The association between health-related quality of life and disease progression in idiopathic pulmonary fibrosis: a prospective cohort study.

Background and objective: Generally, a disease-specific health-related quality of life (HRQOL) measurement is more useful than generic measures in assessing perceived physical and mental health characteristic of a particular disease. The idiopathic pulmonary fibrosis (IPF)-specific version of St. George's Respiratory Questionnaire (SGRQ-I) has been recently developed for patients with IPF. We proposed to evaluate associations between the SGRQ-I and other clinical indices, as well as its prognostic value in patients with IPF.
Methods: Fifty-two patients with IPF were recruited in this prospective cohort study. HRQOL was assessed using the SGRQ-I and the Medical Outcomes Study 36-item Short Form, dyspnea using the modified Medical Research Council (mMRC) dyspnea scale, and psychological status using the Hospital Anxiety and Depression Scale (HADS). We then evaluated the relationship between the SGRQ-I and other clinical measures, as well as one-year clinical deterioration defined as a hospital admission due to respiratory exacerbation or all-cause death.
Results: Stepwise multiple-regression analyses revealed that the mMRC dyspnea scale, the HADS anxiety or depression, and minimum oxygen saturation during a six-minute walk test significantly contributed to the Total and three components of the SGRQ-I. In multivariate Cox proportional-hazards analyses, the Total score of SGRQ-I predicted clinical deterioration independent of forced vital capacity, the six-minute walk distance, or partial pressure of arterial oxygen on room air. Conclusions: The SGRQ-I is a multidisciplinary instrument representing physical, functional and psychological impairments in patients with IPF. The SGRQ-I is a significant predictor of short-term disease progression independent of physiological measurements. (Sarcoidosis Vasc Diffuse Lung Dis 2017; 34: 226-235). Copyright:

Introduction

Idiopathic pulmonary fibrosis (IPF) is a progressive fibrotic pulmonary disease with a poor prognosis (1). IPF has diverse features and various clinical courses. Patients with IPF demonstrate refractory progressive fibrosis, with a median survival of approximately three years, and sometimes develop acute exacerbations (1, 2), whereas some patients with IPF remain stable over the long term (3). In clinical trials, pulmonary function measurements such as forced vital capacity (FVC) often serve as primary endpoints. Recently, however, multidisciplinary and comprehensive assessments including not only pulmonary function measurements but also patient-reported outcomes (PROs) such as health-related quality of life (HRQOL) and dyspnea have been recognized as important in IPF (4), given that chronic progressive symptoms that may not be well controlled by novel IPF drugs adversely affect the HRQOL (5, 6).

The St. George’s Respiratory Questionnaire (SGRQ) is one of the most common instruments for evaluating HRQOL in patients with IPF (7, 8). Although the SGRQ was originally developed for patients with chronic airflow limitation such as chronic obstructive pulmonary disease (COPD) or asthma (9), it has been validated for use in other pulmonary diseases, including IPF (7, 10, 11). Recently, an IPF-specific version of SGRQ (SGRQ-I), a simpler version of the SGRQ, has been developed for patients with IPF, with certain SGRQ items removed because of their lesser relevance to patients with IPF (12). However, the characteristics of the SGRQ-I have not been well established.

Although the SGRQ has been shown to predict mortality in patients with COPD (13, 14), no such relationship has been demonstrated in patients with IPF (15). We previously reported that the relationship between HRQOL and mortality in COPD was partly dependent on the questionnaire used (14, 16). We hypothesized that the SGRQ-I, which is more specific to IPF than the SGRQ, would be associated with disease progression in patients with IPF. In this prospective single-center cohort study of IPF, we sought to identify clinical indices associated with the SGRQ-I cross-sectionally and to determine the association between the SGRQ-I and the short-term progression of IPF longitudinally.

Methods

Patients

We prospectively enrolled 52 consecutive patients with IPF who attended the Department of Respiratory Medicine at Kyoto University Hospital from April 2013 to October 2014. The diagnosis of IPF was made according to the criteria for IPF (1). Patients who were <20 years old and had active malignant disease or cognitive impairment were excluded. This study was approved by the Ethics Committee of Kyoto University (approval number; E1765), and all patients provided written informed consent. This study was registered on UMIN.org (UMIN000011142).

Data collection

At baseline, all patients underwent blood tests including an arterial blood gas (ABG), pulmonary function tests (PFTs), a six-minute walk test (6MWT), and evaluations of HRQOL, dyspnea, psychological status, and sleep quality. Patients who developed acute events such as acute exacerbations or pulmonary infections within four weeks of these baseline measurements were excluded. We evaluated the time from baseline to admission due to respiratory exacerbation or all-cause death within one year, which we arbitrarily defined as clinical deterioration.

Blood tests and physiological measurements

All patients underwent an ABG on room air. The PFT and 6MWT were performed according to published guidelines (17, 18). The diffusing capacity of the lung for carbon monoxide (DLCO) was mandatory wherever possible. When performing the 6MWT, the patients were permitted to use supplemental oxygen at a concentration equal to that available during exertion. The GAP index was calculated by age, gender, and the percentages of the predicted FVC (%FVC) and DLCO (%DLCO) and had a range of 0 to 8 (19). We also evaluated the extent of emphysema according to previous reports (20, 21). Combined pulmonary fibrosis and emphysema (CPFE) was diagnosed on the basis of having 10% or more emphysema area by visual assessment in patients with IPF. The presence of CPFE was independently assessed by two observers (K.I. and T.K.) who were blinded to clinical information, and interobserver disagreements were resolved by consensus.

Patient-reported outcomes

HRQOL was assessed using the Japanese versions of the Medical Outcomes Study 36-item Short Form (SF-36) (22), a generic questionnaire, and the SGRQ-I (12), a disease-specific questionnaire. SF-36 consists of eight subscales: Physical functioning, Role physical, Bodily pain, General health, Vitality, Social functioning, Role emotional, and Mental health. Each subscale is scored with a range from 0-100. Higher scores indicate better HRQOL. The SGRQ-I consists of three components: Symptoms, Activity, and Impacts. Each component and the Total are scored with a range from 0-100. Higher scores indicate worse HRQOL. Dyspnea was evaluated by the Japanese version of the modified Medical Research Council (mMRC) dyspnea scale graded from 0-4 (23, 24), and psychological status was assessed by the Hospital Anxiety and Depression Scale (HADS) (24, 25). Patients’ sleep quality was assessed by the Japanese version of the Pittsburgh Sleep Quality Index (PSQI) (24, 26).

Statistical analysis

Data are expressed as the median and interquartile range, unless otherwise stated. Statistical analyses were performed using JMP 10.0 (SAS Institute Inc., Cary, NC, USA). Correlations between variables were evaluated using Spearman’s rank correlation coefficient tests. Stepwise multiple-regression analyses were performed to identify the clinical variables that significantly contributed to the SGRQ-I scores. Clinical variables whose p-value was less than 0.10 in correlation to HRQOL in univariate analyses were included in stepwise multiple-regression analyses. Univariate and multivariate Cox regression analyses were performed to identify factors predicting clinical deterioration within one year, which was defined as a hospital admission due to respiratory exacerbation or all-cause death. A receiver operating characteristic (ROC) analysis of SGRQ-I (Total score) was performed to determine the threshold for predicting clinical deterioration. All analyses were considered statistically significant when p < 0.05.

Results

Study population

Fifty-two consecutive patients with IPF were enrolled in this study (Table 1). Seven patients (13.5%) were diagnosed by histology, confirming usual interstitial pneumonia pattern. The median age of the patients was 73.0 years, and 44 patients (84.6%) were male. Forty-seven patients (90.4%) had a smoking history. The medians of %FVC and %DLCO were 86.5% and 39.7%, respectively. Seventeen (32.7%) patients were diagnosed with CPFE.

Table 1.

Patient Characteristics (n = 52)

	Median (interquartile range) or number (%)
Age, y	73.0, (66.3-76.8)
Gender, male	44 (84.6%)
Smoking History	47 (90.4%)
BAL, performed	22 (42.3%)
SLB confirmation	7 (13.5%)
CPFE	17 (32.7%)
%FVC, %	86.5, (71.9-99.9)
%Dlco, %	39.7, (33.0-50.0)
6MWD, m	491, (416-529)
Minimum SpO2 during 6MWT, %	86.0, (80.5-89.3)
PaO2 on room air, Torr	83.1, (75.7-91.5)
KL-6, U/mL	784, (625-1190)
SP-D, ng/mL	232, (176-232)
mMRC (0 to 4)	1, (0-2)
HADS (0 to 21)
Anxiety	3, (1-5)
Depression	4, (2-7)
PSQI (0 to 21)	4, (2-8)

Abbreviations: IPF, idiopathic pulmonary fibrosis; BAL, bron- choalveolar lavage; SLB, surgical lung biopsy; CPFE, combined pulmonary fibrosis and emphysema; %FVC, percentage of the predicted forced vital capacity; %DLCO, percentage of the predicted diffusing capacity of the lung for carbon monoxide; 6MWD, six- minute walk distance; SpO2, oxygen saturation measured by pulse oximetry; 6MWT, six-minute walk test; PaO2 partial pressure of arterial oxygen; KL-6, Krebs von den lungen-6; SP-D, surfactant protein-D; mMRC, modified Medical Research Council; HADS, Hospital Anxiety and Depression Scale; PSQI, Pittsburgh Sleep Quality Index.

Baseline HRQOL scores

Baseline HRQOL scores are shown in Table 2. Regarding SF-36, median scores of six among eight subscales were over 70, and 36-49% of patients had the maximal scores (best health) in four subscales. Regarding SGRQ-I, the median of the Total score was 28.4, slightly skewed toward a better score, and 17.3% scored worst on the Activity component, whereas 23.1% scored best on the Impacts component.

Table 2.

Baseline HRQOL scores

	Median (interquartile range)	Patients with minimal score, n (%)	Patients with maximal score, n (%)
SF-36
Physical functioning	80.0 (45.0-88.8)	0 (0.0)	1 (1.9)
Role physical	75.0 (43.8-100)	0 (0.0)	19 (36.5)
Bodily pain	84.0 (64.0-100)	0 (0.0)	25 (48.1)
General health	50.0 (31.3-57.0)	0 (0.0)	0 (0.0)
Vitality	56.3 (43.8-75.0)	1 (1.9)	1 (1.9)
Social functioning	75.0 (50.0-100)	0 (0.0)	20 (38.5)
Role emotional	79.2 (50.0-100)	1 (1.9)	19 (36.5)
Mental health	70.0 (50.0-83.8)	0 (0.0)	3 (5.8)
SGRQ-I
Symptoms	43.7 (18.6-66.8)	4 (7.7)	3 (5.8)
Activity	48.1 (21.9-78.2)	4 (7.7)	9 (17.3)
Impacts	18.5 (3.2-40.4)	12 (23.1)	0 (0.0)
Total	28.4 (14.3-57.9)	0 (0.0)	0 (0.0)

Higher scores indicate better status on the SF-36, and lower scores indicate better status on the SGRQ-I.

Abbreviations: HRQOL, health-related quality of life; SF-36, Medical Outcomes Study 36-item Short Form; SGRQ-I, idiopathic pulmonary fibrosis-specific version of the St. George’s Respiratory Questionnaire

Correlations between SGRQ-I and other clinical measurements

The three components and the Total of the SGRQ-I had significant weak-to-moderate correlations with physiological parameters including %FVC, %DLCO, minimum oxygen saturation measured by pulse oximetry (SpO2) during the 6MWT, partial pressure of arterial oxygen (PaO2) on room air (absolute Spearman’s rank correlation coefficient (Rs)=0.38-0.59, p<0.01) except for the non-significant correlation between 6MWD and the Symptoms of the SGRQ-I (Table 3). The three components and the Total of the SGRQ-I had moderate to strong correlations with the mMRC dyspnea (Rs=0.58-0.82, p<0.001) and weak to moderate correlations with HADS anxiety and depression (Rs=0.39-0.66, p<0.01), except for a non-significant correlation between HADS anxiety and the Activity of the SGRQ-I. Relationships between the SGRQ-I and SF-36 are also shown in Table 3. The Total of the SGRQ-I was moderately to strongly related to subscales of the SF-36 (absolute Rs=0.53-0.85, p<0.05), except for an insignificant relationship with the SF-36 Bodily pain. A similar trend was found for the three components of SGRQ-I.

Table 3.

Correlations between the SGRQ-I and other indicators, including physical measurements and patient-reported outcomes

	SGRQ-I Symptoms	SGRQ-I Activity	SGRQ-I Impacts	SGRQ-I Total
%FVC	-0.44‡	-0.52‡	-0.44†	-0.50‡
%Dlco	-0.38†	-0.56‡	-0.56‡	-0.57‡
6MWD	-0.22	-0.59‡	-0.53‡	-0.51‡
Minimum	SpO2 during 6MWT	-0.42†	-0.55‡	-0.54‡	-0.57‡
PaO2 on room air	-0.44†	-0.48‡	-0.57‡	-0.54‡
mMRC	0.58‡	0.82‡	0.66‡	0.77‡
HADS Anxiety	0.44†	0.27	0.44†	0.39†
HADS Depression	0.50‡	0.54‡	0.66‡	0.64‡
PSQI	0.17	0.15	0.25	0.22
SF-36
Physical functioning	-0.57‡	-0.90‡	-0.80‡	-0.85‡
Role physical	-0.49‡	-0.67‡	-0.64‡	-0.65‡
Bodily pain	-0.24	-0.21	-0.21	-0.24
General health	-0.55‡	-0.70‡	-0.71‡	-0.72‡
Vitality	-0.48‡	-0.63‡	-0.71‡	-0.67‡
Social functioning	-0.47‡	-0.53‡	-0.64‡	-0.59‡
Role emotional	-0.57‡	-0.58‡	-0.67‡	-0.64‡
Mental health	-0.46‡	-0.47‡	-0.55‡	-0.53‡

Data are presented as Spearman’s rank correlation coefficient.

* p<0.05, † p<0.01, ‡ p<0.001.

Abbreviations: SGRQ-I, idiopathic pulmonary fibrosis-specific version of the St. George’s Respiratory Questionnaire; %FVC, percentage of the predicted forced vital capacity; %D, percentage of the predicted diffusing capacity of the lung for carbon monoxide; 6MWD, six- minute walk distance; SpO2, oxygen saturation measured by pulse oximetry; 6MWT, six-minute walk test; PaO2 partial pressure of arterial oxygen; mMRC, modified Medical Research Council; HADS, Hospital Anxiety and Depression Scale; PSQI, Pittsburgh Sleep Quality Index; SF-36, Medical Outcomes Study 36-item Short Form

Factors contributing to SGRQ-I

Stepwise multiple-regression analyses were used to identify those variables that contribute most to the SGRQ-I scores (Table 4). The mMRC dyspnea scale, HADS anxiety or depression, and minimum SpO2 during the 6MWT significantly explained the Total and the three components of the SGRQ-I. PaO2 on room air also weakly but significantly explained the Impact of the SGRQ-I. These factors accounted for 60-83% of the variance. Neither pulmonary function tests (%FVC and %DLCO) nor the 6MWD were significant predictors.

Table 4.

Exploration of factors contributing to the SGRQ-I in patients with IPF using stepwise multivariate analyses

Variables	Symptoms	Activity	Impacts	Total
%FVC
%Dlco
6MWD
Minimum SpO2 during 6MWT	0.24	0.15	0.21	0.21
PaO2 on room air			0.11
mMRC	0.15	0.50	0.23	0.34
HADS Anxiety		0.21
HADS Depression		0.14	0.28	0.23
PSQI
Cumulative R2	0.60	0.79	0.83	0.78

Data are presented as the coefficients of determination (R2).

R2 is shown when p < 0.05.

Abbreviations: SGRQ-I, idiopathic pulmonary fibrosis-specific version of the St. George’s Respiratory Questionnaire; IPF, idiopathic pulmonary fibrosis; %FVC, percentage of the predicted forced vital capacity; %DLCO, percentage of the predicted diffusing capacity of the lung for carbon monoxide; 6MWD, six-minute walk distance; SpO2, oxygen saturation measured by pulse oximetry; 6MWT, six-minute walk test; PaO2 partial pressure of arterial oxygen; mMRC, modified Medical Research Council; HADS, Hospital Anxiety and Depression Scale; PSQI, Pittsburgh Sleep Quality Index.

Relationship of SGRQ-I to clinical deterioration within one year

Eleven patients developed clinical deterioration within one year. Ten patients were hospitalized due to respiratory exacerbation, and one died from septic shock due to a urinary tract infection during the one-year observational period. Among the ten hospitalized patients, three patients died from acute exacerbation, one died from lung cancer, and one died from pulmonary infection, while five patients survived during the observational period. Univariate analyses revealed that PFT measurements, 6MWD, minimum SpO2 during the 6MWT, PaO2 on room air, and the GAP index were significant predictors of clinical deterioration (Table 5). Regarding HRQOL, the Total and the three components of the SGRQ-I significantly predicted clinical deterioration (p<0.001). Five of eight subscales of the SF-36 were also predictive of clinical deterioration (p<0.05). The mMRC was related to clinical deterioration (p=0.002), but the HADS anxiety and depression and the PSQI were not.

Table 5.

Univariate Cox proportional hazards analyses to predict clinical deterioration within one year

Variables	Univariate analysis
	HR	95% CI	p-value
Age	1.01	0.93-1.10	0.861
Male	0.40	0.12-1.82	0.209
Smoking History	0.19	0.05-0.83	0.031
CPFE	0.73	0.16-2.54	0.641
%FVC	0.93	0.89-0.96	<0.001
%Dlco	0.89	0.83-0.96	0.001
6MWD	0.996	0.992-1.000	0.035
Minimum SpO2 during 6MWT	0.88	0.82-0.93	<0.001
PaO2 on room air	0.92	0.87-0.96	<0.001
GAP index	4.68	1.79-13.2	0.002
SGRQ-I
Symptoms	1.04	1.02-1.07	<0.001
Activity	1.05	1.02-1.09	<0.001
Impacts	1.04	1.02-1.06	<0.001
Total	1.05	1.03-1.08	<0.001
SF-36
Physical functioning	0.97	0.95-0.99	0.002
Role physical	0.98	0.96-1.00	0.028
Bodily pain	1.01	0.98-1.04	0.586
General health	0.96	0.93-0.99	0.018
Vitality	0.97	0.95-0.99	0.015
Social functioning	0.98	0.96-1.00	0.040
Role emotional	0.98	0.96-1.00	0.058
Mental health	0.99	0.96-1.01	0.333
mMRC	2.28	1.37-3.95	0.002
HADS Anxiety	0.94	0.73-1.15	0.588
HADS Depression	1.03	0.88-1.18	0.670
PSQI	1.02	0.87-1.16	0.837

Abbreviations: HR, hazard ratio; CI, confidence interval; CPFE, combined pulmonary fibrosis and emphysema; %FVC, percentage of the predicted forced vital capacity; %Dlco, percentage of the predicted diffusing capacity of the lung for carbon monoxide; 6MWD, six-minute walk distance; SpO2, oxygen saturation measured by pulse oximetry; 6MWT, six-minute walk test; PaO2 partial pressure of arterial oxygen; SGRQ-I, idiopathic pulmonary fibrosis -specific version of the St. George’s Respiratory Questionnaire; SF-36, Medical Outcomes Study 36-item Short Form; mMRC, modified Medical Research Council; HADS, Hospital Anxiety and Depression Scale; PSQI, Pittsburgh Sleep Quality Index.

To compare the ability of the SGRQ-I to predict clinical deterioration independent of each physiological measurement previously shown to be related to mortality, multivariate-regression analyses were performed (Table 6). When the %FVC or the GAP index were entered into the analyses with the Total of the SGRQ-I, both were significantly related to clinical deterioration (Models 1 and 7). When the SGRQ-I and the %DLCO were entered as explanatory variables, neither the SGRQ-I nor the %DLCO was significant (Model 2). When the mMRC, 6MWD or PaO2 were used as explanatory variables in addition to the Total of the SGRQ-I, only the SGRQ-I was a significant predictor of clinical deterioration (Models 3, 5, and 6). However, when the minimum SpO2 during the 6MWT was used, only the minimum SpO2 was significant (Model 4).

Table 6.

Multivariate Cox proportional hazards analyses for the SGRQ-I versus other variables to predict clinical deterioration within one year

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7
SGRQ-I Total	1.04(1.01-1.07)*	1.03(0.99-1.06)	1.05(1.02-1.08)*	1.02(0.98-1.06)	1.04(1.01-1.07)*	1.05(1.01-1.09)*	1.05(1.02-1.09)*
%FVC	0.95(0.91-0.99)*
%Dlco		0.94(0.85-1.01)
6MWD			1.000(0.995-1.006)
Minimum SpO2 during 6MWT				0.91(0.83-0.99)
PaO2 at room air					0.97(0.91-1.03)
mMRC						1.16(0.48-2.61)
GAP index							3.28(1.24-9.53)*

Data are shown as the hazard ratio (95% confidence interval).

* p<0.05.

Abbreviations: SGRQ-I, idiopathic pulmonary fibrosis-specific version of the St. George’s Respiratory Questionnaire; %FVC, percentage of the predicted forced vital capacity; %Dlco, percentage of the predicted diffusing capacity of the lung for carbon monoxide; 6MWD, six- minute walk distance; SpO2, oxygen saturation measured by pulse oximetry; 6MWT, six-minute walk test; PaO2 partial pressure of arterial oxygen; mMRC, modified Medical Research Council.

An ROC analysis was performed to determine the cut-off value of the SGRQ-I Total score for predicting clinical deterioration within one year. The area under the curve was 0.88, and the cut-off value was 51.2. When 50 was set as the threshold, patients with an SGRQ-I Total score ≥50 developed clinical deterioration within one year more frequently than patients with a SGRQ-I Total score <50 (hazard ratio, 25.5 [95% confidence interval, 4.8-468]; p<0.001) (Figure 1).

Fig. 1.

The cumulative incidence of clinical deterioration. One-year probability curves for patients with SGRQ-I Total scores ≥50 and <50. Patients with SGRQ-I Total scores ≥50 developed clinical deterioration within one year more frequently than patients with scores <50 (log rank test, p<0.001).

SGRQ-I, idiopathic pulmonary fibrosis-specific version of the St. George’s Respiratory Questionnaire

Discussion

We evaluated the relationships between HRQOL and other clinical measurements cross-sectionally and between HRQOL and future outcomes longitudinally in patients with IPF. One of the characteristics of this study was the use of the SGRQ-I, which is a more disease-specific measure than the original SGRQ. Significant contributing factors to the SGRQ-I included minimum SpO2 during the 6MWT, mMRC dyspnea scale, and HADS, indicating physiological, functional, and psychological impairments, respectively. This finding suggests that the SGRQ-I is an indicator with multidisciplinary properties. Additionally, the SGRQ-I predicted clinical deterioration within one year independent of several important clinical measurements.

One important feature of assessing HRQOL is the ability to predict future outcomes (predictive property) (13-16, 27). We first demonstrated that HRQOL measured with the SGRQ-I predicted clinical deterioration, defined as hospitalization due to respiratory exacerbation or all-cause mortality. The most notable result was that the SGRQ-I predicted clinical deterioration independent of %FVC, the 6MWD, and the GAP index, which are well-known predictors of mortality in IPF (19, 28, 29). This finding indicated that HRQOL is complementary to pulmonary function and the GAP index in predicting outcomes for patients with IPF and that HRQOL should be measured besides physiological measurements such as pulmonary function tests and 6MWT. These findings will be helpful in multi-dimensional assessment in patients with IPF.

The level of dyspnea is related to mortality in several diseases including IPF (30-32) and even in elderly people (33). However, in this study, the SGRQ-I was more significantly related to clinical deterioration than was the mMRC dyspnea scale. This finding may have several causes. First, HRQOL provides a more comprehensive assessment of symptoms other than dyspnea during daily activities. In this study, contributing factors to the SGRQ-I included physiological and psychological factors in addition to the mMRC scale. Second, while the SGRQ-I is more specific to IPF than the respiratory-specific SGRQ, the uni-dimensional five-point mMRC scale may lack the precision and complexity required to detect a correlation between dyspnea and clinical deterioration. A more sophisticated method of assessing dyspnea may be needed. Thus, HRQOL and dyspnea assessments using appropriate measurements are important.

As a respiratory-specific HRQOL measure, the SGRQ is associated with a number of physiological parameters in patients with IPF (7, 8, 34), and changes in the SGRQ scores are partly correlated with changes in physiological parameters and CT indices in IPF (7). Thus, the SGRQ has been evaluated in several clinical trials of IPF (6, 35, 36). The SGRQ-I, modified from SGRQ, was developed as a disease-specific measure to evaluate HRQOL in patients with IPF (12). A previous study revealed that each component of the SGRQ-I corresponding to the SGRQ was significantly associated with %FVC, %DLCO, 6MWD, and all subscales of SF-36, including Bodily pain, except for a non-significant association between the Symptoms of the SGRQ-I and the 6MWD (12). Accordingly, in this study, most components of the SGRQ-I were significantly associated with physiological functions including PFTs, 6MWD, and minimum SpO2 during the 6MWT.

The mMRC contributed to the SGRQ-I score. These results are in accordance with a previous study in which the dyspnea scale, represented by a baseline dyspnea index, was found to be the most significant contributing factor to the SGRQ in patients with IPF (34). Conversely, PFT parameters such as vital capacity and DLCO were not significant (34). The HADS was also a significant factor in the SGRQ-I. Although depression is relatively common in patients with IPF (37), its relationship with HRQOL has not been fully investigated. We found that depression or anxiety was the important determinant of HRQOL in patients with IPF, as with other respiratory diseases (38-40). Sleep quality was also associated with HRQOL and was assessed by the SF-36 in patients with IPF (41). Poor sleep quality is common and results in a poor QOL in patients with IPF (42). In this study, however, sleep quality was not significantly related to the SGRQ-I.

This study has some limitations. First, the cohort size was small and the observational period was short. To investigate the independent influence of the SGRQ-I on clinical deterioration or survival, additional studies with a larger multicenter cohort and longer observational period are necessary. Second, the enrolled patients had relatively mild pulmonary impairment, with a median %FVC of 86.5%, which likely contributed to the low death rate. Therefore, the generalization of the results may be limited. Third, we chose clinical deterioration (arbitrarily defined as hospitalization due to respiratory failure or death) as our measure of disease progression, rather than survival. However, no composite outcome of disease progression specific for patients with IPF has yet been established. Further studies based on other clinical outcomes may be needed.

In patients with IPF, the SGRQ-I is a multidisciplinary instrument representing physiological, functional, and psychological impairment. HRQOL assessed by the SGRQ-I may be an independent predictor of short-term disease progression.