This was a retrospective cohort study that employed both qualitative and quantitative methods of data collection. Study sites constituted five ART treatment centers (three in the NW and two in the SW) regions in Cameroon, serving a large population. These treatment centers were each selected from five different health districts in the two regions of Cameroon, and thus clustering or group effect was minimized. The two regions have been in an armed conflict since 2017, and clashes continue from time to time to this day. The health facilities provided comprehensive healthcare services, including HIV prevention and treatment and were purposefully selected to include urban, semi-urban and rural populations. The records of children and adolescents who were LTFU and those who resumed ART after documented LTFU between January 2018 and December 2021 were assessed.

The study was carried out in five HIV treatment centers, each hosting a pediatric unit with children and adolescents enrolled in HIV care. Records of all children and adolescents who were on treatment or LTFU between January 2018 and December 2021 were reviewed. Children and adolescents 0-19 years with complete information in their records were included, while those with incomplete information in their records on LTFU as the major outcome of interest were excluded from the study. The study also recruited healthcare providers working in the HIV/AIDS pediatric clinics for the qualitative portion. Both quantitative and qualitative studies were conducted from November 2021 to March 2022.

We retrospectively analyzed medical records (n = 1,030) from a cohort of all children 0–9 years (355) and adolescents 10–19 years (675) living with HIV and on ART between January 2018 and December 2021, from the five healthcare facilities, who were followed for treatment outcomes. Treatment centers which were all faith-based institutions were purposively selected to include urban, semi-urban and rural populations. Furthermore, purposive sampling was used to select key informants in the various pediatric units for the in-depth interviews.

The main outcome of interest was the first LTFU defined as not seen or reached for >90 days and not known to have been transferred out of the clinic to another or died during the study period. Secondary outcomes were documented, transferred out and died. We examined these outcomes among all children and adolescents in the cohort.

Quantitative data was inputted into an Excel sheet, cleaned, and exported into SPSS vs. 26 for analysis. Data analysis was also done using the R software. A descriptive analysis of socio-demographic and baseline clinical characteristics was done using counts and percentages for all categorical variables. The person-years of observation (pyo) were also calculated and used to determine the incidence of LTFU, mortality and transferred-out rates per 1000 pyo. Kaplan–Meier curve was used to estimate median survival time. The Cox regression analysis was done to assess the factors associated with time on ART to LTFU among children and adolescents. History of ART adherence was classified into good, fair, and poor according to the number of daily doses missed for children on once-daily doses and twice-daily doses and the percentage summerised. For children on once-daily doses, adherence was classified as Good (≥95% or 1 pill missed per month), Fair (90–95% or 2-3 pills missed per month), or Poor (less than 90% or > 3 pills missed per month). For children on twice daily doses, adherence was classified as Good (≥95% or 2 pills missed per month), Fair (90–95% or 3-6 pills missed per month), or Poor (less than 90% or 3-6 pills missed per month).

After qualitative data collection, all interviews were verbatim transcribed for analysis using thematic areas. The transcribed data was reviewed and cross-validated by another team member. Then, coding and interpretation were done by CWA, followed by validation by EMK for reliability and validity. The coding was done using social-ecological guidelines [11, 12] to assess individual and external parameters of LTFU. Data analyses were conducted using an inductive method of content analysis based on emerging themes (personal, family, community, and health system parameters). The main grouped themes used were: perceptions and experiences of healthcare providers on the causes of attrition rates in children and adolescents and understanding of the importance of ART adherence.

The cohort was observed for 1,030 children/adolescents, with 138 (13.4%) of them LTFU. Out of the 138 participants LTFU, fifty (50) were recovered and reengaged in care. At the end of follow-up, 848 (82.3%) HIV-infected children and adolescents were in care (Fig. 1). More than half (55.5%) of the participants were females. The median age of the children and adolescents was 12 ± 4.82 years. More adolescents (10-19 years) (61.0%) were present in the cohort compared to children 0-9 years (39.0%), with most of the participants, 342(33.0%), being older adolescents of age group 15-19 years. Close to half of the study participants had a primary level of education, 559 (49%). More than half, 604 (59%) of the children and adolescents lived with their parents as their main caregivers, with 377 (37%) of these caregivers having a secondary level of education. About a third of the caregivers, 365 (35%), were farmers (Table 1).

About a third (34.7%) of the children and adolescents in the cohort had WHO clinical stage I, 357 (34.7%). A total of 304(29.5%) had a baseline CD4 count greater than or equal to 500 cells/ml. Close to half of the adolescents (497 [48.3%]) were on a Dolutegravir-based regimen. Many of the study participants, 686 (66.6%), recorded a good history of adherence to ART as seen in records and as reported by healthcare workers. More than half of the caregivers, 586 (56.9%), were aware of their own HIV status as positive (Table 2).

A total of 138 HIV-positive children and adolescents were LTFU. Of the 138, 79(57.2%) were children 0-9 years, while 59 (42.8%) were adolescents 10-19 (Older adolescents (15-19years) and constituted 33 (24%) of those LTFU. A total of 50 (36.2%) children and adolescents recovered and reengaged in care.

The overall incidence of LTFU in children and adolescents was 33 LTFU per 1000 children/adolescent years, while the incidence of death was 3 deaths per 1000 children/adolescent years, and those transferred out (TO) was 7 per 1000 children/adolescent years. Over this period, 138(13.4) were LTFU, 14(1.4%) patients died, and overall attrition was 152 (14.8%). LTFU occurred more in females 76 (7.4%) than in males, while the proportion was more in children 5–9 years old, 44 (4.3%) than in any other age group. Based on the health facility, the proportion was more at health facility 1, 53 (5.1%) in a rural setting than in any other health facility (Table 3).

In this study, all children and adolescents on treatment between Jan 2018 and Sept 2021 were enrolled and followed until December 2021 for the occurrence of the event of interest (LTFU). Children and adolescents who exited or completed the study without the occurrence of the event of interest were censored. The minimum and maximum follow-up period was three months and forty-eight months, respectively (Fig. 2). The overall cumulative survival probability at the end of the 10th month was 0.96 (95% CI: 0.95- 0.97), while it was 0.89 (95% CI: 0.88-0.92) and 0.87 (95% CI: 0.85- 0.89) at the 20th and 30th follow-up month, respectively. In addition, the cumulative survival probability at the 40^th and 48^th follow-up months was 0.86 (95% CI: 0.84-0.88) and 0.86 (95% CI: 0.83, 0.88), respectively (Fig. 1). The survival probability significantly decreased for children 0-4 years, followed by children in the 5-9 years category and older adolescents 15-19 years. There was no statistically significant difference in survival probability for females and males. The survival probability significantly decreased for HF1 and HF3 treatment centers.

Fig. (1). Flow diagram of children and adolescents initiating ART in the NW & SW Region of Cameroon (2018–2021).

Fig. (2). The Kaplan–Meier estimate of the loss to follow-up among children and adolescents attending ART hospitals in the Northwest and Southwest Regions of Cameroon, 2018-2021.

In our study, the following significant variables in univariate analysis were included in the Cox proportional hazard multivariable model: residence, type of caregiver, having a registered phone number for follow-up calls, type of ART regimen, the experience of regimen change, history of ART adherence, duration on ART, and the approximate distance between home and health facility. A P-value of less than 0.05 as a cutoff point was considered statistically significant. Children and adolescents in care at the HF1 treatment center had a two times higher risk of loss to follow-up (AHR: 2.14; 95% CI:1.23-3.76) than their counterparts at HF5. Similarly, children and adolescents receiving care at the HF2 in a rural setting had a two times higher risk of loss to follow-up (AHR: 2.51; 95% CI: 1.40-4.53) compared with their counterparts at the HF5. In addition, children and adolescents in care at HF4 had a two times higher risk of loss to follow-up (AHR: 2.02; 95% CI: 0.98-4.17) than their counterparts at HF5. The risk of LTFU in children of 5-9 years (AHR: 0.55; 95% CI: 0.31-0.98) and adolescents of 10-14 (AHR: 0.25; 95% CI: 0.13-0.49), 15-19 (AHR: 0.34; 95% CI: 0.18-0.64) was lower compared with children of 0-4 years. The risk of LTFU among ART-attending children and adolescents who had non-biological caregivers was two times higher as compared with their counterparts who had primary biological caregivers (AHR: 2.44; 95% CI: 1.59-3.74). The risk of LTFU in children and adolescents on sub-optimal regimens (AHR: 17.83; 95% CI: 11.00-29.00) and other (AHR: 6.83; 95% CI: 3.85-12.14) regimens was very high compared with those on the optimal regimen. Children and adolescents with a history of good adherence had a lower risk of LTFU (AHR: 0.52; 95% CI: 0.35-0.77) than their counterparts with poor or fair adherence. The risk of LTFU in children and adolescents with a prolonged duration on ART, i.e., 13-24 months (AHR: 0.38; 95% CI: 0.19-0.74) and 25+ months (AHR: 0.20; 95% CI: 0.11-0.33) was lower compared with those with ≤12 months duration on ART. ART-attending children and adolescents who had more than fifty kilometers distance between their homes and the health facility had a two times higher risk of loss to follow-up as compared with their counterparts who had less than twenty kilometers distance from home to the health facility (AHR: 2.05; 95% CI: 1.10-3.20) (Table 4).

The true cumulative incidence was 33 per 1000 child/adolescent years of LTFU over the study period, and the periodic prevalence was 13%. The incidence obtained from this study was not different from that obtained in Kenya [13], with 32.9 per 1000 child years of LTFU and a period prevalence of 12.6%. This magnitude of LTFU of 12.6% agreed with findings from previous studies conducted in sub-Saharan Africa cite (ref). About 21% of patients in HIV programmers were LTFU in Africa [1]. LTFU differed radically across populations because authors used different definitions. For instance, a Ugandan study reported 50.1% of children were LTFU [2], while a South African study reported an LTFU rate of 50.2% [3]. Different definitions of LTFU carry the inherent risk of misclassification of active patients as LTFU. Some studies cite the absence of 1 week from a scheduled appointment [4], and the absence of 2 weeks after an appointment date [5], while others use 6 weeks of missed appointment [6]. We employed the WHO [14-17] definition of LTFU as more than 90 days of absence from the missed clinical or drug pick-up appointment without any follow-up visit [7, 8], a definition also used by MINISANTE Cameroon. In Cameroon, previous studies have focused on HIV programs in conflict-free zones. Our study is among the very first looking at LTFU in the context of socio-political crises. In a study conducted in western Kenya, 14.2% of children were LTFU, and for children already initiated on ART, 14.1% dropped out of care [9]. This was, however, slightly greater than the 13% obtained from our study. It has been reported that 10-14% of children on ART get lost to follow-up in HIV care programs in Africa [9].

The cumulative survival probability at the end of the 10th month was 0.96 (95% CI: 0.95- 0.97), while it was 0.89 (95% CI: 0.88-0.92) and 0.87 (95% CI: 0.85- 0.89) at the 20th and 30th follow-up month, respectively. Our findings were similar to that obtained in Ethiopia, where the cumulative survival probability at the end of the 10th, 20th, 40^th, and 60th follow-up month was 0.89 (95% CI:0.87,0.91), 0.82 (95% CI:0.79,0.85), 0.78 (95% CI:0.75,0.81), and 0.74 (95% CI:0.70,0.77), respectively.

Children and adolescents receiving care at HF1 and HF2, all located in rural and semi-rural settings, were more likely to be loss to follow-up compared to those in an urban setting (HF5). Our finding was similar to a systematic study reported from Asia-Pacific, Central, East, West, and Southern Africa, the Caribbean and Central and South America [14] where rural settings (as 1.21, 95% CI 1.13 to 1.29) risk LTFU. Another study showed no significant difference in the outcome of children in rural and urban settings [15].

Infants and young children were more likely to be LTFU compared to other age group categories. Our findings were similar to that obtained in South Africa [16]. A study in Zimbabwe [17] showed that those starting ART as young adolescents (10–14 years) had a 1.5 times higher rate of LTFU after moving into the 15–19 age group; adolescents starting ART aged 15–19 had an increased rate of LTFU compared to adolescents of the same age established on ART at an earlier age; and moving from adolescence into young adulthood increased the risk of LTFU by nearly twofold. Our findings were contrary to that obtained in India [18], where age ≥15 years (adjusted SHR (aSHR) 2.44; 95% CI:1.18 – 5.02) was a risk factor for LTFU.

Children and adolescents with non-biological caregivers had a greater risk of LTFU than those with biological parents as the main caregiver. Similar findings were obtained in Kenya [13]. Our findings were, however, contrary to that obtained in Ethiopia [19], where Children cared for by biological parents were nearly 2.8 at high risk of loss to follow-up as compared to children cared for by non-biological parents.

Children and adolescents on sub-optimal and other ART regimens had a greater risk of loss to follow-up compared with those on the optimal regimens. As of 2018, ART coverage among children was only 28% [5]. A systematic study in sub-Saharan Africa showed that more than half of all children on ART were still receiving suboptimal formulations. The use of suboptimal formulations negatively affects treatment adherence and viral suppression and ultimately increases HIV-related morbidity and mortality [5]. Sub-optimal and other ART combinations in children may result in lower efficacy, drug resistance, and subsequent poor retention.

The risk of loss to follow-up among ART-attending patients who did not have a primary caregiver was two times higher compared with their counterparts who had a primary caregiver (AHR: 2.59; 95% CI: 1.84, 3.66). This study’s finding was similar to the studies conducted in South Africa [16] and Ethiopia [20]. This might be due to family support, which means families may have the greater emotional support to offer than clinic staff. The child's primary caregiver was the parent (mostly mothers) in 59% of the cases, followed by relatives (aunties and uncles) 22%, and grandmother in 7% of cases. This means that a substantial number of the children and adolescents were under the care of grandmothers, relatives, and siblings who often have poor socioeconomic and low educational status contributing to loss to follow-up. No biological parents with poor socioeconomic status as main daily caregivers was identified as one of the challenges in the provision of HIV care in resource-limited settings [10].

Good history of adherence to ART for children and adolescents was a lower risk of LTFU compared with the history of poor or fair adherence on ART. Similar findings were found in Ethiopia [20], where children who had fair/poor adherence to ARV drugs were two and a half times more at risk of being LTFU than children who had good adherence. Our findings were contrary to that obtained in South Africa [21], where adherence was not a risk factor for loss to follow-up. Disruption in HIV care because of missed appointments can undermine clinical outcomes, including assessment of adverse events, ongoing provision of prophylactic medications, clinical and neurodevelopment assessment, and early identification of treatment failure results from LTFU.

Children and adolescents on ART above thirteen months had a lower risk of loss to follow-up compared with their counterparts on ART less than or equal to twelve months. Similar findings were obtained in South Africa [21], where most children and adolescents got lost within the first year of initiation to ART. Counseling on the importance of ART adherence should be done for caregivers whose children have been enrolled in ART within 1 year or less to prevent loss to follow-up.

Long distances had a higher risk of LTFU than children and adolescents with short distances of less than fifty kilometers. Our study was similar to that obtained in Zambia [22], where the long distance to health facilities was a risk factor for poor retention in patients on ART.

Many healthcare providers complained of multiple caregivers for children and adolescents as the cause of LTFU in them. This was consistent with a study in Zambia [22], where the change of caregiver was a reason for LTFU in children on ART. Some studies have recorded inconsistent parental care [20] as a reason for LTFU. At the extreme, parental death had many implications on treatment outcomes [12, 21] in children, including LTFU in care if the children were on ART. The effects are widespread, including multiple separations of couples, movements to new households, educational challenges, and an increased risk for internalizing disorders and post-traumatic stress [4, 2]. Most often, due to this constant switching of caregivers, children find themselves in the hands of caregivers who are unaware that the children are on treatment. There is a need for health facilities to educate caregivers on the proper transitioning of children and for treatment centers to constantly update the demographic information for children and adolescents in case they have relocated to ensure continuity in care.

A serious socio-political crisis started in November 2016 in the Northwest and Southwest Regions of Cameroon, with over 4,000 people killed, 500,000 internally displaced and at least 40,000 externally displaced into neighboring Nigeria [23]. The armed conflict initially started as localized conflicts in certain districts in the NW and SW regions but quickly spread into a generalized armed conflict in both regions leading to disruption of movement, dysfunction of many health facilities, and numerous ghost towns [23]. This indeed led to a drop in health facility visits and retention in care for children and adolescents on lifelong ARVs, as observed from this study. Our findings were congruent with those of Marta et al. [24], who showed that sociopolitical crises in resource-limited settings lead to treatment interruptions and subsequent drug resistance among people with HIV/AIDS.

Adolescence is a stage of transition from childhood to adulthood, associated with specific challenges (including puberty) and vulnerability (such as early sexual debut, HIV, and STI acquisition) [25]. In addition, coping with the clinical and psychosocial impact of HIV imposes a substantial burden on this vulnerable group [25]. These developmental features may hinder an effective adherence to chronic medical therapies.

According to findings from the healthcare workers, some caregivers, due to stigma, do not come to the health facilities with their children nor tell family members of their health condition, which has contributed to poor retention in some children and adolescents. Our findings were consistent with previous findings in a systematic review in SSA [26], Malawi [27], Ethiopia [28], and Kenya [29]. A Qualitative study in Ethiopia [30] also demonstrated stigma as a cause for poor retention among adolescents. They described how they feared being seen by others while taking medication, visiting the clinic for appointments, and collecting ART medication as they were afraid of being stigmatized. Medication times and hospital appointments were not respected due to the fear of shame from friends and the risk of disclosing their status [30]. This highlights the need to intensify interventions to reduce HIV-related stigma and discrimination in all domestic and social circles.

Lack of timely disclosure at the appropriate age also emerged as a theme for poor retention in older children and adolescents. Our findings were in line with a qualitative study conducted among healthcare workers in South Africa [31], where they referred to the regrettable, negative outcomes of delaying disclosure to older children and adolescents that they had observed in their interaction with children and adolescents. When children are not informed of their status on time, this leads to treatment default as they continue to take medication without being informed about their HIV diagnosis. They become rebellious or refuse to take treatment or stop attending the clinic completely. Systematic review shows adherence to antiretroviral therapy (ART) improved post-disclosure in children in resource limited settings [32].

According to healthcare providers, some of the children come to the health facility, but since they do not know their HIV status or the reason why they are taking medications, they tend to shy away from care. The World Health Organization (WHO) also recommends that age-appropriate disclosure and caregivers’ regular support are essential to improve retention in ART care [1].

Lack of understanding of the disease process and treatment by caregivers was also reported as a reason for poor retention in care for children and adolescents. Similar findings were obtained in Malawi [27], where healthcare workers reported attrition as a result of HIV mothers themselves or their children feeling better, i.e., a lack of understanding of the treatment process. Our findings show the need for healthcare workers to constantly educate caregivers and adolescents on the disease process/treatment outcome and the importance of continuous and timely adherence to treatment.

Seeking alternative forms of healthcare with religious leaders and traditional healers was another reason for poor retention in children and adolescents. This decision to seek alternative care appeared to be influenced by religious teachings and traditional beliefs. This finding suggests the need for continued community sensitization about HIV and its treatment. Our findings aligned with those obtained in Malawi [27, 28], where seeking alternative healthcare was associated with LTFU.

In our study, the shortage of ARVs was cited by some key informants as a reason for poor retention in children and adolescents. At the peak of the sociopolitical crises, the hospital gave ARVs just for two weeks when the stock was limited to keep everyone on treatment. This discouraged many caregivers who lived in distant areas and had to pay the high cost of transportation for ARV refills just after two weeks. Similar findings associated with the shortage of commodities, including ARVs in Malawi [27], were reasons for poor retention. Sub-optimal regimens and other regimen combinations may lead to poor efficacy and retention. There is a need to devise emergency strategies to ensure that ARVs are constantly available for children and adolescents in conflict-affected areas. Moreover, for children and adolescents living in far distances and hard-to-reach areas, there is a need for a multi-month dispensation of ART.