14

experiences, Meade and colleagues (1996) developed the Blood Donation Reactions Inventory (BDRI) through which donors rate the severity of the VVR-type symptoms they experience. BDRI scores significantly positively correlate with phlebotomist ratings of VVRs (Meade et al. 1996), but with larger numbers of donors reporting VVR symptoms than typically seen in official incidence rates. Between 14-38% of first-time WB donors self-report vasovagal symptoms using the BDRI (Masser et al. 2013). Further, the BDRI outperforms phlebotomist ratings in predicting donor return (France et al. 2004), suggesting donors’ perceptions of the reactions they experience are a more important determinant of subsequent behaviour than official incidence rates. Regardless of how they are assessed, VVRs impact negatively on donor safety and future donation behaviour. From an operational perspective, VVRs also have a negative impact on donor wait times, the management of appointments, and the number of completed blood collections (France This article is protected by copyright. All rights reserved. 4 et al. 2004; France et al. 2005; van Dongen et al. 2013). As a consequence of this, VVRs are important to BCAs and are the focus of a significant body of donor research. In this narrative review, we consider what is known from the literature with regard to risk factors for VVR and prevention methods. Further, we will also consider how donors and staff manage these types of reactions and how VVRs impact donor retention. Risk factors for vasovagal reactions A multitude of factors are linked with a heightened risk of experiencing a VVR during or after blood donation. These can be broadly divided into three categories – donor characteristics that are generally observable (e.g., gender, ethnicity), donor characteristics that may not be immediately observable without additional questioning or assessment (e.g., prior night sleep duration, fear of needles), and contextual features of the donation experience (e.g., wait time, phlebotomist experience; see Table 2). Generally observable donor characteristics At the univariate level, an array of observable or easily calculable donor characteristics are associated with an increased risk of experiencing a VVR. Age is frequently implicated, with younger WB donors more likely to experience a VVR than older donors (Ogata et al. 1980; Tomita et al. 2002; France et al. 2005; Zervou et al. 2005; Eder et al. 2008; Ditto et al. 2012), as is first-time donor status (vs. more experienced donors; Ogata et al. 1980; Kasprisin et al. 1992; Meade et al. 1996; Newman 2003; France et al. 2005; Zervou et al. 2005; Eder et al. 2008; France et al. 2013), This article is protected by copyright. All rights reserved. 5 and gender, with women being more likely to experience a VVR than men (Ogata et al. 1980; Tomita et al. 2002; France et al. 2005; Eder et al. 2008; Ditto et al. 2012; Takanashi et al. 2012; Vossbeck-Elsebusch and Gerlach 2012; France et al. 2013). Other analyses report higher rates of VVRs among White donors (vs. non-White donors; Newman 2002; France et al. 2005, but see also Newman et al. 2003), lower BMI/weight (vs. higher BMI/weight; Ogata et al. 1980; Kasprisin et al. 1992; Newman 2003; France et al. 2013), and lower estimated blood volume (vs. higher, relative to the proportion being collected; Tomita et al. 2002; France et al. 2013). Of all of these, age, firsttime donor status, and gender are routinely identified as risk factors, with many large-scale analyses evidencing independent effects of these characteristics. This occurs even when the shared variance between these univariate predictors is controlled for in multivariate analyses that consider these factors (and others) in conjunction with each other. For example, in an analysis of 1,776,445 collections from BCAs in the US, Eder and colleagues (2008) showed that 16-17-year-old donors were significantly more likely to experience an officially recorded VVR than 18-19-year-old donors or donors aged over 20 years. Further, in the same analysis, donor status (first-time vs. more experienced) and gender (being female vs. male) were also implicated as significant independent predictors of VVRs. For other characteristics, the evidence base is either less comprehensive or compelling (e.g., there are only a small number of studies that have considered ethnicity as a predictor of VVRs, and conflicting results have emerged within these analyses; e.g., Newman 2002, France et al. 2005, in comparison to Newman et al. 2003). For example, Zervou and colleagues (2005) found an association of lower body weight with officially recorded VVRs only among male donors, whereas Tomita et al. (2002) found the association of circulating blood This article is protected by copyright. All rights reserved. 6 volume and officially recorded VVRs stronger for older

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19