0
Correspondence |

ResponseResponse FREE TO VIEW

Paul W. Jones, PhD
Author and Funding Information

From the Division of Clinical Science, St. George’s University of London.

Correspondence to: Paul W. Jones, PhD, Division of Clinical Science, St. George’s University of London, Cranmer Terr, London, SW17 0RE, England; e-mail: pjones@sgul.ac.uk

Financial/nonfinancial disclosures: The author has reported to CHEST the following conflicts of interest: Dr Jones has received consulting fees and speakers honoraria from GlaxoSmithKline.


Financial/nonfinancial disclosures: The author has reported to CHEST the following conflicts of interest: Dr Jones has received consulting fees and speakers honoraria from GlaxoSmithKline.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2012;142(1):268-269. doi:10.1378/chest.12-0838
Text Size: A A A
Published online

I appreciate the questions posed by Dr Kocks and colleagues following our article in CHEST1 concerning the responsiveness of the minimal clinically important difference (MCID) for the COPD Assessment Test (CAT). They discuss two methods for estimating the MCID: one that is clinical (anchor based) and the other statistical (distribution based).

The definition of the MCID is key to this debate. One of the most comprehensive is “the smallest difference in score which patients perceive as beneficial and would mandate, in the absence of troublesome side effects and excessive cost, a change in the patient’s management.”2 If a patient’s perception of benefit is accepted as a core principle, it is difficult to justify use of distribution-based estimates of the MCID for patient-reported outcomes.

Anchor-based estimates are often made by comparing the patient-reported outcome score with a global change score. Investigators sometimes apply this approach retrospectively, using a global scale that was applied in the study for another purpose (as used by Dr Kocks and colleagues on our data). That is a mistake. Global scales are designed for a specific purpose, in our case to test whether the CAT could measure recovery after a specific period (14 days). Our scales were not worded to identify minimum beneficial change; furthermore, a change judged worthwhile at 14 days may be different from one at 28 days.

No single MCID estimate is sufficient. For the St. George Respiratory Questionnaire (SGRQ), multiple approaches were used.3 Clinically-based methods produced consistent estimates (≈4 units), but distribution-based estimates were quite different and dependent on methodology.

Mapping techniques are also used for MCID estimation and are applicable to the CAT because the relationship between the CAT and the SGRQ is constant across the scaling range.4 The mathematical relationship is CAT =0.4 × SGRQ; so, the estimate for the CAT MCID is 1.6.

Dr Kocks and colleagues suggest using the effect size to estimate treatment effect. This is another statistical approach that compares change within individuals to the distribution of scores between individuals. They did not mention that the effect size for the SGRQ with rehabilitation in the study they quoted5 was very small at 0.2, whereas the measured SGRQ change was 3.9 (ie, very close to the MCID). Distribution-based estimates of change and clinical estimates of benefit clearly measure different things, which takes us back to the reason for using MCIDs: to identify the size of the benefit to the patient.

Using our mapped estimate of the CAT MCID, the mean improvement in CAT score reported by Dodd et al5 was 1.8 times the MCID. This shows that the CAT was very responsive, not unresponsive as suggested by Dr Kocks and colleagues.

References

Jones PW, Harding G, Wiklund I, et al. Tests of the responsiveness of the COPD Assessment Test following acute exacerbation and pulmonary rehabilitation. Chest. 2012;142(1):134-140. [PubMed] [CrossRef]
 
Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials. 1989;10(4):407-415.
 
Jones PW. St. George’s respiratory questionnaire: MCID. COPD. 2005;2(1):75-79.
 
Jones PW, Tabberer M, Chen W-H. Creating scenarios of the impact of COPD and their relationship to COPD Assessment Test (CAT™) scores. BMC Pulm Med. 2011;11:42.
 
Dodd JW, Hogg L, Nolan J, et al. The COPD assessment test (CAT): response to pulmonary rehabilitation. A multicentre, prospective study. Thorax. 2011;66(5):425-429.
 

Figures

Tables

References

Jones PW, Harding G, Wiklund I, et al. Tests of the responsiveness of the COPD Assessment Test following acute exacerbation and pulmonary rehabilitation. Chest. 2012;142(1):134-140. [PubMed] [CrossRef]
 
Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials. 1989;10(4):407-415.
 
Jones PW. St. George’s respiratory questionnaire: MCID. COPD. 2005;2(1):75-79.
 
Jones PW, Tabberer M, Chen W-H. Creating scenarios of the impact of COPD and their relationship to COPD Assessment Test (CAT™) scores. BMC Pulm Med. 2011;11:42.
 
Dodd JW, Hogg L, Nolan J, et al. The COPD assessment test (CAT): response to pulmonary rehabilitation. A multicentre, prospective study. Thorax. 2011;66(5):425-429.
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543