LAUSR

When I was a boy, I used to play a strategy board game called ‘Risk’, the premise of which was to build up an army to conquer the world. The element of risk came through deciding where to place your armies to ensure that there was sufficient defence to hold territory while having enough to capture other squares on the board, with the results determined by a throw of the dice. The game required strategy, tactics and negotiation skills—skills that are useful in clinical diabetes practice. Risk is a well-known concept in diabetes research and clinical care; indeed, the word ‘risk’ appears in the title of nine articles in this month’s issue. In most of these articles, the authors have described risk factors for the development of diabetes [1,2], its complications [3–6] or treatment side effects [7]. By contrast, Rasouli et al. [8] showed that snus (smokeless tobacco or moist snuff) was not a risk factor for Type 2 diabetes or latent autoimmune diabetes of adulthood, suggesting that the reason why smokers are at increased risk of diabetes may not be attributed to nicotine, but to other substances in tobacco smoke. The last of the articles describes how an inexpensive virtual assistance-based lifestyle intervention can reduce diabetes risk factors [9]. Despite risk being in common parlance in clinical care, people are not very good at understanding or communicating risk. In an old but interesting study of how the presentation of information about risk can influence a response, Misselbrook and Armstrong presented data from the 1985 MRC Mild Hypertension Trial to 102 people with hypertension and 207 people without in four different ways and asked about the likelihood that they would then take antihypertensive treatment [10]. When told that treatment would reduce the risk of stroke by 45% (relative risk reduction), 92% of respondents said that they would take treatment, compared with 75% if they were told that their risk was reduced from one in 400 to one in 700 (absolute risk reduction). A total of 68% would take treatment if the number needed to treat was 35 over 25 years (number-needed-to-treat model), but only 44% would take antihypertensive medication if only 3% taking the treatment would benefit (personal probability of benefit from treatment model). Although this study is old, the issues are still pertinent because an understanding of personal risk affects health behaviour. In their systematic review of the perception of experiencing diabetes-related complications, Rouyard et al. [4] found evidence of low risk awareness in most of the dimensions measured and the existence of optimistic bias: ‘it’s not going to happen to me’. Inaccurate risk perceptions are a major barrier to the adoption of self-care behaviours and, as a result, become an additional risk for poor diabetes outcomes per se. People who underestimate their risks appear less likely to embrace recommended behaviours and so a deeper understanding of who is most affected and how to correct these misperceptions is vital if we are to support behaviour change and self-management of diabetes.