Prevalence and incidence of type 1 diabetes in the world: a systematic review and meta-analysis

1Diabetes is referred to a group of diseases characterized by high glucose levels in blood. It is caused by a deficiency in the production or function of insulin or both, which can occur because of different reasons, resulting in protein and lipid metabolic disorders. The long-term effects of hypoglycemia are tissue and organ damage.

2Symptoms of diabetes include polyuria, thirst, vision disorders, and weight loss. In some cases there are more severe forms of diabetic ketoacidosis and hyperosmolar that may lead to stupor and coma. But most symptoms are not severe, which may cause damage or even failure of different organs in the long run and lead to irreparable injuries such as blindness, amputation, stroke and eventually death. Previously, type 1 diabetes was called insulin-dependent diabetes and it could happen at any age but is most common in children and young people.

3People with type 1 diabetes are not able to produce enough insulin. This type constitutes about 5%–10% of all cases of diabetes. In this type, the cellular destruction of beta cells occurs in the pancreas. In type 1 diabetes, the pancreas does not release any insulin. Since there is no epidemiologically accurate information on the prevalence and incidence of type 1 diabetes in the world and in the region, therefore, the present study was designed and implemented as a systematic review and meta-analysis, because of geopolitical map of the policy on the prevention and treatment of this disease can be done better.

4In this systematic review and meta-analysis, a systematic search of resources was conducted by a librarian (N.V.) to investigate the prevalence and incidence of type 1 diabetes (condition) in the people (population) of the world (context). The PICO of study based on the JBI protocol as CoCoPop for prevalence and incidence studies.

5The databases of Medline via (PubMed, Ovid), Embase, Scopus, Web of Science from January 1980 to September 2019 were searched to locate English articles. Also, SID, Magiran, and Barakat databases were searched for Persian studies. The grey literature and ongoing studies were searched using the following: OpenGrey, Google Scholar and for thesis and dissertations ProQuest and studies presented at conferences were also searched. Also, experts and professionals on this subject were reached and their opinions were gathered for information on published and unpublished studies. The search was performed using MESH and free keywords. The keywords selected for the search were: “type 1 diabetes”, “prevalence”, and “incidence” with this search strategy: (((“Diabetes Mellitus, Type 1”[Mesh]) OR ((((((((((((((((((((IDDM[Title/Abstract]) OR T1DM[Title/Abstract]) OR “Type 1 Diabetes”[Title/Abstract]) OR “Autoimmune Diabetes”[Title/Abstract]) OR “Juvenile Onset Diabetes”[Title/Abstract]) OR “Juvenile-Onset Diabetes”[Title/Abstract]) OR “Brittle Diabetes Mellitus”[Title/Abstract]) OR “brittle diabetes”[Title/Abstract]) OR “diabetes mellitus type 1”[Title/Abstract]) OR “diabetes mellitus type I”[Title/Abstract]) OR “diabetes type 1”[Title/Abstract]) OR “diabetes type I”[Title/Abstract]) OR “early onset diabetes mellitus”[Title/Abstract]) OR “insulin dependent diabetes”[Title/Abstract]) OR “juvenile diabetes”[Title/Abstract]) OR “juvenile diabetes mellitus”[Title/Abstract]) OR “type I diabetes”[Title/Abstract]) OR “type I diabetes mellitus”[Title/Abstract]) OR “Insulin Dependent Diabetes Mellitus”[Title/Abstract]) OR “Insulin-Dependent Diabetes Mellitus”[Title/Abstract]))) AND ((((“Prevalence”[Mesh]) OR ((Prevalence[Title/Abstract]) OR Prevalences[Title/Abstract]))) OR ((“Incidence”[Mesh]) OR ((Incidence[Title/Abstract]) OR Incidences[Title/Abstract]))). The complete search strategy of Medline and Embase is in Supplementary file 1.

6Inclusioncriteria for selecting studies include: 1. Articles published between 1980 and 2019; 2. Articles published in English and Persian. The exclusion criteria were: 1. Studies with no reported sample size; 2. Studies that had low quality; 3. Studies that were published before 1990.

7The located articles were screened in multiple levels of title, abstract, and full-text and final studies that met the inclusion criteria were retrieved and included in the study. The studies were critically appraised by 2 subject specialists and low-quality studies were excluded. In cases of disagreements between two experts (M.M. and M.S.) at each stage of selection and appraisal, third person opinion was used.

8Articles were evaluated using the STROBE checklist. In this checklist, the minimum score was 2 and the maximum was 4. Finally, articles that received a score of 4 on checklist questions were included in the research, 128 articles earned 4 score, 46 articles earned 3 score and 19 articles earned 2 score and finally their data were extracted to perform the meta-analysis.

9The information extracted from the articles were entered in the extraction form. Extracted data included: first author, year of publication, country of study, sample size, and incidence of diabetes in the studies.

10Statistical analysis was performed using CMA v.2.0 software and P value less than 0.05 was considered as significant. The binomial distribution was used to calculate the variance. Weighted mean was used to combine the prevalence rate of different studies. Meta-analysis was used to obtain the incidence of type 1 diabetes. The heterogeneity between studies was assessed by Cochran (Q) and I statistics, which expressed the percentage of variation between studies. Random effects model was used to calculate the overall and pooled effect size.

11In a systematic search of sources, 65 765 articles were identified. A total of 58 239 articles were duplicates, and 7107 were excluded after reviewing the title and abstract of the articles. After reviewing the full-text articles, 49 articles were excluded. Finally, 193 studies were included in the systematic review and meta-analysis. Figure 1 shows the identified and retrieved articles in the study. Tables 1, 2 and 3 show the specifications of the articles that were studied.

12Prevalence and incidence of type 1 diabetes were extracted from meta-analysis studies. In type 1 diabetes incidence, the heterogeneity between studies in the meta-analysis was significant (Q = 50.51; df = 16; P < 0.001; I2 = 68.33), but in the prevalence of diabetes 1, the heterogeneity was not significant (Q = 5220; df = 6; P < 0.001; I2 = 99.88). The incidence of type 1 diabetes in Asia was 15 per 100 000 population, which was statistically significant (Incidence = 0.015, 95% CI = 0.010 to 0.021, P < 0.001), and the prevalence of type 1 diabetes was 6.9 per 10 000 people, which was statistically significant (Prevalence = 0.069, 95% CI = 0.020 to 0.214, P < 0.001). Figures 2A and 2B show the forest plot of prevalence and incidence of type 1 diabetes in Asia.

13Prevalence and incidence of type 1 diabetes were extracted from meta-analysis studies. In type 1 diabetes incidence, the heterogeneity between studies in the meta-analysis was not significant (Q = 23.79; df = 6; P < 0.001; I2 = 74.78) and in the prevalence of diabetes 1, the heterogeneity was not significant too, (Q = 4.4; df = 1; P < 0.001; I2 = 77.27). The incidence of type 1 diabetes in Africa was 8 per 100 000 population, which was statistically significant (Incidence = 0.008, 95% CI = 0.003 to 0.021 P < 0.001), and the prevalence of type 1 diabetes was 3.5 per 10 000 people, which was not statistically significant (prevalence = 0.035, 95% CI: 0.022 to 0.055, P < 0.001). Figures 3A and 3B show the forest plot of prevalence and incidence of type 1 diabetes in Africa.

14Prevalence and incidence of type 1 diabetes were extracted from meta-analysis studies. In type 1 diabetes incidence, the heterogeneity between studies in the meta-analysis was significant (Q = 895.56, df = 96, P < 0.001, I2 = 89.28) but in the prevalence of diabetes 1, the heterogeneity was not significant, (Q = 5792.85, df = 15, P < 0.001, I2 = 99.74). The incidence of type 1 diabetes in Europe was 15 per 100 000 population, which was statistically significant (Incidence = 0.015, 95% CI = 0.013 to 0.018, P < 0.001), and the prevalence of type 1 diabetes was 12.2 per 10 000 people, which was statistically significant (Prevalence = 0.122, 95% CI = 0.085 to 0.171, P < 0.001). Figures 4 and 5 show the forest plot of prevalence and incidence of type 1 diabetes in Europe.

15Prevalence and incidence of type 1 diabetes were extracted from meta-analysis studies. In type 1 diabetes incidence, the heterogeneity between studies in the meta-analysis was significant (Q = 18.88, df = 16, P = 0.27, I2 = 15.28) and in the prevalence of diabetes 1, the heterogeneity was significant too, (Q = 1120.79, df = 7, P < 0.001, I2 = 99.38). The incidence of type 1 diabetes in America was 20 per 100 000 population, which was statistically significant (Incidence = 0.020, 95% CI = 0.010 to 0.021, P < 0.001), and the prevalence of type 1 diabetes was 12.2 per 10 000 people, which was statistically significant (Prevalence = 0.093, 95% CI = 0.063 to 0.137, P < 0.001). Figures 6A and 6B show the forest plot of prevalence and incidence of type 1 diabetes in America. A sensitivity analysis was done for Incidence of type 1 diabetes in America based on excluding studies with too wide CIs. Sensitivity analysis’s results show that the incidence of type 1 diabetes in America is 19 per 100 000 population, which is statistically significant (Incidence = 0.019, 95% CI = 0.016 to 0.022, P < 0.001).

16Prevalence and incidence of type 1 diabetes were extracted from meta-analysis studies. In type 1 diabetes incidence, the heterogeneity between studies in the meta-analysis was significant (Q = 1020.30, df = 137, P < 0.001, I2 = 86.57) and in the prevalence of diabetes 1, the heterogeneity was significant too, (Q = 14760.32, df = 32, P < 0.001, I2 = 99.78). The incidence of type 1 diabetes in world was 15 per 100 000 population, which was statistically significant (Incidence = 0.015, 95% CI = 0.013 to 0.017, P < 0.001), and the prevalence of type 1 diabetes was 9.5 per 10 000 people, which was statistically significant (prevalence = 0.095, 95% CI = 0.070 to 0.128, P < 0.001). Figure 7 shows the forest plot of prevalence and incidence of type 1 diabetes in the world.

17In order to assess the publication bias, Eggers Regression test was used. Based on the results, the population bias between studies was not significant (t-value = 1.26, df 93, P = 0.21).

18Meta-regression was used to determine the effect of time on type 1 diabetes incidence. The results showed that the incidence of type 1 diabetes has increased over time. The meta-regression plot is shown in Figure 8.

19The global trend of increasing prevalence of type 1 diabetes, with multiple etiologies, operates through multiple mechanisms. In the present study, data were extracted from 193 articles between 1990 and 2019. The results showed that the incidence of type 1 diabetes in continental subgroups (Asia, Africa, Europe, and America) was 15 per 100 000, 8 per 100 000, 15 per 100 000 and 20 per 100, respectively. Also, the global prevalence of continental subtypes of type 1 diabetes in the above regions was, 6.9 per 10 000, 3.5 per 10 000, and 12.2 per 10 000, respectively.

20Relative differences between obtained results and previous statistics may be due to different research time periods and new global population status. Especially in recent years (social, political and economic migration), the changing global climate coupled with new policies and sanctions that have led to poorer middle-income and low-income countries.

21The pathogenesis of type 2 diabetes is known, which is associated with different genes and the involvement of multiple factors. Type 2 diabetes can be prevented and treated by removing or reducing these factors. Most of the warnings of national and international health bodies and diabetes associations are based on lifestyle changes and stress reduction that can prevent diabetes.

22But in type 1 diabetes, that make up 5 to 15 percent of diabetics and often involve children, Prevention ways have not yet been defined. However, screening of type 1 diabetes in prone families in relation to autoantibodies has recently been proposed. Also, clinical studies on the prevention of type 1 diabetes have been conducted.

23If one foot was amputated every 30 seconds, today it’s every 15 seconds. Need for dialysis equipment will increase. The CCU and ICU beds will be full of stroke and myocardial infarction patients. The population of the blind increases and unfortunately, new, effective, and less complicated treatments become more expensive.

24The disease shows a significant increase in glucose and possibly DKA. These patients definitely need insulin due to the pathogenesis of insulin deficiency. Manufacturing and production of insulin (traditional insulins and analog insulins) and insulin pumps, despite being inexpensive in producing countries, is shipped to low- and middle-income countries for high prices which is a major problem for the managing of type 1 diabetes patients. Certainly, uncontrolled hyperglycemia in type 1 diabetic patients will make all the problems more severe.

25One of the limitations of the study was the poor quality of some articles and, despite a careful search, the lack of access to some of the full text of the published articles.

26According to the results, the incidence and prevalence of type 1 diabetes are increasing in the world. As a result, insulin will be difficult to access and afford, especially in underdeveloped and developing countries. Thus, warnings about this can help international organizations and countries to plan for preventive measures.

27This research was approved by the Local Ethics Committee with No. 61701.

28The authors declare that they have no competing interests.

29This article was supported by the Research Center for Evidence-Based Medicine, and the Research Vice-Chancellor of Tabriz University of Medical Sciences.

30Concept: MM. Study design: MSH and TA. Systematic search: NV. Critical reviews: MM and TA. Data extraction: MSH and MGH. Data analysis: MGH and HHF. Writing: NV, TA and MM. All authors had primary responsibility for the final content of the manuscript and read and approved the final manuscript.

31Special thanks to the Research Vice-Chancellor of Tabriz University of Medical Sciences for financial support for this study.