Prescribing in older people with dementia is, it is notable that large numbers of older people with dementia continueto be treated with antipsychotics. Such prescribing is often in response to the need to manage distressing behavioural and psychological disturbance. Given the lack of highly effective alternative treatments, the correct level of antipsychotic use in this population is unlikely to be zero, although it is almost certainly less than current levels in Scotland [13,31].Implications of the FindingsThis study provides further evidence that risk communications from regulators do change clinical practice, although it raises important questions about how such risk communications should best be designed and disseminated [1]. Although an observational study cannot definitively ascribe causality, we believe that the 2004 risk communication was associated with a large change in prescribing, including a large initial impact most likely because it prompted widescale review of 1527786 patients already being prescribed antipsychotics. While the 2009 risk communication was not associated with any immediate change in prescribing, the rates of prescribing subsequently fell (although whether this is due to the risk communication or other publications and policy activity around the same time cannot be determined in an analysis with pre-specified interventions to examine). The limited dissemination of the 2009 risk communication is of particular note. There were seven other risk 18204824 communications sent Title Loaded From File directly to healthcare professionals in the UK in the first quarter of 2009, relating to efavirenz, temsirolimus, toremifine, bevacizumab, efalizumab, recombinant coagulation factor VIII, and fondaparinux [32]. Additionally, there was a large campaign to publicise new advice about the risks of over the counter preparations for colds andRisk Communications and Antipsychotic Prescribingcoughs in children [33]. Despite being more urgently and more widely disseminated, none of these were associated with the scale of harm of antipsychotics in dementia, which in 2009 were estimated to kill approximately 1800 people in the UK, and to additionally cause a further 1620 cerebrovascular events [13]. We therefore believe that the effectiveness of regulatory risk communications could be improved by better Title Loaded From File attention to the content and method of dissemination of risk communications, tailored to the level of risk and harm involved. Based on the larger observed changes in prescribing associated with the 2004 risk communication, and the smaller observed changes in 2009 despite multiple policy publications as well as the risk communication, it is also plausible that a clear and authoritative recommendation to review patients with dementia prescribed antipsychotics disseminated directly to prescribers would lead to wide-scale, targeted review and significant implementation of the guidance to stop antipsychotics wherever possible.marked differences in their design and method of dissemination. Although impact in this kind of real-world intervention is likely to vary with context, the findings are consistent with previous research examining the impact of FDA risk communications in terms of impact varying with the design of the warning, and in particular of warnings clearly specifying the actions expected of prescribers having greater impact [1]. There is a need for applied research to systematically examine why impact varies in order to understand how better to design and dissemin.Prescribing in older people with dementia is, it is notable that large numbers of older people with dementia continueto be treated with antipsychotics. Such prescribing is often in response to the need to manage distressing behavioural and psychological disturbance. Given the lack of highly effective alternative treatments, the correct level of antipsychotic use in this population is unlikely to be zero, although it is almost certainly less than current levels in Scotland [13,31].Implications of the FindingsThis study provides further evidence that risk communications from regulators do change clinical practice, although it raises important questions about how such risk communications should best be designed and disseminated [1]. Although an observational study cannot definitively ascribe causality, we believe that the 2004 risk communication was associated with a large change in prescribing, including a large initial impact most likely because it prompted widescale review of 1527786 patients already being prescribed antipsychotics. While the 2009 risk communication was not associated with any immediate change in prescribing, the rates of prescribing subsequently fell (although whether this is due to the risk communication or other publications and policy activity around the same time cannot be determined in an analysis with pre-specified interventions to examine). The limited dissemination of the 2009 risk communication is of particular note. There were seven other risk 18204824 communications sent directly to healthcare professionals in the UK in the first quarter of 2009, relating to efavirenz, temsirolimus, toremifine, bevacizumab, efalizumab, recombinant coagulation factor VIII, and fondaparinux [32]. Additionally, there was a large campaign to publicise new advice about the risks of over the counter preparations for colds andRisk Communications and Antipsychotic Prescribingcoughs in children [33]. Despite being more urgently and more widely disseminated, none of these were associated with the scale of harm of antipsychotics in dementia, which in 2009 were estimated to kill approximately 1800 people in the UK, and to additionally cause a further 1620 cerebrovascular events [13]. We therefore believe that the effectiveness of regulatory risk communications could be improved by better attention to the content and method of dissemination of risk communications, tailored to the level of risk and harm involved. Based on the larger observed changes in prescribing associated with the 2004 risk communication, and the smaller observed changes in 2009 despite multiple policy publications as well as the risk communication, it is also plausible that a clear and authoritative recommendation to review patients with dementia prescribed antipsychotics disseminated directly to prescribers would lead to wide-scale, targeted review and significant implementation of the guidance to stop antipsychotics wherever possible.marked differences in their design and method of dissemination. Although impact in this kind of real-world intervention is likely to vary with context, the findings are consistent with previous research examining the impact of FDA risk communications in terms of impact varying with the design of the warning, and in particular of warnings clearly specifying the actions expected of prescribers having greater impact [1]. There is a need for applied research to systematically examine why impact varies in order to understand how better to design and dissemin.
