The underlying risk of MI is continuously changing as a result of many factors influencing particular risk components (e.g. lipid-lowering treatment, diagnosis of diabetes or smoking cessation) and NNH values should not be considered as constant [23,24]. In addition, a delay in the onset of an adverse event may occur after exposure and NNH is not able to capture this effect [41]. Therefore, the most see more appropriate approach would be to assess patients’ risk on a regular basis, according to current guidelines for care of HIV-1-infected patients [42], along with repeated

adjustments for the NNH. Risk assessment should also be made available for patients’ use in terms of communicating risk and increasing adherence to risk-lowering interventions. To facilitate this, an appropriate tool will be made available publicly at the Copenhagen HIV Programme webpage (http://www.cphiv.dk/TOOLS.aspx). With increasing duration of antiretroviral learn more treatment and aging of the HIV-1-infected population, more adverse effects can be observed. It is therefore of great importance to develop methods that incorporate

this information into daily practice. The use of NNH, as presented in this paper, could have a positive impact on patients’ health, as we describe an increase in the NNH with simple lifestyle and/or medical interventions [43–45]. Conclusions regarding the long-term safety and efficacy of antiretrovirals should be drawn based on both clinical trials, typically of a shorter duration, and observational studies, with many years of follow-up [30,46,47]. The development of understandable methods for patients also applies the principles of good clinical practice in terms of delivering informed consent with regard to the treatment offered [48,49]. There are a number of limitations of our study which should be taken into consideration. Firstly, the potential harm of ID-8 the treatment must be weighed against its benefit, which has

not been presented here [12,23]. For the majority of HIV-infected patients, the benefits of antiretroviral treatment far outweigh the potential harm [50,51], which should be taken into account in clinical decision-making [46]. Secondly, the parametric model developed by Anderson et al. [25] used here to determine the underlying risk of MI reflected the Framingham study characteristics, which may be different from those of HIV-1-infected patients. Comparisons of predicted and observed rates of MI in HIV-infected populations suggest that the Anderson equation may overestimate the rate of MI in patients unexposed to antiretrovirals and underestimate it in those exposed to antiretrovirals [52]. Work is ongoing to develop a cardiovascular risk equation for HIV-infected persons, which will address this issue [53].