The EQ-5D (Euroqol) is a valid generic instrument for measuring quality of life in patients with dyspepsia

Ethical approval was obtained by the University Malaya Medical Centre Ethics Committee prior to the commencement of this study. Consecutive adult outpatients with dyspepsia attending the Gastroenterology clinic of this tertiary teaching institution were invited to participate in the study. Most of these patients had functional or non-ulcer dyspepsia (defined as dyspepsia with a normal or minor endoscopic findings), with the remaining attending follow up for organic disease (endoscopic findings of duodenal erosions, peptic ulcer disease and erosive oesophagitis). Patients were interviewed by a trained research assistant using HRQOL and disease severity instruments validated for use in Malaysia (see below) in English or Malay (depending on individual patient's language proficiency). Additional information obtained included period of medical consultation and socio-economic-demographic status. Test-retest reliability of EQ-5D was evaluated by administering the EQ-5D twice to the same subjects, 2-weeks apart, and comparing the scores obtained on these two occasions. The second interview was conducted over the telephone by our trained research assistant.

The EQ-5D comprises five questions on mobility, self-care, pain, usual activities and psychological status with three possible answers for each item (1 = no problems, 2 = moderate problems, 3 = severe problems) [7]. An overall utility score is calculated based on these domains, with a range score from 0 (worse health scenario) to a maximum of 1.0 (best health scenario). An additional visual analogue scale (VAS, scale 0 – 100) is used to assess general health status with 100 indicating the best health status. Malaysian English and Malaysian Malay versions of the EQ-5D were developed by the EuroQoL Group: 2005 (original developers) using their standard translation and linguistic validation process [10] and have been validated for use in Malaysia.

The SF-36 is an established generic HRQOL instrument which comprises 36 questions and eight different sub-scores: physical functioning, physical role limitations, bodily pain, general health perceptions, vitality, social functioning, emotional role limitations, mental health and 2 composite scores – Physical Component (PCS) and Mental Component Scores (MCS) scored on a 0 to 100 scale [11]. The maximum score of 100 indicates the best possible health state. English and Malay versions have previously been validated in Malaysia and shown to be a reliable measure of general QOL health status [12].

The Short Form (SF) Nepean Dyspepsia Index is a disease-specific HRQOL instrument for dyspepsia, comprising a 10-item questionnaire examining the influence of dyspepsia on five elements (sub-scales) of health in patients, namely tension/anxiety, interference with daily activities, disruption to regular eating/drinking, knowledge towards/control over disease symptoms and interference with work/study [13]. Each item is measured by a 5-point Likert scale ranging from 0 (not at all or not applicable), 1 (a little), 2 (moderately), 3 (quite a lot) to 4 (extremely). Individual items are subsequently computed to obtain a score range from 0 (lowest HRQOL score) to 100 (highest HRQOL score) as per the developers' original calculation formula [5]. A total, overall SF-NDI score is obtained by the summation of all sub-scale scores followed by division of five. This instrument has previously been shown to be valid, sensitive and reliable for measuring HRQOL status in Malaysian patients with dyspepsia [14]

The Leeds Dyspepsia Questionnaire (LDQ), is an eight item symptom-based questionnaire relating to the frequency and severity of various upper G.I. symptoms, namely upper abdominal pain/discomfort, heartburn, regurgitation, dysphagia, belching, nausea, vomiting and post-prandial distension/early satiety [15]. The total score ranges from 0 – 40, with lower values indicating less severity and higher values more severe dyspepsia. A score of 15 or more had been defined by the developers as indicative of severe dyspepsia. The questionnaire has previously been validated in our local population and shown to be reliable in assessing dyspepsia amongst Malaysians [16].

Psychometric properties of the EQ-5D were evaluated by assessing its frequency of missing data, ceiling and floor effects (i.e. percentage of samples achieving highest and lowest values), test-retest reliability and construct validity. The reliability of an instrument refers to its ability to yield reproducible and consistent results. Construct validity is an assessment of the degree to which an instrument measures the construct that it was designed to measure. The process involves forming hypothetical models to describe the constructs being assessed and postulating their relationship. An assessment of the relationship between the construct data is then made to confirm or refute prior expectations and hence determine the validity of the instrument being tested [17].

Construct validity of the EQ-5D was assessed by testing 26 a-priori hypotheses related to item or utility scores as follows: a) patients with less favourable demography in relation to dyspepsia would report more problems with relevant EQ-5D dimensions, b) patients with more severe dyspeptic symptoms would have more problems with relevant EQ-5D dimensions and lower utility scores c) subjects reporting problems for any EQ-5D dimensions would have lower scores for relevant SF-36 and SF-NDI scales and d) utility scores for EQ-5D would correlate well with SF-36 and SF-NDI summary scores. Test-retest reliability of EQ-5D was evaluated by administering the EQ-5D twice to the same subjects, 2-weeks apart, and comparing the scores obtained on these two occasions. The first interview was conducted face-to-face and the second interview was conducted over the telephone, both by our trained research assistant.

Hypothesized trends were tested using Chi-square or Mann-Whitney tests, or Spearmans' correlation coefficient where appropriate. Strong, moderate and weak correlations were defined as > 0.60, 0.30 – 0.60 and < 0.30 respectively [18]. Test-retest reliability of the EQ-5D utility score was assessed using intraclass correlation coefficients, with a desired value of > 0.6 [17]. Statistical significance for hypothesis fulfillment was defined as a p value of < 0.05. Data were analysed with SPSS for windows (version 12, SPSS Inc, IL, USA).

A total of 113 consecutive patients with dyspepsia were interviewed between May and October 2006 and their characteristics and basic demography are summarized in Table 1. The mean age of patients was 53.7 ± 14 years, the male:female ratio was 1:1.02 and their ethnicity were as follows: 28 (24.8%) Malays, 42 (37.2%) Chinese and 41 (36.3%) Indians. 90.3% of patients had above secondary level education and most patients (41.6%) were retired at the time of interview. The majority of cases (70.8%) were functional dyspepsia and the median period of medical consultation (either primary care or hospital specialist) was 6 years (range 3 – 18). The median LDQ score in patients was 16 (11 – 22), indicating that most patients had fairly severe or poorly-controlled symptoms despite many years of treatment.

The response rate was 100% with no missing data for all variables. The calculated utility score had a median value of 0.725 (skewness – 1.78), with a range from as low as 0.077 to a maximum of 1.00 (best health scenario). The overall VAS had a median value of 70.0 (skewness – 0.52), with a range from 20.0 to 95.0 (Figure 1).

Ninety four (83.2%) patients participated in the follow-up telephone interview, which was conducted at a median of 16 days (range 13 – 18) after the original interview. Intraclass correlation coefficient for the EQ-5D utility score was 0.66 (95% CI = 0.55 – 0.76, p < 0.0001).

Twenty of the 26 hypotheses relating EQ-5D dimensions/summary scores to other variables were fulfilled. At the attribute level, problems with usual activity and pain were associated with worsening dyspeptic symptoms but these were not statistically significant. However, significantly lower values for the EQ-5D utility scores and VAS were demonstrated with severe dyspepsia (Table 2). All hypotheses relating EQ-5D dimensions to relevant SF-36 sub-scales were supported (see Additional file 1). For example, patients who reported problems with mobility in the EQ-5D had significantly lower scores in relevant SF-36 sub-scales (i.e. physical functioning and role physical), whilst those reporting problems with anxiety had significantly lower scores for SF-36 mental health and role emotional sub-scales (see Additional file 1). Five out of the 7 hypotheses relating EQ-5D dimensions to relevant SF-NDI sub-scales were supported (see Additional file 1). In particular, patients reporting problems with usual activities and pain demonstrated significantly lower scores in the "interference" and "eating" sub-scales, whilst those reporting problems with anxiety had significantly lower scores in the "tension" sub-scales. Utility scores of the EQ-5D demonstrated good correlation with the SF-36 physical component summary score (r = 0.45), the SF-36 mental component summary score (r = 0.49) and the cumulative SF-NDI score (r = 0.47) (see Additional file 1).