Table 1 Purpose of psychometric tests

Item convergent validity

To assess an item's correlation with its own hypothesized sub-scale score (satisfied if correlation achieved is ≥ 0.40)

Item discriminant validity

To assess whether an item considered in isolation has a higher correlation with its hypothesized scale than with other scales in the questionnaire

Internal consistency reliability

To evaluate the extent to which individual items of the instrument are consistent to one another and reflect an underlying scheme or construct (satisfied if Cronbach's alpha coefficient = 0.70 is achieved)

Test-retest reliability

Assesses the extent to which the measure yields the same results in repeated applications in an unchanged population. The intra-class correlation coefficient (ICC) was used as a measurement of test-retest reliability, and was assessed in patients who reported their health status to be stable between baseline (week 0) and study end (7 to 10 days later) (satisfied if an ICC coefficient = 0.70 is achieved)

Floor and ceiling effects

Refer to a high percentage of patients scoring the lowest score possible and a high percentage of patients achieving the highest score possible, respectively. High baseline floor or ceiling effects are indicative of a scale that is limited in its responsiveness to clinical change. Minimal floor and ceiling effects are therefore recommended. For the EAPIQ scales a percentage of 20% at floor or at ceiling was considered a significant effect

Scale-scale correlations

To determine whether the concepts measured in the individual scales (domains) of the EAPIQ were distinct and that none of the domains were redundant

Concurrent validity

Concurrent validity was supported if the EAPIQ sub-scales were substantially correlated (≥ 0.40), with miniRQLQ sub-scales measuring similar concepts. Conversely, sub-scales measuring unrelated concepts should be poorly correlated. As a generic measure of health status the HUI2/3 was expected to be less strongly correlated with the EAPIQ scales

Known-group validity

Differences in EAPIQ scores were expected among groups of patients known to differ in their patient-evaluated health status

Clinical validity

Clinical validity assesses the ability of scores to discriminate among groups of patients defined according to clinical severity. Patients who have a good clinical status at baseline should score well in the questionnaire, and patients who have a poor clinical status at baseline should score poorly

Responsiveness

Responsiveness refers to the ability of a measure to reflect underlying change. Preliminary responsiveness of the EAPIQ was assessed by comparing EAPIQ scores in those patients who report a change in their health status over the two-week period. Patients who were assessed at baseline and two weeks later were stratified by their report of worsening, no change and improvement in their 'overall health', 'all allergies', and 'eye allergy' symptoms, over the 7 to 10 days