Too much, too little, too soon, too late? Transfusion and long‐term survival in children

Too much, too little, too soon, too late?Transfusion and long-term survival in children H ow blood transfusions impact patient outcome is of interest to health economists, insurers, transfusion medicine specialists, prescribing physicians, families, and most importantly, vulnerable patients.There are two different approaches to evaluating impact within the health care system: micro and macro.At the micro level, attention is focused on a single patient or group of patients, one provider at a time.Population-based clinical trials, clinical effectiveness studies, meta-analyses, and clinical experience are used to inform physician decision making on a micro level.Discussion of risks and benefits are implicit in the consent process and occur at the bedside with the physician, parent, and child (if age appropriate) reviewing how transfusion can alleviate symptoms and/or support a child's hematopoietic system.In these cases, the physician's allegiance to the patient is measured as an "n of 1" study, and there is no role for the discussion of cost implications, resource limitations, or societal burden.At the macro level, care is provided to populations of patients.In this setting, care that benefits the greatest number of patients while minimizing risks, complications, and costs in the aggregate is the focus.Measurement strategies like mortality and complication rates, proportions, incidence, and prevalence become important tools that help to inform health care policy decisions affecting access, cost, and quality.Those technologies and treatments, be they mechanical or biologic, that result in better outcomes at lower costs will have the greatest value to populations of patients.Admittedly, there is considerable interplay between the two care systems just defined.In reality, they are not discrete, but rather porous to effects in both directions and have important implications for transfusion medicine.In this issue of TRANSFUSION, Gauvin and colleagues 1 describe a single institution's experience with the long-term follow-up of patients who received blood transfusions between 1990 and 1992.Their demographic findings provide some insight regarding practice patterns in the early 1990s.From a diagnostic perspective, cardiac disease, malignancies, and operative procedures, specifically cardiac and orthopedic procedures, were the diagnoses most frequently requiring transfusion.Age was an important consideration for transfusion; nearly a third of