Examining Associations between Baseline Health-Related Quality of Life and Depression and Physical Functioning Improvement Following Pulmonary Rehabilitation

Abstract

Purpose:

This study examined whether health-related quality of life (HRQL) and depression assessed prior to pulmonary rehabilitation (PR) participation (i.e., at baseline) predicted change in 6-minute walk distance (6MWD) from baseline to end of PR.

Methods:

Pulmonary patients were consecutively referred/enrolled in a PR program between 2009–2022 (N=503). Baseline 6MWD was assessed along with self-report measures of HRQL (St. George’s Respiratory Questionnaire [SGRQ]) and depression (Geriatric Depression Scale [GDS]). The SGRQ total score was used to assess overall HRQL, and SGRQ subscales assessed pulmonary symptoms, activity limitations, and psychosocial impacts of pulmonary disease. Multiple linear regression was used to examine whether baseline SGRQ scores and depression predicted Δ6MWD.

Results:

Baseline SGRQ total score (F(1, 389)=8.4, P=.004) and activity limitations (F(1, 388)=4.8, P=.03) predicted Δ6MWD. Patients with an SGRQ activity limitations score ≤ 25^th percentile showed the most 6MWD improvement (M=79.7m, SE=6.7), and significantly more improvement than participants scoring between the 50–75^th percentiles (M=54.4m, SE=6.0) or > 75^th percentile (M=48.7m, SE=7.5). Patients scoring between the 25–50^th percentiles (M=70.2m, SE=6.1) did not differ significantly from the other groups. The SGRQ symptoms and impacts subscales were unrelated to Δ6MWD (F(1, 388)=1.2–1.9, P>.05), as was depression (F(1, 311)=0.0, P>.85).

Conclusions:

Patients with greater HRQL at baseline may experience greater physical functioning improvement following PR. Additional support for patients with lower HRQL (e.g., adjunctive self-management interventions) may enhance PR outcomes, particularly for patients who report greater activity limitations. Alternatively, early referral to PR (i.e., when less symptomatic) may also benefit physical function outcomes.

Condensed Abstract

Associations between both health-related quality of life (HRQL) and depression and changes in 6-minute walk distance (Δ6MWD) were assessed in patients attending pulmonary rehabilitation (PR). The HRQL, but not depression, was associated with Δ6MWD. Patients with higher HRQL prior to PR may experience greater physical functioning improvement upon completing PR.

A primary goal of pulmonary rehabilitation (PR) is to lessen disease-related limitations and improve physical functioning through individualized exercise training, education, and psychosocial support. The effectiveness of PR is well-established, and patients with pulmonary disease at varying levels of severity can expect to see improvements in physical functioning and subjective well-being following PR completion.^1–6 Given the established effectiveness of PR for improving patient outcomes following pulmonary deterioration, research has shifted from examining whether PR improves patient health and well-being to examining potential avenues for optimizing PR. One such avenue is to examine patient characteristics that may predict PR responsiveness and inform potential adjunctive interventions.

Baseline physical functioning has been shown to predict physical functioning improvement in response to PR.^7,8 For example, one study demonstrated that among patients with interstitial lung disease, lower baseline 6-min walk distance (6MWD) was associated with greater improvement in 6MWD following PR,⁷ and similar observations have been made in patients with chronic obstructive pulmonary disease (COPD).⁸

Associations between baseline psychosocial characteristics such as health-related quality of life (HRQL) (i.e., subjective health impairment encompassing perceived symptom severity, daily activities limited by symptoms, and psychosocial impacts of disease), depression, and PR outcomes have been somewhat inconsistent. Observational studies on patients with COPD have demonstrated that improvements in HRQL correlate with improvements in 6MWD,^9,10 and similar correlations have been observed in patients with non-cystic fibrosis bronchiectasis.¹¹ Relatedly, research has demonstrated that compared to PR program non-completers (i.e., completed ≤ 50% of PR sessions), program completers (i.e., completed > 50% of PR sessions) may have better HRQL prior to starting PR.¹² Given the correlational nature of these studies, research controlling for important covariates (e.g., baseline 6MWD) is needed to make stronger inferences about whether baseline HRQL predicts PR outcomes.^9–11

Several studies also have examined associations between depression and physical functioning over the course of PR. In one study, depression negatively correlated with 6MWD at baseline and follow-up, but associations between baseline depression and Δ6MWD following PR were not reported.¹³ Two other studies found no association between depression and Δ6MWD following PR after controlling for anxiety.^14,15 An additional small study (n=39) found no association between depression and 6MWD at PR entry or at exit.¹⁶ It is possible that patients with depression may be less likely to complete and therefore benefit from PR,^8,17 although some research suggests depression may only predict PR non-completion in women,¹⁸ or not at all.¹⁹ Further research is needed to examine the potential impact of depression on Δ6MWD following PR, particularly as it may affect patients’ capacity to complete PR.

To our knowledge, no studies have simultaneously examined how baseline HRQL and depression associate with Δ6MWD following PR while controlling for relevant covariates. Accordingly, the purpose of this study was to examine associations between the independent variables, baseline HRQL (overall and domains, described below) and depression, and the primary dependent variable, Δ6MWD following PR, controlling for relevant covariates. Additionally, as a supplemental analysis, baseline HRQL and depression were examined as predictors of a secondary dependent variable, probability of completing PR.

METHODS

Data were collected prospectively from patients consecutively enrolled in an American Association of Cardiovascular and Pulmonary Rehabilitation (AACVPR) certified PR program located in Vermont between 2009 and 2022 (Figure 1). Inclusion criteria were based on AACVPR’s “Guidelines for Pulmonary Rehabilitation Programs, Fifth Edition”.²⁰ All patients were ≥ 18 yr, referred to PR with an appropriate diagnosis (e.g., COPD, interstitial lung disease), symptomatic, free of unstable medical problems (i.e., no hospitalization or need for medication for an exacerbation for ≥ 3 wk, with co-morbid conditions corrected or stabilized), cognitively able and willing to participate in PR, able to safely exercise independently and participate in group exercise, and not currently smoking. This study was considered exempt from the University of Vermont IRB due to exclusive use of unidentifiable/de-identified data.

Figure 1.

Flow diagram illustrating the number of patients who were referred, enrolled, and completed pulmonary rehabilitation, reasons for not completing pulmonary rehabilitation, and the number of patients who were included in the main and supplemental analyses.

The PR program was based at an academic network hospital, ran 2 d/wk, 3 hr/d (i.e., 2 hr of PR and 1 hr of education), for 8 wk, and was coordinated by a respiratory therapist and supported by physical therapists. The emphasis of PR was on endurance training and breathing retraining, with individualized treatment based on comorbidities of patients.²⁰ Educational components included medications, energy conservation, nutrition, bronchial hygiene, coping with chronic disease, interpreting lung function tests, and advance care planning. In March of 2020 the program switched to a virtual format to accommodate COVID-19 restrictions and returned to in-center in February of 2021. Note that 84% of 31 patients who participated during the COVID-19 period were ultimately excluded from the analyses for missing data on HRQL and depression; however, 97% of these patients completed the program.

Participants completed assessments at baseline and upon PR completion. A battery of questionnaires and physical tasks were administered. The present study focused on examining relationships between the following variables:

Baseline health-related quality of life (HRQL).

The 50-item St. George’s Respiratory Questionnaire (SGRQ) was used to assess the independent variable baseline HRQL.²¹ The SGRQ total score measures overall HRQL, and three SGRQ subscales assess distinct yet related HRQL domains including subjective symptom severity (symptoms), activities limited by symptoms (activity), and psychosocial impacts of pulmonary disease (impacts). Item formats include Likert-type (e.g., “I cough…” [not at all] to [most d/wk]) and dichotomous response options (e.g., “Coughing hurts” [true/false]). The SGRQ composite and subscale scores range from 0 (no health impairment) to 100 (maximum health impairment).

Baseline depression.

The Geriatric Depression Scale (GDS) Long Form was used to assess the independent variable baseline depression.²² The GDS is a 30-item questionnaire measuring the presence of depression symptoms using dichotomous response options (e.g., “Do you feel that your situation is hopeless?” [yes/no]). GDS scores ranging from 0–9 indicate no depression, 10–19 indicate mild depression, and ≥ 20 indicate severe depression.

Δ6MWD.

The primary dependent variable, Δ6MWD, was assessed using the results of a baseline 6-min walk test (6MWT) subtracted from the results of a 6MWT administered after completing of PR. At each assessment the test was performed once, and the result was recorded in ft and then converted to m. During the COVID-19 remote monitoring period (March 2020-February 2021), we used the most recent 6MWT performed in the pulmonary clinic whenever available. A video 6MWT was performed if a recent clinic-based 6MWT was unavailable (i.e., patients would report to staff the distance of a home-based track, either a loop through the house or back and forth in a hallway, and staff would conduct the test remotely using secured video chat). Aside from this temporary deviation, 6MWT were consistent with European Respiratory Society/American Thoracic Society (ERS/ATS) standards in accordance with AACVPR guidelines.²⁰

Covariates included age, sex assigned at birth, baseline 6MWD, body mass index (BMI), and spirometry tests (i.e., forced expiratory volume and forced vital capacity [% predicted], performed according to European Respiratory Socitey/American Thoracis Sociey standards following AACVPR guidelines),²⁰ time (d) from referral to enrollment, and % PR sessions completed.

Statistical Analyses

Descriptive statistics and bivariate correlations were used to summarize participant characteristics and examine preliminary associations. For the main analysis, multiple linear regression was used to examine associations between the independent variables, HRQL (overall and domains) and depression, and the primary dependent variable, Δ6MWD (Figure 1). As a supplemental analysis, multiple logistic regression was used to examine associations between the independent variables and the secondary dependent variable, probability of completing PR (defined as attending ≥ 10 billable sessions over a 12-wk period, in accordance with AACVPR guidelines)²⁰ (Figure 1). Participants with incomplete data were coded as missing and excluded from the regression analyses.

Because the SGRQ total score (i.e., overall HRQL) encompasses the SGRQ subscale scores (i.e., HRQL domains), it would be redundant to include the SGRQ total and subscale scores in the same models. Therefore, separate models were constructed to examine associations between the dependent variables and SGRQ total and subscale scores. This procedure can help determine whether idiosyncratic relationships exist between HRQL domains and PR outcomes and could potentially inform targeted adjunctive interventions.

For the primary dependent variable, Δ6MWD, three outliers were identified and winsorized to values one unit lower/higher than the next lowest/highest values. All regression models controlled for age, sex assigned at birth, baseline 6MWD, body mass index (BMI), and spirometry tests, and time from referral to enrollment. Additionally, % PR sessions completed was included as a covariate in the linear regression models used for the main analysis, but it was not included in the logistic regression models as it is redundant with the outcome (i.e., probability of completing PR). Backward stepwise elimination, using stay significance level of 0.05, was used to eliminate predictors and identify the most parsimonious models (elimination criteria were not applied to HRQL domains as they represent conceptually similar variables assessed using a single measure). Lastly, variance inflation factors were generated to examine multicollinearity (generally, a variance inflation factors ≥ 4 indicates multicollinearity may exist). Statistical analyses were conducted using SAS 9.4 (SAS Institute, Inc.), and significance for all models was set at 0.05.

RESULTS

Patients were primarily older adults, and more than half of the sample reported female sex assigned at birth (Table 1). The majority of patients entered PR with a diagnosis of COPD. Baseline SGRQ scores and 6MWD generally indicated moderate impairment, although baseline depression symptoms were minimal. Spirometry tests indicated moderate airflow limitation. Bivariate correlations were mostly in line with expectations (Table 2); however, older age was generally correlated with less baseline impairment, which may suggest that older patients with more impairment and/or younger patients with less impairment were less likely to be referred/enrolled.

Table 1.

Patient characteristics and pulmonary rehabilitation program outcomes.

	Mean ± SD [range] or %
Baseline characteristics
Age, yr (n=503)	68.8 ± 11.3 [22–91]
Sex assigned at birth, % female (n=502)	56.5
Pulmonary diagnosis (n=498)
Chronic Obstructive Pulmonary Disease	68.4
Interstitial Lung Disease	15.7
Other (e.g., asthma, bronchiectasis)	14.9
Health-related quality of life (n=457)
SGRQ Total	46.4 ± 15.8 [10.2–100]
SGRQ Symptoms	49.1 ± 21.4 [0–97.7]
SGRQ Activity	67.5 ± 18.4 [11.2–100]
SGRQ Impacts	33.3 ± 16.8 [1.6–86.3]
GDS depression (n=371)	9.6 ± 6.5 [0–29]
6MWD, m (n=476)	274.4 ± 89.6 [36.6–496.8]
BMI, kg/m2 (n=484)	29.2 ± 7 [14.3–55.6]
Spirometry (n=480)
FEV1% predicted	52.2 ± 21.3 [9–128]
FVC% predicted	72.4 ± 18.4 [16–180]
Time from referral to enrollment, d (n=464)	210 ± 180 [2–1258]
Program outcomes
Δ6MWD, m (n=402)	62.4 ± 63 [-262.7–294.4]
% of enrolled patients who completed PR (n=485)	82.9
Mean % of 16 PR sessions completed (n=485)	84.5 ± 18 [0–100]

Note: N=503; the number of patients contributing data to each variable is noted in parentheses next to the variable “(n=###)”. Abbreviations: 6MWD, 6-min walk distance; BMI, body mass index; FEV1, forced expiratory volume during first second; FVC, forced vital capacity; GDS, Geriatric Depression Scale; SGRQ, St. George’s Respiratory Questionnaire.

Table 2.

Bivariate correlations between study variables.

Study Variables	1	2	3	4	5	6	7	8	9	10	11	12
1. Age	.
2. SGRQ total	−.22***	.
3. SGRQ symptoms	−.19***	.73***	.
4. SGRQ activity	−.09	.81***	.45***	.
5. SGRQ impacts	−.26***	.93***	.60***	.64***	.
6. Depression (GDS)	−.15**	.52***	.36***	.37***	.54***	.
7. Baseline 6MWD	−.13**	−.33***	−.17***	−.41***	−.26***	−.12*	.
8. Baseline BMI	−.02	.09	.09	.09*	.07	.05	−.12**	.
9. Baseline FEV1% predicted	.26***	−.26***	−.22***	−.25***	−.20***	−.16**	.19***	.25***	.
10. Baseline FVC% predicted	.26***	−.21***	−.13**	−.21***	−.20***	−.08	.17***	.06	.59***	.
11. % PR sessions completed	.02	−.05	−.11*	−.02	−.03	−.13*	.11*	−.11*	−.01	.03	.
12. Time from referral to enrollment	.01	.08	.12*	.03	.04	.06	−.08	.06	−.10*	.03	−.02	.
13. Δ6MWD, m	−.11*	−.04	.05	−.04	−.04	−.01	−.18***	.03	.01	−.00	.04	−.02

Note:

^*P<.05,

^**P<.01,

^***P<.001. Column heading numbers (1–12) indicate the corresponding study variable (leftmost column). Abbreviations: 6MWD, 6-min walk distance; BMI, body mass index; FEV₁, forced expiratory volume during first second; FVC, forced vital capacity; GDS, Geriatric Depression Scale; PR, pulmonary rehabilitation; SGRQ, St. George’s Respiratory Questionnaire.

Parameter estimates for the final linear regression models are presented in Table 3. There was a significant main effect of overall HRQL on Δ6MWD (F(1, 389)=8.4, P=.004) (Table 3, Model 1). Thus, for every 1-point increase in SGRQ total score at baseline (indicating greater overall subjective impairment), patients who completed PR, on average, experienced 0.61 m less improvement in 6MWD. To probe this association, patients were categorized according to baseline SGRQ composite score quartiles, and a post-hoc analysis of covariance was conducted to examine between-group differences in Δ6MWD. Group averages in Δ6MWD are presented as Least Squares Means (LSM) ± standard errors with Bonferroni adjustment for multiple comparisons (Figure 2). Patients scoring in Q1 experienced the most improvement in 6MWD (LSM=79.0 ± 6.0 m), and significantly more improvement than participants scoring in Q3 (LSM=55.1 ± 6.0 m, P=0.03). Patients scoring in Q2 (LSM=64.3 ± 5.9 m), Q3, and Q4 (LSM=55.9 ± 6.7 m) experienced similar improvement in 6MWD (P>.08).

Table 3.

Main analysis; final multiple linear regression models examining predictors of change in 6-min walk distance from baseline to end of pulmonary rehabilitation.

	B ± SE	P	VIF
Model 1: Overall health-related quality of life		<0.0001	NA
SGRQ total	−0.61 ± 0.21	0.004	1.19
Age	−1.10 ± 0.29	0.0001	1.11
Baseline 6MWD	−0.17 ± 0.04	<0.0001	1.12
Model 2: Health-related quality of life domains		<0.0001	NA
SGRQ symptoms	0.20 ± 0.18	0.27	1.60
SGRQ activity	−0.51 ± 0.23	0.03	1.89
SGRQ impacts	−0.37 ± 0.27	0.17	2.22
Age	−1.08 ± 0.29	0.0002	1.13
Baseline 6MWD	−0.19 ± 0.04	<0.0001	1.23

Note: N=390. Abbreviations: 6MWD, in 6-min walk distance; SGRQ, St. George’s Respiratory Questionnaire; VIF, variance inflation factor.

Figure 2.

Analysis of covariance comparing differences in Δ6MWD by SGRQ total score quartiles. Bars represent Least Squares Means and Standard Errors with adjusted for multiple comparisons (Bonferroni). The difference between Q1 and Q3 was significant at P=.03.

Regarding HRQL domains (Table 3, Model 2), there was a significant main effect of the SGRQ activity subscale on Δ6MWD (F(1, 388)=4.8, P=.03). For every 1-point increase in SGRQ activity score at baseline (indicating greater symptom-related limitation during daily activity), patients who completed PR, on average, experienced 0.51 m less improvement in 6MWD. A second post-hoc analysis of covariance was conducted to examine between-group differences in Δ6MWD. For this analysis, patients were categorized according to baseline SGRQ activity subscale score quartiles. As above, group averages in Δ6MWD are presented as LSM ± standard errors with Bonferroni adjustment for multiple comparisons (Figure 3). Patients scoring in Q1 experienced the most improvement in 6MWD (LSM=79.7 ± 6.7 m), and significantly more improvement than patients scoring in Q3 (LSM=54.4 ± 6.0 m, P=0.05) and Q4 (LSM=48.7 ± 7.5 m, P=0.04), although patients scoring in Q2 did not differ significantly from the other groups (M=70.2 ± 6.1 m, P>.20). The SGRQ symptoms and impacts subscales were unrelated to Δ6MWD (F(1, 388)=1.2–1.9, P>.05) (Table 3, Model 2), as was depression (F(1, 311)=0.0, P>.85) (Supplemental Table 1).

Figure 3.

Analysis of covariance comparing differences in Δ6MWD by SGRQ activity score quartiles. Bars represent Least Squares Means and Standard Errors with adjusted for multiple comparisons (Bonferroni). The difference between Q1 and Q3 was significant at P=.05, and the difference between Q1 and Q4 was significant at P=.04.

Supplemental Analyses (Probability of Completing PR)

Odds ratio estimates for the final logistic regression models are presented in Table 4. There was a significant association between overall HRQL and PR program completion (OR = 0.98, Wald CL, 0.96–1.00, P=.02), such that for every 1-point increase in SGRQ total score at baseline, patients were 2% less likely to complete PR (Table 4, Model 1). There was also a significant association between the SGRQ symptom subscale and PR program completion (OR = 0.97, Wald CL, 0.96–0.99, P=.003), indicating that for 1-point increase in SGRQ symptom score at baseline (indicating greater subjective symptom severity), patients were 3% less likely to complete PR (Table 4, Model 2). The SGRQ activity and impacts subscales were unrelated to PR program completion (P>.40) (Table 4, Model 2), as was depression (P>.30) (Supplemental Table 2).

Table 4.

Supplemental analysis; final multiple logistic regression models examining predictors of pulmonary rehabilitation completion.

	OR (95% Wald Confidence Limits)	P	VIF
Model 1: Overall health-related quality of life		0.01	NA
SGRQ total	0.98 (0.96–1.00)	0.02	1.01
Baseline BMI	0.96 (0.92–1.00)	0.04	1.01
Model 2: Health-related quality of life domains		0.002	NA
SGRQ symptoms	0.97 (0.96–0.99)	0.003	1.60
SGRQ activity	0.99 (0.97–1.01)	0.48	1.70
SGRQ impacts	1.01 (0.99–1.04)	0.40	2.12
Baseline BMI	0.96 (0.92–1.00)	0.05	1.01

Note: N=454. Abbreviations: BMI, body mass index; SGRQ, St. George’s Respiratory Questionnaire; VIF, variance inflation factor.

DISCUSSION

This study demonstrated that baseline overall HRQL was associated with Δ6MWD following PR, and this association may be driven primarily by the degree to which patients report their pulmonary symptoms negatively impact daily activity. More specifically, compared to patients who reported more symptom-related limitation during daily activity, patients who reported less symptom-related limitation experienced a greater increase in 6MWD following PR. This observation is generally consistent with evidence from cross-sectional and observational studies demonstrating similar correlations between HRQL assessed using the SGRQ and 6MWD.^9–11

Overall, the present findings suggest that patients with lower HRQL prior to starting PR could potentially benefit from additional support to help improve HRQL early in treatment. In particular, patients with higher levels of symptom burden may benefit from behavior change (e.g., self-management) interventions aimed at increasing their motivation, competence, and confidence to engage in healthy behavior and disease management.^23,24 Given evidence that such interventions can improve both HRQL and physical functioning in patients with COPD,^23,24 PR programs may consider integrating self-management interventions or other evidence-based therapeutic techniques to enhance self-efficacy to help maximize success of patients.

Although integrating interventions to enhance the self-efficacy of patients could potentially improve PR outcomes, it may be more practical to refer patients to PR earlier (i.e., before impairment becomes moderate or severe). The benefits of PR in less symptomatic patients have been described elsewhere. For example, a meta-analysis of four randomized control trials demonstrated that patients with COPD who had no or mild dyspnea, and participated in PR, experienced greater improvement in 6MWD and HRQoL compared to patients who received usual care only (i.e., medical treatment with or without smoking cessation advice).²⁵ Similar effects were observed in a quasi-experimental study examining the impact of PR on patients with mild COPD.²⁶ The present findings indicate that patients with less symptom burden may experience greater increases in 6MWD following PR completion and may also be more likely to complete PR. Therefore, efforts to refer/enroll patients in PR before their symptoms begin to negatively impact their daily functioning may be warranted.

Although some research has demonstrated that depression may negatively correlate with physical functioning at a single timepoint¹³ and predict PR non-completion,^8,17 the present study found no association between baseline depression and either outcome.^14–16,19 Depression screening and support are standard of care in AACVPR-certified PR programs, and the present results suggest that to experience the full benefits of PR, patients with symptoms of depression may not need additional services that go above and beyond these standards. PR participation is itself likely to confer clinically significant benefits on depression symptoms,²⁷ and future prospective studies examining whether changes in depression mediate the association between baseline HRQL and PR outcomes could be informative. Also note that in this study depression was not assessed at time of PR referral, and future studies examining whether depression at time of PR referral interferes with enrollment may help inform the referral process.

There were limitations to this study and the findings. First, although data were collected prospectively, the present findings were based on a retrospective analysis of a single-center database, predominantly within a COPD population. Relatedly, race and ethnicity were not assessed; however, it is reasonable to speculate that the race and ethnicity characteristics of the present sample reflect that of the surrounding community, which is approximately 94% non-Hispanic, White.²⁸ Replicating the present findings in more diverse samples will be important for informing practice. Second, the older patients in this study may have had less health impairment than the younger participants, which suggests research to examine potential barriers to PR referral/enrollment for older, more symptomatic and/or younger, less symptomatic patients is warranted. Third, the GDS was only completed by a subset of the sample. Although the findings suggest that the GDS is unlikely to be related to Δ6MWD or PR program completion in the present sample, more complete data would be needed to increase confidence in this assertion. Fourth, a number of potentially important covariates were not assessed, such as smoking history (e.g., number of smoking years), comorbid conditions (e.g., heart disease, anxiety), and medications, and future studies on this topic should control for these variables whenever possible. Finally, although it is unlikely that remote PR monitoring adversely affected the outcomes of this study, we were unable to include the COVID-19 remote monitoring period as a covariate in our analyses due to the high amount of missing HRQL and depression data among patients who participated in PR during COVID-19.

CONCLUSION

In summary, baseline HRQL may be an important predictor of PR outcomes, including Δ6MWD and program completion. That is, patients with less subjective impairment (e.g., lower self-reported symptom burden) may show greater improvement in physical functioning compared to patients who have more subjective impairment, and patients with less subjective impairment may also be more likely to complete PR. These findings suggest that strategies to help optimize PR outcomes might include early program referral/enrollment and/or adjunctive psychosocial (e.g., self-management) interventions to help support patients with lower baseline HRQL at the start of PR. Unfortunately, patients who qualify for PR face many barriers to referral such as lack of knowledge of providers of the benefits of PR.²⁹ Long wait lists may prohibit timely access to PR, and qualification and insurance coverage for PR are often based on objective measurements of disease severity and medical necessity rather than subjective symptomatology.³⁰ Efforts may be needed to promote routine assessment of HRQL, particularly along domains of symptom burden, and to hasten the process of PR referral/enrollment for less symptomatic patients in order to avoid missing a potentially favorable treatment window.

Key Perspective.

What is novel?

• Baseline health-related quality of life (HRQL) as indexed by the St. George’s Respiratory Questionnaire (SGRQ) may predict treatment response to pulmonary rehabilitation (PR), with patients scoring lower on SGRQ scales showing greater improvement in 6-min walk distance (6MWD) after completing PR.

• Depression as indexed by the Geriatric Depression Scale (GDS) was unrelated to changes in 6MWD among patients who completed PR.

What are the clinical and/or research implications?

• Research examining strategies to provide additional support to patients with lower HRQL (e.g., adjunctive self-management interventions) or refer/enroll less symptomatic patients could inform efforts to enhance PR outcomes.