Dilemmas and directions in the care of the diabetic teenager: the Arnold Bloom Lecture 1999

Arnold Bloom died on 22 November 1992. He was for many of us one of the guiding lights in diabetes during the 1970s. A warm and kindly man, he had a tremendous understanding of the issues of living with diabetes. His words of encouragement were particularly appreciated in the early 1980s when in the face of much scepticism we set out with an idea that became the Youth Diabetes Project. At that time, there was a clear gap in targeted services for the care of the teenager with diabetes. Health professionals were attempting to come to terms with the special problems encountered in this age group. However, some members of the diabetes team had abandoned even attempting to understand the complexity of the problem. We had heard that a prominent diabetologist once described young people with diabetes as ‘uncommitted, impervious to instruction, cheating and defying advice’. It is important to ask whether we regard this pattern of behaviour in a teenager as normal or abnormal. We, as physicians, may have been partly responsible for creating an air of mystery and confusion about this special group of patients. In many respects, we have turned their management into an academic exercise and taken refuge in somewhat strict rules of diabetes care, rather than listening to clear messages from young people with diabetes themselves. It is likely that we have gone through a period of realisation of our misconceptions. This commentary examines some of the problem areas and introduces the views of young people with diabetes. The ability to listen and understand may provide insight into more appropriate strategies for diabetes care and future research. The problem is of course not great numerically. Prevalence data from Tayside, Scotland1, 2 suggest that most United Kingdom clinics deal with fewer than 200 teenagers with type 1 diabetes. Nonetheless, they are perceived as presenting special difficulties. The first problem for many is the marked deterioration in glycaemic control during this time. Data from our own clinic clearly demonstrate worsening of control in the late teenage years. Most worrying is the sub-group of very poorly controlled teenagers, often with a high proportion of young women, who track within the top quartile of glycated haemoglobin. This is particularly evident if we compare the 75th centiles of glycated haemoglobin (Fig. 1). Perhaps just as important is the clear evidence from our own data and from Pounder in Nottingham3 of settling of glycaemic control as these patients reach their early 20s. Mean and 75th centiles of HbA1c in 175 teenagers with diabetes in 1995 The second problem is that patients who suffer extremes of poor control, or so called ‘brittle diabetes’, are most often in their adolescent years. Robert Tattersall defined ‘brittle diabetes’ to exist in one whose ‘life is constantly disrupted by episodes of hypo or hyperglycaemia whatever their cause’4. One such patient within our own clinic was hospitalised for eight months during one year as a result of recurrent diabetic ketoacidosis. Like so many others, she encountered difficulties over vascular access and during this period had a dramatically large number of unnecessary investigations and questionable therapeutic interventions. We discovered clear evidence that she was self-puncturing tunnelled Hickman lines using an autolet stylo, and subsequently was found to be injecting presumed isophane insulin from a vial that contained almost pure distilled water5. Since Tattersall's original description, investigation by Geoff Gill and George Alberti in Newcastle and David Schade in the University of New Mexico have implicated the psycho-social/factitious nature of the ‘brittle’ diabetes problem6, 7. Nonetheless, many professionals involved in the care of the teenager with type 1 diabetes are still pre-occupied that both ‘brittle diabetes’ and ‘chronic poor control’ are problems caused by a biochemical defect and that sophisticated methods of insulin delivery are the therapeutic answer8. The results of the DCCT remind us of the impact that poor control during adolescence may have in causing complications in the longer term. Although only 18% of subjects in the study were teenagers, it is worth emphasising that the mean length of follow-up was only six years, in other words equivalent to the duration of teenage life9. Furthermore, in addition to this evidence of a direct relationship between control and overt microvascular complications, there are now data showing sub-clinical abnormalities of endothelial function during teenage diabetic life long before the development of clinical complications10. The advent of ‘pen’ injection devices and more convenient injection equipment during the 1980s led to a move to more intensified insulin regimens as a theoretical solution to poor control. Although, a number of studies have recognised the greater flexibility conferred by such devices, there are no data to suggest that changes in equipment have ameliorated the problem of glycaemic chaos during teenage years. For example, Henrik Mortenson in Denmark studied three different insulin requirements in children and teenagers over a two year period: (i) twice daily split and mixed regimen; (ii) intensified regimen using four times daily injection and (iii) twice daily insulin but changed to intensified regime during the study period. There were no differences in glycated haemoglobin between the study groups after two years; in all groups the control deteriorated and weight increased11. More blood glucose testing is likewise cited as a potential solution to our control problems. However, on ‘self-report data’ there is no evidence that blood testing leads to better glycaemic control12. Furthermore, most doctors are aware that ‘the book of blood test results can create conflict during the medical consultation process’. Many doctors sit judgementally mulling through blood test results making meaningless suggestions for changes in insulin dosage. More objective data from the DARTS database in Tayside has looked at the encashment of pharmacy scripts for blood testing strips. This suggests that over an 18 month period only 50% of the teenage group ever even obtain testing strips. Interestingly, the same applies to those patients on insulin in their 20s13. However, in those who are obtaining blood testing strips regularly there is a clear correlation between glycaemic control and frequency of testing. We must continue to take an extremely positive view of the need for home blood glucose monitoring but be mindful of the reality if we are to avoid becoming over-judgemental of those who are not testing! Increasing preoccupation with education about blood testing and self-care is not the sole answer to the problems of the teenager with diabetes. There is a danger that it is a ‘let-out’ that at least we appear to have done something professionally in circumstances where we should be looking at alternative support strategies. Jim Farquhar, formerly Professor of Paediatrics at the University of Edinburgh, died on 6 June 1998. Jim was by any standards a visionary in the care of the adolescent with type 1 diabetes. In the early 1980s when there was clear evidence of a gap in the provision of care for the teenager with diabetes and growing strife between paediatricians and adult diabetologists as to who should best care for this age group, Jim had the foresight to propose a novel model of care, which directly addressed the special needs of the adolescent with diabetes. This model still forms the framework for our approach today. In 1983 the first annual Firbush Course for Young People with Diabetes took place under the auspices of the Youth Diabetes Project14, and two years later the first Conference for Young Diabetics took place in Birmingham15. At the same time, the Scottish Study Group for the Care of the Young Diabetic was formed with the clear philosophy of collaboration between those paediatricians and adult diabetologists with an interest in the care of the young diabetic. This year will be the 18th Consecutive Firbush Young Diabetic Course by Loch Tay in Central Scotland. The original philosophy was to encourage young people with diabetes to come together to share experiences of diabetic life. At a personal level, young people were able to gain confidence to participate in outdoor activities and push themselves to the limit in a protected environment. It was believed that bonding processes and the sharing of experiences would lead to development of both social and self-management skills in diabetes. However, over the past 17 years the educational opportunity afforded not only for people with diabetes but also for professionals has become apparent. The opportunities for carers to live with people with diabetes is now fundamental to the philosophy of Firbush; for example, the opportunity to discuss the problems of living with diabetes in a protected environment and the chance to speak with young people openly has provided major clues to the reasons for control difficulties during teenage years. Young people have repeatedly challenged: ‘doctors just study us… they should listen and understand!’. Firbush has coincided with new avenues of research that have focused on psycho-social factors that are associated with poor control. These include anxiety, confidence, locus of control, depression, empowerment, education and more recently self-efficacy in problem solving16. We believe we had identified the clear behaviour patterns that appear to be important determinants of biomedical outcomes, for example, the clear association of alcohol intake as a cause of diabetic ketacidosis17 and ‘sexual abuse’ as an association with chronic poor control in the teenage girl. More fundamentally the self-care issues of ‘eating problems’, ‘abnormal injection treatment behaviour’ and ‘fear of hypoglycaemia’ are all important if we are to seek new directions for the future. Evidence from the Young Diabetes Conference in 1987 indicated that 71% of young people with type 1 diabetes ‘binge’ eat, and that this is often associated more with feelings of extreme guilt15. In addition, UK data have demonstrated clear associations between formal clinical eating disorders and type 1 diabetes in teenage girls18, which have important long term implications. More recently, Rydall and colleagues in Toronto19 demonstrated a striking association between established eating disorders in diabetes and diabetic retinopathy. Most of us who are involved in clinics for young people are aware how important dietary compliance is as a factor in poor diabetic control. Even more intriguing is the re-emergence of interest in abnormalities of gastric motility in relation to chronic hyperglycaemia. It is clear that abnormalities of gastric emptying do not necessarily imply irreversible autonomic neuropathy but may be transient in relation to hyperglycaemia20, 21. This may in turn contribute to poor control by causing mismatch of insulin action and absorption of carbohydrate. Furthermore, are our own assumptions about high prevalence of bulimia and eating disorders in our patients correct or should we question whether some recurrent vomiting and gastrointestinal symptomatology not only relates to chronic hyperglycaemia but may in turn contribute to poor control—a ‘vicious cycle’? It is important, however, to recognise that such eating abnormalities are common in the non-diabetic, and that it must be regarded as almost normal teenage behaviour. For example, data suggest 79% ‘binge’, 70% ‘consider themselves fat’ and ‘half of these will be actively dieting’, 15% ‘meet the diagnostic criteria for bulimia’ and ‘as many as 11% may induce vomiting’22, 23. Perhaps the more significant clue from discussions with young diabetic people as to the aetiology of poor diabetic control is abnormal insulin treatment behaviour. Discussions over several years have highlighted that most young people with type 1 diabetes deliberately miss insulin injections at some stage. In some cases the aim is to lose weight, but other reasons include experimentation, trial as a manipulative gesture or occasionally needle phobia. Intermittent use of insulin is much more a reality than was ever believed previously. Some indirect evidence of this came from a study performance in Dundee24, which looked at the clinical characteristics of 122 consecutive episodes of proven diabetic ketoacidosis. The young adults with diabetic ketoacidosis had significantly worse glycaemic control, much lower plasma glucose concentrations at presentation, shorter time to recommencing subcutaneous insulin following admission, fewer infections and markedly reduced duration of hospital stay. Most important, however, was the likelihood of finding a proven precipitating cause. In the vast majority of those under 25, there was no identifiable cause apart from previous high alcohol intake or the likelihood of insulin withdrawal24. From the Youth Diabetes Project, it was possible to hypothesise that, for many patients who are chronically poorly controlled, insulin is given only when the patient is symptomatically intolerably hyperglycaemic. There are long spells of missed injections and occasionally delay in intervening and responding to hyperglycaemia may precipitate a rapid progression to diabetic ketoacidosis. The DARTS Collaboration in Tayside has now provided much more direct evidence for this hypothesis25. In a study of 89 young patients with type 1 diabetes the recommended insulin dose according to the clinic was compared with the amount of insulin obtained at community pharmacies over an 18 month period. Insulin administered during hospital admission was also taken into account. An ‘adherence index’ was then calculated, which reflected the total days in insulin covered in one year. Unsurprisingly, 28% of these patients obtained less than the prescribed dose of insulin, but, more significantly, their mean deficit was 115±68 days per annum, i.e., at least a quarter of patients missed the equivalent of four months of insulin per year. There was, in addition, an inverse relationship between the adherence index and glycated haemoglobin, and 90% of patients who suffered episodes of diabetic ketoacidosis obtained insufficient insulin to account for their daily dose. Most important of all, in those patients in the top quartile of HbA1c (>10%) the main insulin coverage was significantly less than 365 days (Fig. 2). Clearly, those patients with glycated haemoglobin level in the high range are omitting insulin. If the adherence index is correlated with five year age bands in the clinic, the non-adherent group are predominantly in the 15–20 year age band (Fig. 3). Correlation of glycaemic control (HbA1c) with adherence (days of insulin supplied per annum) in 89 subjects with type 1 diabetes. •, hospital admission for diabetic ketoacidosis. Reproduced with permission from the Lancet25 Morris AD, Boyle DIR, McMahon AD, Greene SA, MacDonald TM, Newton RW. Adherence to insulin therapy, glycaemic control and ketoacidosis in insulin-dependent diabetes. Lancet 1997; 350: 2505–1510. Relationship between age, glycaemic control (lower panel) and adherence to insulin (upper panel). Reproduced with permission from the Lancet25 Morris AD, Boyle DIR, McMahon AD, Greene SA, MacDonald TM, Newton RW. Adherence to insulin therapy, glycaemic control and ketoacidosis in insulin-dependent diabetes. Lancet 1997; 350: 2505–1510. These data clearly show that omission of insulin is a major determinant of glycaemic control. It is important to consider how to use this information in our clinical practice. Does it become a further interface for conflict, accusation and judgement, or should it be more reasonably regarded as normal behaviour? It is tempting to speculate that, in the very poorly controlled group, the simplest possible insulin regime to achieve ‘cover’ would be preferable to an intensive four times daily regime. Even a insulin may be appropriate for who insulin. is that, rather than that more education is for this group, alternative strategies of support and encouragement should be has been a significant problem in diabetes at We studied the requirements and use of insulin in subjects between and We in insulin of at least on the first of outdoor which a with the there were and episodes in the most not have the confidence to their insulin by The main for one was only The to on two in a significant to only the in the daily dose of insulin by the of the was between and 1). In a of less than in the total daily dose and a large insulin dose were clear of in this environment of We believe that directly to exercise is regarded as less of a problem by teenagers it is at the of insulin action and is less In that it is worth in that may be regarded as such exercise and data from the Youth Diabetes Conference has indicated that as many as one in six young people with diabetes have had during The of teenagers is or early the following It often after but the with exercise to be and and not by alcohol all episodes of which medical between and 1995 at Firbush after or during the following The DCCT is regarded as a study by all care professionals in diabetes. In we the messages of the DCCT to the teenagers with diabetes in our We a which in terms of the association between control, in complications and also the in in the control group. In to the question would about blood at lower to the majority of subjects were most about more and more were about of their and the majority about possible weight Furthermore, the demonstrated that at least 50% of patients who are at the level and only to control at the from the Youth Diabetes Conference that as many as one in five teenagers with diabetes are of and their insulin in to avoid this The adult diabetes view of the of care of the teenager is of a with and issues of and and the of Furthermore, the of the adult diabetologist was that such patients were only at a time that diabetic control was the other the view of adult diabetes is an with for complications and a care that a of environment and in the early of that all of these problems are important, care of this age group should not be the for a adult diabetologist There is a for a philosophy of collaboration between the and adult services to care of the The views of young people to other important issues in their and are It is that professionals have an insight into the problems of living with type 1 diabetes. a long term and relationship can be is This may have a major impact in clinic with approach on and of patients has us about the care in our We have recently studied a large number of young people and it has us much about the consultation In the of has been and leads to a long term An of the of such is the control’ often by less clinic more has a greater in control and clinical further We have further of the of the clinic environment. A clinic is as but this should not mean chaos in terms of Young people with diabetes more able to be of a and less to We have a in the of clinical However, in our with for complications we must take into account a to about the future and the for this group of patients. A of follow-up and of the is it is clearly it may be in high in the long term. long as study in of was regarded as one of the best of in people with type 1 The views of young people with diabetes that we have been to have been in many of the problems for this age group but they do us with we with different and of type 1 some professionals more are we with same in different For treatment and control difficulties are to psycho-social and For it may self-care an behaviour pattern during Perhaps the is that diabetes care is clearly a high for it is often not so for the teenager with diabetes. It is to poorly controlled teenagers and recognise their special problems. We openly the problems of eating but more particularly that of abnormal insulin treatment behaviour and insulin injection as fundamental to the problem. We believe this to be the for an for the clinical of professionals by patients will only be if a consultation of blood of and insulin dose is For a it may be important to the insulin at least to achieve some insulin Furthermore, we are to at new strategies for problem by Most important is that we should clinical we reach the time of the more 20s. In we of the young people in our Diabetes would most to The answer was as as We would to all of the medical who have the Youth Diabetes but particularly our colleagues who have had such a fundamental in guiding and care the of young people with diabetes has been most important of

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.