Whole-blood donors provide around 110 million donations worldwide for transfusions used in the management of general surgery and severe trauma, high-dose chemotherapy regimens, and stem cell and organ transplantation.^1,2 The NHS Blood and Transplant (NHSBT) service is an essential part of the health service in England, and in 2015–16 issued 1.594 million units of red cells at a production cost of approximately £160M.³ The overall demand for red blood cells for the NHS has fallen by about 17% since 2004–5, due partly to advances in technology and education of health-care professionals, which have reduced red cell use per surgical procedure.³ The NHSBT has managed the required reduction in the overall supply of whole blood; between 2012 and 2016 the number of registered whole-blood donors fell by around 15%⁴ (C Wickenden, NHSBT, 2016, personal correspondence). There is interest in identifying cost-effective changes to blood collection services that can be implemented in mobile (temporary) blood collection venues, where around 85% of whole-blood donors currently donate in England, but also at static (permanent) blood collection centres. A future challenge for NHSBT is to maintain the requisite supply of whole blood not just overall, but according to particular donor subgroups.

Demand for some blood types, in particular O negative (O–) (the universal blood type), A negative (A–), B negative (B–) and other rare blood types more common in black and Asian minority ethnic (BAME) donors, is increasing. A major concern is to ensure sufficient supply of those blood types that are in relatively high demand, and there may be times when increased stocks are required to meet higher demand. The O– blood type is essential for those transfusions required by emergency trauma patients if blood types are unmatched, and demand for this blood type accounts for around 13% of all hospital requests.⁵ However, only around 7% of the donor population are blood type O–, and so the NHSBT encourages these donors to donate blood more often. A further concern is that patients with genetic blood disorders such as sickle cell disease and thalassaemia disease require multiple transfusions with extensively matched blood to reduce adverse reactions. These transfusions require blood subtypes, such as Ro, which are relatively prevalent in donors from BAME ethnic groups, but BAME donors represent only 5% of the overall donor population. If the required blood types are unavailable, O– blood is used instead, further increasing demand for this blood type. Hence, a key policy objective for the NHSBT is to collect more blood from BAME groups and those with high-demand blood types such as O–.³

The NHSBT could increase investment in marketing strategies to attract new donors, but finding and retaining donors is costly, and it is more efficient to increase donation frequency among existing donors, in particular those whose blood type is in high demand.⁶

A key challenge for the blood service is therefore to develop strategies that can increase the frequency of donation from those donors whose blood type is in relatively high demand, at low additional cost. A central pillar of the 2013–17 Blood Donation Strategy⁴ is to improve the experience of the 1.3 million registered voluntary blood donors to help encourage existing donors to donate whole blood at the requisite frequency. The NHSBT has invested in strategies to improve the donation experience; for example, donors can book appointments online, the 24 static donor centres across England offer free Wi-Fi, and the NHSBT sends text messages to remind donors about their appointment and then to say when and where the donation has been used. However, it is unclear whether or not these and other future changes to the blood service are clinically effective or cost-effective.

The NHSBT surveys donors to help understand which aspects of the donor experience warrant improvement, but it does not elicit donor preferences using the formal techniques that are required to recognise any trade-offs between the different aspects of the service (e.g. additional travel time to a donor centre vs. improved appointment availability). Hence, these surveys do not provide an adequate basis for predicting the effects of potential service changes on the frequency of whole-blood donation. More generally, there is little evidence on donors’ relative preferences for alternative types of blood donation service.^7–12 The extant literature suggests that donors prefer shorter waiting times before whole-blood donation and convenient locations,^13,14 and that non-monetary incentives are more effective than monetary incentives for encouraging blood donation.^15–17 None of these studies have used appropriate formal techniques for preference elicitation, evaluated strategies of direct relevance to the NHSBT or considered subgroups of current policy relevance, such as donors whose blood type is in high demand, BAME donors or those donor subgroups who are less likely to continue donating (younger or less experienced donors).

Economic evaluations of alternative ways of organising the blood service are required, ones that recognise that donation frequency will be driven by donors’ experiences of, and preferences for, alternative features of a blood donation service. Such studies are also required to recognise that donors’ relative preferences for alternative strategies may differ according to the individuals’ characteristics and constraints. There is limited evidence on the costs and cost-effectiveness of alternative ways of organising the blood donation service.^18–28 Previous studies have estimated the effect of previous policy changes on the volume and costs of whole blood collected,^18–28 and predicted the efficient location and staffing for static donor centres.^19,29–31 None of these studies estimated the cost-effectiveness of possible changes to the blood donation service that are of direct relevance to future NHSBT strategies, nor have they recognised donors’ relative preferences for alternative changes to the blood service.

A strategy of potential interest to the NHSBT is to invite existing whole-blood donors to donate blood more often, particularly those donors whose blood type is in high demand. In England, the minimum donation interval is currently 12 weeks for men and 16 weeks for women.³² It is currently unknown if reducing the minimum interdonation interval is clinically effective or cost-effective.³³ Shorter recall intervals may lead to an increased risk of iron deficiency, higher rates of donation deferral (temporary suspension of donors giving blood) and lower health-related quality of life (QoL) for donors.^34–36 Increased rates of deferral may lead to higher costs and encourage donors to leave the register.^37,38 International variations in blood service policy reflect this uncertainty about the optimum minimum interdonation interval. For example, the minimum intervals between blood donations are every 8 weeks (both sexes) in the USA,³⁹ and every 8 (men) and 12 weeks (women) in France and Germany.^40,41 The INTERVAL trial was designed to establish whether or not reducing the minimum recall interval for donors attending static donor centres in England would increase the frequency of whole-blood donation while maintaining donor health.⁴² However, the INTERVAL trial alone will not provide sufficient evidence to inform whether or not the NHSBT should reduce the minimum donation interval. In particular, research is also required on the relative costs and cost-effectiveness of reducing the minimum donation interval, in particular for subgroups of prime policy relevance.