Development of the patient Harvey Bradshaw index and a comparison with a clinician-based Harvey Bradshaw index assessment of Crohn's disease activity

Abstract

Goals and background: The objective is to develop a patient-based Harvey Bradshaw Index (P-HBI) of Crohn's Disease (CD) activity and to compare it with the clinician-based HBI of CD activity in CD outpatients.

Study: Consecutive patients with CD randomly completed the P-HBI either before or after the consultation. The gastroenterologist assessed patient's CD activity on the same day. Overall agreement between HBI and P-HBI was calculated with Spearman's ρ and Mann-Whitney U test. Agreement regarding active disease versus remission and agreement at item level was calculated by percent agreement and Cohen's κ.

Results: One hundred eighty-one (response rate 88.3%) CD patients participated. P-HBI and HBI showed a large correlation (rs=0.82). The medians (interquartile range) of the total HBI (2; 0 to 4) and P-HBI (4; 1 to 7) were statistically significantly different (z=-8.411; P<0.001). Fortunately, in 82.6% of the cases this difference between clinicians and patients was not clinically significant (<3.2). The percentage agreement between clinician and patient, judging CD as active or as in remission, was 77%, rs=0.56, κ=0.52, indicating a moderate agreement. P-HBI and HBI on frequent extraintestinal manifestations in CD varied from less than chance (κ=-0.02) to a perfect agreement (κ=1). Patients tended to report more symptoms while completing the patient-based questionnaire compared to what they reported to the clinician during consultation.

Conclusions: The P-HBI is the first step in developing a potential promising tool given its adequate agreement with the original HBI and its feasibility, especially in patients with low scores. Future research is necessary to develop a validated patient-based version studied in several patient populations.