Minimally Clinically Important Difference in Breathlessness
Minimally Clinically Important Difference in Breathlessness
Objectives The aim of the study was to determine the minimally clinically important difference (MCID) for breathlessness due to chronic heart failure (CHF).
Background The measurement of breathlessness is difficult because it is subjective and multifactorial. Statistically significant changes in assessment may not be clinically meaningful. This is the first determination of MCID in chronic breathlessness in CHF using patient-rated data.
Methods Measurements were made as part of a randomized, controlled, crossover trial of morphine, oxycodone, or placebo for breathlessness in CHF. Breathlessness intensity was assessed at baseline and at the end of each intervention (day 4) using 11-point numerical rating scales (NRS), modified Borg (mBorg) scales, and global impression of change (GC) in breathlessness at day 4. From these data, the change in NRS or mBorg associated with a 1-point change in GC was calculated.
Results Thirty-five patients completed all study interventions, resulting in 105 data sets. We defined MCID as a 1-point change in GC. Regression analysis found that the MCID, including 95% CIs, equaled change in average NRS breathlessness per 24 hours of 0.5 to 2.0 U (P < .001), change in worst NRS breathlessness per 24 hours of 0.4 to 2.9 (P < .001), change in average mBorg score of 0.2 to 2.0 (P < .001), and change in worst mBorg score as between 0.3 and 1.9 (P < .001). Corresponding effect size calculations lay within the 95% CIs for the regression analysis for each measure.
Conclusions A 1-point change in NRS or mBorg score is a reasonable estimate of the MCID in average daily chronic breathlessness in CHF.
Chronic breathlessness is a daily experience for many people with chronic heart failure (CHF). The assessment of breathlessness, particularly as an end point in clinical trials, is challenging because it is both subjective and multifactorial. Breathlessness has been used more often as a primary end point in clinical trials in acute decompensated failure rather than in CHF. There have been recent calls for a more systematic approach to its measurement so that this symptom, which is highly relevant to patients and caregivers, can be given more prominence in CHF research. Validated unidimensional tools such as the 11-point numerical rating scale (NRS), visual analogue scale (VAS), and modified Borg (mBorg) scale have been used in studies that have demonstrated statistical improvements in breathlessness intensity with study interventions. However, we do not know whether the statistical significance equates to a clinically meaningful improvement for the patient.
The minimally clinically important difference (MCID) is commonly reported as the change in scores on the measurement scale in question that relates to the smallest change in symptom intensity subjectively recognized by the patient. With regard to patient self-report global rating, this has been variously defined as "a little worse/better" and "somewhat worse/better," with some studies including the category "almost the same, hardly any worse/better. Minimally clinically important difference estimations should be derived from subjective patient-rated information. Previous studies in chronic obstructive pulmonary disease have suggested that estimated MCID should be based on patient-derived global impressions of change (GCs); for breathlessness, this would be the change in breathlessness intensity score that the patient was able to discern as an improvement or deterioration.
The MCID has been estimated in acutely decompensated CHF and acute asthma, where a difference of 21.1 and 22 mm, respectively, on a 100-mm VAS was equivalent to subjectively "feeling a little less breathless." However, the qualitative experience of acute breathlessness, with its accompanying distress (often severe), is likely to be different to that of daily unremitting breathlessness. The MCID for one experience should not be extrapolated to the other.
A consensus statement in chronic breathlessness has suggested that the MCID for breathlessness is a 1-point improvement in the Borg score and 10 mm on a VAS regardless of the etiology of breathlessness. Ries recommends an MCID of 1 point in the Borg scale and 10- to 20-mm improvement on a VAS for patients with chronic lung disease using a "distribution"-based method. Rather than using "anchor-based" measurement scales to report acute breathlessness and comparing the result with the smallest patient-reported description of change in breathlessness, the MCID calculations are derived from the change in score from baseline divided by the SD of the baseline scores. The use of effect sizes in this way is a statistical method of analyzing the data to determine the lowest meaningful detectable difference based on the underlying variability or distribution of the reported data.
The MCID has never been assessed in patients with CHF using either method, that is, comparison with patient-reported impression of change or estimated effect size. We believe that this is an important omission for both researchers and clinicians working with patients with CHF in order for them to understand what a meaningful change is for the patient. We report the first study assessing MCID for chronic breathlessness in CHF using patient-rated change. We will use both anchor- and distribution-based methods for determining MCID. Given that the question under study was the change in score that relates to a clinically noticeable difference, the mechanism whereby the intervention effected that change is not the focus of this article. There are useful reviews that discuss this interesting topic.
Abstract and Introduction
Abstract
Objectives The aim of the study was to determine the minimally clinically important difference (MCID) for breathlessness due to chronic heart failure (CHF).
Background The measurement of breathlessness is difficult because it is subjective and multifactorial. Statistically significant changes in assessment may not be clinically meaningful. This is the first determination of MCID in chronic breathlessness in CHF using patient-rated data.
Methods Measurements were made as part of a randomized, controlled, crossover trial of morphine, oxycodone, or placebo for breathlessness in CHF. Breathlessness intensity was assessed at baseline and at the end of each intervention (day 4) using 11-point numerical rating scales (NRS), modified Borg (mBorg) scales, and global impression of change (GC) in breathlessness at day 4. From these data, the change in NRS or mBorg associated with a 1-point change in GC was calculated.
Results Thirty-five patients completed all study interventions, resulting in 105 data sets. We defined MCID as a 1-point change in GC. Regression analysis found that the MCID, including 95% CIs, equaled change in average NRS breathlessness per 24 hours of 0.5 to 2.0 U (P < .001), change in worst NRS breathlessness per 24 hours of 0.4 to 2.9 (P < .001), change in average mBorg score of 0.2 to 2.0 (P < .001), and change in worst mBorg score as between 0.3 and 1.9 (P < .001). Corresponding effect size calculations lay within the 95% CIs for the regression analysis for each measure.
Conclusions A 1-point change in NRS or mBorg score is a reasonable estimate of the MCID in average daily chronic breathlessness in CHF.
Introduction
Chronic breathlessness is a daily experience for many people with chronic heart failure (CHF). The assessment of breathlessness, particularly as an end point in clinical trials, is challenging because it is both subjective and multifactorial. Breathlessness has been used more often as a primary end point in clinical trials in acute decompensated failure rather than in CHF. There have been recent calls for a more systematic approach to its measurement so that this symptom, which is highly relevant to patients and caregivers, can be given more prominence in CHF research. Validated unidimensional tools such as the 11-point numerical rating scale (NRS), visual analogue scale (VAS), and modified Borg (mBorg) scale have been used in studies that have demonstrated statistical improvements in breathlessness intensity with study interventions. However, we do not know whether the statistical significance equates to a clinically meaningful improvement for the patient.
The minimally clinically important difference (MCID) is commonly reported as the change in scores on the measurement scale in question that relates to the smallest change in symptom intensity subjectively recognized by the patient. With regard to patient self-report global rating, this has been variously defined as "a little worse/better" and "somewhat worse/better," with some studies including the category "almost the same, hardly any worse/better. Minimally clinically important difference estimations should be derived from subjective patient-rated information. Previous studies in chronic obstructive pulmonary disease have suggested that estimated MCID should be based on patient-derived global impressions of change (GCs); for breathlessness, this would be the change in breathlessness intensity score that the patient was able to discern as an improvement or deterioration.
The MCID has been estimated in acutely decompensated CHF and acute asthma, where a difference of 21.1 and 22 mm, respectively, on a 100-mm VAS was equivalent to subjectively "feeling a little less breathless." However, the qualitative experience of acute breathlessness, with its accompanying distress (often severe), is likely to be different to that of daily unremitting breathlessness. The MCID for one experience should not be extrapolated to the other.
A consensus statement in chronic breathlessness has suggested that the MCID for breathlessness is a 1-point improvement in the Borg score and 10 mm on a VAS regardless of the etiology of breathlessness. Ries recommends an MCID of 1 point in the Borg scale and 10- to 20-mm improvement on a VAS for patients with chronic lung disease using a "distribution"-based method. Rather than using "anchor-based" measurement scales to report acute breathlessness and comparing the result with the smallest patient-reported description of change in breathlessness, the MCID calculations are derived from the change in score from baseline divided by the SD of the baseline scores. The use of effect sizes in this way is a statistical method of analyzing the data to determine the lowest meaningful detectable difference based on the underlying variability or distribution of the reported data.
The MCID has never been assessed in patients with CHF using either method, that is, comparison with patient-reported impression of change or estimated effect size. We believe that this is an important omission for both researchers and clinicians working with patients with CHF in order for them to understand what a meaningful change is for the patient. We report the first study assessing MCID for chronic breathlessness in CHF using patient-rated change. We will use both anchor- and distribution-based methods for determining MCID. Given that the question under study was the change in score that relates to a clinically noticeable difference, the mechanism whereby the intervention effected that change is not the focus of this article. There are useful reviews that discuss this interesting topic.