Analysis of two questionnaires on quality of life of Chronic Obstructive Pulmonary Disease patients*

Objective: to evaluate the efficacy of quality of life questionnaires St. George Respiratory Questionnaire and Chronic Obstructive Pulmonary Disease Assessment Test in patients with chronic obstructive pulmonary disease based on correlation and agreement analyses, and identify the most effective tool to assess their quality of life. Method: cross-sectional cohort study with patients hospitalized in a Spanish hospital for exacerbation of chronic obstructive pulmonary disease. Health-related quality of life was assessed with both questionnaires. The correlation and the agreement between the questionnaires were analyzed, as well as the internal consistency. Associations were established between the clinical variables and the results of the questionnaire. Results: one hundred and fifty-six patients participated in the study. The scales had a correlation and agreement between them and high internal consistency. A higher sensitivity of the Chronic Obstructive Pulmonary Disease Assessment Test was observed for the presence of cough and expectoration. Conclusion: the questionnaires have similar reliability and validity to measure the quality of life in patients with acute chronic obstructive pulmonary disease, and the Chronic Obstructive Pulmonary Disease Assessment Test is more sensitive to detect cough and expectoration and requires a shorter time to be completed.

influence the gathering of information, especially if the patient needs to complete the questionnaire.
According to the recommendations of the Spanish Society of Respiratory Diseases (SEPAR) and international research (15) , the most used questionnaires are the St.
George Respiratory Questionnaire (SGRQ) and the

Chronic Obstructive Pulmonary Disease Assessment
Test (CAT). The SGRQ (16) validated in Spanish (17)(18) is the most used questionnaire in the population with respiratory diseases (17,19) , and it is also validated for administration via telephone call (20) . The SGRQ has 50 items distributed into three categories -symptoms, activity, and impact -with 76 weighted responses, and requires 10 minutes for completion (17) . Each item has an empirically derived weight, and a score has to be calculated.
The CAT questionnaire, recommended by the SEPAR, is aimed at evaluating HRQL in patients with a diagnosis of COPD. The instrument had initially 21 items (21) which were later reduced to 8; a total score is obtained from the sum of these items (22) .
The Cronbach's alpha of the SGRQ is 0.94 (symptoms: 0.72, activity: 0.89, impact: 0.89) and of the CAT is 0.88, and the intraclass correlation coefficient of the SGRQ and CAT is 0.9 and 0.8, respectively (22) .
A literature review revealed a significant correlation between the SGRQ and the CAT in a population with a diagnosis of COPD in primary care centers (21)(22)(26)(27) .
Likewise, it was observed that there was a correlation between the two scales in the hospital setting in patients with stable COPD (23) , where there is also a correlation between the two questionnaires, although CAT is of much faster and easier application.
No studies were found to investigate the best questionnaire to evaluate HRQL in the hospital setting in patients with exacerbated COPD, as confirmed in the Spanish guide to COPD patient care (28) . This is one of the most important moments to evaluate HRQOL, that is, to verify the effectiveness of the treatment provided, as well as to administer the help and resources necessary to train patients before discharge.
Despite the interest generated by the study of HRQOL, there is only consensus in the literature about the use of the CAT in non-acute stages (29) and primary care centers (22) , while no consensus exist on the most  A calculator is used for calculation (31) .
The CAT (13) consists of eight questions related to cough, phlegm, chest tightness, breathlessness in activities of daily living, activity limitation at home, confidence leaving home, sleep, and energy. The score interval of each element varies between zero and five, with a maximum score of 40 (22) . According to the total CAT scores and the revised literature, patients can be classified into the following categories: 1-10 low impact; 11-20 average impact; 21-30 high impact; 31-40 very high impact (13,27) .  (32) , tcough, expectoration, wheezing, drowsiness, fever, need to sit, and edema (presented To perform the calculation, it was considered necessary to multiply the CAT score by 0.25 to make it directly comparable with the total SGRQ score (27) . In order to observe the differences between the health status of A summary of the characteristics of the sample is presented in Table 1.   Table 2 shows the correlation matrix of the spheres of the two questionnaires. The Pearson correlation coefficient showed a correlation between the new spheres created in the CAT and the existing ones of the SGRQ. A strong correlation was obtained globally and in the spheres of activity and impact, as well as a moderate correlation in the sphere of symptoms, as shown in Table 3. It was also observed that the two questionnaires presented adequate internal consistency in the sample studied, with Cronbach's alpha coefficients of 0.843 for the SGRQ and 0.799 for the CAT (Table 3). The Bland and Altman graph (Figure 1) showed that the mean scores of each of the questionnaires were within the limits of agreement, confirming the agreement between the two questionnaires.   presented an internal reliability above 0.7, similar to another review (25) on the attributes of the two questionnaires. Thus, the existence of correlation between the results of the two questionnaires has also been valued by several researchers (13) and even a correlation of the SGRQ spheres with the total CAT result has been reported (26) , but no study correlated all questions with the SGRQ spheres. There was a correlation in the questions attributed to the sphere of symptoms, activity or impact of the SGRQ with the questions related to these items in the CAT.
Similarly, the results of this study demonstrated that HRQOL in COPD patients is associated with dyspnea (7,42-46), pain (7), anxiety and depression (7,45,47) , limitation in the performance of activities ( 48) , wheezing, would be interesting to evaluate is the filling time of each of the questionnaires whose data were not studied.

Conclusions
The CAT scores were correlated with the ones of the SGRQ, in total, according to spheres and questions.