Table 3

Adjusted threshold models taking into account characteristics (gender, age, practice and internship) also testing for gender-age and gender-internship interactions Results of final models obtained via backward selection based on AIC (Akaike Information Criterion)

ORP valueThreshold
UTI: test threshold
 Intercept0.580.036Overall12.8 (1.05 to 20.7)
 Disease probability1.53<0.001
EstimateP valueThresholds
UTI: treatment threshold
 Intercept−1.25<0.001Gender: male
Internship: yes
75.0 (48.1 to 100)
 Disease probability1.18<0.001Gender: male
Internship: no
8.7 (0.0 to 51.4)
 Gender: female0.630.090Gender: female
Internship: yes
100 (79.5 to 100)
 Internship: no3.03<0.001Gender: female
Internship: no
36.4 (0.0 to 73.7)
ORP valueThresholds
GIO: test threshold
 Intercept1.250.509Practice: GP, IM or ECC
Internship: yes
0.00 (0.00 to 5.63)
 Disease probability1.90<0.001Practice: GP, IM or ECC
Internship: no
15.7 (1.14 to 27.9)
 Practice: ‘other’0.170.007Practice: other
Internship: yes
24.1 (3.14 to 43.6)
 Internship: no0.290.008Practice: other
Internship: no
43.3 (20.8 to 68.3)
EstimateP valueThresholds
GIO: treatment threshold
 Intercept0.00<0.001Practice: IM, ECC or other89.0 (84.0 to 95.7)
 Disease probability2.33<0.001Practice: GP76.2 (65.0 to 88.3)
 Practice: GP2.960.042
  • Results of final models obtained via backward selection based on Akaike Information Criterion.

  • ECC, emergency and critical care; GIO, gastrointestinal obstruction; GP, general practice; IM, internal medicine; UTI, urinary tract infection.