Screening for type 1 diabetes: are we nearly there yet?

• Users may freely distribute the URL that is used to identify this publication. • Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. • User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) • Users may not further distribute the material nor use it for the purposes of commercial gain.

symptoms [8]. Currently, these investigations are used for research purposes, to understand the natural history and development of type 1 diabetes, and to identify individuals in whom prevention therapies can be tested.
However, cases like those of Peter Baldwin, and the significant proportion of children who present with DKA at diagnosis, make us question whether these screening tests should be rolled out to the general population. This is an emotive topic and one we should explore carefully. The current thinking on this subject suggests we should not be routinely screening for type 1 diabetes because there is no therapy currently proven to prevent or significantly delay the onset of this condition. This is supported by guidelines from respected authorities in this area [9]. However, this does not consider the benefits associated with early detection of type 1 diabetes, not least of which is prevention of death by DKA.
So, what are the benefits of screening for type 1 diabetes? A number of research studies have now outlined what these may be (Table 1). These research studies have either screened the general population or been more focused and screened people deemed at risk because they have a family member with the disease. Screening has been undertaken through genetic and autoantibody tests followed by glucose tolerance tests for those deemed at risk. To date, screening studies have largely studied children and not explored adults (the age group in which almost 40% of type 1 diabetes presents). The benefit most consistently reported across these studies is avoidance of DKA. This is a significant benefit, with a reduction from a quarter of new type 1 diabetes being diagnosed in DKA down to 3%. Also, presumably because they are diagnosed at an earlier stage of disease, there is more residual beta cell function (as measured by C-peptide), lower insulin requirements and lower HbA 1c at the time of diagnosis [10][11][12][13][14][15]. The paper by Lundgren and colleagues [4] adds to the evidence. They report from follow-up of the Diabetes Prediction in Skane (DiPiS) study, where almost 40,000 children in southern Sweden were screened for type 1 diabetes. Of these, 6000 were deemed to be at some degree of risk and offered  The psychological consequences of awareness of high risk of a chronic disease for which there is no cure is not known The natural history of the condition, including development from latent to declared disease, should be adequately understood.

Not known
Natural history remains to be fully elucidated, different rates of progression remain to be understood. Age, ethnicity and environment appear to influence natural history and these effects remain to be elucidated There should be an agreed policy on whom to treat as patients

Yes
People fulfilling standard WHO criteria for diabetes will be treated as diabetic The cost of case-finding should be economically balanced in relation to possible expenditure on medical care as a whole Not known Case-finding should be a continuing process and not a 'once and for all' project Yes A long-term programme can be implemented nationally follow-up. About two-thirds accepted this offer of follow-up. Lundgren and colleagues report on the outcome of the 51 children who developed type 1 diabetes in the follow-up group, compared with the 78 who developed type 1 diabetes but had not accepted the offer of follow-up. Children who chose follow-up had a lower frequency of DKA (2% vs 18%), and lower HbA 1c (9 mmol/mol [0.8%] lower) at diagnosis. Importantly, HbA 1c remained significantly better up to 5 years after diabetes diagnosis. A potential caveat is that a greater proportion of the participants who opted for follow-up had mothers of Swedish origin, and a greater engagement in research may reflect a propensity for greater involvement in diabetes care following diagnosis. That said, this is the first time that such a prolonged HbA 1c benefit has been reported, and this HbA 1c benefit has a clear clinical and economic benefit. Furthermore, since HbA 1c generally tends to rise in the first years after diagnosis, and then stabilise and 'track' after about 5 years [16], we could postulate that the lower HbA 1c levels in the follow-up group persist over the long term. Do these benefits make a workable case for screening? There are those who would put forward a scientific argument to support this case [17,18]. To provide some clarity, Table 2 presents the benefits of screening against the WHO guidelines for screening, originally proposed almost 50 years ago [19]. Here, the benefits relate to early detection, and not to prevention of the disease. There are a number of criteria that remain to be satisfied.
First, we do not know the psychological consequences of alerting a person to a condition for which there is no current cure. Granted, this time may usefully be spent in educating and preparing the person for managing type 1 diabetes. However, concurrent work on the psychological impact of informing patients of high risk of type 2 diabetes suggests that, even with a condition that can be significantly delayed, there can be a negative psychological impact. These include negative markers of mental health, reduced motivation and lack of engagement with behaviour change [20].
Second, the natural history of type 1 diabetes is still not clearly understood, and the influence of age, ethnicity and environment remain to be elucidated. The environment may influence the rate of development, as evidenced by migration studies, where populations migrating from areas of low incidence to high appear to adopt the risk of the host population [21]. Whilst we previously believed that rates of beta cell loss were faster at a younger age, more recent work suggests that the rate may remain the same across the age spectrum [22], despite islet histology changing with age of presentation [23]. Importantly, all major screening studies to date have focused on children.
Lastly, the cost benefit needs to be established for a formal screening programme. Since 2015 the Bavarian Fr1da study has been screening children aged 3-4 years for type 1 diabetes [24]. The aim is to screen 200,000 children, with each screening roughly costing 20 Euros per child. If DKA and hospitalisation is prevented in 200 children, this cost saving will in itself cover a third of the cost of the study. Furthermore, patients presenting with DKA tend to have an HbA 1c that is up to 1.4% higher than those who do not over the long term [3]. If the lower HbA 1c reported by Lundgren et al persists to reduce the incidence and economic impact of diabetic complications, combined with the saving on DKA cost, we may be a significant way to covering the economic cost of screening. Further work is required in this area.
Until some of the issues above are resolved (and there is significant work ongoing) the way forward is in public and healthcare education, and raising awareness. As a direct result of the efforts of Peter Baldwin's family, the Welsh government recently (October 2018) discussed ten recommendations around raising awareness of type 1 diabetes. These recommendations include adopting the Diabetes UK 4Ts campaign [25], and a recommendation that all cases diagnosed in DKA are reviewed for shared learning [26]. These measures are critical if we are going to make a meaningful change to the devastation caused by death by DKA.