditions in addition to HIV, including tuberculosis, diabetes, hypertension, hepatitis B/C, and other clinically diagnosed illnesses. Pre-ART HIV-TB coinfection was defined as a documented history of active tuberculosis prior to initiation of antiretroviral therapy, as recorded in medical records. This variable reflects prior disease burden and treatment complexity rather than current TB status. Data were obtained from medical records and supplemented by patient selfreport when necessary. Spouse HIV status was self-reported by participants.
Assessment of Adherence: Adherence was assessed using the pill count method, a validated objective approach, calculated as: Adherence (%) = (Total doses taken/Total doses prescribed) ×100. Data were collected by the principal investigator, who received standardized training in interview techniques, research ethics, confidentiality, and questionnaire administration. Pill counts were performed using a standardized protocol. Based on adherence levels, patients were categorized as optimal (≥95%), suboptimal (80–94%), or poor (<80%). For analysis, adherence was dichotomized into adherent (≥95%) and nonadherent (<95%) groups. Additionally, participants reported missed doses over the past week, month, and six months, with reasons for nonadherence elicited through open-ended and structured questions.[4] Patients were also asked about reasons for missing medications through open-ended questions, with responses categorized into predefined themes aligned with WHO’s adherence framework. Adherence was assessed primarily through pill counts. Self-reported missed doses were collected only to supplement and triangulate adherence behavior, without inclusion in statistical classification.
Ethical Considerations: The study was approved by the Institutional Ethics Committee of Netaji Subhash Chandra Bose Medical College (IEC/2023/7335-94, dated 18/08/2023). Written informed consent was obtained from all participants before their inclusion in the study. Participants were informed that refusal to participate or withdrawal from the study at any stage would not affect their ongoing treatment or services received at the ART center. Interviews were conducted in a private setting to ensure confidentiality. Data were de-identified using unique study IDs and stored securely with access restricted to the investigators.
Statistical Analysis: Data were entered into Microsoft Excel 2010. Analysis was performed using IBM SPSS Statistics version 23.0. Data were checked for completeness at the time of collection. Records with missing key variables related to adherence or major predictors were excluded from analysis (complete case analysis). There were no missing data for the study variables, and no participant was excluded from the analysis due to incomplete records. Therefore, all 1,067 enrolled participants were included in the final analysis. Descriptive statistics included frequencies, percentages, means, and standard deviations for continuous variables. Categorical variables were presented as frequencies and proportions. Chi-square test was used for categorical variables. Yates’ continuity correction was applied where appropriate for small cell counts. Fisher’s exact test was used when expected cell counts were less than 5. The independent t-test was used for comparison of continuous variables. Variables with a P value < 0.25 in univariate analysis were entered into a multivariate logistic regression model to identify independent predictors of ART nonadherence. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were computed. Multicollinearity among predictor variables was assessed using variance inflation factors (VIF). All VIF values were below 2.0, indicating no significant multicollinearity.
RESULTS
Characteristics of patients: A total of 1,067 HIV-positive patients attending the Jabalpur ART Centre were included in the study. The mean age was 34.8 ± 10.7 years, with a predominance of males (61.9%). Most participants had primary education (68.1%) and belonged to lower (41.4%) or lower-middle (33.5%) socioeconomic classes. The majority were married (87.5%), resided in rural areas (55.7%), and lived in joint families (61%). Substance use was common—smoking (43.1%), alcohol (42.9%), and smokeless tobacco (26%). Comorbidities (8.4%) included diabetes (5.2%), HIV/TB (3.7%), and hypertension (1.9%). Heterosexual contact was the predominant mode of transmission (96.1%), with a small proportion belonging to high-risk groups (3.2%), including injection drug users, men who have sex with men, and sex workers. Most participants were in WHO clinical Stage I (65.3%), with opportunistic infections present in 6.5%. Family support was reported by the majority (99.2%), while perceived stigma and NGO linkage were low (0.6% and 3.1%, respectively). Nearly all participants had disclosed their HIV status (99.4%). The majority of participants (97.5%) were receiving a first-line ART regimen, while a small proportion were on second-line (0.5%) and third-line regimens (2.0%). Due to the limited number of patients on alternative regimens, treatment regimen was not included as a variable in further analysis.
Figure 1 illustrates the distribution of ART adherence among study participants. Overall, adherence was high (n = 880, 82.5%), with a smaller proportion of patients (n = 187, 17.5%) classified as nonadherent.
Figure 1.
Level of adherence among HIV patients (n = 1067)
Age ≤35 years was significantly associated with nonadherence (OR = 1.62, P < 0.05). No significant associations were observed with gender, education, occupation, socioeconomic status, marital status, type of family, or locality (P > 0.05 for all) [Table 1].
Table 1.
Sociodemographic factors associated with ART adherence among HIV patients (n=1067)
| Sociodemographic factors | Nonadherent (n=187) | Adherent (n=880) | OR (95% CI) | P | ||||
|---|---|---|---|---|---|---|---|---|
| Age group (years) | ||||||||
| ≤35 | 130 (69.5%) | 515 (58.5%) | 1.62 (1.15 to 2.27) | P=0.007 | ||||
| >35 | 57 (30.5%) | 365 (41.5%) | ||||||
| Gender | ||||||||
| Male | 113 (60.4%) | 547 (62.2%) | 0.93 (0.67–1.28) | P=0.719 | ||||
| Female | 74 (39.6%) | 333 (37.8%) | ||||||
| Education | ||||||||
| Up to high school | 186 (99.5%) | 868 (98.6%) | 2.57 (0.33–19.90) | P=0.177 | ||||
| Above high school | 1 (0.5%) | 12 (1.4%) | ||||||
| Occupation | ||||||||
| Unemployed | 59 (31.6%) | 265 (30.1%) | 1.07 (0.76 to 1.50) | P=0.505 | ||||
| Employed | 128 (68.4%) | 615 (69.9%) | ||||||
| Socio economic status as per modified BG Prasad scale (2023) | ||||||||
| Class I, II, III | 151 (80.7%) | 648 (73.6%) | 0.67 (0.45 to 0.99) | P=0.428 | ||||
| Class IV, V | 36 (19.3%) | 232 (26.4%) | ||||||
| Marital status | ||||||||
| Single, Divorcee, Widowed | 50 (26.7%) | 226 (25.7%) | 1.06 (0.74 to 1.51) | P=0.629 | ||||
| Married | 137 (73.3%) | 654 (74.3%) | ||||||
| Type of Family | ||||||||
| Nuclear, Living alone | 63 (33.7%) | 353 (40.1%) | 0.76 (0.54 to 1.06) | P=0.356 | ||||
| Joint | 124 (66.3%) | 527 (59.9%) | ||||||
| Locality | ||||||||
| Rural | 112 (59.9%) | 482 (54.8%) | 1.23 (0.89 to 1.70) | P=0.857 | ||||
| Urban | 75 (40.1%) | 398 (45.2%) |
Patients belonging to high-risk groups were significantly more likely to be nonadherent (9.1%) compared to adherent patients (1.9%) (P < 0.001). Patients whose route of transmission was non-heterosexual were significantly more likely to be nonadherent (9.1%) compared to adherent patients (2.8%) (P = 0.001). Dissatisfaction with family: patients with poor family support had significantly higher odds of nonadherence to ART (OR of 43.61, 95% CI: 5.23–363.78; P < 0.001). However, the estimate for dissatisfaction with family support should be interpreted cautiously due to very small numbers in the exposed category, resulting in a wide confidence interval and potential statistical instability. Perceived stigma was reported exclusively by nonadherent patients (3.7%), with none among adherent patients (P < 0.001). Alcohol consumption, smoking, smokeless tobacco use, disclosure status, and spouse’s HIV status did not significantly affect ART adherence. (P > 0.05 for all) [Table 2].
Table 2.
Patient-related factors associated with ART adherence among HIV patients
| Patient-related factors | Nonadherent (n=187) | Adherent (n=880) | OR (95% CI) | χ2 and P | ||||
|---|---|---|---|---|---|---|---|---|
| Alcohol consumption | ||||||||
| Yes | 87 (46.5%) | 371 (42.2%) | 1.19 (0.87 to 1.62) | P=0.773 | ||||
| No | 100 (53.5%) | 509 (57.8%) | ||||||
| Smoking | ||||||||
| Yes | 88 (47.1%) | 372 (42.3%) | 1.22 (0.89 to 1.68) | P=0.715 | ||||
| No | 99 (52.9%) | 508 (57.7%) | ||||||
| Smokeless tobacco | ||||||||
| Yes | 59 (31.6%) | 218 (24.8%) | 1.41 (0.98 to 2.03) | P=0.437 | ||||
| No | 128 (68.4%) | 662 (75.2%) | ||||||
| Risk group based on community transmission categories | ||||||||
| General population | 168 (89.8%) | 857 (97.4%) | NA | P<0.001 | ||||
| High-risk group | 17 (9.1%) | 17 (1.9%) | ||||||
| Bridge population | 2 (1.1%) | 6 (0.7%) | ||||||
| Sexual Risk Behavior | ||||||||
| Heterosexual | 169 (90.4%) | 856 (97.3%) | 3.44 (1.82 to 6.52) | P=0.001 | ||||
| Others | 17 (9.1%) | 25 (2.8%) | ||||||
| Family Support | ||||||||
| Dissatisfied | 8 (4.3%) | 1 (0.1%) | 43.61 (5.23 to 363.7) | P<0.001 | ||||
| Satisfied | 179 (95.7%) | 879 (99.9%) | ||||||
| Stigma | ||||||||
| Yes | 7 (3.7%) | 0 (0%) | NA | P<0.001 | ||||
| No | 180 (96.3%) | 880 (100%) | ||||||
| Disclosure Status | ||||||||
| Yes | 184 (98.4%) | 877 (99.7%) | 5.33 (0.85 to 33.45) | P=0.998 | ||||
| No | 3 (1.6%) | 3 (0.3%) | ||||||
| Spouse/Partner Affected With HIV | ||||||||
| Yes | 155 (82.9%) | 756 (85.9%) | 1.26 (0.91 to 1.75) | P=0.164 | ||||
| No | 32 (17.1%) | 124 (14.1%) |
Patients with a CD4 count <200 cells/mm3 had higher odds of being nonadherent to ART compared to those with a CD4 count >200 (OR = 2.99, 95% CI: 1.93–4.65; P < 0.001). However, comorbidities, pre-ART HIV-TB coinfection, clinical stage at the time of diagnosis, CD4 count before ART initiation, and opportunistic infections did not significantly affect ART adherence (P > 0.05 for all). [Table 3].
Table 3.
Clinical condition-related factors associated with ART adherence among HIV patients
| Clinical condition-related factors | Nonadherent (n=187) | Adherent (n=880) | OR (95% CI) | χ2 and P | ||||
|---|---|---|---|---|---|---|---|---|
| Comorbidity | ||||||||
| Yes | 30 (16%) | 139 (15.8%) | 1.02 (0.67 to 1.55) | P=0.820 | ||||
| No | 157 (84%) | 741 (84.2%) | ||||||
| Pre ART HIV-TB coinfection | ||||||||
| Present | 5 (2.7%) | 35 (4%) | 0.65 (0.24 to 1.68) | P=0.673 | ||||
| Absent | 182 (97.3%) | 845 (96%) | ||||||
| Clinical stage at the time diagnosis as per WHO | ||||||||
| I (Asymptomatic) | 126 (67.4%) | 571 (64.9%) | 0.90 (0.64 to 1.28) | P=0.548 | ||||
| II, III & IV | 61 (32.6%) | 309 (35.1%) | ||||||
| Opportunistic infections (OI) | ||||||||
| Present | 10 (5.3%) | 59 (6.7%) | 0.78 (0.39 to 1.53) | P=0.730 | ||||
| Absent | 177 (94.7%) | 821 (93.3%) | ||||||
| CD4 count before initiation of ART | ||||||||
| <200 | 58 (31%) | 274 (31.1%) | 0.99 (0.70 to 1.41) | P=0.917 | ||||
| >200 | 129 (69%) | 606 (68.9%) | ||||||
| Present CD4 count | ||||||||
| <200 | 38 (20.3%) | 70 (8%) | 2.99 (1.93 to 4.65) | P<0.001 | ||||
| >200 | 149 (79.7%) | 810 (92%) |
Side effects did not significantly impact adherence. Patients with treatment duration >5 years had higher odds of nonadherence than those with ≤5 years of treatment (17.1%) (OR: 1.39, 95% CI: 1.08–1.81; P = 0.010). Patients with NGO linkage had a significantly higher likelihood of being nonadherent (OR: 4.74, 95% CI: 2.32–9.70; P < 0.001). [Table 4].
Table 4.
Therapy and health system-related factors associated with ART adherence among HIV patients
| Therapy-related factors | Nonadherent (n=187) | Adherent (n=880) | OR (95% CI) | χ2 and P | ||||
|---|---|---|---|---|---|---|---|---|
| Side effect* | ||||||||
| Yes | 1 (0.5%) | 13 (1.5%) | 0.35 (0.04 to 2.82) | P=0.237 | ||||
| No | 186 (99.5%) | 867 (98.5%) | ||||||
| Duration of treatment | ||||||||
| >5 years | 155 (82.9%) | 683 (77.6%) | 1.39 (1.08 to 1.81) | P=0.010 | ||||
| ≤5 years | 32 (17.1%) | 197 (22.4%) | ||||||
| NGO link# | ||||||||
| Yes | 16 (8.6%) | 17 (1.9%) | 4.74 (2.32 to 9.70) | P<0.001 | ||||
| No | 171 (91.4%) | 863 (98.1%) |
*Side effects include nausea, vomiting, dizziness, fatigue. #NGO linkage refers to association with community-based organizations
Multivariate analysis of factors associated with ART nonadherence
Variables with P < 0.25 on univariate analysis (n = 7) were included in the multivariate logistic regression model. The final model demonstrated good fit and discrimination (Hosmer–Lemeshow χ² = 8.64, P = 0.373; Nagelkerke R² = 0.253; AUC-ROC = 0.762), with an overall classification accuracy of 84.8% (sensitivity 28.9%, specificity 96.8%). Although the model demonstrated high specificity (96.8%), the low sensitivity (28.9%) indicates limited ability to correctly identify nonadherent individuals, suggesting that additional unmeasured factors may influence adherence. Five variables emerged as independent predictors of ART nonadherence (P < 0.05). Younger age (≤35 years), non-heterosexual mode of transmission, dissatisfaction with family support, current CD4 count <200 cells/µL, and linkage to NGOs were significantly associated with higher odds of nonadherence. Overall, the model explains 25.3% of the variance in ART nonadherence and highlights the critical role of family support, immunological status, age, and psychosocial vulnerability in adherence outcomes [Table 5].
Table 5.
Multivariate analysis of factors associated with ART nonadherence
| Predictor | Adjusted OR | 95% CI | P | Interpretation | ||||
|---|---|---|---|---|---|---|---|---|
| Age ≤35 years | 1.51 | 1.06–2.14 | 0.021 | Younger patients 51% more likely to be nonadherent | ||||
| Sexual Risk Behavior (Others) | 3.09 | 1.51–6.32 | 0.002 | Non-heterosexual transmission 3× higher risk | ||||
| Family Support (Dissatisfied) | 38.61 | 4.78–312.07 | 0.001 | Strongest predictor-nearly 39× higher risk | ||||
| Present CD4 <200 | 2.90 | 1.84–4.57 | <0.001 | Low immunological status increases risk 3-fold | ||||
| NGO Linkage (Yes) | 3.82 | 1.68–8.68 | 0.001 | Paradoxically, NGO linkage associated with higher nonadherence |
DISCUSSION
This cross-sectional study evaluated factors influencing ART nonadherence among 1,067 PLHIV in Jabalpur, Madhya Pradesh and revealed a nonadherence prevalence of 17.5%. This finding aligns with previous Indian studies reporting nonadherence rates ranging from 10.9% to 28%.[6,7,8,9,10,11] The observed nonadherence (17.5%) in the present study was lower than the pooled estimate of 45.9% reported by Basu et al.[12] in a recent systematic review and meta-analysis from India. The adherence rate of 82.5% in our study, while encouraging, falls short of the WHO-recommended ≥95% threshold necessary for optimal viral suppression and prevention of drug resistance.
Sociodemographic predictors
A recent study from Jhansi similarly identified younger individuals as being at higher risk of poor adherence.[6] This may be attributed to several factors, including denial of chronic illness, peer-related stigma, competing life priorities such as education and career progression, irregular lifestyles, and a lower perceived sense of vulnerability. These challenges highlight the need for targeted, age-appropriate interventions that emphasize lifelong treatment adherence.
In contrast, variables such as gender, education, occupation, socioeconomic status, marital status, family type, and locality were not significantly associated with ART adherence in the present study. This differs from earlier research, which has identified demographic and socioeconomic disadvantages—such as male gender, lower educational attainment, unemployment, and low socioeconomic status—as important predictors of nonadherence.[6] The lack of such associations in our study may reflect improvements in HIV care delivery, including strengthened counseling services, gender-sensitive approaches, flexible ART delivery systems, and enhanced social and community support, which may have mitigated traditional barriers and reduced the impact of stigma.
Patient-related factors
In the present study, non-heterosexual modes of transmission were associated with higher levels of nonadherence. Key populations—including injection drug users, men who have sex with men, and transgender individuals—demonstrated greater challenges with adherence. These groups often experience intersecting vulnerabilities such as stigma, discrimination, unstable living conditions, frequent mobility, mental health comorbidities, and substance use, all of which may adversely affect treatment continuity.
However, this finding should be interpreted with caution, as the relatively small number of participants in non-heterosexual categories may have resulted in unstable estimates. The observed association is therefore unlikely to reflect a direct effect of transmission category, but rather underlying social and structural determinants, including limited social support and barriers to healthcare access. While Sharma et al.[6] did not report a similar association, differences in study populations and representation of key groups may explain this variation. Overall, this finding underscores the importance of addressing broader social vulnerability rather than attributing nonadherence to transmission category alone.
The strong association between family support and ART adherence observed in this study is consistent with recent evidence highlighting the critical role of social support systems in sustaining long-term treatment adherence. Family support not only facilitates practical aspects such as medication reminders and healthcare access, but also provides emotional reassurance, which is essential for coping with chronic illness. Recent studies have further demonstrated that social and family support can mediate the effects of psychological stress, stigma, and mental health challenges on adherence behavior. Conversely, lack of family support has been identified as a key barrier, often leading to poor treatment continuity and increased risk of nonadherence. These findings reinforce the importance of integrating family-centered approaches into HIV care to improve adherence outcomes.[6,8,13,14]
Perceived stigma emerged as an important factor influencing ART adherence in this study. Stigma can adversely affect adherence through multiple pathways, including fear of disclosure, social isolation, reduced healthcare engagement, and psychological distress. Individuals experiencing stigma may avoid taking medication in the presence of others, miss clinic visits, or disengage from care altogether.[15] This is consistent with existing evidence, which demonstrates that stigma is a key determinant of ART nonadherence and negatively impacts treatment continuity and health-seeking behavior.[1,7] These findings underscore the need for targeted interventions addressing stigma at individual, family, and community levels to improve adherence outcomes.
In contrast, substance use (including alcohol, smoking, and tobacco) was not significantly associated with adherence in the present study. This differs from findings reported by Alvi et al.[7] and Shukla et al.,[8] who identified substance use as a major barrier to adherence. This discrepancy may reflect the impact of integrated de-addiction services and adherence counseling within the study setting. In India, de-addiction services are delivered through government facilities, medical colleges, District Mental Health Programme initiatives, and targeted interventions under NACO, particularly for high-risk groups such as people who inject drugs. Strengthening linkages between ART centers and these services—through counseling, opioid substitution therapy, and community-based support—may help address substance use–related barriers and improve adherence outcomes in similar settings. Our findings are consistent with those of Basu et al.,[12] who reported forgetfulness, alcohol use, travel-related barriers, stigma, non-disclosure of HIV status, and depressive symptoms as the major contributors to ART nonadherence in India.
Condition-related factors
In the present study, lower CD4 counts were associated with nonadherence; however, this relationship should be interpreted with caution. Given the cross-sectional design, temporality cannot be established, and the association is likely bidirectional. While advanced immunosuppression may impair adherence due to increased opportunistic infections, treatment burden, adverse effects, and reduced functional capacity, poor adherence itself can lead to declining CD4 levels, indicating the possibility of reverse causation. Although this finding may appear counterintuitive—since individuals with more severe illness might be expected to adhere better due to increased perceived need for treatment—advanced disease is often accompanied by treatment fatigue, depression, and general malaise, which may adversely affect adherence. Similar observations have been reported by Alvi et al.,[7] underscoring the need for closer monitoring and targeted adherence support among immunocompromised individuals.
In the present study, No significant association was found between WHO clinical stage and adherence, differing from Sharma S et al.[6] and Alvi Y et al.,[7] who noted poorer adherence in stage IV cases. Likewise, comorbidities and opportunistic infections showed no relationship with adherence, unlike Sharma M et al.,[6] who identified TB coinfection as a major barrier.
Depression and pill burden were not directly evaluated; however, earlier studies found depression predictive of nonadherence.[7,8] The minimal impact of pill burden in current settings may reflect the widespread use of simplified fixed-dose combinations under NACP.
Therapy-related factors
In this study, treatment duration >5 years was associated with increased nonadherence, suggesting possible treatment fatigue or reduced motivation. Long-term adherence maintenance presents unique challenges including medication monotony, accumulated side effects, normalization of health status leading to reduced perceived risk, and psychological exhaustion from lifelong regimen requirements.[14]
The absence of significant association between medication side effects and adherence likely reflects the improved tolerability of modern ART regimens. The widespread adoption of dolutegravir-based regimens under NACP Phase-V, which offer once-daily dosing with minimal side effects, has substantially reduced treatment burden compared to earlier regimens.[2] In contrast, Alvi Y et al.,[7] who reported side effects as a major barrier (OR: 2.99).
Health system-related factors
In the present study, a significant negative association was found between NGO linkage and ART adherence, with 48.5% of NGO-linked individuals being nonadherent compared to 16.5% of those not linked (P < 0.001). This likely reflects reverse causality, as patients with adherence difficulties are more often referred to NGOs for support. Consistent with prior evidence, NGO programs typically serve vulnerable populations—those with poor adherence, social instability, or limited family support—creating a selection effect rather than a causal relationship. Reliance on external support without strengthening self-management and family mechanisms may further contribute to suboptimal adherence. Thus, NGO linkage in this context is better understood as a marker of vulnerability rather than a determinant of nonadherence.
Although waiting time and service satisfaction were not directly assessed, previous studies highlight their importance. Alvi Y et al.[7] reported higher adherence among patients with shorter waiting times (<30 minutes) (OR: 2.08, 95% CI: 1.22–3.54) and greater satisfaction with ART services (OR: 2.75, 95% CI: 1.29–5.85). Similarly, Shukla M et al.[8] found that dissatisfaction with ART center services strongly predicted nonadherence (AOR: 5.63, 95% CI: 1.88–16.84). These findings emphasize the need for efficient, patient-centered care to enhance treatment continuity and trust in the healthcare system.
The findings of this study align with broader evidence indicating that ART adherence is influenced by a complex interplay of individual, social, and health system-level factors. A recent systematic review by Aung S et al.[1] highlighted that barriers such as stigma, inadequate social support, and structural challenges within healthcare systems significantly contribute to poor adherence, while facilitators include effective counseling, community support, and accessible services. This supports the multifactorial nature of adherence observed in the present study, where both psychosocial and programmatic factors played a significant role. These findings emphasize the need for integrated and context-specific interventions that address not only individual behavior but also the broader social and health system determinants of adherence.
Limitations
This study has certain limitations. The cross-sectional design limits causal inference, as temporal relationships between factors and nonadherence cannot be definitively established; longitudinal studies would better elucidate causality. Reliance on self-reported adherence data, even when validated through pill counts, may introduce social desirability bias and recall bias, potentially underestimating true nonadherence. The single-center setting limits generalizability to other populations and healthcare contexts. Additionally, factors such as mental health status, counseling quality, provider relationships, and detailed ART regimen characteristics were not assessed. Viral load, an important indicator of treatment response, could not be included in the analysis due to incomplete data availability.
CONCLUSION
This study identified a nonadherence prevalence of 17.5% among PLHIV receiving ART in Jabalpur. Younger age, non-heterosexual transmission routes, inadequate family support, and low current CD4 counts were significant independent predictors of nonadherence. These findings underscore that adherence is not solely a patient responsibility but a complex phenomenon requiring comprehensive, multilevel interventions. Achieving the national 95-95-95 targets necessitates strengthening patient-centered care models, expanding social support systems, reducing HIV-related stigma, and implementing targeted interventions for vulnerable populations.
Recommendations
This study identified specific gaps in ART adherence requiring targeted interventions. Younger patients and those with poor family support emerged as highrisk groups, underscoring the need for age-specific counseling and family-centered support. Nonheterosexual transmission categories highlight the importance of stigma reduction and inclusive healthcare services. Patients with low CD4 counts should be prioritized for enhanced monitoring and followup. The observed association with NGO linkage suggests that such patients may already represent a high-risk subgroup; therefore, better integration and coordination between ART centers and community-based organizations is needed to provide continuous and structured adherence support. Overall, interventions should move beyond generalized approaches and focus on tailored, context-specific strategies for vulnerable populations.
Declaration
ChatGPT (OpenAI, GPT-5 version) was used only for language editing and grammar improvement of the manuscript. The content, data interpretation, and scientific conclusions are entirely the authors’ original work.
Conflicts of interest
There are no conflicts of interest.
Acknowledgments
The authors express sincere gratitude to all the study participants who willingly shared their time and information, without whom this research would not have been possible. We acknowledge the staff of the ART center at Netaji Subhash Chandra Bose Medical College, Jabalpur, particularly the counselors, nurses, pharmacists, and medical officers, for their cooperation and support during data collection.
Funding Statement
Nil.
REFERENCES
- 1.Aung S, Hardy N, Chrysanthopoulou SA, Kyaw A, San Tun M, Aung KW, et al. Stigma determines antiretroviral adherence in adults with HIV in Myanmar. J Acquir Immune Defic Syndr. 2022;89:19–26. doi: 10.1097/QAI.0000000000002813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.National AIDS Control Organisation (NACO) New Delhi: Ministry of Health and Family Welfare, Government of India; 2023. HIV estimates 2023: Factsheets. Available from: https://naco.mohfw.gov.in/division/strategic-information-management/surveillance-epidemiology/factsheet/. [Last accessed 2024 Nov 10] [Google Scholar]
- 3.National AIDS Control Organisation. 7th. New Delhi: Ministry of Health and Family Welfare, Government of India; 2026. Sankalak Booklet. Available from: https://naco.mohfw.gov.in/wp-content/uploads/2026/06/Sankalak-Booklet_Seventh-Edition-As-per-NACO-website-as-on-30012026.pdf. [Last accessed 2026 Jul 10] [Google Scholar]
- 4.World Health Organization (WHO) Geneva: World Health Organization; 2003. Adherence to long-term therapies: Evidence for action. Available from: https://iris.who.int/handle/10665/42682. [Last accessed 5th Aug 2024] [Google Scholar]
- 5.Joint United Nations Programme on HIV/AIDS (UNAIDS) Geneva: UNAIDS; UNAIDS fact sheet 2023. Available from: https://www.unaids.org/en/resources/fact-sheet. [Last accessed on 2025 Jan 05] [Google Scholar]
- 6.Sharma S, Bano M, Malhotra AK. Factors influencing adherence to anti-retroviral therapy in HIV patients registered at ART Center Jhansi, Uttar Pradesh. Natl J Community Med. 2022;13:851–7. [Google Scholar]
- 7.Alvi Y, Khalique N, Ahmad A, Khan HS, Faizi N. World Health Organization dimensions of adherence to antiretroviral therapy: A study at antiretroviral therapy centee, Aligarh. Indian J Community Med. 2019;44:118–24. doi: 10.4103/ijcm.IJCM_164_18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shukla M, Agarwal M, Singh JV, Tripathi AK, Srivastava AK, Singh VK. Nonadherence to antiretroviral therapy among people living with HIV/AIDS attending two tertiary care hospitals in district of Northern India. Indian J Community Med. 2016;41:55–61. doi: 10.4103/0970-0218.170970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Patel AA. Study to assess the adherence to antiretroviral therapy among HIV positive patients attending art centers in haryana and its effect on their quality of life. Int J Acad Med Pharm. 2022;4:652–6. [Google Scholar]
- 10.Aytenew TM, Demis S, Birhane BM, Asferie WN, Simegn A, Nibret G, et al. Non-adherence to anti-retroviral therapy among adult people living with HIV in Ethiopia: Systematic review and meta-analysis. AIDS Behav. 2024;28:609–24. doi: 10.1007/s10461-023-04252-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bhagat VK, Chellaiyan DVG, Kumar G, Bhatnagar A, Adhikary M. Anti retroviral therapy adherence and its determinants among patients attending ART centre, Bhopal. Int J Community Med Public Health. 2018;5:4566–72. [Google Scholar]
- 12.Basu S, Marimuthu Y, Garg S, Saravanakumar V, Ganesh B. Anti-retroviral therapy adherence in India (2012-18): A systematic review and meta-analysis. Indian J Sex Transm Dis AIDS. 2024;45:2–7. doi: 10.4103/ijstd.IJSTD_28_20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rajasekar V, Thirunaaukarasu D, Surendar R. Socio-demographic profile and adherence to anti-retroviral treatment among HIV/AIDS patients attending the ART centre in Tamil Nadu. J Med Sci. 2023;9:163–8. [Google Scholar]
- 14.Claborn KR, Meier E, Miller MB, Leffingwell TR. A systematic review of treatment fatigue among HIV-infected patients prescribed antiretroviral therapy. Psychol Health Med. 2015;20:255–65. doi: 10.1080/13548506.2014.945601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Casale M, Cluver L, Boyes M, Toska E, Gulaid L, Armstrong A, et al. Bullying and ART nonadherence among South African ALHIV: Effects, risks, and protective factors. J Acquir Immune Defic Syndr. 2021;86:436–44. doi: 10.1097/QAI.0000000000002574. [DOI] [PubMed] [Google Scholar]