Screening of pre-type 1 diabetes in high-risk paediatric population: a pilot study in Latvia

Abstract

The incidence of type 1 diabetes (T1D) among the paediatric population in Latvia is gradually increasing, and ~ 50% of patients with newly diagnosed T1D present with diabetic ketoacidosis (DKA). This study aimed to initiate T1D screening for high-risk children to reduce the rate of DKA and enable early pre-symptomatic intervention. This prospective observational study was conducted at Children’s Clinical University Hospital (CCUH) from March 2024 to February 2026. Siblings aged 2–17 years of patients with T1D were invited to participate. The parents completed a short questionnaire regarding participants’ health status and family history. Screening was performed in two stages: (1) 3-screen ELISA and insulin autoantibody (IAA) ELISA testing; if ≥ 1 positive then (2) ELISA testing for all four autoantibodies (AAs) separately: IAA, insulinoma-associated antigen-2 (IA-2A), zinc transporter 8 (ZnT8A) and glutamic acid decarboxylase 65 (GAD-65A) antibodies, glycaemic status assessment, and screening for celiac disease and autoimmune thyroiditis. Participants with ≥ 2 AAs in the second screening underwent a 2 h-oral glucose tolerance test (OGTT). Parents of participating children were screened for anxiety and depression using GAD-7 (general anxiety disorder 7) and PHQ-9 (patient health questionnaire-9) instruments. Of the 83 participants in the first screening, 26 (31.3%) had ≥ 1 positive test results. This group more frequently had ≥ 1 family member with T1D than those with negative results (30.8 versus 5.3% respectively, p = 0.003). In the second screening stage, 13 participants (50%) tested negative for all 4 AAs, 9 (10.8%) had one positive AA (AA +), one participant had two AA + , and three children had three AA + . At the time of the second screening, all participants had normal glycaemic measurements, negative results for celiac disease and autoimmune thyroiditis screening. During observation period, two participants developed stage 3 T1D. Following the OGTT, another participant was diagnosed with stage 2 T1D, but two other participants with ≥ 2 AA + remained normoglycemic. The overall prevalence of depression and anxiety was similarly high among parents of children with both positive and negative T1D screening results; however, parents in the positive screening group more frequently exhibited severe anxiety.

Conclusion: Positive T1D screening with ≥ 2 AAs was observed in 5 (6.02%) participants. Three participants progressed to stage 2 or stage 3 T1D during follow-up, corresponding to a positive predictive value of 60%, within a relatively short observation period of 10–24 months. These findings support the implementation of T1D screening in Latvia, particularly in high-risk populations.

What is Known: • Islet AA testing for several years has been used to screen for T1D in multiple international studies and national screening programs. • Timely detection of islet AA can prevent from DKA development and ensure presymptomatic insulin treatment with fewer complications.

What is New: • These data are of particular significance, because no prior screening initiatives for T1D have been reported in Latvia. • There is an urgent need to improve recognition of T1D in Latvia to decrease rates of DKA and other complications, and screening could be an effective tool for that.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00431-026-07278-7.

Introduction

T1D is one of the most common chronic diseases in childhood [1]. The incidence of T1D is gradually increasing worldwide at an average rate of 3–4% per year [2]. Depending on the region, 15–70% of patients (on average 30%) initially present with diabetic ketoacidosis (DKA) [1, 3].

A significantly increased risk of developing T1D is observed in individuals with a positive family history of T1D, especially when first-degree relatives are affected [1]. Other high-risk groups include individuals with other autoimmune diseases that may co-occur with T1D, such as autoimmune thyroiditis, celiac disease, and others [2].

It is well established that pancreatic islet autoantibodies (AAs) can be detected months or even years before symptomatic diabetes, therefore they are widely used in T1D screening [4]. The risk of developing T1D and the rate of disease progression depend on multiple factors including age, type, and number of autoantibodies detected. A single persistently positive AA repeatedly carries a ~ 15% risk of developing T1D within 10 years. The risk is lower when the detected antibody is IAA or GAD-65 (13.1% and 12.9% respectively), but higher when IA-2 are the only positive AA (40.5%) [2]. Individuals with ≥ 2 positive AA have an 80–90% risk of developing symptomatic diabetes within 10 years, and their lifetime risk is nearly 100%. Based on these findings, the pathophysiology of T1D has recently been described in four stages: stage 1 (≥ 2 positive AA and euglycemia), stage 2 (≥ 2 positive AA and dysglycemia), stage 3 (≥ 2 positive AA and diabetes criteria ± symptoms of T1D), and stage 4 (established life-long diabetes). A pre-stage 1 phase, characterized by a single positive or transiently positive AA, may also be defined. [4]. Although no treatment currently halts progression to stage 3 T1D, screening offers several benefits: (1) reducing the risk of initial presentation with DKA, thereby lowering mortality and complication rates, subsequently also medical costs at the initial hospitalization; (2) enabling early insulin therapy, which may preserve beta-cell function and improve diabetes compensation; (3) providing with time to learn about T1D management in less stressful circumstances; (4) reducing the risk of misdiagnosis as type 2 diabetes; and (5) where available, – a monoclonal anti-CD3 antibody course may be prescribed in order to delay progression from stage 2 to stage 3 [5]. Several studies have demonstrated a significantly lower prevalence of DKA following the implementation of T1D screening programs [6, 7].

Current situation in Latvia. The incidence of T1D in Latvia is steadily increasing, exceeding the global average incidence of 14 per 100′000 annually [1]. The most recent data from 2022 indicate that the incidence of T1D in Latvia has reached 28.12 per 100′000 annually [8]. Moreover, a substantial proportion of the newly-diagnosed patients initially present with DKA. National data from 2017 and 2023 show that 48.7% of patients presented with DKA and 28.5% required admission to a paediatric intensive care unit (CCUH data). These findings highlight the need to improve early recognition of T1D, including education of society and implementation of T1D screening programs. Another important issue is the limited availability of islet AA testing in Latvia, which is currently state-funded only for patients with new-onset diabetes. Testing includes only GAD-65 and islet cell antibodies (ICA), with the latter no longer recommended for diabetes diagnostics [9]. Considering these factors, it was decided that T1D screening should be initiated in an increased risk group of – siblings of T1D patients. The aim of this study was to identify children with preclinical diabetes (pre-stage 1 – stage 2) in order to prevent the manifestation of DKA and start early insulin treatment.

Methods

This prospective observational study was conducted at CCUH from March 2024 to February 2026. Siblings aged 2–17 years of patients with T1D (diagnosed before the age of 24 years) were invited to participate. Families were informed about the screening through social media platforms (Facebook, WhatsApp, Latvian Radio, and the CCUH homepage), during outpatient visits and hospital admissions. The initial target sample size was 150 children; this number was determined based on economic considerations and the availability of financial support for autoantibody (AA) ELISA reagents. This study protocol was reviewed and approved by Riga Stradiņš University Ethics Committee (approval number 2-PĒK-4/53/2024) and institutional review board of the Children’s Clinical University Hospital.

Screening scheme

Prior to participation, written informed consent was obtained from the parents of each participant. Participants and parents were also asked to complete a questionnaire regarding their health status, comorbidities and family history of T1D and other autoimmune diseases (autoimmune thyroiditis, celiac disease, etc.).

The overall screening and follow-up protocol was based on the Consensus Guidance reported on June 24, 2024, in Diabetes Care [4]. The screening was organized in two stages. During the first stage (March 2024 to May 2025), all participants were tested using a three-screen ELISA (detecting the presence of IA-2, GAD-65, and ZnT8 antibodies without differentiating between them) and IAA antibody ELISA. Participants with a positive result in either test were invited to the second stage, conducted at the outpatient department of CCUH where all four antibodies (IAA, IA-2, ZnT8 and GAD-65) were measured individually. During the second stage, glucose metabolism was assessed by measuring the HbA1c and random blood glucose levels. In addition, the participants were screened for autoimmune thyroiditis (TSH and anti-TPO antibodies) and celiac disease (total IgA and tTg IgA antibodies). Parents of patients with a positive first screening were instructed about further testing plans and actions in case of symptom development.

Participants with a single persistently positive AA in the second stage were advised to undergo annual glycaemic monitoring by their general practitioner and were informed about indications for earlier reassessment. To determine glycaemic status, an OGTT was performed in patients who had at least two positive AA in the second screening. Depending on the timing of enrolment, participants with positive screening results were followed up for 10–24 months (until March 2026). After receiving the results of the first screening, parents were asked to complete two surveys: (1) the GAD-7 anxiety scale and (2) the PHQ-9 depression scale. GAD-7 measures severity of anxiety, and consists of 7 short questions, asking about experienced symptoms over the past two weeks. Result may vary between 0 and 21 points, graduating mild anxiety symptoms between 5 and 9 points, moderate between 10 and 14 points, and severe anxiety – 15 points or higher [10]. PHQ-9 is a screening tool for depression, that consists of 9 questions about frequency of depression symptoms over the past 2 weeks. Amplitude of the final score varies between 0 and 27 points. Screening result is “positive” with 5 or more points and can be further categorized: mild depression symptoms between 5 and 9 points, moderate 10–14 points, moderately severe 15–19 points and severe depression – 20 points or higher [11]. Results from GAD-7 and PHQ-9 were analysed and compared between families with positive and negative diabetes screening results. The screening scheme is illustrated in Fig. 1.

Fig. 1.

Screening scheme for performed pilot study. Note. AA = autoantibodies, GAD-65 = glutamic acid decarboxylase 65 antibodies, GAD-7 = General Anxiety Disorder 7, IAA = insulin autoantibodies, IA2 = insulinoma-associated antigen-2 antibodies, OGTT = oral glucose tolerance test, PHQ-9 = Patient Health Questionnaire-9, T1D = type 1 diabetes mellitus, ZnT8 = zinc transporter 8 antibodies

Sample collection and testing

Venous blood samples were collected in EDTA tubes (BD, catalogue No 367525) for the first screening at CCUH between March 2024 and May 2025. Samples were centrifuged to separate approximately 2000 µL of plasma, aliquoted and stored at −70 °C. Laboratory analyses were performed at the Scientific Laboratory of Molecular Genetics, Riga Stradiņš University, in accordance with the manufacturer’s instructions. Three-screen ELISA was conducted using a commercial kit (BioVendor R&D, catalogue No. R3ZGI/96R), each sample was analysed in triplicate, and the final result was calculated from the mean absorbance value. The cut-off value for a positive three-screen result was defined as an index value of ≥ 30. Results were classified as borderline positive when the index value was 30–100, highly positive when the index exceeded 100. IAA antibodies were tested, using Sebia kit (catalogue No ORG 520), each sample was analysed in duplicates. Cut-off value of positive IAA test was ≥ 10 U/ml.

Venous blood samples for the second screening were collected in EDTA tubes (BD, catalogue No 367525) at CCUH between October 2025 and November 2025. Sample preparation procedures were identical to those used in the first screening. All AAs were analysed using ELISA method, following manufacturer’s instructions: IAA antibodies were tested using same method as in the first screening; ZnT8 (Euroimmun, catalogue No EA 1027–9601, cut-off value for positive result ≥ 15 RU/ml), GAD-65 (Euroimmun, catalogue No EA 1022–9601 G, cut-off value for positive result ≥ 10 IU/ml) and IA-2 antibodies (Euroimmun, catalogue No EA 1023–9601 G, cut-off value for positive result ≥ 10 IU/ml). Each sample was tested in duplicate, and the final result was calculated from the mean absorptions. The results were calculated, using a 4-parameter fit method.

Statistical analysis

Descriptive statistics were used to characterize the study population. For parametric data, descriptive statistics were reported as mean ± SD. For nonparametric data, variables were reported as medians and interquartile ranges (Q1–Q3). Categorical variables were presented as proportions. Data normality assumptions were assessed using the Kolmogorov–Smirnov test.

To compare categorical variables (e.g., sex and family history of T1D) between groups, the Chi-Square Test or Fisher's Exact Test was performed. For comparison of parametric data, the independent samples t-test was used. The Mann–Whitney U test was used to compare continuous nonparametric data. Statistical significance was set at P < 0.05. In the case of missing data, the results were calculated only from the known data. All statistical analyses were conducted using SPSS version 31.0.

Results

A total of 83 participants applied and were enrolled in this study between March 2024 and May 2025. The highest enrolment activity occurred at the beginning of the first notification and then after repeated reminders closer to the end of the first screening. Another factor that may have contributed to increased enrolment during the spring is the higher incidence of newly diagnosed T1D cases at that time, as families were invited to enrol their healthy children while attending the inpatient department. The screening entrance activity is shown in Fig. 2.

Fig. 2.

Number of children who underwent the islet cell screening per month (March 2024 – May 2025)

Characteristics of participants are shown in Table 1.

Table 1.

Characteristics of participants and their family history, n = 83

Parameter	Value
Age, years, mean (SD)	9.8 (± 4.6)
Age groups, n (%):
2–4 years old	14 (16.9)
5–9 years old	31 (37.3)
10–14 years old	23 (27.7)
15–17 years old	15 (18.1)
Sex – boys, n (%)	43 (51.8)
Autoimmune comorbidities:
Autoimmune thyroiditis, n (%)	3 (3.6)
Celiac disease, n (%)	3 (3.6)
Non-autoimmune comorbidities:
Bronchial asthma, n (%)	6 (7.2)
Congenital heart disease, n (%)	1 (1.2)
Gilbert syndrome, n (%)	1 (1.2)
Sibling with type 1 diabetes:
Brother, n (%)	44 (53.0)
Sister, n (%)	39 (47.0)
Mean age at T1D diagnosis in sibling, years	7.5 (± 3.6)
Another family member with T1D besides sibling:
Mother, n (%)	1 (1.2)
Father, n (%)	4 (4.8)
Second degree relative, n (%)	5 (6.0)
Third degree relative, n (%)	2 (2.4)
Other autoimmune diseases in family, n (%)
No	45 (54.2)
Yes	26 (31.3)
No information about other family members	12 (14.5)

SD standard deviation, T1D type 1 diabetes mellitus

Among all participants, eight (9.6%) tested positive for IAA antibodies, with titres ranging from 10.06 to 31.19 U/ml (assay range 0–100 U/ml). Whereas 21 participant (25.3%) showed positive three-screen ELISA results. Most participants had borderline positive results, but six (7.2%) had an index value > 100. Both tests were positive in 3 (3.6%) patients (in all cases – low titre). Overall, 26 (31.3%) participants had at least one positive test, and all those children were invited for the second screening. One participant (girl, aged 13 years at the time of the first screening) developed diabetes symptoms and was diagnosed with T1D 11 months after the first screening. Retrospectively she had highly-positive three-screen test (Index 1186.7). A comparative analysis was performed between participants with positive and negative screening results (Table 2). We observed that patients with a positive first screening test more frequently had multiple family members who were diagnosed with T1D.

Table 2.

Comparative analysis between participants with negative versus positive first screening result, n = 83

	Negative screening n = 57	Positive screening n = 26	p-value
Age, years, median (IQR)	8.58 (5.5–13.75)	10.92 (7.7–14.2)	0.35
Sex, n (%)
Boys	27 (47.4)	16 (61.5)	0.23
Girls	30 (52.6)	10 (38.5)
Autoimmune comorbidities, n (%)
Autoimmune thyroiditis	1 (1.8)	2 (7.7)	0.23
Celiac disease	2 (3.5)	1 (3.8)	1.00
Sibling with T1D, n (%)
Brother	31 (54.4)	13 (50.0)	0.71
Sister	26 (45.6)	13 (50.0)
Relatives with T1D, n (%)
Only siblings	54 (94.7)	18 (69.2)	0.003
Multiple relatives	3 (5.3)	8 (30.8)
Other autoimmune comorbidities in family, n (%)
No	26 (56.5)	19 (73.1)	0.13
Yes	20 (43.5)	6 (23.1)
No information about other family members	11 (19.3)	1 (3.8)

IQR interquartile range, T1D type 1 diabetes mellitus)

Immediately after the first screening, the families were identified and invited to participate in the second screening. One participant who had a negative first screening was also invited for the second screening, because during the waiting period before test results, HbA1c was measured independently after parents wish and was increased (6%). During the second screening, the participants were also questioned about possible diabetes symptoms, which were denied in all cases. In addition to blood samples for AA, tests for glycaemic status and screening for AIT and celiac disease were performed. At the time of testing, all participants had normal glycaemic profiles. The results of these investigations are summarized in Table 3.

Table 3.

Evaluation of glycaemic status and screening of autoimmune thyroiditis and coeliac disease during the second screening, n = 26

Parameter	Value
Age, years, mean (SD)	10.3 (± 4.4)
BMI, n (%):
Underweight	2 (7.7)
Normal weight	15 (57.7)
Overweight	6 (23.1)
Obese	3 (11.5)
Random glucose, mean (SD), range	5.0 (± 0.6), range: 3.7–6.7
HbA1c, mean (SD), range	5.1 (± 0.2), range: 4.6–5.6
TSH, uU/ml, mean (SD)	2.2 (± 1.0)
Positive anti-TPO, n (%)	2 (7.7)
Total IgA, mean (SD), range	1.5 (± 0.7), range: 0.4–3.0
Positive tTg IgA, n (%)	0 (0.0)
Second diabetes screening result, n (%)
Negative	13 (50.0)
1 AA positive	9 (34.6)
2–4 AA positive	4 (15.4)

AA autoantibodies, Anti-TPO thyroid peroxidase antibodies, BMI body mass index, HbA1c haemoglobin A1c, IgA immunoglobulin A, SD standard deviation, TSH thyroid-stimulating hormone, tTg IgA tissue transglutaminase immunoglobulin A antibodies

ELISA analysis for the second screening showed that results were negative in 13 (50%) patients, single-positive AA was detected in nine (34.6%) patients, two AA were detected in one (3.9%) patient, and three highly positive AA (ZnT8, GAD-65 and IA-2) were detected in three (11.5%) participants. One of them (a girl, 13 years old) developed diabetes symptoms and mild DKA 1.5 months after the second screening despite normal glucose tests at the time of the second screening. Before the onset of diabetes symptoms, the patient had a viral infection that could have accelerated beta cell destruction. The AA test results of the second screening are presented in Table 4.

Table 4.

Islet autoantibody test results in the second screening (range of autoantibody titre reflected in participants with positive test results

Positive autoantibody (-ies)		Number of patients, n (%)
1	IAA (range 10.99–38.38 U/ml)	5 (19.2)
	GAD-65 (range 10.64–328.52 IU/ml)	2 (7.7)
	ZnT8 (range 20.99–162.74 RU/ml)	2 (7.7)
2	IAA (13.89 U/ml) + GAD-65 (13.454 U/ml)	1 (3.9)
3	IA2 (> 2000 IU/ml) + ZnT8 (range 1368.9–1857.3 RU/ml) + GAD-65 (range 104.48–392.73 IU/ml)	3 (11.5)
Total number of patients with positive AA		13 (50)

AA autoantibodies, GAD-65 glutamic acid decarboxylase 65 antibodies, IAA insulin autoantibodies, IA2 insulinoma-associated antigen-2 antibodies, ZnT8 zinc transporter 8 antibodies

Participants with 2–3 positive AA were recommended to undergo an OGTT and to repeat HbA1c. Glucose tolerance disturbances (OGTT: glucose in 120 min 8.46 mmol/l) were observed in one patient who had 3 positive AA at a high titre.

AA screening results showed that nine participants (10.8%) had a single positive AA: five (6%) IAA +, two (2.4%) GAD-65 +, and two (2.4%) ZnT8 +. At least two positive AA were recognized in 5 (6%) patients.

Final results of T1D screening are reflected in Table 5.

Table 5.

Comparative analysis of both screening stages and final results

First screening result	Number of patients, n (%)	Second screening	OGTT	Final result	Number of patients, n (%)
Negative	57 (68.7)	-	-	Screening negative	69 (83.1)
IAA +	1 (1.2)	negative	-
3-screen + (Index 30–100)	10 (12)	negative
3-screen + (Index 30–100) and IAA +	1 (1.2)	negative	-
IAA +	3 (3.6)	IAA +	-	Single positive AA	9 (10.8)
3-screen + (Index 30–100)	1 (1.2)	GAD-65 +	-
3-screen + (Index 30–100)	1 (1.2)	ZnT8 +	-
3-screen + (Index 30–100) and IAA +	2 (2.4)	IAA +	-
3-screen + (Index > 100)	1 (1.2)	GAD-65 +	-
3-screen + (Index > 100)	1 (1.2)	ZnT8 +	-
IAA +	1 (1.2)	IAA + and GAD-65 +	Euglycemia	Stage 1 diabetes	2 (2.4)
3-screen + (Index > 100)	1 (1.2)	IA-2 +, ZnT8 +, GAD-65 +
3-screen + (Index > 100)	1 (1.2)	IA-2 +, ZnT8 +, GAD-65 +	Dysglycaemia	Stage 2 diabetes	1 (1.2)
	1 (1.2)	IA-2 +, ZnT8 +, GAD-65 +	Was not performed (diabetes symptoms)	Stage 3 diabetes	2 (2.4)
	1 (1.2)	-

AA autoantibodies, GAD-65 glutamic acid decarboxylase 65 antibodies, IAA insulin autoantibodies, IA2 insulinoma-associated antigen-2 antibodies, ZnT8 zinc transporter 8 antibodies, OGTT oral glucose tolerance test

Three of five children with ≥ 2 autoantibodies progressed to stage 2 or stage 3 diabetes during follow-up, showing a positive predictive value of 60%, even within a relatively short observation period of 10–24 months. An overview of both screening stages is shown in Fig. 3.

Fig. 3.

Overview of screening results and glycaemic status. Note: AA = autoantibodies, GAD-65 = glutamic acid decarboxylase 65 antibodies, IAA = insulin autoantibodies, IA2 = insulinoma-associated antigen-2 antibodies, ZnT8 = zinc transporter 8 antibodies

A comparative analysis between the positive and negative screening groups after the second test is shown in the supplementary material, Table 1.

Psychological assessment

Following the announcement of the first screening result, parents were asked to complete two questionnaires: the GAD-7 and PHQ-9 depression screening tools. These data were analysed and compared between families with positive and negative screening results. A high prevalence of both anxiety and depression was observed among parents regardless of the screening outcome. Notably, mild anxiety was more frequently reported in the negative screening group, whereas moderate anxiety was more common among parents in the positive screening group. A higher proportion of parents of children with positive screening results reported symptoms of depression than those with negative results; however, this difference was not statistically significant. A comparison of parents’ psychological assessments between positive and negative screening groups is presented in the supplementary material, Table 2.

Discussion

This pilot study of T1D screening in high-risk groups demonstrates that autoantibody screening can be an effective method for detecting the early stages of T1D. Three of five children with ≥ 2 autoantibodies progressed to stage 2 or stage 3 diabetes during follow-up, indicating a high positive predictive value of the screening. Overall, the screening program increased awareness and vigilance among the families with positive screening results. The prevalence of autoantibody positivity in this cohort was higher than that in other large-scale screening programs targeting high-risk groups. For example, in the international screening study TrialNet Pathway to Prevention (TN01), conducted in the United States, Canada, Europe, and Australia, involving more than 250,000 relatives of patients with T1D aged 3–45 years, only 5% had a single positive autoantibody, and 2.5% had at least two autoantibodies [12]. Similarly, the INNODIA study, which included both relatives and the general population in Europe (> 4,400 participants), reported 6% with a single positive autoantibody and 2.6% with ≥ 2 autoantibodies [12]. The Bart’s Oxford (BOX) Family Study in the United Kingdom reported that 6% of participants had a single positive autoantibody and 2% had ≥ 2 autoantibodies [12]. A screening approach similar to that used in the present study was applied to the German Fr1da study conducted in children aged 2–5 years in Bavaria [13]. In that study, a three-screen ELISA (for GAD65, IA-2, and ZnT8 antibodies) was used in the first stage, followed by a second-stage assessment using individual autoantibody testing (IAA, GAD65, IA-2, and ZnT8) with a radiobinding assay (RBA) in participants with a three-screen ELISA result above the 97.5th percentile [13]. For high-risk patients with negative screening results, it should be considered whether testing should be repeated in older age, as islet autoantibodies may develop later. Some experts recommend repeating the screening at 10 years of age in children who were initially screened negative between 2 and 6 years and belonged to higher-risk groups (family history of T1D, presence of autoimmunity) [14].

The psychological assessment of the parents revealed a notably high prevalence of anxiety and depression. An important contributing factor may be that these parents already have a child with diabetes, which may increase concerns about the risk in other children. A Latvian study involving 502 participants (patients with diabetes and their parents), which did not include screening but used the same GAD-7 and PHQ-9 instruments, reported anxiety symptoms in 62.9% and depression in 61% of parents, which was significantly higher than that in parents of somatically healthy children [15]. Compared to these findings, the prevalence of anxiety and depression in the present cohort was similar. International data indicate that mild to severe depression is observed in 32.7% of parents of patients with diabetes, with higher rates observed in mothers than in fathers [16].

This study has several limitations. First, participation in the screening program was less than expected, resulting in an extended recruitment period to achieve a sufficient sample size. Most families who refused to participate in the screening justified it with “not wanting to know” or logistical difficulties, as screening was available only at CCUH. Additionally, repeated venous blood sampling may be particularly burdensome for younger children; therefore, capillary blood testing should be considered for initial screening. This approach has been shown to be reliable, safe, and acceptable for families, particularly in population-based screening programs [17–20]. Similar challenges have been reported by several screening programs, that link this to the lack of understanding of reasoning for screening, anxiety, and fear of positive test results [12, 21, 22]. Second, the testing methodology required gathering all the samples and performing the analysis together, which prevented rapid reporting of results. Consequently, the principle of timely feedback has not been achieved. Nevertheless, participation in screening increased parental awareness, and some families initiated regular glucose monitoring or dietary modifications while awaiting results. The delay in receiving results may have also contributed to increased parental anxiety, as reflected in the GAD-7 scores. Ideally, testing method for the second screening should have been the RBA, as it is considered the gold standard for all four AA detections [23, 24]; however, this was limited by the availability of equipment and reagents. As recommended in ISPAD guidelines [25], repeated screening should be further suggested for participants with initially negative results. However, AA testing is currently not available in Latvia and also is not state-funded. All the participants were informed about diabetes symptoms and laboratory tests (glucose, HbA1c) that should be performed at general practitioner in case of those symptoms. Authors hope that results of this study would promote nationwide T1D screening.

The findings of this study highlight the need to improve access to high-quality islet autoantibody testing in Latvia, not only for patients with newly diagnosed diabetes, but also for high-risk groups. Population-based screening in Latvia may also be warranted, considering high incidence of T1D and initial presentation in DKA. [12] Finally, the implementation of nationwide screening could provide a valuable foundation for future therapeutic interventions, such as treatment with monoclonal anti-CD3 antibodies (e.g., teplizumab), aimed at delaying or preventing progression to stage 3 T1D [26].

Conclusion

A pilot screening program for type 1 diabetes in a high-risk paediatric population revealed a high proportion of autoantibody-positive children, particularly among families with multiple members diagnosed with T1D. Of all the participants with positive screening, two had developed stage 3 diabetes, and one had – stage 2 diabetes during the short observation period (from 10 to 24 months), showing the positive predictive value of these screening tests. The results of this pilot study, epidemiological situation, and high prevalence of DKA in Latvia prove that T1D screening in Latvia should be initiated and at least be available in high-risk groups. However, the organization of screening should be adjusted and optimized for better compliance, faster screening results, and effectiveness. Nationwide screening should also raise awareness of diabetes among society and general practitioners. Psychological support should be increased for families that raise children with diabetes to decrease the rates of depression and anxiety.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AA

Autoantibody

DKA

Diabetic ketoacidosis

GAD-65

Glutamic acid decarboxylase 65 antibodies

GAD-7

General Anxiety Disorder 7

IAA

Insulin autoantibodies

IA2

Insulinoma-associated antigen-2 antibodies

ICA

Islet cell antibodies

OGTT

Oral glucose tolerance test

PHQ-9

Patient Health Questionnaire-9

T1D

Type 1 diabetes mellitus

ZnT8

Zinc transporter 8 antibodies

Author contributions

LB, LG, IDK and JP contributed to conception and design of the study. LB organized database. LB recruited the patients and collected the data. DR, KR, and LB performed the laboratory analysis. AKU, EV, and MZ supported the study design and contributed to the data analysis. LB wrote the first draft of the manuscript and made corrections after reviews that was approved by all the authors.

Funding

This research was supported by a doctoral study grant from Riga Stradiņš University for the purchase of ELISA reagents.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics

This study protocol was reviewed and approved by Riga Stradiņš University Ethics Committee (approval number 2-PĒK-4/53/2024) and Children’s Clinical University Hospital. All procedures were performed in accordance with the principles of the Declaration of Helsinki.

Written informed consent was obtained from the parents or legal guardian.

Competing interests

The authors declare no competing interests.