Juntendo Medical Journal, 2026 Original Articles Juntendo Medical Journal 2026 Fear of Recurrence and Medication Adherence in Pediatric Leukaemia Survivors in Low-Resource Settings Sudip SAHA1), Aateka Fatima MOHAMMED2), Deepthi NAGULA3), Pooja CHAKRABARTI4), Makurira VUNYOROHWASHE5), Sirajudheen ANWAR6), Kaushalendra Mani TRIPATHI7, 8), Shaumile Hasan KHAN9), Khwaja Waleed MAQBOOL10), Kanwal SHAZIA11) 1)Department of Surgery, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, United Kingdom 2)Department of Acute Medicine, Ajman University, Ajman, United Arab Emirates 3)Department of Medicine, Chalmeda Ananda Rao institute of medical sciences, Telangana, India 4)Department of Cardiology, Manipal Hospitals, West Bengal, India 5)Department of Acute Medicine, Royal Cornwall Hospital, Cornwall, United Kingdom 6)Department of Pharmacology and Toxicology, College of Pharmacy, University of Hail, Hail, Saudi Arabia 7)Department of Medicine, White River Medical Center - White River Health, Arkansas, United States 8)Department of Medicine, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences (SIMATS), Saveetha University, Tamil Nadu, India 9)Department of Medicine, FMH Medical College, Punjab, Pakistan 10)Department of Biochemistry, Quetta Institute of Medical Sciences (QIMS), Balochistan, Pakistan 11)Department of Medicine, Scarborough General Hospital, Scarborough, United Kingdom Objectives: Childhood leukaemia patients who survived their treatment often experience Fear of Cancer Recurrence (FCR), which can affect Medication Adherence (MA), particularly in low-resource environments. This paper explored the relationship between FCR and adherence among pediatric survivors. Materials and Methods: A cross-sectional study was conducted in tertiary care pediatric oncology clinics in low-resource settings, involving 444 children diagnosed with Acute Lymphoblastic Leukaemia (ALL) or Acute Myeloid Leukaemia (AML). The FCR was assessed using the validated Fear of Cancer Recurrence Inventory-Short Form (FCRI-SF), and adherence was assessed using the validated Medication Adherence Rating Scale (MARS-10). The statistical tests were correlation test, inde­ pendent t-tests, one-way ANOVA, and multiple linear regression. Results: The average FCRI-SF score was 27.0 +/- 2.6, and the average MARS-10 score was 15.1 +/-1.1. Medication adherence was significantly negatively associated with fear of cancer recurrence (r = −0.210, p < 0.001). Women had higher FCRI-SF (27.4 ± 2.9) than men (26.5 ± 2.4), and lower MARS-10 scores (14.8 ± 1.1 vs. 15.4 ± 1.0). The highest FCR (27.8 ± 2.4) and the lowest adherence (14.7 ± 1.0) were found in adolescents (15-17 years). Regression analysis revealed that FCR (B = -0.15, p = 0.001), female gender (B = -0.40, p = 0.001), treatment phase (B = -0.09, p = 0.003), and missed chemotherapy cycles (B = -0.07, p = 0.001) were significant predictors of poor adherence. Conclusion: FCR is a significant psychological risk factor of medication adherence in survivors of paediatric leukaemia. Psycho­ social support and follow-up strengthening can be used to decrease non-adherence and treatment abandonment. Key words: pediatric leukaemia, medication adherence, survivorship, fear of cancer recurrence Corresponding author: Kanwal Shazia Department of Medicine, Scarborough General Hospital Woodlands Dr, Scarborough YO12 6QL, United Kingdom TEL: +44-20-7946-0123 〔Received E-mail: kanwalshazia209@gmail.com Dec. 15, 2025〕 〔Accepted Jan. 23, 2026〕 J-STAGE Advance published date: Mar. 26, 2026 Copyright © 2026 The Juntendo Medical Society. This is an open access article distributed under the terms of Creative Commons Attribution License (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original source is properly credited. doi: 10.14789/ejmj.JMJ25-0069-OA 1 Saha S, et al: Fear of recurrence and medication adherence in pediatric leukaemia Introduction The most common cancer type in children is child­ hood leukaemia, and its onset is due to the abnormal growth of white blood cells in the bone marrow, which interferes with normal haematopoiesis1). Childhood cancers such as leukaemia remain highly morbid and lethal and have few new treatments because of their rarity and biological diversity. In 2021, the reported childhood leukaemia cases were estimated worldwide to be 58,785, and the most signi­ ficant number of cases and deaths were reported in Low Socio-Demographic Index (SDI) regions, with the highest number of cases affecting infants and boys2, 3). Acute leukaemias of childhood, particularly prec­ ursor B-cell Acute Lymphoblastic Leukaemia (ALL), tend to have an aetiological basis during the uterine period, with subsequent influences during the post­ natal period, including infections and immune-asso­ ciated genetic variations, which influence the disease4). L-asparaginase is a fundamental treatment for pedi­ atric ALL. Yet, immunogenicity, allergic reactions, and drug resistance limit its use, fueling the emer­ gence of new variants and the development of indi­ vidualised treatment approaches5). Survivorship also introduces a new dimension of concern, such as fear of cancer recurrence (FCR), that afflicts a significant percentage of childhood cancer survivors; about 43% of childhood cancer survivors report at least once, and 21% report persistently. Higher depressive symptoms and younger age of diagnosis correlate with higher FCR, which in most cases accompany long-term symptoms like pain and fatigue6, 7). Treatment abandonment, which entails the inabi­ lity to finish therapy to control the disease, is a leading factor in treatment failure among children with cancer in low and middle-income countries8). Medication non-adherence exists on a continuum, with treatment abandonment representing its most severe form. Minor lapses, such as missed appoint­ ments or incomplete follow-up, may progressively lead to treatment abandonment, particularly in lowresource pediatric oncology settings9, 10). About 25% of children who have ALL withdraw therapy, signi­ ficant ones being the costs of treatment, the belief in the incurability of the condition, problems in transportation, and the fear of adverse effects of 2 treatment11, 12). In low-resource settings, limited access to followup care, psychosocial services, and health educa­ tion may amplify fear of recurrence. These systemic gaps can exacerbate anxiety-driven non-adherence, increasing the risk of treatment interruption or abandonment9). The interaction between fear of recurrence and treatment abandonment holds essential information to make effective survivor­ ship care strategies. These factors can significantly impair the medical prognosis and mental health in resource-limited settings where psychosocial support is frequently low. Thus, the current research seeks to investigate the relationship between fear of recurrence and treatment abandonment in lowresource settings among pediatric patients who have survived leukaemia to establish the crucial psychosocial and systemic predictors that can be used in future interventions and policy formulation. Objectives The main aim of the research is to investigate the rates and severity of FCR among survivors of childhood leukaemia in low-resource countries and to examine their correlation with treatment adher­ ence. The study also seeks to determine the most important demographic, clinical, and psychosocial variables that may affect fear of recurrence and compliance with follow-up care. With knowledge of these determinants, the study aims to inform how survivors and their families face difficulties and where interventions, support programs, and policy modifications can be applied to curb treatment abandonment and improve overall survivorship in resource-constrained settings. Materials and Methods Ethical considerations The institutional ethics review board of Scarbor­ ough General Hospital, Scarborough, United Kingdom (Approval No.: SGH-REC-2024-017) approved the study. Informed consent was obtained in written form from parents or legal guardians, and children aged seven or older provided assent. The involve­ ment was voluntary, and confidentiality was ensured. Caregivers and children were assured that refusal or disengagement from the survey would not affect their ongoing medical care. Juntendo Medical Journal, 2026 Study approach and contextual setting The cross-sectional study was conducted in tertiary care hospitals and pediatric oncology clinics in low-resource urban and semi-urban regions. The study sought to examine the relationship that exists between FCR and treatment abandonment among survivors of pediatric leukaemia. Study population and sampling The patient population consisted of children aged between 5 and 18 years who had attained remis­ sion after being diagnosed with acute lympho­ blastic or acute myeloid leukaemia. Only survivors who would not have a profound cognitive deficit or comorbidity that would have jeopardised participa­ tion were taken into consideration. A total of 444 participants, together with their primary caregivers, were recruited via convenience sampling to obtain a representative sample of survivors receiving follow-up care in low-resource settings. Conve­ nience sampling was employed due to the absence of centralised survivorship registries, logistical constraints, and the need to maximise participant inclusion in low-resource pediatric oncology settings. Data collection instruments Data were collected using three structured tools: Demographic questionnaire A structured questionnaire was created to docu­ ment age, gender, leukaemia type, treatment history, length of remission, family socioeconomic status, and distance. Medication adherence rating scale (MARS) The MARS was developed by Thompson et al. (2000) to measure treatment adherence and consists of 10 items that assess medication adherence beha­ viours and attitudes. The items are rated using a dichotomy (Yes = 1, No = 0), and the overall score indicates better adherence. In children under 7 years of age, the primary caregiver would complete the MARS on behalf of the child; those 7 years or older would complete it themselves13). Fear of cancer recurrence inventory-short form (FCRI-SF) The FCRI-SF, developed by Simard and Savard in 2009, was used to measure FCR. This 9-item scale measures the severity of worry about cancer experienced in the previous month using a 5-point Likert scale (0 = never, 4 = almost always). The total score ranges from 0 to 36, with higher scores indicating greater fear of recurrence. In younger children, the caregiver provided the information on behalf of the child, but older children provided it themselves14). Procedure Data were collected through face-to-face inter­ views and from medical records. Parents or guard­ ians provided informed consent, and children aged seven or older provided assent. For younger survi­ vors, caregiver-reported responses were collected, as caregivers are primarily responsible for treat­ ment supervision. Caregiver and self-reports were analysed together to reflect real-world adherence behaviours. Psychological support was provided to participants if they felt distressed when discussing recurrence or adherence. Scoring and interpretation MARS: Total scores were summed; higher scores indicate greater adherence. FCRI-SF: Scores were added and classified as low, moderate and high fear of recurrence according to recommended cut-offs. The descriptive and subgroup analyses were conducted based on demographic variables. Data analysis Data analysis was conducted in SPSS version 26. Descriptive statistics summarized demographic characteristics, fear of recurrence, and adherence scores. Pearson correlation was used to examine relationships between continuous variables (FCRISF and MARS-10). Independent samples t-tests and one-way ANOVA were used to compare groups, while multiple linear regression identified predic­ tors of medication adherence. The level of statis­ tical significance was set at p < 0.05. Results Summary of participant demographic characteristics Table 1 summarises the demographics of 444 children with leukaemia. The majority of the respo­ ndents were teenagers aged 15-17 years (297, 3 Saha S, et al: Fear of recurrence and medication adherence in pediatric leukaemia Table 1 Variable Demographic characteristics of participants (N = 444) f (N) % f (N) % 5-7 years 8 1.8 25-34 years 150 33.8 35-44 years 100 22.5 8-11 years 41 9.2 45-54 years 75 16.9 12-14 years 98 22.1 55 years or above 59 13.3 15-17 years 297 66.9 Parent/Caregiver educational level No formal education 108 24.3 Male 257 57.9 Primary (Grades 1-5) 128 28.8 Female 187 42.1 Middle (Grades 6-8) 73 16.4 High school (Grades 9-10) 49 11.0 Child's age Child’s gender Child’s educational level Variable Not attending school 63 14.2 Higher secondary/college 54 12.2 Primary (Grades 1-5) 90 20.3 University/professional 32 7.2 Middle (Grades 6-8) 70 15.8 Leukaemia type High school (Grades 9-10) 106 23.9 ALL (Acute lymphoblastic leukaemia) 176 39.6 Higher secondary/college 115 25.9 AML (Acute myeloid leukaemia) 268 60.4 Urban 195 43.9 < 6 months 153 34.5 Semi-urban 101 22.7 6-12 months 94 21.2 Rural 148 33.3 1-2 years 79 17.8 > 2 years 118 26.6 < 5 km 99 22.3 Current treatment phase 5-10 km 128 28.8 Induction 121 27.3 11-25 km 99 22.3 Consolidation/Intensification 120 27.0 > 25 km 118 26.6 Maintenance 96 21.6 Completed 107 24.1 Residence place Time since diagnosis Distance to treatment center Travel time to hospital < 30 minutes 137 30.9 Number of chemotherapy cycles missed 30-60 minutes 141 31.8 None 191 43.0 1-2 hours 85 19.1 1 therapy 81 18.2 >2 hours 81 18.2 2 therapies 64 14.4 > 2 therapies 108 24.3 Parent/Caregiver relationship to child Mother 200 45.0 Experience of treatment side effects Father 180 40.5 None 32 7.2 Other relative/guardian 64 14.4 Mild 158 35.6 Moderate 110 24.8 Severe 144 32.4 Parent/Caregiver age < 25 years 60 13.5 Note. f = frequency, % = percentage; Values are presented as N (%), N = 444; No statistical comparisons were performed for demo­ graphic variables in this table. 66.9%) and male (257, 57.9%). The children mostly attended higher secondary school or college (115, 25.9% or high school (Grades 910; 106, 23.9%), and their living conditions were in urban (195, 43.9%) or rural areas (148, 33.3%). The number of acute myeloid leukaemia (AML, 268, 60.4%), acute lymp­ hoblastic leukaemia (ALL, 176, 39.6%), and 153 (34.5%) were diagnosed within six months. The 4 distribution of treatment stages was quite balanced, but the response to chemotherapy was not homo­ geneous, and the vast majority of children reported mild to severe side effects. Correlation between fear of cancer recurrence and medication adherence Table 2 shows a significant negative correlation Juntendo Medical Journal, 2026 Table 2 Association between fear of cancer recurrence (FCRI-SF) and medication adherence (MARS-10) (N = 444) Variables r t(df) p Total fear of cancer recurrence inventory-short form (FCRI-SF) - - - Total medication adherence rating scale (MARS) -0.210 -4.52 (442) <0.001** Note. Values represent Pearson correlation coefficients (r) between continuous variables; N = 444; p < 0.001 (2-tailed) was considered statistically significant and is denoted with double asterisks (**). between fear of cancer recurrence (FCRI-SF) and medication adherence (MARS-10) (r = -0.210, t = -4.52, df = 442, p < 0.001). This indicates that higher fear of recurrence was associated with lower medi­ cation adherence. In this study, non-adherence is used as a proxy for treatment abandonment in pedi­ atric leukaemia survivors. Gender-based differences in fear of cancer recurrence and medication adherence Table 3 reveals that there was a significant gender difference in both fear of recurrence of cancer and medication adherence among the 444 respondents. Females reported slightly higher fear of cancer recurrence (M = 27.4 ± 2.9) than males (M = 26.5 ± 2.4), which was statistically significant (t = -3.10, p = 0.002, Cohen’s d = 0.38). Conversely, males showed higher medication adherence (M = 15.4 ± 1.0) than females (M = 14.8 ± 1.1; t = 4.20, p < 0.001, Cohen’s d = 0.50). These results indicate a smallto-moderate gender difference in both FCR and adherence. Visual comparison of gender differences in FCRI-SF and MARS scores Figure 1 shows gender variations in FCRI-SF and MARS scores among the 444 participants. Females reported slightly higher fear of cancer recurrence (FCRI-SF) than males, while males demonstrated Table 3 higher medication adherence (MARS). These diffe­ rences were statistically significant, indicating a small-to-moderate effect of gender on both FCR and adherence. Age-related differences in fear of cancer recurrence and medication adherence Age-related differences in FCR and medication adherence are presented in Table 4. FCR increased with age (F(3, 440) = 11.02, p < 0.001, η2 = 0.070), with adolescents (15-17 years) reporting the highest fear. Medication adherence declined slightly with age (F(3, 440) = 3.90, p = 0.009, η2 = 0.026), sugges­ ting a minimal age effect. These findings indicate that older children and adolescents experience greater fear of recurrence and slightly lower adherence. Predictors of medication adherence based on psychological, demographic, and clinical factors Table 5 presents the results of multiple regres­ sion-based prediction of medication adherence among children with leukaemia. Multiple linear regression (Table 5) showed that higher FCRI-SF scores significantly predicted lower medication adhe­ rence (B = -0.15, β= -0.35, t = -5.00, p < 0.001). Older age predicted greater adherence (B = 0.10, β = 0.20, t = 4.00, p < 0.001), while female gender predicted lower adherence (B = -0.40, β = -0.18, t = -3.33, p = 0.001). Clinical variables also influenced adherence: Gender differences in fear of cancer recurrence (FCRI-SF) and medication adherence (MARS) (N = 444) Variables Male (N=257; 57.9%); (M ± SD) Female (N=187; 42.1%); (M ± SD) t p Total fear of cancer recurrence inventory-short form (FCRI-SF) 26.50 ± 2.40 27.40 ± 2.90 -3.10 Total medication adherence rating scale (MARS) 15.40 ± 1.00 14.80 ± 1.10 4.20 95% CI Cohen's d LL UL 0.002** -1.46 -0.34 0.38 < 0.001** 0.31 0.82 0.50 Note. Values are presented as Mean ± Standard Deviation; Independent samples t-tests were conducted to compare participants of both male and female groups; Group sizes are shown as N (%); Reported statistics include p-values, t-values, 95% Confidence Intervals (CI), and effect sizes (Cohen’s d); A p-value < 0.001 was considered statistically significant, N = 444. 5 Saha S, et al: Fear of recurrence and medication adherence in pediatric leukaemia Figure 1 Gender differences in FCRI-SF and MARS mean scores (N = 444). Figure created by the authors Table 4  ge differences in fear of cancer recurrence (FCRI-SF) and medication adherence (MARS-10) among children A and adolescents (N = 444) Variables 5-7 years (N = 8; 1.8); (M ± SD) 8-11 years (N = 41; 9.2); (M ± SD) 12-14 years 15-17 years (N = 98; 22.1); (N = 297; 266.9); (M ± SD) (M ± SD) Total fear of cancer recurrence inventory-short form (FCRI-SF) 25.00 ± 2.50 26.20 ± 2.60 27.10 ± 2.70 Total medication adherence rating scale (MARS) 15.30 ± 0.90 15.10 ± 1.00 14.90 ± 1.05 F(3,440) p η2 27.80 ± 2.40 11.02 <0.001*** 0.070 14.70 ± 1.00 3.90 0.009** 0.026 Note. Data are presented as Mean ± Standard Deviation (M ± SD); Group sizes are shown as N (%); One-way ANOVA was conducted to examine the effect; All comparisons were significant at p < 0.01, < 0.001; η2 represents partial eta-squared effect size. Table 5  ultiple regression predicting medication adherence (MARS-10) from fear of cancer recurrence, age, gender, and M clinical variables (N = 444) Predictor B SE β t p 95% CI lower 95% CI upper (Constant) 13.50 0.50 - 27.00 < 0.001*** 12.52 14.48 Total FCRI-SF -0.15 0.03 -0.35 -5.00 < 0.001*** -0.21 -0.09 Child age (years) 0.10 0.03 0.20 4.00 < 0.001*** 0.05 0.15 Child gender (female) -0.40 0.12 -0.18 -3.33 0.001** -0.64 -0.16 Leukemia type 0.22 0.08 0.12 2.75 0.006** 0.06 0.38 Current treatment phase -0.09 0.03 -0.12 -3.00 0.003** -0.15 -0.03 Number of chemotherapy cycles missed -0.07 0.02 -0.11 -3.50 < 0.001*** -0.11 -0.03 Note. Multiple linear regression was conducted to identify predictors of Medication Adherence (MARS-10) Scores; Values include unstan­ dardized coefficients (B), 95% confidence intervals (CI), standard error (SE), standardised beta coefficients (β), and p-values; **= p < 0.01, ***= p < 0.001 was considered statistically significant, N = 444. leukaemia type (B = 0.22, β= 0.12, p = 0.006), current treatment phase (B = -0.09, β = -0.12, p = 0.003), and number of missed chemotherapy cycles (B = -0.07, β = -0.11, p < 0.001) were significant predic­ tors. Overall, psychological, demographic, and treat­ ment-related factors contribute to medication adhe­ 6 rence among pediatric leukaemia survivors. Discussion In the current research, the FCR was a signifi­ cant predictor of poor medication adherence among patients with leukaemia, and this meant that the Juntendo Medical Journal, 2026 anxiety related to the disease returning may decrease adherence to medications. This is in line with the earlier studies, which reported fear of recurrence to be strongly correlated with medication beliefs and illness perceptions, which indicate that the psychological aspects can affect the adherence behaviour among all cancer populations15). The observed gender differences suggest that female survivors may experience greater emotional vulnerability, which could partially explain lower adherence behaviors, and the effect was slight-tomoderate, consistent with previous findings of slight but consistent gender differences in FCR16). Conse­ quently, medication adherence was higher in male participants compared to female participants, which agrees with the literature that female cancer survi­ vors tend to report cost-related or overall nonadherence17). Age was also correlated with FCR and adher­ ence. The FCR was highest among adolescents and lowest among the youngest children (5-7 years), which aligns with a developmental approach, in which older children tend to be more aware of potential health risks18). In line with this, compli­ ance was minimally age-dependent, with adoles­ cents less compliant than younger children, sugge­ sting decreased parental supervision and increased autonomy19). Greater FCR was a predictor of worse medica­ tion adherence, with females showing slightly lower adherence, and age affecting adherence patterns, consistent with prior research15, 17, 19). Clinical factors also influenced adherence patterns. The type of leukaemia was a significant predictor of adherence, and some subtypes showed slightly higher MARS10 scores, as previously reported20). The level of medication adherence also decreased in the latter, more intensive stages of treatment, probably because of the increased weight of home-based mainte­ nance treatment and issues with caregivers21). Lastly, lower adherence was linked to absenteeism from chemotherapy sessions, highlighting the clinical significance of continuous oral maintenance therapy in pediatric ALL22). Together, these findings underscore the syner­ gies among psychological, demographic, and clin­ ical factors in medication compliance among chil­ dren with leukaemia. These findings highlight the importance of integrating routine screening for fear of recurrence into survivorship care, particu­ larly for adolescents and female survivors. Targeted psychosocial interventions may help mitigate anxi­ ety-driven non-adherence. Limitations of our study and future recommendations This research has several limitations that must be taken into account when analysing the outcomes. First, the cross-sectional design does not allow for causal inference regarding the relationship between fear of recurrence and adherence. Second, selfreport scales, particularly when an adult completes them on behalf of young children, can be vulner­ able to response and social desirability bias. Third, convenience sampling of hospitals limits the gener­ alizability of the results to all low-resource areas. Lastly, some contextual variables, including finan­ cial burden, cultural attitudes toward cancer and access to transportation, were also not examined but could be significant contributors to adherence and psychological distress. Additionally, medication adherence was self-reported, which may introduce recall or social desirability bias. Finally, the study did not include a qualitative exploration of psycho­ social factors, limiting the understanding of under­ lying barriers to adherence. Future studies ought to use longitudinal designs to monitor fear of recurrence and adherence patterns throughout and after treatment. Psychosocial inter­ vention, including counselling, family-based support programs and structured survivorship education, needs to be considered culturally adapted to decrease FCR and enhance adherence. Furthermore, health systems in low-resource countries should consider approaches such as engaging community health workers, providing telemedicine follow-ups, and implementing financial or logistical support inter­ ventions to reduce treatment abandonment. Addi­ tional information about the family's experience and barriers to regular care could also be obtained through qualitative research. Conclusion Overall, it is concluded that FCR is associated with poor medication adherence among low-resource setting pediatric survivors of leukaemia. The rese­ arch demonstrates the interplay among psycholog­ ical, demographic, and clinical factors in adherence 7 Saha S, et al: Fear of recurrence and medication adherence in pediatric leukaemia behaviour, as well as the importance of integrative survivorship care. The treatment of FCR in child­ hood leukaemia survivors by identifying it early, supporting it using psychosocial means, and impro­ ving the system can potentially decrease the rate of treatment abandonment and improve overall long-term survival. At a policy level, strengthening survivorship programs with integrated psychoso­ cial support and adherence monitoring is essential to reduce treatment abandonment in low-resource pediatric oncology settings.