Donor iron policy: from research to practice

It is now 4 years since Cable and colleagues of the REDS-II Donor Iron Status Evaluation (RISE) multicenter study published their results showing the alarming frequency at which iron deficiency occurs in blood donors.1 Popovsky's2 accompanying editorial raised a “call to action.” Currently there is general agreement in the United States and elsewhere that iron deficiency caused by blood donation needs to be addressed. A few centers have interventions in place, but there is no consensus as to the optimal approach. Iron status can be addressed by some combination of: donor education, donor criteria such as minimum hemoglobin (Hb) concentration, length of Hb deferral periods or minimum interdonation intervals, and direct interventions such as ferritin testing or iron supplementation. In addition to how biologically effective they are, how well strategies can be implemented at a given center and their operational impact necessitate a range of suitable interventions from which to select. The literature provides insight into possible interventions through studies of individual interventions,3, 4 observational studies,5-7 modeling studies,8 and examples of blood programs in which a combination of interventions have been implemented.9 This issue features results of a new study by Mast and colleagues,10 the Strategies to Reduce Iron Deficiency (STRIDE) study. They compare head to head three candidate interventions (two different doses of iron supplements irrespective of ferritin levels and ferritin testing with donor notification of results) in a randomized controlled trial. In their study, iron status improved in all three intervention groups but not in control groups. It did not matter which dose of iron was given (38 or 19 mg ferrous gluconate for 2 months) and supplements were mostly well tolerated. Testing and informing donors of poor iron status was almost as effective because many donors took action as recommended in the notification letter (taking iron supplements or increasing the length of time between donations). Of perhaps equal importance, the study also showed that even under carefully controlled study conditions, these interventions did not eliminate iron deficiency. Roughly two-thirds of donors had low ferritin at enrollment, and approximately half as many in the intervention groups still had low ferritin on the final visit. Well-designed studies such as the current study are an essential basis for policy formulation. Key strengths of the STRIDE study are that participants were randomly assigned to groups (three intervention, two control) so there would be no difference in the composition of donors in the groups. This maneuver allows us to compare the effect of the intervention directly. In addition they had appropriate control groups so that we can see what happens over 2 years if donors continue donating without intervention. However, study conditions rarely mirror operational conditions exactly, and some caution should be exercised in extrapolating expectations from study conditions to implementation. One reason for this concern is that studies apply selection criteria when participants are recruited and results may not be generalizable to donors who do not meet these criteria. The donors were all regular donors when they were recruited, being the group of greatest interest, but new donors’ receptiveness to these interventions has not been tested. Donors were recruited by letter with supplementary recruitment by posters, e-mail, and on-site invitation.11 Thus donors self-selected because they were interested in participating, they received extra information, and they may have been more keen than donors who did not want to participate. Donors were invited from three participating centers and there could be population-specific factors at other centers that would affect the donor response to the intervention. In addition, the impact of the intervention over the life of the donation career could not be tested and behavior changes can be difficult to sustain. In the 2-year study only 57% of donors enrolled completed the final visit. Among participants taking pills about one-third actively withdrew. Some wanted to be taking iron and did not know if they had iron or placebo and presumably they would take the supplements if implemented. Approximately 20% believed that they had side effects and were less likely to comply. In the ferritin testing arm, active withdrawal from the study was less frequent, but then again some were not taking any action. The study shows the impact of these interventions in regular donors that wanted to be in the study, but the compliance of donors who were not eligible or did not want to participate remains unknown. A second reason that outcomes may be different is that the policy implemented is frequently not the same as that of the study. In the study, donors received a letter after each donation informing them of their ferritin result and advising them of what to do, thus continuous engagement. Implementation would likely take the form of an algorithm that maximizes benefit while reducing the number of samples tested. Possibilities include focusing on high-risk groups such as young female donors or testing the first plus every fifth or 10th donation. In such scenarios, donors would not receive communication about their ferritin after every donation, so the percentage taking action and even the actions taken may not emulate the study. Providing iron supplements to every donor is unlikely to be cost-effective if a significant percentage do not take them. Many centers who choose to provide iron supplements may therefore decide to apply some form of donor selection. Which donors are selected and how the pills and information are delivered will affect the outcome. Finally, the study did not include a group receiving both iron supplements and ferritin test results, but this combination could have a greater effect than either on its own. Donors are already provided with basic information about iron and donation, but these communications have very limited impact on donor action. There are several reasons for this lack of effect. First, donors are less likely to follow advice from education materials if they do not see any personal relevance. Anemia is associated with symptoms such as fatigue and pica but in a large study of blood donors who met the Hb threshold to donate, no association between low iron stores and a quality-of-life index was identified.12 Those who feel well have little motivation to change. Second, a Hb level acceptable for donation is often seen as proof that all is well. Third, deferral for low Hb creates confusion especially for females because the lower limit of the normal range (120 g/L) is unacceptable for donation. Furthermore, short deferral periods (varying from 1 day to 3 months in the United States)13 are not consistent with encouraging donors to take iron depletion seriously and send the wrong message. Fourth, even when written in simple terms, donor comprehension is imperfect.14 Fifth, many people have strong views on diet and health and may be at least somewhat resistant to suggestions that challenge these beliefs. Thus the inadequate response to donor education materials is likely due to a combination of not attending to them, misunderstanding or resistance to the message, and low motivation to change. In the study by Mast and colleagues,10 providing donors with evidence of their low iron made it personally relevant, which prompted some form of action in many donors. There may be some room for improvement here because many donors are confused when told their Hb (interpreted as “iron”) is good, but their ferritin (also interpreted as “iron”) is low. Providing donors with iron supplements probably made it easier for donors to take action and possibly sent a tangible message that they should. The study shows that ferritin testing and iron supplementation both can overcome donor inertia to some degree, at least in regular donors who wanted to participate in the study. However, the way donors are encouraged to take action will make a difference. More research is still needed to determine the most effective way of informing donors and/or encouraging them to take supplements. A fourth reason that results in the real world of blood center collection may not be identical to the study is that criteria are also changing, making the baseline donor iron policy different from that of the study. In the United States the minimum Hb level for males will increase from 125 to 130 g/L by May 201615 to be consistent with the lower limit of normal male Hb. Ideally blood centers should also consider increasing the deferral period for donors with low Hb to something consistent with the messaging of the overarching policy. Longer interdonation intervals (reducing the number of donations per donor) as is done in Europe16 are more consistent with the time it takes to recoup iron lost from donation.17 The STRIDE study10 is an important development in the journey toward robust, effective donor iron policy because it shows that more than one approach can produce similar results and blood centers can reasonably choose between iron supplementation or ferritin testing interventions to alleviate iron deficiency. The choice hinges on whether the objective of the policy is to strictly counteract the iron deficiency caused by blood donation or a broader mandate to provide a service that encourages donors to take their health issues into their own hands. The former favors iron supplementation without assessment of iron status, and the latter ferritin testing and notification of donors. Randomized controlled trials must control conditions so that interventions can be studied, compared, and understood. If either iron supplements or ferritin testing with notification were to be implemented, an improvement in iron status would be expected but it could be less than reported in the study. This limitation is because we do not know the outcome for donors who were either ineligible or did not want to participate, bearing in mind that it is very difficult to get people to make changes in their daily life. In addition, the policy implemented will likely be a little different from the one studied and the backdrop of other aspects of donor iron policy on which the intervention will be superimposed will be different. Some degree of iron deficiency is inevitable, and some iron-deficient donors may be at risk of adverse health outcomes. The fact that approximately one-third of donors receiving iron supplementation or ferritin testing with notification were still iron deficient even under study conditions underscores the need to address iron deficiency in blood donors with a multipronged approach. It should be acceptable for policies to vary from center to center to address specific needs and issues in their donor base and to be operationally practical. The author has disclosed no conflicts of interest. Sheila F. O'Brien1,2 e-mail: sheila.obrien@blood.ca 1Canadian Blood Services 2School of Epidemiology, Public Health and Preventive Medicine University of Ottawa Ottawa, ON, Canada

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.