Glycemic profile of children and young adults living with diabetes in Lubumbashi and associated factors with poor glycemic control

¹University Clinics of Lubumbashi, Department of Pediatrics, Faculty of Medicine, University of Lubumbashi, Lubumbashi, Democratic Republic of the Congo, ²Abass Ndao Hospital, Cheikh Anta Diop University, Dakar, Sénégal, ³Department of Family Medicine and Primary Health Care, Protestant University in the Congo, Kinshasa, Democratic Republic of the Congo, ⁴School of Public Health, University of Lubumbashi, Lubumbashi, Democratic Republic of the Congo

^&Corresponding author
Maguy Omoy Ngongo, University Clinics of Lubumbashi, Department of Pediatrics, Faculty of Medicine, University of Lubumbashi, Lubumbashi, Democratic Republic of the Congo

Introduction    

Once diabetes is diagnosed, achieving optimal glycemic control becomes an absolute necessity, as it is the only factor that can prevent or delay the onset of complications, which are responsible for high morbidity and mortality, and ensure the well-being to some one living with diabetes. Large-scale studies such as the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) and others have demonstrated the importance of strict glycemic control to avoid and delay diabetes complications [1]. Based on this evidence, for the past 20 years, scientific societies such as the International Society for Pediatric and Adolescent Diabetes (ISPAD), the American Diabetes Association (ADA), and the National Institute for Health and Care Excellence (NICE) have established and regularly updated glycemic targets. They currently recommend optimal glycemic targets, namely a glycated haemoglobin (HbA1c) level ≤ 6.5% for those who have access to advanced diabetes technology, such as continuous glucose monitoring systems (CGM), automated insulin delivery systems (insulin pump and hybrid closed-loop), newer insulin (analogues), and structured therapeutic education, and < 7% for those without access to these technologies. Lower glycemic targets are associated with improved glycemia [2-5]. These new diabetes technology become the gold standard of care for T1D in many countries, particularly for children, adolescents, and young adults [4,6].

As a result, developed countries have observed significant improvements in both quality of life and life expectancy among children with diabetes, due to a marked reduction in morbidity and mortality. However, in developing countries, health system constraints, widespread poverty, limited access to insulin and quality healthcare, and a shortage of qualified healthcare professionals make achieving recommended glycemic targets a major challenge. This has been reported by several African studies conducted in Senegal, Nigeria, Ethiopia, Tanzania, and elsewhere, which have consistently shown very high glycemic levels among children and young adults with diabetes [7-12].

In the Democratic Republic of the Congo (DRC), no study had analyzed the glycemic profile of children and young adults living with diabetes before this one. Therefore, the objective was to identify the risk factors for not achieving optimal glycemic control that could be targeted and improved, in order to achieve better glycemic balance and ensure better survival among children and young adults living with diabetes in resource-limited settings.

Methods     

Study design: a single-group prospective cohort study was conducted to determine the glycemic profile of children and young adults living with diabetes in Lubumbashi.

Study setting: two health facilities located in the center of Lubumbashi served as the setting for this study. There were two facilities: the Diabetes Center, which provides outpatient care, often free of charge, to people living with diabetes, depending on the availability of supplies; and the Lubumbashi University Clinics, a tertiary-level university hospital, the highest level of referral in the healthcare system in the DRC. This institution, dedicated to university training, offers several medical specialties, including pediatric endocrinology, and is therefore the referral center for the management of children and young adults with diabetes during critical and intercritical phases. However, patients bear the full cost of their care at this university hospital.

Participants: a total of 71 patients were included in the study. They were recruited through a review of medical records (patient files, registers) from which their contacts were identified (phone numbers, WhatsApp, email, etc). So, parents or guardians of the children, as well as the young adults living with diabetes, were contacted. A therapeutic education session was conducted to emphasize the importance of optimal glycemic control, recommended glycemic targets, and related requirements. The study objectives and the potential benefits for the participants, the science, and the community were explained. Afterward, the children's parents or guardians and the young adults were invited to provide informed consent; assent was obtained from older children. Each participant was followed for a minimum period of 18 months and a maximum of two years. Participant attendance at follow-up appointments during the study was ensured by programmed telephone alerts (alarms), telephone reminders, or home visits when appropriate; transport costs to health facilities were covered.

inclusion criteria: children aged 0 to 17 years living with diabetes and young adults aged 18 to 25 years who had been diagnosed with diabetes before the age of 18, who were followed at the Diabetology Center of Lubumbashi and the University Clinic had participated in the study from October 2022 to September 2025. All patients were receiving treatment with either insulin or oral antidiabetic drugs (OADs), depending on the type of diabetes.

Exclusion criteria: participants who were unreachable, lost to follow-up, traveling, or unwilling to participate were not included, and deceased patients were excluded. Patients aged over 25 years living with diabetes and followed at both health facilities were excluded.

Sampling and sample size determination: the sampling was exhaustive of cases, the minimum sample size was calculated by the Cochran's formula, recommended basic formula in biostatistics, in single-group cohort study (descriptive cohort) [13]:

With Z: confidence level (95%, Z = 1.96), p: estimated proportion or prevalence (prevalence of childhood diabetes in DRC = 2.156%, p = 0.02156) [14], and e: acceptable margin of error or degree of precision (5%, i = 0.05). After calculation, the minimum sample size obtained was 32.4 patients, to which 10% was added to account for potential loss to follow-up, resulting in N = n + n/10 = 35.7 patients, approximately 36 patients expected. This study included 71 patients, almost twice the expected sample size of patients. This was the study size.

Variables: they were constituted by the dependente variables which included the poor glycemic control and the independent variables which included, the sociodemographic characteristics (age, sex, nutritional status, socioeconomic status), the clinical characteristics (type of diabetes, duration of diabetes, regularity at follow-up visits, daily glycemic monitoring, possession of the glucometer and glucose test strips, symptoms of diabetes), and the therapeutic characteristcs (current treatment, therapeutic modality, therapeutic adherence, dietetitian consultation, carrying out sports activities, psychologist consultation).

Data collection procedures: data collection was conducted prospectively, based on a pre-established questionnaire. Sociodemographic, clinical, and therapeutic characteristics were collected through interviews and patient clinical examinations. The cardinal syndrome or clinical signs of diabetes consisted of: polyuria, polydipsia, weight loss or weight gain depending on the type of diabetes, asthenia, polyphagia [14]. Nutritional status was assessed using body mass index (BMI), calculated according to the Quetelet formula [weight (kg)/[height (m)]² and interpreted based on World Health Organization (WHO) age-specific standards [15]. Socioeconomic status was classified according to World Bank standards, which define the extreme poverty threshold as the minimum income or consumption level required to meet basic needs (food, housing, healthcare), set at < USD 2.15 per person per day (based on 2017-2022 price levels). Socioeconomic status was categorized as low (< USD 2.15), middle (= USD 2.15), or high (> USD 2.15) [16]. Glycated haemoglobin (HbA1c) levels were measured every 6 months. Annual and then overall averages for the total follow-up period were calculated. According to ADA and ISPAD recommendations, HbA1c levels were classified as: optimal, if <7%; suboptimal, if between 7-9%; and high risk if >9% [4,6,17].

Bias: information bias related to daily blood glucose measurement, which was not standardized for all patients; the glucose cycles recorded in their logbooks were unreliable, and some patients did not have logbooks. Some lacked equipment for daily glucose monitoring, and the current automated continuous glucose monitoring system was unavailable. To overcome this difficulty, we opted for monitoring patients' glycemic profiles by measuring HbA1c every six months and extending the follow-up period to standardize the data. Considering the irregularity of their attendance at follow-up appointments, programmed telephone alerts (alarms), telephone reminders or home visits when appropriate, and transport costs to health facilities were covered to ensure adherence to appointments. In the data analysis, logistic regression was used to identify the actual risk factors for poor glycemic control in patients.

Statistical analysis: data were entered into an Excel 2010 database and exported to IBM SPSS for Windows version 25 for statistical analysis.

Quantitative variables: quantitative variables were presented according to their mean and standard deviation, or their median and interquartile range (IQR), depending on whether they followed a normal or an asymmetric distribution. Categorical variables were presented according to absolute and relative frequencies [proportions (%)].

Statistical methods: paired t-tests (or Wilcoxon signed-rank tests) were used to compare HbA1c levels between the first and second year of follow-up. Student's t-test, Mann-Whitney U test, and Pearson's chi-square test (or Fisher's exact test) were used to compare means, medians, and proportions between two groups, respectively. Simple linear regression was used to assess the relationship between HbA1c and patient age. Logistic regression analysis was performed to identify factors associated with poor glycemic control. Adjusted odds ratios (aORs) and their 95% confidence intervals (95% CIs) were calculated to estimate the strength of associations. Statistical significance was set at p < 0.05 for all tests.

Ethical considerations: ethical approval was obtained from the Medical Ethics Committee of the University of Lubumbashi (UNILU), under number: UNILU/CEM/014/2023, dated 24-04-2023. Authorization was granted by the General Directorate of University Clinics of Lubumbashi and the Diabetology Center. Informed consent was obtained from parents or guardians of children and from young adults. Assent was obtained from older children. Confidentiality was ensured. Electronic data was stored on a laptop and protected by a password.

Results     

Participants: based on the inclusion criteria, 86 children and young adults living with diabetes were initially identified as potentially eligible: 2 had died just before the follow-up period (2.3%), 4 had traveled (4.7%), and 6 were unreachable (6.9%). Thus, 74 patients were included (86.0%). Among them, 3 died during the study and were excluded (3.5%). A total of 71 children and young adults living with diabetes were ultimately included and followed in the study, including 56 from the Diabetology Center (78.9%) and 15 from the University Clinics of Lubumbashi (21.1%) (Figure 1).

Descriptive data

Sociodemographic characteristics: the mean age was 16.0 ± 5.1 years (range: 3-25 years), with a higher proportion of children aged 0-17 years (54.9%). The male-to-female ratio was 1.3. Patients were underweight (64.8%), 4.2% were obese. A low socioeconomic status was observed in 69% of patients. Clinic characteristics: type 1 diabetes represented.8%, and the median duration of diabetes was 4 years (IQR: 2.0-5.0 years; range: 1-12 years). Patients were irregular in follow-up visits (94.4%), mainly due to schooling (71.6%), work (6%), and lack of transportation costs (59.7%). Daily glycemic monitoring was poor in 93% of patients due to fear of needle pricks (91.5%), lack of a glucometer (32.4%), lack of glucose test strips (56.3%), and forgetfulness (67.6%). Clinical symptoms of diabetes were present in 90.1% of patients.

Therapeutic characteristics: insulin therapy was the main treatment (95.8%). However, 21.1% of patients experienced intermittent insulin shortages due to stock-outs (14%) or financial constraints (21.1%). 4.2% of patients were on oral antidiabetic drugs. Regarding care, 78.9% of patients received partial support from the Diabetology Center, while 21.1% were fully supported by their families. Therapeutic adherence was poor in 93.0% of patients, mainly due to fear of injections (56.9%), fear of hypoglycemia (49.3%), financial constraints (45.1%), forgetfulness (24.6%), lack of meals (14.1%), stock-outs (14.0%), and school-related reasons (19.7%). Patients had never consulted a dietitian (90.1%), mainly due to lack of information (76.1%) and financial limitations (14.1%). 88.7% of patients did not engage in sports activities due to fear of hypoglycemia (52.4%), insufficient food intake (29.6%), and lack of information (47.6%). None of the patients had consulted a psychologist.

Main results

Glycemic profile: the overall average HbA1c throughout a minimum period of 18 months and maximum of two years follow-up was 13.4 ± 3.6% (range: 7.5-22%) (Table 1). The distribution was highly heterogeneous (Figure 2), with a very high proportion of patients having high-risk HbA1c, > 9% (91.5%), followed by those with suboptimal HbA1c (7- 9%; 8.5%); none achieved optimal HbA1c levels (<7%). Each participant had been followed for a minimum period of 18 months and a maximum of two years. During the first year, the average HbA1c was 13.0 ± 2.7% (range: 7.5-20%), with 88.7% of patients having high-risk HbA1c and 11.3% having suboptimal HbA1c. During the second year, the average HbA1c increased to 13.7 ± 4.0% (range: 8.5-22%), with 94.4% of patients having high-risk HbA1c and only 5.6% having suboptimal HbA1c. The average HbA1c during the first year was significantly lower than that of the second year (p = 0.030). There was a statistically significant and positive linear correlation between HbA1c and patient age (p < 0.001), with a correlation coefficient of 42.5% (r = 0.425). Patients aged 18 years and older had a significantly higher average HbA1c compared to those under 18 years (p = 0.048).

Associated factors with poor glycemic control: in the univariate analysis, patients aged 18-25 years had a significantly higher risk of poor glycemic control compared to those aged 0-17 year, Female sex was associated with higher risk than male sex, Disease duration >5 years increased the risk, Low socioeconomic status was also a significant factor, as was irregular attendance to follow-up visits. Poor daily glycemic monitoring and poor therapeutic adherence were also significant. Patients receiving partial medical care had markedly higher risk. Lack of dietetic consultation and absence of physical activity were additional univariate predictors. After multivariate adjustment, independent predictors of poor glycemic control included: age 18-25 years (adjusted OR [aOR]: 4.01; 95% CI: 2.79-5.43; p = 0.012), disease duration >5 years (aOR: 4.50; 95% CI: 2.83-6.81; p = 0.016), low socioeconomic status (aOR: 2.20; 95% CI: 1.80-6.95; p = 0.036), irregular follow-up visits (aOR: 3.45; 95% CI: 2.63-7.69; p = 0.004), poor daily glycemic monitoring (aOR: 2.86; 95% CI: 1.91-9.10; p = 0.024), poor therapeutic adherence (aOR: 2.36; 95% CI: 1.89-5.51; p = 0.001), partial medical care (aOR: 7.02; 95% CI: 3.56-9.40; p = 0.011), and lack of sports activities (aOR: 3.20; 95% CI: 2.76-5.83; p = 0.041) (Table 2).

Discussion     

Participants: the majority of patients (78.9%) came from the diabetology center, where care was largely free of charge.

Sociodemographic characteristics: the mean age was 16.0 ± 5.1 years [range: 3-25 years], consistent with juvenile diabetes [18]. The sex ratio was 1.3 in favor of males. Most patients had a low BMI (64.8%), suggestive of either poor management of diabetes as a chronic disease or undernutrition due to poverty [7,19]. Indeed, 69.0% of patients had a low socioeconomic status, which limited access to technology and quality healthcare [12,19,20].

Clinical characteristics: type 1 diabetes predominated (95.8%), with a median disease duration of 4 years (IQR: 2.0-5.0) [range: 1-12 years], consistent with literature reporting early onset of the disease [21,22]. Almost all patients were irregular in follow-up visits (94.4%), mainly due to school obligations (71.6%), work (6%), and lack of transportation (59.7%), highlighting poor continuity of care. The distance to the care center posed a real accessibility problem, the use of telecommunication tools was limited, family poverty, patients’ school age, the need for dose adjustments during growth and development particurary during adolescence, and the demands of school and work, training healthcare staff as community-based relays could improve close follow-up, dose adjustments, continuous patient education, and rapid assistance when necessary, provided there is close monitoring and support from the central level. Daily glycemic monitoring was poor (93%), due to fear of injections (91.5%), lack of glucometers (32.4%), lack of test strips (56.3%), and forgetfulness (67.6%), consistent with other African studies [9,23]. This predicted poor glycemic control, evidenced by the presence of clinical signs of diabetes in 90.1% of patients.

Therapeutic characteristics: insulin therapy predominated (95.8%), reflecting the high proportion of T1D in the cohort. There was poor adherence to treatment because some patients experienced periods without insulin (26.8%) due to stock-outs at the diabetology center (14%) or in pharmacies for those managed only by parents, and financial constraints (26.8%), increasing the risk of complications, as reported in a Senegalese study [23]. Even when insulin was available, therapeutic adherence was also poor (93.0%) due to fear of injections (56.9%), fear of hypoglycemia (49.3%), missed meals (14.1%), forgetfulness (24.6%), school and work obligations (19.7%); consequences of limited access to diabetes management technology innovations, which facilitates automated insulin delivery, painless continuous glucose monitoring, use of newer insulins with more physiological pharmacokinetics and improve time spent in glycemic targets. The associated carbohydrate counting and the strengthening of therapeutic education allow for painless functional insulin therapy, reduced hypoglycemia risk, offering freedom with regard to meals. Hence, an improved quality of life [2,3,24,25].

In our context, available insulins (short-acting insulin (regular), and intermediate-acting insulin) require daily injections, causing higher peaks and greater dependence on meals, which are often insufficient in quantity or quality, increasing hypoglycemia risk. Capillary glucose monitoring for a minimum of 2-3 times per day further increases injection burden. Conventional insulin therapy combined with insufficient therapeutic education increases stress, dysglycemia risk, complications, and reduces quality of life [11,23]. Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications studies have demonstrated the superiority of functional insulin therapy over conventional regimens in adherence, glycemic control, hypoglycemia prevention, reduced risk of acute and chronic complications, and improved survival; some patients in this study had switched from conventional to functional insulin therapy [1]. A small proportion (4.2%) were on oral antidiabetic drugs (Metformin), corresponding to T2D patients, mostly obese. Insulin resistance secondary to increasing pediatric obesity underlies T2D emergence in children [26,27]. Oral therapy was less prone to stock shortages and easier to administer, yet adherence and glycemic monitoring remained poor due to forgetfulness and fatigue; their blood glucose monitoring was also poor for the same reasons.

Most patients had not consulted a dietitian (90.1%), due to lack of information (76.1%) or financial constraints (14.1%). In fact, even with the practice of counting meal carbohydrates, the diet is no longer recommended for children living with T1D, nutrition counseling remains essential to ensure accurate carbohydrate counting, to avoid glycemic excursions, to adapt the need for optimal glycemic contole to individual preferences, family food availability and sociocultural factors. This ensures adequate energy intake for normal growth and satisfactory pubertal development [28,29]. Sports activities were insufficient (88.7%), mainly due to fear of hypoglycemia (52.4%), insufficient meals (29.6%), or lack of information (47.6%), reflecting gaps in therapeutic education. Regular sports activities are crucial for achieving glycemic control, provided it is properly supervised [30]. No patient consulted a psychologist due to lack of referral or additional costs. Psychological support is important to help patients and families cope with the psychosocial and economic burden of chronic disease, as diabetes significantly impacts mental health, causing anxiety, depression, or stress due to stigma, daily management, and complications [19,31,32]. Multidisciplinary care remains essential.

Glycemic profile: according to current recommendations, good glycemic control is defined as maintaining HbA1C within target ranges: ≤6.5% or <7%, depending on technology access [2-5]. In our cohort, the overall average HbA1C was 13.4 ± 3.6% [range: 7.5-22%] throughout a minimum period of 18 months and a maximum of two years, indicating very poor control and an excessive exposure to hyperglycemia. Glycated haemoglobin (HbA1c) distribution was highly heterogeneous, with 91.5% of patients having high-risk HbA1C and 8.5% suboptimal HbA1C; none achieved optimal HbA1C (Table 2, Figure 2). Control worsened over time: the first-year average HbA1C (13.0 ± 2.7% [7.5-20%]) was significantly lower than the second-year average (13.7 ± 4.0% [8.5-22%]; p=0.030). In the first year, 88.7% had high-risk HbA1c; in the second year, 94.4% had high-risk HbA1c. Glycated haemoglobin (HbA1c) correlated positively with age (r=0.425; p<0.001), with higher values in young adults ≥18 years compared to children <18 (p=0.048). Studies in resource-limited settings found similar results [7-9,33-35], some of their patients achieved glycemic targets due to direct support from support organizations, which is lacking in Lubumbashi; hence, insufficient training and knowledge on diabetes management in children and young adults, difficult access to insulin, and gaps in therapeutic education. Currently, with the advent of new technology, there is a large gap in glycemic contrôle betwen those who have access and those who do not. Even among those who have access, the difference is noted depending on the equipment used [36].

Associated factors with poor glycemic control: parental supervision declines with age and disease duration, and adolescents often seek autonomy without adequate self-management skills, a situation that also persists in young adults. With a long disease duration, weariness often sets in with the constraints, obstacles, and very difficult conditions of diabetes management in resource-limited settings. Low socioeconomic status limits regular appointments, access to insulin, follow-up, and adherence, while insufficient sports activities reflect gaps in therapeutic education and trained personnel [7-9,35,37].

Study limitations: quarterly HbA1C measurements as recommended were not always feasible due to financial constraints. Patient numbers decreased due to loss to follow-up, relocation, or death. Nevertheless, these limitations did not prevent the successful completion of the study.

Conclusion     

Most children and young adults with diabetes in Lubumbashi, DRC did not achieve glycemic control, and it gets worse over time. Their exposure to hyperglycemia is too high and alarming. Diabetes complications could be very frequent. Low socioeconomic status creates significant barriers to therapeutic adherence and glycemic control. Therapeutic gaps are mainly due to insufficient patient education and lack of trained personnel in juvenile diabetes. Urgent measures include training healthcare staff for effective patient therapeutic education, strengthening proximity healthcare services, introducing health insurance, ensuring regular access to insulin and glycemic monitoring supplies through direct organizational support, and encouraging multidisciplinary care to improve their glycemic profile, prevent diabetes complications, and thus improve their quality of life and life expectancy.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

All authors have read and approved the final version of this manuscript.

Acknowledgments     

We thank the management of the University Clinics of Lubumbashi and the Diabetology Center for authorizing and facilitating this study, as well as all patients, parents, and guardians.

Tables and figures     

Table 1: glycemic profile of children and young adults living with diabetes, recruited at the diabetology center and university clinics of Lubumbashi in the Democratic Republic of Congo, from October 2022 to September 2025 (N=71)

Table 2: associated factors with poor glycemic control in children and young adults living with diabetes, recruited at the diabetology center and university clinics of Lubumbashi in the Democratic Republic of Congo, from October 2022 to September 2025 (N=71)

Figure 1: flow diagram of children and young adults living with diabetes, recruited at the diabetology center and university clinics of Lubumbashi in the Democratic Republic of Congo for their glycemic monitoring, from October 2022 to September 2025 (N=71)

Figure 2: flow diagram of children and young adults living with diabetes, recruited at the diabetology center and university clinics of Lubumbashi in the Democratic Republic of Congo for their glycemic monitoring, from October 2022 to September 2025 (N=71)

References