Letter to the Editor: What Do Orthopaedists Believe is Needed for Incorporating Patient-reported Outcome Measures into Clinical Care? A Qualitative Study

To the Editor,

Patient-reported outcome measures (PROMs) play a crucial role in ensuring representation of the patient perspective during evaluation of the interventions we deliver. We therefore commend the work of Whitebird et al. [14], which investigates barriers to incorporating PROMs in clinical practice.

The use of appropriate, valid, and reliable PROMs can be an extremely powerful tool in assessing the efficacy of interventions. However, as so eloquently described in the adage popularized by a favorite comic superhero, “with great power comes great responsibility.” In today’s era of evidence-based medicine, our clinical practice is guided by outcomes reported in high-quality research studies [6]. Incorrect use or flawed interpretation of PROMs can therefore have a catastrophic “domino effect” on our decision-making and clinical practice [4].

Very broadly speaking, two types of outcomes are used in orthopaedic research: clinician-assessed outcomes such as measurements made using various imaging modalities or objective measures of strength, and patient-reported outcome measures, namely PROMs. The most robust and high-quality research studies make the effort to control for confounding factors and limit sources of bias that may influence the measurements made or the data captured. For example, clinician-assessed outcomes are often controlled so that the clinician assessing the outcome is blind to the specific intervention performed. Because a provider’s knowledge of the intervention used can influence his or her assessment of patients afterward, blinding is considered a key quality-control step in robust research designs, and is considered a marker of higher-quality research in all of the widely used tools that assess quality and risk of bias in studies, including MINORS, ROBINS-I, and The Newcastle Ottawa Scale [11-13]. Essentially, risk of measurement bias is reduced through controlling for the state of mind of the outcome assessors and any underlying conscious or unconscious biases they may hold.

However, blinding to reduce observer bias is more challenging when dealing with PROMs. Although blinding patients to intervention choice may reduce some patient bias, this is difficult in orthopaedic surgery and fails to account for a major source of patient bias: the patient’s own state of mind and the presence or absence of emotional distress. Mental health conditions such as depression and anxiety are common in the general population, and may be more common among patients undergoing orthopaedic surgery [1, 5, 7, 9, 10]. These conditions are associated with symptoms such as low mood, anhedonia, and fatigue, and are therefore by their very nature likely to bias patients toward reporting more pain and poorer function on PROMs [7]. An association between conditions such as depression and anxiety and worse postoperative PROMs has been demonstrated after several orthopaedic interventions [2, 3, 8].

While many studies control for potential confounding factors like patient demographics and physical comorbidities when evaluating differences in PROMs, the potential confounding effect of pre-existing mental health conditions is rarely considered. Of course, it is unreasonable to expect authors to consider every single possible confounding variable, but since PROM responses are so inherently intertwined with patients’ mental states, controlling for pre-existing depression and anxiety are crucial. This may be done through the co-administration of mental health assessment questionnaires such as the Patient Health Questionnaire-9 (PHQ-9) for depression and the Generalized Anxiety Disorder Assessment (GAD-7). The situation is complicated by the fact that underlying musculoskeletal pathology may contribute to patients’ depression/anxiety and be relieved by intervention. It is therefore important that mental health is assessed and controlled alongside both pre- and postoperative PROMs. Failing to control for these factors may lead to invalid research conclusions and less-effective clinical approaches.