State of the world’s children with diabetes


Denis Daneman, MB BCh FRCPC
The Hospital for Sick Children
555 University Avenue
Toronto, Ontario
Canada M5G 1X8.
Tel: 416 813-6122;
fax: 416 813-7479;


Daneman D. State of the world’s children with diabetes.
Pediatric Diabetes 2009: 10: 120–126.

‘Insulin is not a cure for diabetes, it is a treatment.’

Frederick Banting, Nobel Prize Address, 1925

The discovery of insulin at the University of Toronto in 1922 remains one of the sentinel milestones in biomedical research: almost immediately, type 1 diabetes as we know it today went from being a uniformly fatal disorder to one with decreasing morbidity and mortality at disease onset but with important long-term ramifications (1). The lives of two youngsters who received insulin in 1922, Leonard Thompson and Elizabeth Hughes, illustrate the enormous variability in outcome of people with type 1 diabetes in the ‘insulin era’. Both of these young teens were dying from type 1 diabetes when they received early insulin preparations. On 11 January 1922, 14-yr-old Leonard Thompson was the first to receive injections of insulin (‘7.5 cc of McLeod’s serum into each buttock’ according to his medical chart). He died 13 yr later at age 27 yr of the chronic complications of diabetes. Elizabeth Hughes, also aged 14 yr, came from Boston to be treated by Frederick Banting. She wrote to her mother of the ‘unspeakably wonderful’ effects of insulin on her health. Elizabeth lived 59 more years dying at the age of 73 yr.

From the early days of insulin discovery and availability until about the mid-to-late 1970s, pediatricians and pediatric diabetologists had a very limited view of diabetes, the vast majority being because of juvenile diabetes (which changed into insulin-dependent diabetes mellitus and now into type 1 diabetes), with only a few ‘other types’. Over the past 30 yr, diabetes in childhood has become much more complicated: (i) the emergence of the childhood obesity ‘epidemic’ has begun to uncover increasing numbers of youth with type 2 diabetes, especially in high-risk ethnic groups; (ii) the presence of insulin resistance associated with obesity superimposed on type 1 diabetes produces what some have called ‘double diabetes’; and (iii) recently, magnificent advances have been made in defining the molecular basis of both neonatal diabetes and some of the conditions previously labeled MODY or maturity diabetes in the young, now more appropriately termed monogenic diabetes (Fig. 1) (2).

Figure 1.

Classification of diabetes in youth. MODY, maturity diabetes in the young. © 2002 American Diabetes Association From Diabetes Care®, Vol. 25, 2002; 731-736. Reprinted with permission from The American Diabetes Association.

This review will focus on the global issues of type 1 diabetes in youth addressing the magnitude of the problem as well as some of the dilemmas and responsibilities that confront health care professionals committed to the care of children with diabetes. I will consider these in the context of the Ecological Perspectives Model, which postulates that health care outcomes depend on factors in five domains: societal, community, institutional, interpersonal, and intrapersonal (3). It is my hypothesis that the outcome of youth with type 1 diabetes is very much dependent on the macro- (societal and community) and micro- (institutional, interpersonal and intrapersonal) environments in which they ‘find’ themselves (Fig. 2). Some of these are predetermined and others modifiable. This review assumes that our approach to the treatment of type 1 diabetes is informed by the findings of the landmark Diabetes Control and Complications Trial (DCCT) and its long-term follow-up and the Epidemiology of Diabetes Interventions and Complications (EDIC) (4–6) (i) that there is a close, somewhat curvilinear relationship between the glycemic control achieved and the onset and progression of both microvascular and macrovascular complications, and (ii) that the level of control early in the course of the diabetes influences long-term outcomes, a phenomenon termed ‘metabolic memory’. As a barometer of success, I will use the Declaration of Kos, 1993, as the yardstick because this was the ambitious articulation of the International Society for Pediatric and Adolescent Diabetes’ (ISPAD) targets for the year 2000 (Table 1). Have we been successful or are we still far off the mark? (7).

Figure 2.

Five domains of the Ecological Perspectives Model that predict health care outcomes. These are divided into macroenvironments and microenvironments.

Table 1.  Declaration of Kos, 1993: targets set by International Society for Pediatric and Adolescent Diabetes for achievement by the year 2000 (7)
To make insulin available to all children and adolescents with diabetes
To reduce the morbidity and mortality rate of acute metabolic complications or missed diagnosis related to diabetes mellitus
To make age-appropriate care and education available to ALL children and adolescents with diabetes as well as to their families
To increase the availability of appropriate urine and blood self-monitoring equipment for ALL children and adolescents with diabetes
To develop and encourage research on diabetes in children and adolescents around the world
To prepare and disseminate written guidelines and standards for practical and realistic care and education of young people with diabetes – and their families – emphasizing the crucial role of healthcare professionals – and not just physicians – in these tasks around the world

Magnitude of the problem of global type 1 diabetes

The Diabetes Atlas published in 2007 by the International Diabetes Federation (IDF) provides the most up-to-date statistics of type 1 diabetes in youth under age 14 yr (8). Of the world’s 1.8 billion children in this age-group, approximately 440 000 have type 1 diabetes, representing a prevalence of 0.02%, with about 70 000 new cases diagnosed annually and an average annual increment in incidence of 3%. Figure 3 demonstrates the annual incidence of type 1 diabetes in countries around the world: this varies from <4/100 000 population in much of Asia and Central and South America to >20 in Scandinavia (Finland remains the highest with >50), Canada, the UK, Australasia and Kuwait. The incidence rates in most African countries remain unknown but are likely still quite low. Of note, these incidence rates almost exactly parallel the life expectancy rates at birth of United Nations members states also published in 2007, suggesting that type 1 diabetes is a disorder of modern society (Hygiene versus Accelerator Hypothesis?) (9). The rapid increase in incidence of type 1 diabetes in certain populations suggests the role of environmental factors rather than genetic drift, for example the incidence in Kuwait has increased from <4 to >20/100 000 since the end of the first Gulf War in 1991. The difference in incidence of type 1 diabetes between two genetically identical, but socioeconomically disparate societies, is further testimony to the potential role of environmental factors, for example Finland vs. Russian Karelia (10).

Figure 3.

Incidence rates of type 1 diabetes in children-0–14 years (cases per 100,000 population per year). Diabetes Atlas third edition © International Diabetes Federation 2006.

These changing incidences of diabetes demand that at least part of our research into etiology includes populations at both ends of the incidence spectrum and especially those in whom diabetes becomes much more prevalent when they migrate to higher incidence areas, for example individuals from the Horn of Africa. Furthermore, even though incidence rates in areas such as South-East Asia remain low, the actual number of children with diabetes is considerable, given the large population base. Statistics from the IDF suggest that in many regions of the world, precious little funding is available from local governments for direct support of diabetes care. There is a clear need for training of health care professionals as well as for the affordable availability of insulin and testing equipment in these underserviced areas (2).

The supply of insulin and blood or urine glucose testing equipment into remote areas of the world remains uneven. This is especially problematical in parts of Africa. In an era when limitless supplies of insulin can be produced by recombinant DNA technology, its unavailability to any child with diabetes ought to be intolerable. The barriers to availability include cost, quality of the insulin products being provided, transport to remote locations, lack of critical mass of individuals with diabetes in these regions, and in some cases, lack of political will (2).

It is tragic that in 2008, the most common cause of death in youth with type 1 diabetes globally is simply the lack of access to insulin. In the poorest parts of the world, insulin and testing equipment are so expensive that families must choose between insulin for one child and starvation for the other children or inevitable death of the child with diabetes so that the rest of the family can survive. It has been estimated, for example, that the life expectancy of a child with type 1 diabetes in rural Mozambique is as few as 7 months.

It is the responsibility of governments, international aid agencies, and pharmaceutical companies to ensure that these tragedies be averted. One company, NovoNordisk, has pledged to provide insulin at 20% cost to the 50 poorest nations, of which at least 33 have accessed this program.

Outcomes of children with diabetes from the Ecological Perspectives Model

The following sections highlight some of the factors in each domain of the Ecological Perspectives Model that impact on the outcome of children with type 1 diabetes. This cannot and does not represent an exhaustive listing of all the potential influences on outcome.

Societal factors

The enormous discrepancy between developed and developing countries is starkly defined by the balance between ‘visible’ and ‘hidden’ threats to child health. In much of the developing world, this balance is overwhelmed by poverty, malnutrition, and the multiple infections including HIV/AIDS, tuberculosis, and malaria, while the problems of type 1 diabetes and childhood obesity remain relatively less evident. This is especially true of subSaharan Africa. In developed countries, the new threats of overnutrition and inactivity are much more evident, while underprivilege or disadvantage (neglect, poverty, handicap, and chronic disease) remain an issue in most societies, although relatively hidden in more affluent regions.

How then does one start to address the issues of type 1 diabetes in the developing world? These governments are often overwhelmed with not only the issues of poverty, malnutrition, and widespread infectious disease, but also regional conflicts suck up much of the resources necessary to fund national health care projects. One solution proposed by Professor Francois Bonnici of the University of Cape Town is to ‘identify AND resource individual physicians who have the passion for caring for children with diabetes’ (personal communication). With relatively limited resources, many have been able to establish networks of care and build on what to me are the five fundamental requirements for diabetes care:

  • (i) Availability of food and clean water;
  • (ii) Availability of insulin;
  • (iii) Availability of urine and/or blood testing equipment (the latter is much more expensive in fact than the insulin, even in developed countries);
  • (iv) Prevention of both diabetic ketoacidosis (DKA) and hypoglycemia;
  • (v) Protection against harm.

There is a more than fivefold difference in the incidence rates of DKA at disease onset (from 15–20% in Sweden and Canada to over 70% in parts of Africa and Asia) in different countries around the world (inversely proportional to the incidence of diabetes in that region) and a more than 40-fold difference in DKA-related mortality reported (from 0.15% in Canada to about 4–6% in Sudan and Ethiopia) (11). Campaigns have begun in many countries to raise awareness of diabetes and to prevent DKA. These are largely modeled on the highly successful program in Parma, Italy, where incidence rates of DKA at disease onset fell from 78% to under 12.5% over a 6-yr period (12).

Because many countries are either unable to unwilling to address the issues of diabetes in childhood, a number of volunteer groups have developed to raise funds for supplies and services for these children. For example, Life for a Child, a project of the IDF run by Dr Graham Ogle in Sydney now supports over 900 children with diabetes in 17 countries (13): Azerbaijan, Bolivia, Congo, Ecuador, Fiji, India, Mali, Nepal, Nigeria, Papua New Guinea, The Philippines, Rwanda, Sri Lanka, Sudan, Tanzania, Uzbekistan, and Zimbabwe. This program meets the children’s immediate needs (insulin, syringes, monitoring, and education), builds local capacity, and lobbies governments to establish sustainable solutions. Ron Raab, also Australian, heads Insulin for Life, which is a not-for-profit organization that collects and distributes insulin and other diabetes supplies that would otherwise be wasted (14). These superb humanitarian efforts are deserving of our attention and support.

Community factors

The approach to childhood diabetes has differed from one country to another. In Italy, for example laws enshrine the rights of children with diabetes to excellent care in diabetes centers; in the USA, there is a law ensuring access of children with chronic diseases to safe care in the school system; in the province of Ontario, Canada, recent legislation provides pumps and supplies to those with type 1 diabetes wishing to use this approach to management. Many other examples exist.

In some countries, national surveillance registries exist, for example in Germany and Scotland, to allow diabetes centers to benchmark against one another. In Ontario, Canada, where there is a single payer province-wide health care system, the Network of Ontario Pediatric Diabetes Programs has been established. This includes 5 tertiary and 29 secondary multidisciplinary health care teams that ensure as uniform care as possible to the almost 7000 children in the province with diabetes and the 750–900 who develop type 1 diabetes each year.

Staffing of diabetes centers differs enormously around the world with consensus staffing complements in different countries. In the UK and Sweden, for example, there is one diabetes nurse specialist per 70–100 children with type 1 diabetes, compared to one per 200–250 in the USA and Canada. Information on staffing of diabetes centers in the developing world is largely lacking. Through its Africa Initiative, ISPAD has begun to attempt to address care of childhood diabetes in Africa.

Institutional factors

The Hvidore Study Group has provided convincing data over the past 13 yr that outcomes in children with type 1 diabetes differ between diabetes centers around the world (15–17); however, the reasons for these differences remain elusive. Some of the factors associated with better outcomes (defined by mean A1c levels) include

  • (i) Availability of a 24-h hotline to assist patients and their families with urgent needs;
  • (ii) Clear and lower targets for glucose and A1c;
  • (iii) Greater frequency of visits with the diabetes physician;
  • (iv) Balanced approach to parental involvement: avoidance of extremes of neglect and over-dependence;
  • (v) Children from immigrant (or minority status), lower socioeconomic status, and single-parent families tend to have higher A1cs.

It is likely that institutional attitudes to both the dietary management and the incorporation of physical activity into the treatment regimen also impact on these outcomes. Of note, clinic/center ranking in terms of A1c means has remained relatively stable since inception of the series of studies in 1995. In addition, there has been a close correlation demonstrated between mean A1c levels in the early stages of the diabetes (first 3 yr) and those later in the course (beyond 3 yr). This suggests that there may be an important impact of the initial management (both education and treatment goals) on later outcomes, emphasizing the importance of a good start to diabetes management.

Interpersonal and intrapersonal factors

In this section, the focus is on three factors: attachment theory, quality of life studies, and the profile of youth in chronic poor metabolic control. Ciechanowski et al. (18) have studied attachment theory in adults with diabetes, and, in a two-by-two design, have shown that there is a strong interaction between the patient’s ‘model of self’ and ‘model of other’ with significant impact on A1c levels (Fig. 4). Subjects falling into the ‘Secure’ attachment style had, on average, the lowest A1c levels, those with ‘Dismissive’ the highest, and ‘Preoccupied’ and ‘Fearful’ the intermediate levels. These data underline the importance of the relationship of the individual with diabetes and his/her health care team and other health care providers and family members.

Figure 4.

Attachment style categories and model for self and other [Ciechanowski et al. (18)].

In a large multinational cohort of teens with type 1 diabetes, the Hvidore Study Group demonstrated both a strong relationship between quality-of-life measures and metabolic control (A1c) and a deterioration in the adolescents’ perceived quality of life with progression through the teen years, perhaps more prominent in girls than in boys (16). In contrast, the parents and health care professionals involved in their care perceived a decreasing burden of diabetes at the very same time the adolescents report declining quality of life. This suggests that as parents and health care providers transfer more of the responsibility for self-care to the teens, the teens do not respond to his new responsibility in the most positive way.

Finally, in an ongoing study, teens (more than two-thirds female) with persistently poor metabolic control defined by A1c levels above 10% for more than 6 months (The ‘Over 10’ Program), we have found common intrapersonal and interpersonal barriers to good metabolic control, most often multiple occurring in the same teen (19). Among intrapersonal barriers, mental health issues were present in more than 80% of the teens including weight and shape concern in the majority of the girls (either fullblown or subthreshold eating disorders), depression and/or anxiety in about half, and substance abuse and oppositional behavior in smaller numbers. Fear of hypoglycemia was identified in about 20%, a learning or attention problem in 13%, but no teen had an isolated deficit in knowledge about their diabetes. Interpersonal barriers included inadequate or ineffective parental support in almost all, family systems difficulties in >80%, single-parent families, mental health issues in one or both parents, and financial stress in 30–40% each. This information underlines the extremely strong psychosocial distress in these families. Interventions in these youth have been minimally successful in improving metabolic control into the target ranges.

What are the solutions to diabetes care in the developing world?

The International Insulin Foundation (20) has proposed that responsibility for ensuring good diabetes care globally should be spread between many governmental and non-governmental agencies, organizations, and individuals. The following are the modification of their position: (i) International Donors and Non-Governmental Organizations must recognize childhood diabetes as a health problem in developing countries and allocate and advocate appropriately. This especially applies to increasing insulin and testing equipment availability to those who cannot afford it; (ii) National governments in developing countries should be encouraged to adopt policies to alleviate the financial burden of diabetes in youth; (iii) National governments in developed countries must address the problem of diabetes in developing countries early before it becomes overwhelming; (iv) The private sector should control costs of insulin, other medicines, glucose-monitoring equipment, and syringes for the world’s poorest. This requires humanitarian corporate policies that challenge the ‘bottom lines’ of such companies; and (v) Individuals worldwide should include childhood diabetes in the list of important threats to child (and adult) health globally and advocate with appropriate agencies.

It is clear that the target of the Declaration of Kos, 1993, to address global deficiencies in childhood diabetes care by the year 2000 has fallen far short. Progress has been made on many fronts, diabetes has started to receive the global attention it demands (e.g., United Nations Resolution 2007) (21), standards and guidelines are available (ISPAD 2007/2008) (22), and organizations are supplying insulin and testing equipment to ever more needy children.

The challenges remaining

Since the discovery of insulin in 1922, there have been a number of important milestones in research that have improved the outcomes for children with type 1 diabetes, including refinements in insulin products, development of blood testing strips and meters, measurement of A1c, surveillance for, and early intervention in long-term diabetes-related complications. Multicenter and multinational research groups [e.g., TrialNet and Trial to Reduce Insulin-dependent diabetes in the Genetically at Risk (TRIGR)] are studying interventions to prevent diabetes onset, and the new age of molecular and stem cell science holds unlimited promise of advances still to come that have the potential to revolutionize diabetes care.

Some words of caution, however, are in order here. First, new technologies are invariably more costly and available, at least initially, only to the minority thereby enhancing the ‘gap’ between the haves and the have-nots and the developed and the developing world. Second, we often place unrealistic expectations on the pace at which new medical advances reach clinical application. To quote Roger Kornberg, Nobel Laureate 2006, who, at a recent graduation ceremony at Stanford University, stated that ‘The success of medical science has become, in a way, its undoing. We are dazzled by the knowledge we have acquired and rush to apply it to medical problems. This is understandable, but often premature.’(23). We have only begun to scratch the surface of understanding for example, stem cell biology. Its clinical application remains beyond the horizon for the present.

Summary and conclusions

As one considers the near future of diabetes in children, some very sobering thoughts are in order. First, with respect to the magnitude of the problem, the worst is yet to come! The emergence of new economic giants such as China and India will bring with it potential massive increases in type 1 diabetes (2–5% per annum increases in incidence in the world’s most populous countries) and childhood obesity (with its associated insulin resistance and type 2 diabetes). This will demand the training of enormous numbers of health care professionals and delivery of insulin and testing equipment that is of high quality and reasonable cost. Second, the majority of children with diabetes presently do not and, likely in the foreseeable future, will not achieve and maintain levels of metabolic control that provide protection from microvascular and macrovascular complications. In fact, an analysis by the Centers for Disease Control in the USA (24) suggested that a child of 10 yr of age developing diabetes in the year 2000 would live a further approximately 50–55 yr, losing about 18–20 yr of life, this in the world’s richest nation!

Finally, Edwin Gale in 2005 issued the following warning: ‘The individual and communal legacy of poor glucose control will remain with us for the next 30 years, EVEN if an effective means of preventing new cases of the disease were to be introduced tomorrow’(25).

There is much work to be done and little room for complacency. Only when every single child with diabetes has ready access to experienced health care professionals, insulin, and other supplies, food water and protection will the first part of the job be done.


The author is the RS McLaughlin Foundation Chair in Paediatrics at the University of Toronto and Hospital for Sick Children (SickKids) and a Senior Associate Scientist in the SickKids Research Institute. The contributions of members of the Sickkids Diabetes Team and the Hvidore Study Group to the data presented in this review are greatly appreciated. Nonetheless, the opinions expressed in this manuscript represent those of the author alone. The author is indebted to Dr Kuben Pillay and the Organizing Committee of ISPAD 2008 for the invitation to present this lecture.