A New Tool to Capture Patients' Perceptions of the Effects of Lung Transplantation.

For patients with many incurable, life-shortening lung conditions, lung transplantation offers “curative” therapy and the only therapeutic option with a reliable chance to improve their quality of life (QOL). But what is really meant by the term “QOL,” and how do we measure this abstract construct?

For individuals, QOL refers to their perceptions of how well their needs and wants are met across dimensions of life that matter most to them. Thus, accurate assessments of QOL require knowledge of patients’ perceptions, their needs and wants, and the dimensions of life they care about. Carefully crafted questionnaires, developed with systematically collected input from patients with the condition of interest, can capture all these things.

Until now, investigators and other stakeholders interested in examining the effect of lung transplantation on a person’s QOL have had to rely on existing questionnaires (1–3). However, none of those questionnaires adequately address all of the domains that are important to patients after lung transplantation, including, among other things, emotional well-being, pulmonary and extrapulmonary symptoms, effects of immunosuppression, and neurocognitive status. In fact, some experts say that many existing questionnaires—even ones that call themselves “QOL” or “health-related QOL” questionnaires—are really measures of health status. They measure certain domains (and do it well), but by confining themselves to assessing physical and/or emotional health, they fail to capture other domains that contribute to a person’s QOL, including physical and emotional well-being as well as material comforts; relationships with a spouse/partner, family, and friends; being able to help other people; learning; independence; working; understanding oneself; expressing oneself; and socializing (4, 5).

In this issue of the Journal, Singer and colleagues (pp. 1008–1019) describe the development and initial testing of a questionnaire that aims to assess QOL after lung transplantation (6). Rather than developing their own items de novo, these lung transplant and questionnaire-development experts used clusters of items from existing questionnaires to capture aspects of life after lung transplant. They initially identified 126 items comprising clusters from eight questionnaires. Next, they conducted 43 cognitive interviews with lung transplant recipients who were asked whether the words and grammar used for individual items were clear, and whether there was redundancy in the items or clusters. For clusters that tapped the same construct (e.g., respiratory symptoms), interviewees were asked to select which item cluster more clearly reflected their definition of health-related QOL. Forty-two items were dropped after the interviews, leaving 84 items for field testing. Results from an exploratory factor analysis led the investigators to drop another 24 items. This left their questionnaire, which they call the Lung Transplant Health-Related Quality of Life (LT-QOL) survey, with 60 items comprising 10 scales (pulmonary symptoms, gastrointestinal symptoms, neuromuscular symptoms, treatment burden, worry about future health, cognitive limitations, sexual problems, anxiety/depression, health distress, and general QOL).

Analyses showed that the LT-QOL has acceptable-to-excellent psychometric properties, ranging from internal consistency to floor and ceiling effects. In support of its validity, its scores correlated in expected directions with relevant domains from other questionnaires completed in the same sitting. Its scores also correlated, as hypothesized, with concurrently collected spirometric and walk-test data, and LT-QOL scores were significantly worse for respondents with severe chronic lung allograft dysfunction than for those without. These analyses suggest that the LT-QOL is a reliable questionnaire, and it would seem to assess domains that are meaningful to patients after lung transplantation. In short, it has easily passed the first psychometric hurdle. In my opinion, the methods used to develop this questionnaire were sound, and the numerous cognitive interviews with patients in the target population ensure its relevance.

The LT-QOL has some limitations, and there is more work to be done on it. Although the authors describe how the LT-QOL is more comprehensive than existing questionnaires, its 10 scales cover only five broad constructs: 1) physical health, 2) emotional/mental health, 3) treatment burden, 4) sexual health, and 5) general QOL. Measurement experts may argue that the LT-QOL does not measure health-related QOL. As alluded to above, inquiring about the frequency or intensity of symptoms (physical or emotional/mental) gets at health status, but that line of inquiry is subtly different from specifically asking how much those symptoms—or the condition in all its aspects—affects the full spectrum of life domains that feed into QOL. I would call the LT-QOL a hybrid health-status/QOL questionnaire developed specifically for the post–lung transplant population. Its last two items—which I really like (“I am able to enjoy life” and “I am content with the quality of my life right now”)—tap general QOL, but even they do not reveal whether or how the transplant (and all it involves and impacts) has affected the respondent’s QOL or ability to enjoy life. It is important to emphasize what makes this questionnaire distinct, as I suspect many investigators who will use the LT-QOL will not have expertise in measurement and will not review its content in detail. Typically, they would, I suspect, see its name and assume it captures a comprehensive view of QOL after lung transplantation. They should know what they’re getting by using it.

There may be some hidden redundancy in the LT-QOL: 10% (n = 6) of the items (from three different scales) include the term “worry,” “worried,” or “worrying” (e.g., “I worry that my lung transplant will not work . . . about getting infections . . . that my health will get worse . . . about not being able to stop or control worrying,” “worrying too much about different things,” and “is health a worry in your life?”). But redundancy only lengthens the questionnaire; it does not detract from its psychometric soundness.

This work suggests that the LT-QOL is a nice new tool that is capable of capturing patients’ perceptions of the effects of lung transplantation, and it provides a foundation upon which to conduct additional analyses to support the validity of this tool. Remember that establishing the validity of a questionnaire is an ongoing process, not a threshold phenomenon, and involves using well-formulated hypotheses to test that questionnaire in many studies, under multiple conditions, to understand what its scores are able to reveal about respondents. The LT-QOL is primed and ready for use in observational and interventional studies. The investigators have placed the LT-QOL in the public domain, so others are free to use it. They suggest administering a generic QOL questionnaire alongside it. The next phase of analyses should assess how the LT-QOL tracks the post–lung transplant course, establish its responsiveness to changes in health status as time passes after transplantation, and determine its sensitivity for detecting differences in the health-status/QOL trajectory between groups over time.