The status of care for youth with type 1 diabetes within and coming from humanitarian crises settings: a narrative review

Abstract

Background

Humanitarian crises bring unique, and potentially growing challenges to people with type 1 diabetes (T1D). We aimed to determine, in youth with T1D (mean age (± 1SD) 0–17.9 years) within and coming from humanitarian crises settings (HCS), the reported prevalence that meet international consensus targets for glycaemic, blood pressure and lipid management, and incidence of severe hypoglycaemia or diabetic ketoacidosis.

Methods

A narrative review of quantitative data was conducted, using a systematic process. MEDLINE (Ovid), Global Health, Web of Science, Scopus, Embase, CINAHL, APA PsycINFO, Cochrane trials, and the reference lists of eligible records were searched (January 2014-February 2024); ten records covering ten separate studies were retrieved.

Results

Glycaemic management was consistently suboptimal in HCS. However, among individuals coming from HCS, glycaemia varied. Across both groups, data relating to blood pressure, lipids, severe hypoglycaemia or diabetic ketoacidosis were either unavailable or limited.

Conclusion

Findings expose the dearth of data relating to defined youth with T1D within and coming from HCS, leaving the status of this population largely uncharacterised. With limited data indicating suboptimal T1D management, there is a pressing need for the development of a consensus guideline on, and core indicators relating to such youth within and coming from HCS, plus monitoring systems and outcome data.

Background

Diabetes is one of the most pressing public health problems in the world due to its high and rising prevalence and widespread morbidity, which collectively has an adverse effect on people, healthcare systems, and national economies [1, 2]. When considering children and adolescents with type 1 diabetes aged < 15 years, for example, 108,300 were estimated to have been diagnosed worldwide, a number rising to 149,500 when the age range extends to < 20 years; the ratio of incidence in 15–19 years compared to those aged 0–14 years is particularly high in some countries in sub-Saharan Africa, North Africa/Middle East, and in Mexico [3, 4]. It has been well documented that the management of glycaemia (blood glucose) and cardiometabolic risk factors, supply of necessary medications, and reduction of acute and chronic consequences can all help people with diabetes to have improved quality of life and to live longer [2, 5,6,7,8]. However, in humanitarian crisis settings, there can be unique challenges that constitute a particular threat to people with diabetes, including in youth.

Humanitarian crises occur as a singular event or a series that threaten the health, safety or well-being of a community or large group of people, and are caused by factors such as armed conflicts, epidemics, famine, natural disasters, and other major emergencies. With an estimated 110 million people having been forcibly displaced worldwide at mid-2023, of which 36.4 million were refugees [9], such crises are a pressing global health concern [10]. This concern is especially the case in lower- and middle-income countries, where more than 80% of the displaced population is located, and 80% of the world’s diabetes population resides [11, 12]; thus, the regions of the world that are most affected by humanitarian crises also host most of the people with diabetes globally. Because each humanitarian crisis is brought on by a particular set of circumstances, each one brings about a different combination of challenges to diabetes management and calls for a different kind of response that is tailored to impacted sectors. Such challenges include food insecurity, inadequate access to medications and testing supplies, and a shortage of providers with expertise in diabetes care; access to insulin represents a major challenge during a crisis, especially for people with type 1 diabetes, for whom the interruption of insulin is life threatening. The need to reduce diabetes risk and for people with diabetes to have equitable access to comprehensive and affordable care regardless of setting, has been the focus of the Global Diabetes Compact; stimulating implementation of the Global Action Plan for the Prevention and Control of Non-Communicable Diseases 2013–2020, endorsed at the 74th World Health Assembly Resolution [13]. When considering humanitarian crisis settings, strengthening integration of non-communicable diseases in humanitarian responses has been the focus of a recent global technical meeting, hosted by the Government of the Kingdom of Denmark and co-organized by the World Health Organization and United Nations High Commissioner for Refugees [14].

Despite the potential breadth and severity of type 1 diabetes in humanitarian settings, there has been limited assembly or synthesis of data on this problem. Population-based studies conducted outside humanitarian crises settings show that both the access to and the delivery of evidence-based care is suboptimal. For example, in a cross-sectional analysis of pooled, individual data from 55 nationally representative surveys in low-and middle-income countries, fewer than one in ten people with diabetes receive coverage of guideline-based comprehensive diabetes treatment [15]. Problems exist even in well-resourced health systems such as Australia; data from the Australasian Diabetes Data Network have indicated suboptimal blood glucose, blood pressure and lipid control amongst adolescents and young adults with type 1 diabetes [16,17,18]; increasing risks for premature morbidity and mortality [19, 20]. Multicomponent quality-improvement initiatives have shown sustained benefits in achieving diabetes-care goals and reducing vascular complications, even in low-resource settings, but have had limited global reach [21, 22].

Our review aimed to determine, in defined youth with type 1 diabetes within and coming from humanitarian crises settings, the reported prevalence that meet consensus targets for glycaemic, blood pressure and lipid management, and incidence of severe hypoglycaemia or diabetic ketoacidosis.

Methods

We conducted a narrative review of quantitative data using a systematic process. To promote rigour in review methods, processes adapted from established review methods set out by the Centre for Reviews and Dissemination were used, and reporting standards derived from the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) applied [23]. Standards appropriate to a narrative review were applied [24]. The review protocol is available on request, and the review was not registered.

Outcome definitions and recommended measurement methods

Definitions and criteria for ‘best practice’ HbA1c, blood pressure, total cholesterol and low-density-lipoprotein targets were determined. Detailed recommendations were available in American [25], Australian [26], Canadian [27] and International Society for Pediatric and Adolescent Diabetes [28] guidelines. Displacement was taken to include defined youth of second generation. This was because such youth may have been in-utero at the time of displacement.

Literature search methods

MEDLINE (Ovid), Global Health, Web of Science, Scopus, Embase, CINAHL, APA PsycINFO, and Cochrane trials were searched by the first author (SJ) using Medical Subject Headings and keywords (Supplementary material I) to identify relevant records published from January 2014 to February 2024. The reference lists of all eligible records were hand searched for further relevant records.

Inclusion criteria were:

• people with type 1 diabetes in and coming from humanitarian crisis settings. ‘Coming from humanitarian crisis settings’ was taken to include individuals displaced within the country of the crisis, or displaced/as a refugee in another country;

• mean age (± 1SD) 0–17.9 years, or where the results for this age range were reported separately from other age groups; a population that will be referred to as youth, acknowledging that definitions of youth vary worldwide [29]; and

• humanitarian crises settings, as defined by the places in which the World Health Organization or Médecins Sans Frontières has previous, or current projects running [30, 31]—Supplementary material II. Examples of such countries and regions include Afghanistan, Iraq, Nigeria, Syria, Ukraine and Yemen, and the Greater horn and Sahel regions of Africa. To help ensure comprehensiveness, an independent search was undertaken to determine whether the World Health Organization or Médecins Sans Frontières had provided or was providing a role in protecting health in humanitarian crisis settings known to the research team; either or both organisations were found to have been, or are currently present in such settings. The search included countries such as Indonesia and the Republic of Türkiye.

Exclusion criteria were:

• records published pre-2014. This year was chosen in view of the changing landscape of diabetes management, to ensure that the review is focused on the most recent and relevant evidence available, help manage the review process workload, and allow for a more efficient and timely dissemination of research findings to inform decision-making.

There was no restriction on the language of publication or use of grey literature. Where records were not in the English language, these were translated by research team members fluent in the respective language, or through use of software.

Search outcome

A total of 2171 records were identified, downloaded to EndNote 20™ software, and then uploaded to Covidence™; of these, 647 records were excluded as duplicates. The titles and abstracts of the remaining 1,524 records were screened, of which 1417 were considered to be irrelevant. The full text articles of the remaining 107 records were screened, of which 97 were considered to not meet study eligibility criteria. The reference lists of the remaining ten records [32,33,34,35,36,37,38,39,40,41] were searched, and four additional eligible records were identified as potentially relevant but later excluded. The search process and outcomes are summarised in Fig. 1.

Fig. 1

PRISMA flow chart [23]

Quality appraisal

To determine the strength of evidence of included records, quality appraisal was undertaken using the JBI checklists for analytical cross-sectional studies and case reports [42] (Supplementary material III and IV). Records were identified, screened and eligibility determined by the first author, although were independently appraised, and data extraction compared in duplicate by another author (SBJ); agreement was reached for all records.

Data extraction and synthesis

Data were extracted to a purpose-designed spreadsheet in Microsoft Office Word™ (Tables 1 and 2), developed based on research team consensus. Findings have been stratified according to whether the sample and assessments were taken within active humanitarian crisis settings, or taken among individuals coming from humanitarian crises settings.

Table 1 Type 1 diabetes within humanitarian crises settings

Table 2 Type 1 diabetes coming from humanitarian crises settings

Results

A total of ten records (six articles [of which two were case reports] and four abstracts) covering ten separate studies reported data from Cameroon (× 2) [37, 38], Haiti [40], Israel [35], Jordan [34], Germany/Austria [36], the Republic of Türkiye [32], Tunisia [41], Ukraine [33], and an unspecified hospital in a United Nations Protection of Civilians camp in Sub-Saharan Africa [39] (Supplementary material V); these were published between 2014 and 2023, and appraisal of records revealed they were of quality (Supplementary material III and IV).

Youth with type 1 diabetes within humanitarian crises settings

Six records provided data relating to glycaemic assessment and management amongst the study population [32, 33, 37, 38, 40, 41] (Table 1); these were published between 2015 and 2023. Glycaemic management was consistently suboptimal, with few records collecting data on the achievement of targets according to consensus guidelines. In the study by Tarçın et al. [32] which focused on the effect of continuous glucose monitoring technology on glycaemic management of Kahramanmaraş Doublet Earthquake Survivors in Adana, the Republic of Türkiye, the median HbA1c for all participants (n = 134) decreased from 8.9 to 8.6% three months after the earthquake. However, this finding was not statistically significant. In those who started using continuous glucose monitoring technology a month after the earthquake (n = 58), HbA1c statistically significantly improved (from 8.9 to 8.3%; p = 0.001), whereas those without continuous glucose monitoring experienced a statistically significant deterioration in glycemia (from 8.9 to 9.6%; p = 0.03). Suboptimal glycaemic management was also reported by Yenhovatova et al. [33], who examined glycemic management, frequency of acute and chronic complications and different treatment modalities for youth in the city of Dnipro, Ukraine. Overall, 33.9% of youth (n = 354) were reported to have a HbA1c < 7.5% (58 mmol/mol); the mean ± SD HbA1c was 7.8 ± 0.8%. Similarly, in a study by Kahan et al. [40], who examined type 1 diabetes management in a reported resource poor setting in Haiti, only one (7.0%) of the 14 participants was reported to have had a HbA1c < 7.5% (58 mmol/mol); this finding was similar to that reported by Sap et al. [38] who studied factors influencing the glycemic management of patients with type 1 diabetes (n = 100) in regions affected by war in Cameroon. Here, only 19.0% of participants were reported to have achieved undefined glycemic targets; median HbA1c was 10.1 [8.7–11.63]%. Finally, elevated blood glucose was also detailed in a case-report by Sap et al. [37], in which a HbA1c of 12.0% (108 mmol/mol) had been documented in a girl living in a refugee camp in Cameroon that had presented with diabetic ketoacidosis and malnutrition, and a case-report by Karrouri et al. [41], in which a HbA1c of 14.0% (130 mmol/mol) at presentation had been documented in a Libyan boy in Tunisia who manifested with type 1 diabetes after severe psychological trauma.

No articles provided data relating to blood pressure or lipid targets, or severe hypoglycaemia amongst youth with type 1 diabetes within humanitarian crises settings. However, one double-case report by Siese [39] provided data relating to diabetic ketoacidosis in the study population. In this, the cases were detailed of: [1] a girl that had been treated for diabetic ketoacidosis with subcutaneous insulin injections, but had remained an inpatient throughout the following two months; and [2] a girl admitted with fever and respiratory distress (treated on admission for severe malaria), with high random glucose, polyuria, polydipsia and weight loss who was discharged after two weeks. Both young people were being treated in an unspecified hospital in a United Nations Protection of Civilians camp (in Sub-Saharan Africa).

Youth with type 1 diabetes coming from humanitarian crises settings

Three records provided data relating to glycaemia amongst the study population [34,35,36] (Table 2); these were published between 2019 and 2022. Glycaemic outcomes varied, with no records having collected data on the achievement of glycaemic targets according to consensus guidelines. Prinz et al. [36] studied diabetes care in the first year of stay in Germany/Austria among pediatric refugees (n = 584 [n = 539 with type 1 diabetes; n = 13 type 2 diabetes; and n = 19 with type 3 diabetes]) from the Middle East and Africa, and to compare the results to local patients to describe current diabetes treatment and outcome. Although not specific to those with type 1 diabetes, HbA1c’s of 8.8% ± 0.1% (72.3 ± 1.0 mmol/mol) [those from the Middle East] and 9.0 ± 0.1% (75.0 ± 1.4 mmol/mol) [those from Africa] were reported. Suboptimal glycaemic management was also reported by Odeh et al. [34] in their study looking at children with type 1 diabetes who were migrants/refugees in Jordan. A greater proportion of migrants/refugees were reported to have a HbA1c ≥ 7.5% (58 mmol/mol) when compared to Jordanians (n = 28, 66.7% vs. n = 63, 60.6%, p = 0.49); the overall HbA1c was 8.9 ± 2.0% (74 ± 22 mmol/mol) for migrants/refugees (n = 42) and 9.0 ± 1.93% (75 ± 21 mmol/mol) for Jordanians (n = 104). Conversely, in a study by Elkon-Tamir et al. [35] which looked at type 1 diabetes outcomes of Eritrean status-less children living in Israel and native-born Israeli children (n = 35 [n = 7 Eritrean; and n = 28 Israeli]), current HbA1c among the Eritrean children was 7.5 ± 0.9% (58.4 mmol/mol), best HbA1c was 7.0 ± 0.8% (52.6 mmol/mol), and peak HbA1c was 8.2 ± 1.2% (65.8 mmol/mol), without statistically significant differences between the Eritrean and Israeli groups. HbA1c at diagnosis was 10.5 ± 1.6% (91.3 mmol/mol).

Limited data were available relating to blood pressure and lipid targets, severe hypoglycaemia, and diabetic ketoacidosis amongst youth with type 1 diabetes from humanitarian crises settings [33,34,35,36]. When considering blood pressure and lipid management, although not specific to those with type 1 diabetes, the study by Prinz et al. [36] reported that 26.1% of pediatric refugees from the Middle East and 26.8% from Africa had arterial hypertension (defined as taking anti-hypertensive medication and/or having elevated blood pressure above the 95th percentile of the reference population). Further, 28.7% of pediatric refugees from the Middle East and 31.4% from Africa were reported to have dyslipidaemia (defined as taking lipid-lowering drugs or having permanently [≥ 2 consecutive measurements] increased total cholesterol, low-density cholesterol, triglycerides or permanently decreased high-density cholesterol). Thresholds were as follows: total cholesterol > 5.2 mmol/l, low-density lipoprotein > 3.4 mmol/l (131 mg/dL), high-density lipoprotein < 0.9 mmol/l (35 mg/dL), and triglycerides > 1.5 mmol/l (133 mg/dL).

Prinz et al. [36] also provided data relating to incidence of severe hypoglycaemia; per 100 patient years this was reported to be 17.8 ± 4.3 amongst pediatric refugees from the Middle East, and 25.4 ± 8.7 amongst pediatric refugees from Africa; the incidence of hypoglycaemia with coma per 100 patient years was reported to be 5.1 ± 1.1 amongst pediatric refugees from the Middle East, and 4.1 ± 1.6 amongst pediatric refugees from Africa. In the study by Elkon-Tamir et al. [35], one Eritrean girl (of n = 7) was reported to have had two episodes of hypoglycemia with seizures. Finally, in the study by Odeh [34], migrant/refugee children with type 1 diabetes had recurrent hypoglycemia more than Jordanian children with type 1 diabetes (33, 78.6% vs. 56, 53.8%; p = 0.006).

Four articles provided data relating to diabetic ketoacidosis [33,34,35,36]. In Prinz et al. [36], the incidence of diabetic ketoacidosis per 100 patient years was reported to be 3.0 ± 1.1 (n = 346) amongst pediatric refugees from the Middle East, and 3.7 ± 1.9 amongst pediatric refugees from Africa, whereas in Elkon-Tamir et al. [35], 42.9% of Eritrean children (n = 3: one mild, one moderate and one severe) were reported to have had diabetic ketoacidosis at the presentation of type 1 diabetes; with no episodes occurring in any of the study patients during the follow-up period. In Yenhovatova et al. [33], in Ukraine, diabetic ketoacidosis was statistically significantly more frequent in internally displaced youth with type 1 diabetes, compared to permanent residents (33.3% vs 5.9%; p < 0.05). Finally, in Odeh [34], a higher proportion of migrants/refugees were reported to have had diabetic ketoacidosis at presentation (n = 29, 69.0% vs. n = 55, 52.9%; p = 0.07) and recurrent diabetic ketoacidosis (n = 13, 31.0% vs. n = 18, 17.3%; p = 0.07) compared to Jordanians. However, these differences did not reach statistical significance.

Discussion

Our review sought to determine, in defined youth with type 1 diabetes within and coming from humanitarian crises settings, the reported prevalence that meet consensus targets for glycaemic, blood pressure and lipid management, and incidence of severe hypoglycaemia or diabetic ketoacidosis. Findings expose the dearth of data relating to youth with type 1 diabetes within and coming from humanitarian crises settings, leaving the status of this vulnerable population largely uncharacterised. Limited available data do, however, indicate suboptimal type 1 diabetes management.

Retrieved data were collected in a limited number of countries (n = 9, plus an unspecified hospital in a United Nations Protection of Civilians camp in Sub-Saharan Africa; a topical geopolitical term [43, 44]); there was a distinct lack of data from well-publicised crises settings such as Bangladesh, Myanmar, Sudan, and Gaza. Only nine records provided data relating to glycaemic assessment and management [32,33,34,35,36,37,38, 40, 41]. There were also limited data available relating to blood pressure and lipids, severe hypoglycaemia, and diabetic ketoacidosis [34,35,36, 39]. While the lack of available data is perhaps unsurprising, data collection within humanitarian crises settings are hindered by factors including the availability of healthcare professionals, resources and time that are often problematic in a crisis. This is exacerbated by the unavailability of core indicators for surveillance and monitoring systems [45]. Many opportunities around data collection and outcomes in such populations are available, especially in view of the large number of type 1 diabetes registries worldwide. The Global Action Plan for the Prevention and Control of Non-Communicable Diseases 2013–2020 [13] calls for formulation of specific coverage and treatment targets to drive action, and assess global and national progress, with attainment by 2030. Further, it expands upon recommendations from the 2021 Lancet Commission on Diabetes [2], which emphasised the importance of better measurement at all levels as a part of health policy to lessen worldwide diabetes burden [11]. Key diplomatic, political and intergovernmental stakeholders such as the United Nations, the United Nations International Children’s Emergency Fund, the Organisation for Economic Co-operation and Development, the European Union and the World Health Organization could be instrumental in supporting and/or financing such action.

With respect to glycaemic management, for which most data were available, despite being suboptimal, variations appeared to exist. For example, the prevalence of an HbA1c < 7.5% (58 mmol/mol) was 33.9% in the study by Yenhovatova et al. [33] and 7.0% (n = 1) in the study by Kahan et al. [40], whereas Sap et al. [38] reported a prevalence of 19.0% for an undefined glycaemic target. This is not unlike data from non-humanitarian crisis settings worldwide, where levels of variation and inequalities in the prevalence of meeting glycaemic targets exist across youth population subgroups, regions, and countries. In an analysis of data from the Australasian Diabetes Data Network, for example, suboptimal glycaemia has been reported in children and adolescents, with only 18.0% meeting the recommended international glycaemic target of HbA1c < 7.0% (53 mmol/mol) [46]. This is compared to 12.3% reported amongst adolescents and young adults [18]. Across other populations with diabetes, the proportions of people in high-income countries who meet recommended targets around complication risk factors including blood glucose or pressure management, ranges from 50 to 70%, and only about 20% of high-income countries meet all recommended targets [47,48,49]. These proportions are lower in low- and middle-income countries, where only about half of people with a diabetes diagnosis have optimal glycaemia, and about one in four have optimal blood pressure [50,51,52]. Possible solutions to improving diabetes care in crisis settings have been described [45]; focusing on ensuring a reliable supply of life preserving medications and diagnostics, restoring, and maintaining access to health care, and adapting service design to the context. A consensus guideline relating to management of the child, adolescent, and young adult with diabetes in limited resource settings is available [53], however limited reference is provided to humanitarian crises settings.

Findings relating to glycaemic control involving people with type 1 diabetes coming from humanitarian crisis settings were of particular interest. Although data were limited, there is some indication that glycaemic outcomes vary between those displaced/who are refugees from a humanitarian crisis setting. For example, in the study by Elkon-Tamer et al. [35], Eritrean status-less children living in Israel and native-born Israeli children appeared to have similar glycaemic control. Conversely, in the study by Yenhovatova et al. [33], diabetic ketoacidosis was significantly higher in internally displaced people in the Ukraine, compared to residents. Such variation may reflect possible weaknesses, discrimination and strengths in health systems for people with type 1 diabetes coming from humanitarian crisis settings.

Based on estimations around the number of people who have been forcibly displaced and/or are refugees worldwide [9] and review findings, the need for a consensus guideline on defined youth with type 1 diabetes in humanitarian crises settings and core indicators, together with monitoring systems and outcome data are pressing [10, 54]; guidelines relating to non-humanitarian crises settings are already available [28, 55]. The guideline for humanitarian crisis settings could be developed by key stakeholders such as the International Society for Pediatric and Adolescent Diabetes, the American Diabetes Association, and the European Association for the Study of Diabetes, whereas core indicators could perhaps be formulated by key stakeholders such as the International Alliance of Diabetes Action [56]; a multi-sectoral partnership of international organisations with an interest in diabetes that are developing concrete collaborative initiatives to increase access to and quality of care for people with diabetes within humanitarian settings. Outcome data should consider the mental health of youth with type 1 diabetes. This is important considering humanitarian crises and displacement can contribute to the development and exacerbation of mental illness; living with diabetes already carries inherent risks for psychosocial problems [57]. In Moria camp on the island of Lesbos, Greece, for example, although not specific to type 1 diabetes populations adverse conditions have been reported to negatively influence the mental health of inhabitants [58].

Some limitations apply to the current review. There is a need for caution when considering generalisation of review findings as much literature had small sample populations. Also, data heralded from varying demographic, socio-economic, clinical and humanitarian contexts, and subject matter experts were not contacted A review strength, however, is the comprehensive search strategy, the high number of databases from which data were obtained and the methodology employed, alleviating publication and selection bias risk, and its potential impact on the results of the review.

Conclusion

In conclusion, findings have demonstrated that limited data are available on defined youth with type 1 diabetes within and coming from humanitarian crises settings, relating to glycaemic, blood pressure and lipid management, severe hypoglycaemia, and diabetic ketoacidosis, leaving the status of this vulnerable population largely uncharacterised. Limited available data do, however, indicate suboptimal type 1 diabetes management. There is a pressing need for the development of a consensus guideline on, and core indicators relating to, youth with type 1 diabetes within and coming from humanitarian crises settings, plus monitoring systems and outcome data. Key diplomatic, political and intergovernmental stakeholders could be instrumental in supporting and/or financing such action. Unless change is made, this vulnerable population will continue to have suboptimal diabetes management, increasing their risk of complication onset and progression.