183 GENERAL DISCUSSION answered. Historically, study populations consisted mainly of middle-aged Western men with few comorbidities and with a relatively high socioeconomic status. Obviously, homogeneity among study participants has methodological advantages and evidence for these people is also needed, but society is clearly much more diverse than this population group. Moreover, results from these studies cannot and should not be generalised uncritically to population groups that were not included, as has become clear from the results of the FRAIL-AF trial described earlier.7 To clarify the maldistribution in research, a quick scoping review was carried out in PubMed yielding the following remarkable results, also shown in Figure 1: of all RCTs recorded in PubMed before the 17th of February 2024, only 0.1% had been performed in frail older patients and 4.4% in older patients. This is anything but a representative percentage when compared to the percentage in Dutch society defined as frail and old (over 4%) or old (about 20%) in 2020.31–33 Note that these last two percentages are probably even higher if only patients actually receiving healthcare are considered. The unequal distribution of participants was also seen in the four pivotal non-vitamin K antagonist oral anticoagulant (NOAC) trials.8–11 The median age of participants in these trials ranged from 70 to 73 years and the mean CHA2DS2-VASc score from 2.1 to 2.8 (with the only exception being the ROCKET AF trial where the mean CHA2DS2-VASc score was 3.5). These trials were thus conducted in a relatively vital population of barely 70-plus with nearly no comorbidities. Consequently, these pivotal trials did not answer the clinically relevant question whether it is safe for frail older atrial fibrillation (AF) patients to initiate a NOAC or to switch from a vitamin K antagonist (VKA) to a NOAC, while in these patients the prevalence of AF reaches its maximum (38%).34 The FRAILAF trial does provide an answer to this last question, because the study population of the FRAIL-AF trial was very different from the pivotal NOAC trials with a mean age of 83 years and a median CHA2DS2-VASc score of 4. Moreover, 88% of FRAIL-AF participants used ≥4 different types of medication, 38% had memory complaints, 17% was unable to walk around the house or to a neighbour, 44% had problems in daily life due to impaired vision, and 55% had problems in daily life due to impaired hearing. Importantly, the fact that the FRAIL-AF population was so different from the pivotal NOAC trials, does not even make it that surprising that the results of the FRAIL-AF trial were so different from the pivotal NOAC trials. The FRAIL-AF trial shows that results (from the pivotal NOAC trials in this example) cannot and should not be generalised to frail older AF patients and, moreover, that it is feasible to conduct an RCT in frail older patients. 9
RkJQdWJsaXNoZXIy MTk4NDMw