Evaluation of the minimal important difference for the feeling thermometer and the St. George's Respiratory Questionnaire in patients with chronic airflow obstruction

https://doi.org/10.1016/S0895-4356(03)00115-XGet rights and content

Abstract

Background and objective

The chronic respiratory questionnaire (CRQ), the St. Georges Respiratory Questionnaire (SGRQ), and the feeling thermometer (FT) evaluate change in health-related quality of life (HRQL) in patients with chronic airflow limitation (CAL). Although the interpretability, and in particular the minimal important difference (MID) in score changes, is well established for the CRQ, this is not the case for the SGRQ and FT. The objective of our study is to explore the interpretation of the SGRQ and FT.

Methods

We analyzed data from 84 patients who completed the CRQ, SGRQ, and FT before beginning pulmonary rehabilitation and 3 months later. We calculated correlations between the four CRQ domains (dyspnea, fatigue, emotional function, and mastery) and the three SGRQ domains (symptoms, activities, and impact), the SGRQ total score, and the FT. When Pearson's correlations were ⩾0.5, we constructed regression equations and used the slope to calculate the change in SGRQ and FT score that corresponded to a change in CRQ score of 0.5 (the MID). Having established MID for SGRQ we than used a similar approach to examine the relation between the SGRQ and FT results.

Results

Comparison with the CRQ dyspnea domain suggested the MID in SGRQ total score is approximately 3.05 with a 95% confidence interval (95% CI) ranging from 0.39 to 5.71 and a change of 5.67 (95% CI 3.43–7.92) represents a moderate change (1.0 on the CRQ dyspnea domain). The MID for the FT based on the CRQ fatigue domain was 6.1 (95% CI 1.87–10.28). The FT MID based on the SGRQ activities domain, impacts domain, and total score were, respectively, 7.4 (95% CI 3.44–11.35), 5.6 (95% CI 1.6–9.64), and 5.9 (95% CI 1.97–9.78).

Conclusion

An MID for the SGRQ approximates the previously suggested estimate of 4 on a scale of 0 to 100. The MID for the FT in patients with CAL is approximately 5 to 8 units on the 0 to 100 scale. These MID estimates should facilitate interpretation of clinical trials in which outcome measures include the SGRQ or FT.

Introduction

Interpreting changes in health-related quality of life (HRQL) remains a challenge to both investigators and clinicians [1], [2], [3], [4], [5]. Ensuring widespread acceptance of HRQL as a patient-important outcome will require clear guidelines for establishing the significance of changes in HRQL [2].

One way of describing and interpreting changes in HRQL is through the minimal important difference (MID), the smallest difference in score in the domain of interest that patients perceive as important, either beneficial or harmful, and that would lead the clinician to consider a change in the patient's management [1]. Approaches to establishing the MID include distribution-based methods, reliance on experts (opinion-based methods), and approaches that rely on sequential hypothesis formation and testing (predictive or data driven) [6]. A particular form of the last approach (also characterized as anchor-based methods) relies on examining the associations between scores on the instrument that is under investigation and an anchor, typically an independent measure of HRQL that clinicians can easily interpret [1].

The Chronic Respiratory Questionnaire (CRQ), which measures HRQL in patients with chronic airflow limitation (CAL), has proven valid, responsive, and useful in the hands of many investigators, in many settings, in a number of countries [7], [8], [9], [10], [11], [12], [13], [14], [15]. A substantial body of evidence suggests that the MID of the CRQ, which uses seven-point Likert-type scales in four domains [16], is approximately 0.5 on the seven-point scale. Changes of 1.0 and 1.5 correspond, respectively, to moderate and large improvement or deterioration [17], [18], [19].

Another instrument investigators have used frequently in CAL patients is the St. Georges Respiratory Questionnaire (SGRQ) [20]. Jones et al. suggested a change of 4 in the total score, ranging from 0 to 100 as the MID for the SGRQ [21], with differences for moderate and large changes in HRQL of 8 and 12, respectively [3]. Little empiric evidence supporting the interpretability of the SGRQ has, however, emerged in the peer-reviewed literature.

In theory, if both instruments measure HRQL in similar domains and their scores correlate, one should be able to map changes in one instrument on to changes in the other. Hence, changes of 0.5 on the CRQ should correspond to changes of 4 on the SGRQ. Thus, if there is a sufficiently high correlation between these two instruments, one could confirm the MID of the SGRQ using the MID from the CRQ, for which there is greater empirical support.

The feeling thermometer (FT), a visual analog scale (VAS) shown as a thermometer, has recently become a focus of increasing investigation [22]. The FT provides preference scores on a scale from 0 (dead) to 100 (full health), and is simpler and generally more efficient than the standard gamble (SG), the gold standard instrument for utility measurement [22], although in one study respondents found the SG easier to complete than the FT [23]. Accumulating evidence suggests that the FT works well as an evaluative instrument in various groups, including patients with CAL [24], [25], [26], [27], [28], [29]. As with many other HRQL instruments, however, interpretability of the FT remains largely unexplored. We hypothesized that we can use the CRQ and the SGRQ to establish the MID for the FT in patients with CAL.

Thus, the purpose of this study was twofold. First, we compared the MID of the CRQ and the SGRQ. Second, having validated the MID of the SGRQ, we used both the CRQ and SGRQ to establish the MID for the FT.

Section snippets

Data and study design

The data for these analyses come from a study investigating measurement of HRQL in patients participating in respiratory rehabilitation programs at the University of Toronto and McMaster University in Hamilton, Ontario, Canada [29], [30]. Eligible patients included all inpatients and outpatients with CAL enrolled in the rehabilitation programs. We excluded patients with the following diagnoses: α1-antitrypsin deficiency, silicosis, sarcoidosis, asbestosis, lupus, or cancer, and those unable to

Baseline characteristics of study participants

Forty-six of the 130 patients initially enrolled did not complete the study. The reasons for not completing the study were: refusal to continue the interviewing process (n = 24), patients' statement “too sick” to complete the second interview (n = 7), failure to complete the rehabilitation program (n = 6), inability to contact the patient because they were out of the region (n = 4), and development of new severe symptomatic illness or cognitive impairment (n = 5). The patients who did not complete the

Discussion

Determining what constitutes important differences in HRQL scores is important for interpreting intervention effects on HRQL. We compared the MID for the SGRQ and CRQ and our results indicate an MID for the SGRQ that is within the range of the previously suggested value (approximately 3.1 compared to 4 on the 100-point scale) [21]. We also evaluated the MID for the FT, a preference-based HRQL instrument for patients with CAL, by comparison with the CRQ and the SGRQ. Based on linear regression

Acknowledgements

This work was supported by a grant from the Medical Research Council of Canada to Gordon H. Guyatt, and by a Buswell Fellowship to Holger J. Schünemann.

References (40)

  • P.W Jones et al.

    The St George's Respiratory Questionnaire

    Respir Med

    (1991)
  • S.D Mathias et al.

    Health-related quality of life and functional status of patients with rheumatoid arthritis randomly assigned to receive etanercept or placebo

    Clin Ther

    (2000)
  • H.J Schünemann et al.

    A randomized controlled trial to evaluate the effect of informing patients about their pretreatment responses to two respiratory questionnaires

    Chest

    (2002)
  • G.H Guyatt et al.

    Measuring functional status in chronic lung disease: conclusions from a randomized control trial

    Respir Med

    (1991)
  • E.J Gallagher et al.

    Prospective validation of clinically important changes in pain severity measured on a visual analog scale

    Ann Emerg Med

    (2001)
  • P.W Jones

    Interpreting thresholds for a clinically significant change in health status in asthma and COPD

    Eur Respir J

    (2002)
  • M.N Lassere et al.

    Foundations of the minimal clinically important difference for imaging

    J Rheumatol

    (2001)
  • P.J Wijkstra et al.

    Reliability and validity of the chronic respiratory questionnaire (CRQ)

    Thorax

    (1994)
  • A.J Busch et al.

    Effects of a supervised home exercise program for patients with chronic obstructive pulmonary disease

    Phys Ther

    (1988)
  • K Simpson et al.

    Randomised controlled trial of weightlifting exercise in patients with chronic airflow limitation

    Thorax

    (1992)
  • Cited by (150)

    • Competition between friends and foes

      2022, European Economic Review
      Citation Excerpt :

      Higher scores indicate feeling closer to the individual member of the respective social group. The other measure of attitude – the feeling thermometer – is commonly used in polling (e.g. American national election studies), political sciences (Greene, 1999; Kaid et al., 1992; Miller and Wlezien, 1993) and also in medicine (Patrick et al., 1994; Jacobson et al., 1992; Schünemann et al., 2003). The feeling thermometer asks subjects to imply how warm they feel towards a specific group or person.

    View all citing articles on Scopus
    1

    Deceased.

    View full text