Responsiveness is not new to the field of psychometrics. Responsiveness refers to the ability of an instrument to discriminate clinical changes as the result of an intervention.1,2 Responsiveness has been described as "an essential element of health status questionnaires".3^(p.1)

It has been used in many areas, including depression, spinal cord injury, stroke, traumatic brain injury, cardiac rehabilitation and pain, to name a few.4-9 Over the years, there has not always been consensus about its place as a psychometric property. Although it has been viewed both as a unique psychometric property and a function of validity, it has always been held as part of the quantitative realm of research.3,10

If we are going to take time to define a psychometric property and subsequently measure it, then we should find a meaningful place for it in order to inform practice. We need to decide where and how responsiveness should be taken into account.

We often see responsiveness reported in studies validating a new or revised instrument.6,11,12 In addition, it is reported when comparing instruments that measure the same construct or function.4 Quantifying responsiveness may also help researchers select the most robust instrument when planning an intervention study. Thus, the importance of selecting and using an instrument with strong psychometric properties cannot be stressed enough.

Responsiveness has the potential to provide greater insight and understanding if we carefully consider the value and meaning of "clinically important change". Statistically significant differences and clinically significant differences are distinctive entities; each has a role in measuring how an intervention can be viewed and evaluated. Statistical significance has been held as the gold standard when it comes to evaluating interventions. Statistical significance can tell us how sure we are that a relationship or difference exists, and whether to accept or reject the null hypothesis. By itself and without further information, statistical significance does not help us make clinical decisions. Therefore, if responsiveness can be used in clinical decsion-making, this would be a more practical approach to determine whether an effect truly exists.

If an instrument has demonstrated that it is responsive and can detect clinical change, then results indicating clinical change tell us that the intervention has had an impact on the function being measured, whether it is depression, pain, motor movement or strength. We are continuously searching for interventions to improve the lives of others and need to consider if/how responsiveness findings help us make clinical decisions. Of course, we need to interpret findings in context, that is, an intervention may only be meaningful for the group being studied. In addition, we need to look at all study findings to see if clinical significance aligns with the overall study findings. By itself, responsiveness only tells us that an instrument has the ability to measure change. But when it does, we need to figure out if the measured change has meaning for clinicians/researchers and/or patients receiving the intervention, or both.

What if we entertain the idea that responsiveness, by its own definition, is really giving us a glimpse of the patient's voice, in effect, informing qualitative inquiry? Are we merely looking at numbers or what those numbers represent? Does clinical change give us an inkling of that change from the patient perspective? What does this clinical change mean to these patients? Is change really saying that the intervention being tested is making a measureable difference in their life?

If statistical significance tells us how sure we are that a relationship or difference exists as a result of an intervention, then maybe clinical significance can be considered a surrogate for "patient voice", indicating that an intervention has made a difference in their life because, for example, they feel better, have less pain or have improved range of motion. This is different than shared decision-making because shared decision-making is an active exchange between a practitioner and patient, with the intention of making healthcare decisions.13 However, information gleaned from impactful clinical changes, as mentioned above, may lead to increased dialogue in shared decision-making regarding interventions or treatments. Treatments/interventions that have been studied and indicate promising outcomes may be considered in these collabrative encounters.

As responsiveness is commonly reported and used in studies measuring physical or psychological function, we need to consider how we should interpret clinically important differences. Similarly, like other statistics, responsiveness cannot be viewed as a stand alone property but needs to be examined in the context of the study and overall study results. Therefore, if responsiveness can discriminate clinically important changes, as the result of an intervention and/or from the patient's perspective, then it may have a place in informing clinical decision-making.

References