Quality and quantification: Is it time to rethink?

Often, I read in an online ultrasound forum a call to improve workforce shortages and long waiting times by using abbreviated scanning or so-called targeted studies. This is the justified realm of point-of-care ultrasound (POCUS) where a clinician answers a clinical question at the bedside allowing for fast triage into appropriate care. I am a fan of POCUS: I have seen instances where care was appropriately expedited. I am also a fan of the targeted scan performed by sonographers in situations where patients have had recent examinations and monitoring is all that is needed. But most patients still need a full study.

Complex protocols are efficacious and safe and the hallmark of sonography practice. Sonographer-led protocol-driven examinations, if performed completely, rule out other pathology, while confirming or excluding a differential diagnosis. In addition, while sometimes the differential is all that matters, its exclusion in an abbreviated scan may mean the true pathology is missed because it was not raised as a possibility by the referrer or recognised as a potential alternative diagnosis by the sonographer. When protocols are discarded, mistakes occur, pathology is missed, and additional (and potentially unnecessary) imaging may be performed. Protocol-driven studies are also a supportive approach for sonographers early in their career trajectory but take time to perform. In this issue of AJUM, Deslandes et al¹ show that to follow published guidelines for endometriosis diagnosis, a longer appointment time is needed. In essence, a one-slot booking system where every patient gets the same amount of time is not going to work. In the absence of a fee per service model, each sonographer would take as long as needed to get the images they required, but the fee per service model encourages practices to ‘squeeze’ more patients in and therefore does not adequately support full protocol-driven studies. Protocol-driven sonography is ideal. The question, though, is which protocol?

Many professional societies have guidelines for scanning but appropriately stop short of mandating their use. As health professionals, we should make individual decisions about how, and when, it is appropriate to offer an abbreviated scan. Yet, that takes a level of skill and expertise gained through years of experience, and therefore, the protocols and guidelines we use help to maintain quality care for all practitioners and patients. So who decides on these guidelines? Often, they are determined by international societies and are influenced by experts who typically work in well-resourced healthcare systems. Is it time we had local guidelines? Or are they too costly to develop for smaller countries? Also, in our region, should Australia and New Zealand have different guidelines from the rest of the world? We could adopt a shared approach, and that would be ideal given the transient nature of both our patient and professional populations.

If we are sharing the same protocols, should we be using the same normal reference values? The simple answer is yes, and the correct answer is, probably no. In this issue of AJUM, Dry et al² evaluate different fetal growth charts and show that, indeed, this is the case. Depending on which chart is used results in different percentages of abnormal fetal growth. Perhaps not surprisingly, the charts developed from a Western Australian population work best when applied to a Western Australian population. In addition, although this may not be surprising, it is an important study because developing any reference chart or reference values needs both a derivation cohort and a validation cohort. The contemporary validation of these charts is therefore necessary and useful. In fact, the World Health Organisation (WHO) calls for local validation.³ When populations are relatively homogeneous, charts developed in one ethnic group are largely applicable to other countries of similar homogeneous ethnicity. Increasingly, our populations are not homogeneous.

For almost all ultrasound measurements, larger people have larger measurements, but it does not stop there. Non-obese individuals of similar height and weight may have very different normal body composition based on ethnicity.⁴ The definition of pathology or disease is therefore impacted by the cohort in which the reference measurements were derived and those in which they are applied. Essentially, they should be similar. Rarely, this occurs. We have to do better. Protocols to guide scanning are immensely helpful, but they should be localised and incorporate local reference values. In addition, practices need to recognise the value of full and careful examinations. Patients deserve this.