Psychometric validation of the Dysphagia Symptom Questionnaire in patients with eosinophilic esophagitis treated with budesonide oral suspension

Journal of Patient-Reported Outcomes

Table 3 Stage 1: Item analysis of the DSQ (version 4.0) to assess quality of completion by patients, item score patterns, item distribution and item discrimination at baseline (visit 2) in patients with EoE. Floor and ceiling effects at the end of the 12-week treatment period are also reported

Parameter	All patients (n = 93)
Quality of completion, % (n/N)
Missing data per item (questions 2 & 3)	15.1% (197/1302)
Missing items per patient (questions 2 & 3)	76.3% (71/93)
Item score pattern ^a (n, number of responses)	Total possible responses (n = 1302)
Missing items	197
No dysphagia (question 2: ‘No’)	301
Dysphagia 1 (question 2: ‘Yes’; question 3: option 1)	238
Dysphagia 2 (question 2: ‘Yes’; question 3: option 2)	401
Dysphagia 3 (question 2: ‘Yes’; question 3: option 3)	133
Dysphagia 4 (question 2: ‘Yes’; question 3: option 4)	31
Dysphagia 5 (question 2: ‘Yes’; question 3: option 5)	1
Item distribution (question 3), % (n/N)
Floor effects at baseline	8.6% (8/93)
Ceiling effects at baseline	6.5% (6/93)
Floor effects at week 12	54.8% (46/84)
Ceiling effects at week 12	8.3% (7/84)
Item discrimination ^b
Question 2^c	‘No dysphagia’	‘Dysphagia’
High DSQ scores, n (%)	2 (1.9)	104 (98.1)
Moderate DSQ scores, n (%)	33 (20.5)	128 (79.5)
Low DSQ scores, n (%)	24 (28.6)	60 (71.4)
Question 3^c	‘No action’	‘Moderate action’	‘Extreme action’
High DSQ scores, n (%)	7 (6.6)	99 (93.4)	0 (0.0)
Moderate DSQ scores, n (%)	89 (55.3)	71 (44.1)	1 (0.6)
Low DSQ scores, n (%)	52 (61.9)	32 (38.1)	0 (0.0)

DSQ Dysphagia Symptom Questionnaire, EoE eosinophilic esophagitis
^aItem scoring patterns defined as number of instances where a patient selected the responses indicated (question 2: ‘Yes’/‘No’; question 3: options 1–5) during the baseline period
^bItem discrimination – based on the 25th and 75th percentiles of the DSQ domain score, patients were assigned into three groups: patients who had high (>75th percentile), moderate (25th to 75th percentile) and low (<25th percentile) DSQ scores. Patient action was defined as the attempt to alleviate symptoms as per answers to question 3: options 1–5 (no action taken = answer option 1, moderate action taken = answer options 2–4, extreme action taken = answer option 5)
^cHigh DSQ scores, N = 106; moderate DSQ scores, N = 161; low DSQ scores, N = 84