An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Jeff Swigris, DO, MS, Susan Baird, PhD, Mark McCormick, Tom Vierzba, Fred Wamboldt, MD, Elisabeth Root, PhD, John Osborne, Dolly Kervitsky, RCP, CCRC, Lyn Cole, Kaci Chacon, and Brenda Crowe.
Author Information and AffiliationsPulmonary fibrosis (PF) is a life-shortening condition marked by progressive shortness of breath, fatigue, and ever-declining quality of life.1,2 For many patients with PF, supplemental oxygen (O2) is prescribed to maintain normoxia and stave off complications of hypoxemia, in the hopes of improving symptoms and maintaining day-to-day physical functioning. Very little is known about O2 in patients with PF—whether and how patients benefit by using it and how they perceive its benefits and hardships.3
To enhance understanding of O2 and its utility in and adoption by PF patients, by examining how they perceive it and by comparing how perceptions and a range of outcomes change from before to after daily O2 use was initiated.
We partnered with PF patients and other relevant stakeholders to conduct a mixed-methods research program. We collected quantitative data at 4 time points (enrollment [T0], the week before O2 [T1], 1 month after O2 [T2], and 9-12 months later [T3]) in a nationwide, pre-post longitudinal study, and we collected qualitative data via semistructured telephone interviews with the aim to enrich findings resulting from the quantitative data.
Of 300 subjects enrolled, 43 were started on O2 by their treating practitioners. Most were White men who had idiopathic pulmonary fibrosis. On average, O2 did not improve dyspnea ratings from immediately before the start of supplemental oxygen to 1 month later (primary end point). However, on secondary analyses, nearly a third of subjects' dyspnea ratings improved by an amount greater than the questionnaire's minimum important difference. The mean 7 months from T0 to T1 showed a trend toward increasing fatigue (slope for Fatigue Severity Scale score 0.41 ± 0.32 points per month; P = .2) that improved significantly by 1 month after O2 (slope −2.5 ± 1.2; P = .03) but was not sustained to 9 to 12 months later. Physical functioning declined significantly among subjects whose need for O2 increased (from exertion only to continuous) from T2 to T3, but not among subjects whose O2 need remained stable (exertion only) during that time.
In the qualitative work, patients and loved ones experienced benefits and hardships from O2 and described ways to make the process of using O2 better. Caregivers described how O2 helped their patient loved ones (PLOs). They also explained how it created several mental and physical challenges for them and their PLOs, stating it forced them to take on more of the physical work around the home; constrained their social participation; and, for many, created strain on their relationship with their PLO.
Supplemental oxygen may improve certain outcomes in PF patients, but significant hardships are associated with its use. Additional research should identify ways to make the process of obtaining and using O2 easier for these patients and their loved ones.
The primary limitation of the longitudinal study was small numbers. We needed 83 subjects to have 80% power to detect a 5-point difference in dyspnea score from T1 to T2. We had only 25% power to detect this difference. We were not equipped to determine if O2 was prescribed as we believe best or if subjects used it as prescribed. Because the study lacked a control group, results should be interpreted with caution.
Pulmonary fibrosis (PF) is a progressive, irreversible disease. While shortening patients' survival, PF insinuates itself into their lives, leaving them breathless and unable to perform physical activities.4 Although nearly every PF patient will be prescribed supplemental oxygen (O2), current knowledge of its effects on patients is almost nil. Limited data, collected from only a handful of PF subjects, reveal that O2 induces immediate improvements in laboratory-based tests of exercise capacity.5,6 The paucity of data has left an expansive gap in evidence for patients and practitioners who need to make informed decisions about O2: No studies have been conducted to examine what patients and prescribers expect PF patients to gain by using O2; to determine whether O2 use creates durable, meaningful improvements in PF patients' daily lives; or to discern whether such putative improvements outweigh patients' perceptions of being “tied to [their] hoses [oxygen cannulas].”7-9 Because thousands of PF patients are prescribed O2 despite a globally insufficient understanding of whether or how it affects them, there is a vital need to conduct studies that aim to discern whether O2 improves symptoms and enhances well-being, to find out whether O2 is—as a patient in 1 of our recently conducted focus groups stated—a “tether” or a “lifeline.” Our overall objective was to enhance understanding of O2—its utility in and adoption by PF patients—by examining how patients perceive it and by comparing how perceptions and outcomes change from before to after daily-use O2 is initiated.
Although understanding of the pathogenesis of PF remains murky, what is clear is that increasing age is a risk factor for PF.10-12 Because the US population is aging, PF will be an expanding health problem for the foreseeable future. Given its estimated US prevalence of about 50 per 100 000 persons,13 PF may not be a rampant burden on the US health care system right now, but it absolutely and unrelentingly weighs down patients. Patients with PF suffer. They have poor quality of life,7,8 and as the disease progresses (and in nearly every case, it will) their ability to perform physical activities dwindles. Many are forced to stop working, and 50% of PF patients face death within 3 years of diagnosis.14
Two drugs are approved by the Federal Drug Administration for a certain type of PF called idiopathic pulmonary fibrosis (IPF).15,16 These drugs slow the disease's progression but do not improve how IPF patients feel or function. To date, no therapy has been indisputably proved to benefit PF patients. Nonetheless, a search of websites from US centers with expertise in caring for patients with PF shows that O2 is a recommended treatment for patients whose peripheral oxygen saturation falls below 89%. Studies to support this recommendation, or studies that systematically examine the effects of supplemental oxygen on outcomes meaningful to patients with PF, such as those proposed above, do not exist.17 As a result, prescribers are severely limited in their ability to confidently inform patients about how O2 will change their lives. In clinics where patients with PF receive treatment, discussions about supplemental oxygen occur daily. In those discussions, prescribers are forced to draw from the chronic obstructive pulmonary disease (COPD) literature—a cache of data that is controversial and limited, and contains dubious applicability to PF patients.18 Perhaps because of the limited information available, most patients with PF rate themselves as “uninformed” on managing supplemental oxygen19—they do not know what to expect while using it, and there is a paucity of data to substantiate the benefits prescribers (the principal investigator [PI] included) hope their PF patients will realize from using O2.
We proposed partnering with patients to define a new standard of knowledge about O2's effects on patients with PF. The interstitial lung disease (ILD) field should not be satisfied with depending entirely on the COPD literature or with making clinical decisions about O2 in PF patients based on assumptions that have been formulated on extrapolations of limited data from patients with a different disease. In this study, we aimed to systematically collect and analyze data from PF patients across the country, to more fully elucidate the effects of O2 on patients' lives. We hoped these data would reveal the benefits of—or call attention to misconceptions about—supplemental O2 therapy for PF patients. Consequently, we expected the ILD field to gain confidence that current conceptions about and prescribing practices for O2 were correct or that conceptions and practices needed to be substantially modified to improve patients' lives. For example, we hypothesized that O2 would improve activity space (ie, expand patients' worlds by allowing them to do more and go more places), but we allowed that finding the opposite could be true: Because of the burdens it imposes (delivery devices are heavy and do not allow patients to stray too far from home), O2 could shrink patients' worlds even more. Either way, we believed conducting a mixed-methods project would generate useful data for PF patients and the practitioners who care for them.
Through qualitative research, 10 years of caring for patients with PF, and working closely with the patients and other stakeholders who formed the research team for the current proposal, we have learned that, by far, the 2 greatest factors impairing quality of life among patients with PF are dyspnea-induced limits on functioning and the burdens of O2.4,7-9,20 The questions this proposal sought to address are patient centered and apply directly to each of the 4 key questions mentioned in PCORI's definition of patient-centered outcomes research.
The central hypothesis of this proposal, synthesized by our team of stakeholders, was that O2 use improves outcomes meaningful to patients with PF. Without this research, prescribers and patients will linger uninformed about the effects of this universally prescribed therapy, and when PF patients and their practitioners set out to make critical decisions about O2 together, they will remain hamstrung by the absence of data to inform their choices. We used mixed methods to pursue 3 aims and achieve our objective: (1) Launch P4f (Patient Participation Program for Pulmonary Fibrosis) from which to recruit subjects for this project; determine the effects of O2 on a full range of outcomes meaningful to PF patients; and (3) identify primary supporters' and prescribers' expectations and perceptions of O2, and for patients, their expectations and perceptions of O2 before and after it was prescribed. This research embraces PCORI's mission to involve key stakeholders, targets PCORI's interest in addressing care for patients with rare conditions and aligns with PCORI's emphasis on studies conducted in “typical clinical populations” and “considering the full range of patient-centered outcomes.” Our research will at long last create a foundation of information that can be used to improve patient/prescriber communication and decision making about O2.
Patients are the foremost key stakeholders for the research we conducted. They live with PF. They suffer with PF. When O2 enters the home, it can affect the relationships PF patients have with their primary supporter and the dynamics of family life. Thus, PF patients' primary supporters are key stakeholders. Other stakeholders include practitioners who prescribe O2 to PF patients, nurses, and other health care team members who work with patients over the course of their disease; PF patient advocacy groups like the Pulmonary Fibrosis Foundation and the Coalition for Pulmonary Fibrosis, which have a keen interest in educating and disseminating information about a host of disease-related topics; PF support group leaders, who have the advantage of reaching patients in person; O2 supply companies; and Medicare, whose O2 prescribing criteria are followed by all US health care payers.
Patients and other key stakeholders (primary supporters, patient advocacy group representatives, support group leaders, physicians, and a nurse with expertise in caring for PF patients) were meaningfully involved since this project's conception (see Table A).
Patient Partner Involvement.
The end-users of our research will most certainly include PF patients, their caretakers, PF patient advocacy groups, PF support group leaders, O2 supply companies, and Medicare and other national health care representatives.
The key stakeholders on our research team were identified and recruited in the following manner: (1) In an attempt to include a broad array of PF patients, we recruited patient partners from the National Jewish Health (NJH) Interstitial Lung Disease clinic, the Joe Walsh Memorial PF Support Group at NJH, and the Wheat Ridge PF Support Group; (2) Dr Swigris leveraged his membership on the Pulmonary Fibrosis Foundation's Medical Advisory Board and a strong working relationship with Ms Kervitsky and the PFF's president and CEO, Dr Dan Rose, to secure the PFF's involvement; we reached out to the National Home Oxygen Patients Association (NHOPA) for information and guidance, and by teaming with NHOPA we ensured that our study will reach the greatest number of patients possible; and (4) we recruited Ms Brenda Crowe, a local PF support group leader.
During support group meetings, O2 is by far the topic that raises the most interest—and questions. Identifying the research question flowed naturally from support group interactions. We used a portion of support group and research-specific meetings to construct the research question and objectives, shape the study design, define the essential characteristics of study participants, and select the specific outcome measures. We selected the PROs we used by reviewing candidate questionnaires for item relevance, missing themes, length, and available response options. The instruments used in this project were selected by consensus and approved by all members of the research team.
Each month, the research team met in person at NJH to discuss progress, review enrollment, and decide what, if anything, needed to change. Stakeholders participated in 6 months of conference calls to PCORI. One example of how our team worked together was, because we were slow to recruit participants, we created a plan to tweak our recruitment strategy. Initially, we had described the study as an investigation of O2, which is true. However, we found out that PF patients who were not on O2 (the ones we were trying to recruit) did not think they were eligible. So, we changed our advertising and recruitment picked up considerably, allowing us to reach our goal.
We have already published 2 methods papers and some of the data from our study. Patient partners are (and perform the requisite work to be placed) on the author line for our manuscripts. We have presented data at international meetings of the American Thoracic Society (ATS) and the American College of Chest Physicians (ACCP). The greatest example of PCORI's influence on our research program occurred when our patient partners presented posters at the ATS and ACCP meetings. After deliberation, the ATS and ACCP determined how our patient partners could register for the meetings. Meeting attendees were surprised to discover our presenters were patients. It gave our patient partners a better appreciation for how these meetings work, and it gave meeting attendees a look into PCORI and patient-centered/patient-active research. There is more dissemination work to do (eg, website posting, grand rounds), and patients will remain actively involved in these efforts.
We used a mixed-methods approach to pursue our aims: We collected quantitative data via questionnaires and other sources, and we collected qualitative data via semistructured telephone interviews and used those data to enrich findings resulting from the quantitative data. First, we launched P4f (the name changed to P3f) and our study website (www.PFresearch.org [link no longer works]) and used the site's database to recruit subjects for the studies. We used a pre-post design and longitudinal study methodology to determine how O2 affected outcomes meaningful to PF patients. We conducted individual, in-depth phone interviews with PF patients, their primary supporters, and their physicians to learn from each of them their perceptions of O2. We conducted multiple interviews with patients to learn how their perceptions of O2 changed over time.
For the longitudinal, pre-post study, we enrolled patients with PF of any cause who had not yet been prescribed O2 for daily use (see Figure 1).
Overview of the Various Facets of the Research Program.
CONSORT Flow Diagram for the Longitudinal Study.
We collected primary data at 4 time points: (1) at enrollment; (2) just before supplemental oxygen was prescribed by subjects' treating physician; (3) after 1 month (± 1 week); and (4) after 9 to 12 months of daily O2 use.
Participants for the longitudinal study were recruited from across the United States via the P4f and internet advertisements for the study on the PF website and its social media outlets, advertisements in ILD clinics across the country, and letters to pulmonary physicians, as well as by posting the study on www.clinicaltrials.gov. Our goal was to recruit a sample representative of the general PF population; this included people of both genders, from several ethnic/racial groups, and across the range of socioeconomic status.
Potential subjects obtained consent forms either from the P4f website or via US mail. Research coordinators conversed with potential subjects via telephone. Potential subjects signed an authorization form that allowed the research team to contact their physician's office to acquire records, including results of pulmonary function tests and imaging studies. We discussed with each patient's physician the methods for the study; specifically, the plan to collect data at the second time point (just before oxygen is started). The physician consented to delaying the patient's start of O2 for 1 week after it is prescribed. This allowed us to collect data for the second time point. The patients we aimed to enroll were those whose PF progressed to the point at which their peripheral oxygen saturation was acceptable at rest (eg, >88%) but fell with activity—and a decision had been made for them to begin daily O2 use. We did not recruit PF patients who were prescribed oxygen for an acute illness. Inclusion criteria included subjects' self-report diagnosis of PF (that was confirmed by the study team via review of chest imaging and medical records), those who were not prescribed O2 for use during the day, those who were able to complete English-language questionnaires, and those with forced vital capacity <75% and diffusing capacity <65% of predicted values. We excluded subjects if their physician did not allow them to wait 1 week before starting O2.
We conducted the study in subjects' home environments. Subjects were recruited from across the United States, and data collection occurred via email and by data capture devices. Devices were mailed to participants at data collection time points, and participants returned them via prepaid delivery envelopes.
There were no study interventions. Decisions about whether and when to start O2 were made by subjects' treating physicians. Patients prescribed O2 were instructed to follow the directions of the prescribing practitioner. Study personnel gave no directions on how O2 was to be used.
Remote data collection occurred via email at 4 time points: (1) at enrollment; (2) just before O2 was started; (3) after 1 month (± 1 week); and (4) after 9 to 12 months of daily O2 use. Questionnaires were emailed to subjects, and wearable devices were mailed to subjects via a shipping service at each time point.
The primary end point was change in dyspnea (as measured by the University of California San Diego Shortness of Breath Questionnaire [UCSD]21) from baseline (pre-O2) to after 1 month of daily O2 use. Secondary end points included change in quality of life (QOL), fatigue, cough, day-to-day functioning (as measured by an accelerometer), and activity space (as measured by a portable global positioning system, or GPS) from baseline to 1 month, and change from the start of O2 to 9 to 12 months later for all of these outcomes.
Outcome measures included the following:
For the longitudinal study, we generated summary statistics for baseline characteristics. We used paired t tests to compare outcomes from just before starting O2 (baseline for the primary end point) to 1 month later and from 1 month on O2 to 9 to 12 months later (secondary analysis). For GPS, accelerometer, and questionnaire outcomes over time, secondary analyses included the development of linear mixed models using visit, forced vital capacity (FVC), age, and gender as key predictors. To account for repeated measures, we included for subjects a random intercept (which induces a compound symmetric covariance structure on repeated measures). We considered a spatial power structure to account for the unequally spaced measures among subjects, but it did not improve the model fit, so we retained the random intercept model. Because FVC measurements were not necessarily collected on the same day as the outcome variables, we matched outcome and FVC values based on nearness of dates of measurement. Models yielded point estimates at the 4 study time points as well as slopes over time.
In ongoing work, we are examining regression calibration to predict FVC for exact dates on which other outcomes were measured. In these models, predicted FVC values are used in place of measured FVC values in the outcome models. We used gender, race, height, weight, age (time varying), and smoking status to predict FVC. We are still refining the predictors, including possibly adding time as a class variable, in addition to age.
For all qualitative analyses, we used grounded theory and a team analytic approach to identify themes, code text, and generate conceptual frameworks for answers to questions posed.
The final protocol is found in the Appendix. Originally, we had planned to enroll only subjects with FVC <75% and DLCO <65% of predicted values, but we amended our approach to include any patient with diffuse pulmonary fibrosis, for many reasons: (1) Recruitment was slow; (2) some patients with combined pulmonary fibrosis and emphysema would be excluded under the initial criteria; and (3) PF is unpredictable and can progress at any time or rate, and we did not want to miss the opportunity to recruit those subjects. The work for this project was approved by the National Jewish Health IRB (HS-2789 and HS-2790).
Results are organized to align with our 3 specific aims as displayed in Figure 1: Aim 1—Launch P4f from which to recruit subjects for this project; Aim 2—Determine the effects of O2 on a range of outcomes meaningful to PF patients; and Aim 3—Identify perceptions of O2 before and after it is prescribed.
Our study website (www.PFresearch.org) became the face of our research program, replete with a physician blog, a patient/caregiver forum, and a conduit for enrollment.
Three hundred subjects consented to participate. The CONSORT diagram shows subject flow throughout the study. Forty-three subjects were prescribed O2 by their physician a median 7 months after enrollment. Of those 43 subjects, 30 contributed data at T1 (the week before starting O2); 5 died or were lost to follow-up; and an additional 13 (who contributed baseline but not T1 data) contributed data at T2. Twenty-five subjects contributed data at both T1 and T2. The 25 subjects who contributed T3 data also contributed data at T2.
The cohort prescribed O2 was predominantly White, and idiopathic pulmonary fibrosis was the most common diagnosis (Table 1).
Baseline Characteristics of Subjects Prescribed O2 for Use During the Day.
There was no difference in the primary outcome of UCSD score (range 0-120) from just before O2 to 1 month later (difference, 0.9 ± 8.7; P = .6). Responder analyses revealed that of the 19 subjects with UCSD scores available at T1 and T2, 6 improved by greater than 5 points (the minimum important difference for the UCSD).
There were trends toward significant improvements in fatigue (difference, 4.4 ± 11.9; P = .1) and the vitality domain of the SF-36 (Table 2). Among subjects with non-IPF diagnoses, O2 improved fatigue and the role emotional domain of the SF-36. Results for other secondary analyses are presented in Tables 3 and 4.
Differences in Outcomes From Pre-O2 to Post-O2.
Modeled Slopes for Change in Outcomes Over Time.
Differences in Outcomes From T2 to T3.
Table 3 shows results for the slope analyses: age-adjusted, gender-adjusted, and FVC-adjusted modeled slopes for change over time are displayed for each outcome for all subjects with available data (30 contributed data at T1, 38 at T2, and 25 at T3). Dyspnea and physical functioning (as assessed by the physical functioning domain of the SF-36 whose range is 0-100) declined from enrollment to just before starting O2. Supplemental O2 improved fatigue (adjusted FSS slope −.5 ± 1.2, P = .03) over 1 month. Fatigue increased from enrollment to T1; after O2 was started, fatigue declined (difference in slopes T0-T1 vs T1-T2 = 2.9 ± 1.4; P = .03). For the entire cohort, from 1 month after starting O2 to 9 to 12 months later, worsening dyspnea trended toward statistical significance. Cough-specific quality of life, general health ratings, and vitality declined. There were no differences in steps per day. On subgroup analyses, there were differences in some outcomes between men and women and between subjects older than 65 vs those younger than 65 (Tables 2a and 2b).
Differences in Outcomes From Pre-O2 to Post-O2 Between Men and Women.
Differences in Outcomes From Pre-O2 to Post-O2 Between Subjects Older Than 65 Years vs Those Younger Than 65 Years.
Progression from needing O2 only with exertion to needing it all the time (even at rest) 9 to 12 months later (ie, T3), was associated with significantly worsened physical functioning and strong trends toward statistically significant worsening general health, mental health and social functioning, vitality, and the role physical domain of the SF-36 (Table 4). There were no such differences or trends among subjects whose O2 needs did not progress (ie, all SF-36 domains remained stable during the 9-12 months from T2 to T3). Dyspnea and fatigue worsened over time regardless of whether O2 needs changed.
To understand how patients viewed O2 before and after starting it, we conducted serial interviews (at enrollment, T1, T2, and T3) with 5 subjects. Subjects had formulated (before O2 was prescribed) and realized (after O2 was prescribed) expectations about the benefits and hardships associated with O2 (Table 5). All 5 expected O2 to help with symptoms, and 4 of the 5 derived the benefit they expected. The hardships associated with O2 use included cumbersome equipment and social stigma attached to being seen with a cannula. Although all 5 subjects had a pulse oximeter, they were not instructed on its use, and self-reported desaturations were frequent.
Hopes/Expectations and Reality of Benefits of Using O2 During the Daytime.
We also interviewed 20 informal caregivers (ICs) of pulmonary fibrosis patients who had been using O2 during the day for at least 8 months to better understand the following: (1) how having O2 in the home affected other members of the household/family; (2) relationships between patient loved one (PLO) and caregiver; and (3) nonpatient views on O2 as helping or holding back their PF PLO. Their baseline characteristics are found in Table 6. IC's initial reactions to PLOs being prescribed O2. Most ICs viewed their PLOs being prescribed O2 in a negative light. Reflecting back, many ICs recalled themselves being “devastated” or “shocked”—it was a time when “everything change[d]” for them and their PLO. One IC said, “[N]either of us ever thought [O2] would be part of our daily lives, and it felt like a really big change in things.” The wife of a patient who had been using O2 for 3 years recalled the first time O2 was delivered to their home as “one of our most depressing days.” Most other ICs saw the prescribing of O2 to their PLO as “stressful,” adding an “emotional toll” and “constant worry” for them. Suddenly, ICs found themselves worried about things like a power outage (a common theme) or running out of O2 when they were away from home.
Demographics of Informal Caregivers.
Some felt “mad” at the disease, or “sad” or “bad” for their PLO because “it [O2] is part of his life now.” The daughter of a patient with PF who had been using O2 for 2 years recalled how “vulnerable” needing O2 made her mother seem to her. Although some ICs mentioned they were initially scared about having O2 in the house or car (because it's flammable), for many, fear arose from “what it [being prescribed O2] meant in terms of his lung status”—it was an “in your face” reminder that “meant he was sick”; an ever-present “thing that just sits there all the time … her need for oxygen.” A few patients were started on O2 at the visit where they were also given the diagnosis of pulmonary fibrosis; this was a particularly “overwhelming” circumstance.
The general theme that emerged was that O2 made PLOs “feel better.” The comment from a husband of a patient on O2 for 3 years captured the theme best: “[W]ith the oxygen, it makes it better. Not perfect, but better.” ICs perceived O2 as allowing PLOs to be more active—“[she] wouldn't be able to do anything [without oxygen]”—even to exercise and participate in activities that they once enjoyed but had given up before starting to use O2 (eg, play golf). ICs also noted improvements in fatigue (“[she] doesn't get as tired as quick”; “[he can] function without … feeling totally exhausted all the time”; “[he's] not … falling asleep all the time”) and energy (“[O2] helps him feel a little more energetic”; “[O2] helped [it] seems like a little bit with his energy”). To ICs, PLOs appeared “more comfortable” when using their O2: They were visibly less short of breath and had “pink cheeks” or “color in his face”—things they didn't have before O2. In 2 cases, PF progression seemed to outpace the benefits that O2 might have conferred, and ICs perceived no improvements at all in their PLOs.
Although most ICs perceived benefits from O2 for their PLOs, they also saw it as limiting. PLOs were better off using O2, but they still had to “drag” or “lug” their O2 around with them. As the daughter of a female patient on O2 for 2 years explained, “it limits her … in where she can go, how long she can go, how far she can go.” Inside, patients were “tied,” “tethered,” or on a “leash,” which was a source of frustration and irritation for them. Cannulas frequently got stuck on something (eg, furniture) and in a couple cases did not allow patients to reach certain areas in their house. Limitations extended outside the home as well, some because of the equipment and some the PLOs self-imposed. The wife of a patient on O2 for 18 months recalled a time when her husband was using a portable oxygen concentrator and their 3-year-old granddaughter “came up to him and she goes, ‘Pops, run with me!’ and he says, ‘Well, I can’t I have to have this [O2],' and she goes, ‘Well, take it out and run with me!’” Outside the home, O2 made PLOs self-conscious (“sometimes people notice it [O2] and that might make him feel uncomfortable”), even to the point of not using it when needed: The wife of 1 patient commented, “He would not be caught dead walking around [grocery store name] hauling a tank behind him.” O2 was a visible reminder to PLOs that they were sick; it made them “feel very frail and fragile.” Some PLOs believed their O2 delivery device was too noisy, so they avoided public places (eg, movie theater) because they did not want to disturb other people. This usually meant ICs did not go, either.
For most ICs, the addition of O2 in the house meant a few extra physical duties related to O2, including filling and/or carrying tanks, loading tanks into the car before leaving home, helping care for and clean equipment, and making sure items (eg, O2 tanks) were ordered on time. Typically, duties were met without resistance, but 1 daughter described being “annoyed” at having to carry tanks for her patient mother (she never mentioned this to her mother), and 1 husband IC had difficulty carrying tanks because his balance was poor. For some ICs, including the husband of a patient who rapidly declined and died after less than a year on O2, duties were more involved: “It was a 24/7 job that consumed all my time I didn't think about doing anything but maintaining, maintaining her oxygen.” The wife of a patient who had been on O2 for 9 years summed up her experience thusly: “[Y]ou have to work everything in your life around the oxygen it's changed [my] life a lot, but definitely not for the better.”
ICs were relegated to literally (and figuratively, in many cases) doing the heavy lifting around the house because they did not want their PLOs to overexert or “exert extra energy.” And they fell into the role easily, as 1 IC mentioned: “I mean, it's my job, you know?” A handful of ICs commented that they had been ill earlier in life and their PLOs took care of them, so it was their turn to take care of their PLOs.
O2 caused IC–PLO pairs to change their lifestyles as couples: O2 “slowed [both of us] down.” It forced the couples to move at a slower pace but also put unwanted limitations on several aspects of their lives.
Spontaneity was no longer an option. Some ICs were silently “frustrated” because their PLOs could not “keep up,” but most had accepted the change. Most pairs, even those in which the PLOs used a portable oxygen concentrator, traveled far less (or not at all) since O2 was started. For pairs in which PLOs required high-flow systems (ie, at least 6 liters per minute of continuous flow), out-of-home activities were particularly constrained. The wife of a patient mentioned how she and her husband brought their dog of 11 years to the veterinarian's office to be euthanized; they had to leave before the dog was put down because the patient's oxygen was going to run out.
Leaving home took some extra time and planning—greater oxygen needs required more thought and preparation. “You can't just fly out the door,” said the wife of a patient on O2 for 18 months. Another IC recalled, “Everything was an expedition if we were going out.” Before leaving, they had to consider where they had to go, how long it would take, and “figure out what your [O2] needs are going to be while you're gone.” Most patients who used a portable oxygen concentrator (POC) took back-up tanks of compressed gas in case the POC malfunctioned or the battery ran out. Three ICs said leaving home with a patient on O2 reminded them of running errands when they had a baby (packing up diapers, bottles, etc).
ICs' homes were different after O2 was brought in. Concentrators were noisy, and it was annoying to have “cords [cannulas]” throughout the house that got tangled and curled up and were “always in the way.” Several ICs referred to cannulas in the home as a “tripping hazard.” In the first few weeks after O2 was prescribed, some ICs worried about the O2 (feared it would “explode”; were concerned because it was “highly flammable”) or power outages, but these concerns did not persist for long.
When O2 was first prescribed, ICs and PLOs found themselves attempting to simultaneously digest what the need for O2 was telling them about the disease's status (it was serious); educate themselves about some of the practical issues about O2, like learning what they needed and figuring out how to navigate the system to get what they needed in a timely fashion; and begin the process of “embracing the new normal.” For the patient, ICs saw how O2 induced great uncertainty, representing a “major change in my health status and [left them questioning] how I'm going to be living my life.” Over time, ICs accepted and adapted to the change and developed strategies for living. But getting to this point took time and effort. ICs used several methods to cope, adapt, and “deal with whatever need[ed] to be dealt with.” A wife of a patient on O2 for 2 years commented, “It was hard at first, just getting used to it, just the adjustment period at first.” Although only 1 IC mentioned using a “team effort,” it was clear that ICs and PLOs worked together, relying on each other to adapt to the new life with O2. IC–PLO pairs had to find balance in their lives—it was impossible for them to completely disregard the presence of O2, but they realized they desperately wanted and needed to continue to live.
The leash and limitations never left; couples just learned to “not fight it,” “relax into it,” and deal better emotionally and physically with the limitations: “He has to live too, so it's balancing things out.” Couples felt a strong motivation to maintain some sense of normalcy in their lives by getting out, remaining socially active, “enjoy[ing] every day,” and “enjoy[ing] life as much as you can.” A wife of a patient on O2 for 3 years said, “Well, we don't ever think of it as good or bad; it just is what it is and you deal with it.” Putting forth a conscious effort to maintain a positive outlook was vital. Some mentioned that keeping a sense of humor, laughing, and joking were important as they moved toward acceptance of O2. Acceptance required “chang[ing] the way you do things” or “taking baby steps” to prove to themselves that they could still live their lives as a couple. They—often subconsciously—adjusted routines around the home. One IC mentioned she did the cooking and her PLO did the preparing (eg, chopping, mixing) of food. Ultimately, most ICs found that dealing with O2 eventually became “so routine and so much a part … of the way we live now” and although “it's certainly different and somewhat cumbersome … it's doable.” ICs of patients with progressive pulmonary fibrosis were required to reaccept and readjust as the PF worsened and oxygen needs increased. Increasing O2 requirements added physical limitations for PLOs, various duties and restrictions in social activities for ICs, and the need to conform to the new normal for them as couples. As 1 IC stated, “Before he was on it [O2] 24/7, it didn't seem like a big deal.” ICs of patients who used O2 24/7 but whose needs could be met with pulsed flow still faced challenges, complexities, and constraints, but seemingly not as many as ICs of patients who required continuous high-flow O2. For most ICs, the prospect of their PLO's health worsening was ever looming, as these quotes from ICs depict: “You're kind of looking down the road … how long will we be able to continue to [use a pulsed flow regulator], and so … it changes your lifestyle, that's for sure”; “When I think about people who have much higher O2 needs than he does, that the chance of them leaving the house would be slim … that's a little frightening.”
In this mixed-methods research program, we collected a cache of qualitative and quantitative data from patients with pulmonary fibrosis to advance understanding of the benefits and hardships associated with O2 use. To our knowledge, this is the first study to assess O2-associated change over time for a range of outcomes in patients with PF. The quantitative data confirmed variability in the response to O2—both the benefits and the hardships. Even in the underpowered longitudinal study, we observed how O2 was associated with improvements in certain symptoms (fatigue) and domains of functioning (social functioning). In many cases, O2 conferred no benefits, but if the disease remained stable, despite O2, outcomes did not worsen. In other cases, disease progression outpaced the benefits O2 could deliver, thus magnifying the hardships of O2.
The rich qualitative data generated give a granular, personal view of the myriad ways O2 affects PF patients. These data also reveal that few, if any, patients use O2 exactly as prescribed and that O2 prescribers' goal of having their patients maintain normoxia at all times may be unrealistic. Patients' and caregivers' perceptions show how big of a deal O2 is—a life-changer for both. Trying to manage O2 (particularly early on) frequently left patients and their caregivers feeling ignorant and overwhelmed with nowhere to turn.
Perhaps the greatest benefit of the data from this research program is that they shed light on a better path forward: Patients and their loved ones would benefit from more readily available, trustworthy, lay-friendly education about O2; O2 prescribers and suppliers could ease the transition to O2 through more detailed conversations about what to expect from O2 (ie, the possible benefits and hardships patients and their loved ones might encounter); and reliable support resources that O2 users can turn to in times of need should be developed.
In isolation, the results of the various facets of this research program will not (and are not meant to) convince PF patients that O2 is a magic bullet against their disease. Nor can the results speak to whether O2 can prevent or forestall the onset of conditions associated with chronic hypoxemia, like pulmonary hypertension and cognitive dysfunction. But, in aggregate, they can help guide PF patients and their practitioners through informed decision making about O2. In the following paragraphs, we use PCORI's 4 key questions for patient-centered research to discuss the specific ways this research advances the field.
“Given my personal characteristics, conditions, and preferences, what should I expect will happen to me?” The longitudinal study did not have adequate power to allow us to make definitive conclusions about the benefits of O2 at the group level. We needed 83 subjects to have 80% power to detect a 5-point difference in dyspnea (UCSD score) from immediately before starting O2 to 1 month later. We had pre-post UCSD data for 19 subjects, giving us only 25% power to detect this difference. However, 6 of those 19 had a 5-point or greater improvement in dyspnea (see Figure 3), and 7 had a greater-than-10-point improvement in fatigue after O2 was started.
Abbreviation: UCSD, University of California, San Diego.
In contrast to our hypothesis, O2 did not lead to increased day-to-day activity (as measured by triaxial accelerometer). It may be that the improved oxygenation that comes with O2 use (something we did not measure in the longitudinal study) may be offset (or overshadowed) by the added physical weight of the O2 delivery apparatus. Additional research is needed to clarify this and identify ways to mitigate it if true.
Because this was a real-world study, we had no control over O2. Thus, it is unknown whether O2 was prescribed or used as scientific rationale would suggest is best (ie, to maintain peripheral oxygen saturation >89% at all times). Additional research is needed to settle this issue.
Over the first year after O2 was started, subjects with IPF—the most common form of PF and, unfortunately, the most lethal—were the ones most likely to have their O2 needs escalate (eg, needed O2 at rest). That progression was associated with significantly worse physical functioning and trends toward worsening in several other domains. In contrast, physical functioning did not worsen among subjects whose O2 needs remained stable during the first year. The qualitative data suggest PF patients (and their loved ones) adapt reasonably well to O2; the frustration and worsening outcomes lie in the need to constantly readapt as the disease progresses. For IPF, there are now 2 FDA-approved antifibrotic drugs that slow disease progression (as defined by decreasing the negative slope for forced vital capacity over 52 weeks). As more effective drug therapies emerge (eg, those that halt the disease), we can anticipate fewer cases in which disease progression outstrips the ability of O2 to maintain how PF patients feel and function.
So, in response to the question posed, PF patients who decide to start using O2 have a decent chance of realizing improvements in various domains, particularly fatigue—one of the most troublesome but underappreciated symptoms of PF. But PF patients must also recognize—and the medical community should work to communicate more clearly—that O2 will likely change things for them, their loved ones, and their home lives.
“What are my options and what are the potential benefits and harms of those options?” Currently, O2 is typically recommended to patients with PF if oxygen saturation falls below 89%, but patients can (and some do) refuse to use it at all—predominantly because of the stigma they perceive is attached to using O2 in public. To our knowledge, no harms are associated with O2 use. Aside from the potential benefits described above, the theoretical benefits of preventing long-term complications of hypoxemia remain. One of the subjects we interviewed before and after starting O2 captured the sentiment most subjects had about balancing the hardships and the benefits: “The tradeoff is well worth it.”
Subjects and caregivers spoke loud and clear about the challenges of using O2.
Although not physically harmful, the use of O2 brought physical challenges for patients (who had to carry delivery devices) and their caregivers who chose—or were forced by circumstance—to embrace more physically active roles around the home. The use of O2 also strained the patient–caregiver relationship and had untoward mental health effects on some patients: They felt stigmatized and did not like how wearing the cannula created a constant reminder of their disease.
“What can I do to improve the outcomes that are most important to me?” By enlisting the input of patients in all facets of this research, including the selection of outcome metrics, this study was guaranteed to assess outcomes important to PF patients. Although more research is needed, and the longitudinal study was underpowered, its data, in aggregate, suggest O2 offers PF patients the possibility to feel and function better.
On slope analysis—which, for multiple reasons, many would argue is a more sophisticated and robust method for analyzing the type of data we collected in the longitudinal study—O2 was associated with improvement in fatigue. For many PF patients, fatigue is the most intrusive, bothersome symptom.9
A key point that PF patients and O2 prescribers must keep in mind is that the manifestations of disease progression can (and often do) overshadow the benefits that O2 might confer. When PF progresses, and oxygen desaturation becomes frequent and/or profound, O2 needs rise. As flow requirements increase, portable oxygen concentrator batteries and compressed gas tanks run out sooner. Patients and loved ones who are accustomed to traveling or being out and about become acutely aware of the constraints.
And the blame is often placed on the O2. True, longer-lasting POC batteries and alternative delivery devices would help matters, but the root of the problem is the disease, not the O2. Getting better at halting disease progression—an incredibly lofty goal—would go a long way toward solving many problems related to O2 use.
“How can clinicians and the care delivery systems they work in help me make the best decisions about my health and health care?” Concerning O2 in PF, prescribers must discuss realistic expectations and potential hardships and the medical community must have adequate educational and support resources for PF patients contemplating or starting O2.
The published literature on O2 for PF is extremely limited. A 2016 Cochrane review identified 3 studies (enrolling a total of 98 subjects, all with IPF) that met its inclusion criteria.32 All studies assessed the effect of O2 on exercise capacity. The Cochrane review identified no study in which investigators examined symptoms or quality of life over the short or long term. To our knowledge, ours is the first study to do so.
In a recently published, landmark study of O2 in patients with COPD who had moderate oxygen desaturation at rest or with exercise (the LOTT trial), the investigators found that O2 did not prolong survival or provide durable improvements in measures of submaximal exercise, symptoms, or quality of life.33 However, these results must not be extrapolated to patients with PF, who typically desaturate more rapidly and more profoundly than COPD patients.
Prescribing O2 for a PF patient only to maintain SpO2 above 88% for 2 minutes of walking—as was the protocol in the landmark LOTT trial—is an unlikely safeguard against profound desaturation in ensuing minutes.
In previously published work, our group coined the term “shrinking world syndrome” to describe the effects of O2 on PF patients and their caregivers.3 The results from the current study build on that work and highlight how many issues connected with O2 use might be improved.
Not applicable.
One of the strengths of this research program is that it enrolled patients from across the United States. Inclusion criteria were left broad to enroll patients across the country and along the continuum of disease severity. Because PF is a relatively rare condition, most subjects were recruited from the centers of excellence where they receive their care. In the end, some degree of tertiary referral bias may be present, but based on prior observational studies and drug trials, the cohort is likely representative of the PF population at large.15,16,34-40
Given the small number of subjects, subpopulation analyses should be considered exploratory. Recognizing this, women's fatigue scores improved to a statistically significant degree in response to O2 initiation, whereas men's fatigue scores did not change, and the difference in fatigue change scores between women and men was statistically significant. Despite a decline of more than 1000 steps per day, younger subjects (those under 65 years) experienced improvements in certain quality of life domains.
There are limitations to the inferences that may be drawn from the results of this study. The primary cause of those limitations is subject number. To account for death and dropout, we enrolled 300 subjects, aiming to observe 83 whose physicians would prescribe O2 for use during the day. Fortunately for the patients, but unfortunately for us, the actual number was far lower, thus significantly limiting the power of our analyses. Among subjects with data at any time point, only 19 had questionnaire data available for the primary analysis and for secondary analyses using t tests. Thirteen subjects had T2 but not T1 data (ie, we had data from 1 month after O2 was started but not from immediately before starting O2). We attempted to keep track of patients' clinic visits, so we could check whether O2 had been prescribed (or was being contemplated). Unfortunately, in these subjects, O2 was started without our knowledge. We elected to keep them in the study, knowing they would be deleted case-wise from t test analysis for the primary end point but could contribute data to the secondary slope analyses. All 43 subjects who were prescribed O2 contributed data for at least 1 time point for the slope analyses.
Based on clinical experience, scientific rationale, and a limited supply of published data, we believe the benefits of O2 are best realized when it is used to maintain a peripheral oxygen saturation >89%. We had no control over how O2 was prescribed or what, if anything, subjects were told about how to use it (eg, whether to use a pulse oximeter to adjust flow). Nor did we possess the capability to assess whether O2 was used as prescribed. This was not a study of “perfect use” O2—if there is such a thing. A meter/detector that tracks O2 flow into the nose combined with an apparatus capable of real-time flow adjustments to maintain a prespecified SpO2 goal would be required.
By design, the longitudinal study was unblinded; subjects knew they started O2, and they also knew when they were wearing accelerometers and GPS units. Knowing about either or both elements could have influenced results. The pre-post design allows for subjects to serve as their own controls, so bias related to the wearables was less of a concern. However, we had no formal control group per se.
We did not track drug-related side effects during the study. In October 2014, 2 drugs were approved by the FDA to treat IPF. Both have significant potential side effects, the presence of which could have negatively affected certain symptoms (ie, fatigue) and/or quality of life. Seven subjects with IPF were taking 1 of the 2 FDA-approved antifibrotic drugs, but whether and how those drugs affected outcomes is unknown. Likewise, subjects with non-IPF pulmonary fibrosis may have been receiving treatment with potent, immunosuppressive/modulatory drugs that, like the antifibrotics, have potential side effects.
Future research should seek to discover ways to make O2 easier for patients to use. A delivery device that integrates real-time, biometric data and automatically adjusts to maintain a prespecified heart rate or SpO2 target should be developed and tested, as should remote controls for regulators and concentrators.
In this pioneering research effort, we observed that, among patients with PF, O2 prescribed for use during the day was not associated with improved dyspnea, cough, quality of life, day-to-day physical activity, or activity space. However, on slope analysis, O2 was found to significantly improve 1 of the most distressing symptoms of PF—fatigue. Within the cohort were subjects for whom O2 improved dyspnea and the other patient-reported outcomes, but the small sample size limited our ability to confidently identify predictors.
Subjects with IPF appeared to derive the least benefit from O2; that likely stems from the progressive nature of their disease. Most subjects with IPF were started on O2 only with exertion, but over the course of 9 to 12 months, they required it at rest as well. The qualitative data we collected reinforces that the leap from needing O2 only with exertion to needing it at rest is quite large.
The qualitative data also remind us that O2 affects patients' loved ones and changes the entire home environment—for most PF patients, O2 is a very big deal. Prescribers should better manage patients' expectations about O2; it will not alleviate dyspnea, but it may help considerably, and it has the potential to positively affect several other domains important to PF patients.
For the individual patient (and his or her caregiver), the decision of whether to use O2 when a prescriber deems it necessary should be made in partnership with the prescriber and after a thoughtful discussion of the potential benefits and hardships. The results of this study tell us that the benefits of O2 may outweigh the hardships, but not for some patients. Patients could consider a trial of O2 to make the determination. As a system, improvements could come through better educational and support resources. Future research should focus on whether strict adherence to maintaining normoxia leads to improved outcomes.
Research reported in this report was [partially] funded through a Patient-Centered Outcomes Research Institute® (PCORI®) Award (CE-12-11-4134). Further information available at: https://www.pcori.org/research-results/2013/evaluating-whether-oxygen-treatment-helps-people-pulmonary-fibrosis-breathe
Swigris J, Baird S, McCormick M, et al. (2018). Evaluating Whether Oxygen Treatment Helps People with Pulmonary Fibrosis Breathe Easier. Patient-Centered Outcomes Research Institute (PCORI). https://doi.org/10.25302/11.2018.CE.12114134
The [views, statements, opinions] presented in this report are solely the responsibility of the author(s) and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute® (PCORI®), its Board of Governors or Methodology Committee.
This book is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License which permits noncommercial use and distribution provided the original author(s) and source are credited. (See https://creativecommons.org/licenses/by-nc-nd/4.0/
Your browsing activity is empty.
Activity recording is turned off.
See more...
