Table 3

Comparisons within each centre between the four levels of delay

Friedman p value<0.001<0.001<0.001<0.001<0.001<0.001<0.001<0.0010.008<0.001
Comparisons between pairs of delays
 1 vs 2 (1>2)NS<0.001NS<0.001<0.001<0.001<0.001<0.001NS<0.001
 1 vs 3 (1>3, except bold)<0.001<0.005<0.001NSNSNSNSNSNSNS
 1 vs 4 (1>4, except bold)<0.001<0.025<0.001<0.001NSNSNSNSNS<0.001
 2 vs 3 (3>2, except bold)<0.001<0.005<0.005<0.001<0.001<0.001<0.001<0.001NS<0.001
 2 vs 4 (4>2, except bold)<0.001<0.001<0.001NS<0.001<0.001<0.001<0.025NSNS
 3 vs 4 (3>4)NSNSNS<0.001NSNSNS<0.01<0.025<0.001
  • Delay 1: symptom onset to request to see initial healthcare professional (HCP) of contact. Delay 2: request to see initial HCP of contact to assessment by that HCP. Delay 3: initial assessment by initial HCP to referral to a rheumatologist. Delay 4: referral to a rheumatologist to assessment by that rheumatologist.

  • The Friedman test was used to compare the different levels of delay at each centre. A post hoc test was used to compare pairs of levels of delay. p Values are shown. Where p>0.05, this was regarded as NS. Where the differences were significant, the text in parentheses in the first column indicates which of the two levels of delay being compared was longer; ‘except bold’ refers to the figures in bold type.