Prediction of persistent shoulder pain in general practice: Comparing clinical consensus from a Delphi procedure with a statistical scoring system

Table 1 Delphi procedure results

	Round 1		Round 2		Round 3
Predictors	mean (SD)	rank n = 46	sum score	rank n = 65	inclusion agreement	rank
Symptom duration	4.26 (0.78)	1	253	1	100%	1
Pain catastrophizing	4.03 (0.95)	3	160	2	84%	2
Symptom history	4.06 (0.80)	2	95	3	90%	3
Fear-avoidance beliefs	3.91 (1.04)	6	86	4	90%	4
Coexisting neck pain	3.49 (1.12)	22	79	5	100%	5
Baseline severity of shoulder disability	3.49 (1.07)	21	78	6	97%	6
Coexisting psychological complaints/general mental health problems	3.83 (0.95)	9	73	7	84%	-^b
Currently on sick leave because of shoulder pain	3.69 (0.87)	11	70	8	81%	-^c
Multisite pain	n = 10^a	-	68	9	84%	7
Somatisation	4.00 (0.97)	4	67	10	84%	-^d
Age	3.58 (0.73)	17	64	11	19%	8
Shoulder pain intensity	3.49 (1.24)	23	64	12	13%	9
Illness perceptions	3.91 (1.01)	5	62	13	13%	10
Depression/depressive symptoms	3.91 (1.07)	7	61	14	10%	-
Passive coping strategies	3.63 (1.06)	15	60	15	6%	-
Repetitive movements	3.86 (0.81)	8	46	16	3%	-
High physical load at work or leisure time	n = 11^a	-	39	17	6%	-
Strain or overuse due to usual activities in work or leisure time	3.43 (1.06)	25	32	18	10%	-
Patient reports stiffness of the shoulder	3.06 (0.97)	30	32	19	6%	-
Diabetes Mellitus	n = 2^a	-	32	20	10%	-

Predictors (only the 20 most highly ranked from round 2) are ordered by their prognostic importance as assessed in the second Delphi round
mean = mean score for rating individual potential predictor importance, [1 (not at all predictive) - 5 (highly predictive)]
SD = standard deviation
rank = the order of appearance of potential predictors in individual Delphi rounds based on scores for prognostic importance
sum score = the sum of all scores for predictive importance for each potential predictor
inclusion agreement = consensus on the inclusion of potential predictors in the selection of the ten most important predictors for persistent shoulder pain intensity
^a additional potential predictor added by n panel members
^b potential predictor omitted from final selection since it overlaps with 'fear-avoidance' and the item is 'too general'
^c potential predictor combined with predictor 'severity of shoulder disability'
^d potential predictor omitted from final selection since it overlaps with 'multisite pain' and 'fear avoidance'

ISSN: 2731-4553