We read with interest the study of the development of the second systemic lupus erythematosus (SLE)–specific quality of life (QOL) instrument, the LupusQoL, in a recent article in Arthritis Care & Research (1). We offer some observations on the LupusQoL and address the remarks the authors made about our questionnaire, the SLE-specific QOL instrument (SLEQOL) (2).
The authors stressed that the items in the LupusQoL were generated through interviews with patients. Although this approach is correct, we see a few problems in the LupusQoL that may have resulted from this.
Because the patients involved in the qualitative study were women, the activities cited seem to carry less relevance for men; for example, decorating, moving furniture, vacuuming, ironing, shopping, cleaning the bathroom, cooking, opening jars, dusting, childcare, and housework. Men may find difficulties in answering the questions that contain such activities.
The authors did not verify if the items suggested by their patients cover the entire spectrum of QOL issues of lupus patients (content validity). As a comparison, the 12 concepts of relevance to patients with SLE reported in another qualitative survey are uncertainty/unpredictability of lupus, fatigue, pain/symptoms, social support, misunderstood by others, fear, dependence/feelings of inadequacy/loss of self, limitations/restricted activities, personal self-management, medical treatment, emotional stress, and financial issues (3). Some of these concepts (such as financial issues, being misunderstood by others, and medical treatment) are not represented in the LupusQoL.
The review team may not have edited the patients' statements adequately, because some questions express 2 ideas. For example, question 9 states, “I am prevented from performing activities the way I would like to because of pain due to Lupus.” If the patient had been prevented from performing activities because of breathlessness, fatigue, or lack of motivation, this fact would not be captured. Also, if the patient experiences pain that causes psychological but not physical impairment that would be missed as well. Other examples are questions 10, 11, and 15. QOL instrument designers generally try to avoid having dual ideas in a single question (4).
The method for item reduction is not clearly documented. We note that a few pairs of items are so similar that one may be eliminated without compromising the questionnaire. If the authors use a combination of factor analysis and Rasch analysis, they will be able to identify the redundant items. The authors report that Cronbach's alpha indicates good internal consistency within the domains, which is not surprising if the items are close in meaning. Five pairs of similar-meaning items include questions 9 and 10 (both regarding pain-inhibiting activities), questions 23 and 24 (“anxious” and “worried” are very close in meaning in conventional use), questions 13 and 14 (unpredictability of lupus), questions 15 and 16 (both about sex), and questions 17 and 18 (both about stress for others).
Determining test–retest reliability usually involves readministering the test in 2 weeks, while the patients remain in virtually the same disease status. This duration is traditionally chosen because it is sufficiently long enough for patients to forget their original responses and short enough for the disease state to remain unchanged (5). The authors employed a 4-week interval and selected only those respondents who reported that their disease remained unchanged based on an independent 15-point scale. This is part of the process of demonstrating responsiveness (determining the “signal-to-noise” ratio), but is not, strictly speaking, an equivalent to determining test–retest reliability.
The authors raised 2 points about the SLEQOL. They stated that body image is not represented in the SLEQOL. This is not accurate, because question 28 of the SLEQOL deals with self-consciousness, question 33 with being the subject of ridicule, and question 35 with embarrassment. The authors also suggested that the 100 patients who reviewed the SLEQOL items generated by healthcare personnel did not add any items because of their reluctance to challenge the health professionals' ideas. The patients did in fact suggest certain issues, but they were already present in the SLEQOL. Similarly, the 20 patients who assessed the draft LupusQoL also added no new items. In conclusion, we have made a few suggestions on ways of improving the LupusQoL and addressed 2 issues about the SLEQOL raised by the authors.