Skip to main content
. 2017 Apr 26;7(4):e015689. doi: 10.1136/bmjopen-2016-015689

Table 4.

Multinomial logit analysis: higher versus neutral level of person-centredness by attribute and by discrete choice experiment (DCE)

Discrete choice experiment Chronic pain
(DCE1)
Breathlessness
(DCE2)
Chronic Pain
(DCE3)
Estimate 95% CI Estimate 95% CI Estimate 95% CI
Regression coefficient
 Information 0.65 (0.58 to 0.73) 0.60 (0.47 to 0.72) 0.93 (0.79 to 1.07)
 Situation 0.91 (0.82 to 0.99) 0.90 (0.76 to 1.05) 1.43 (1.23 to 1.63)
 Living well 0.85 (0.78 to 0.92) 0.81 (0.68 to 0.93) 1.19 (1.04 to 1.35)
 Communication 0.21 (0.15 to 0.26) 0.21 (0.12 to 0.30) 0.31 (0.20 to 0.42)
 Cost* −0.06 (−0.07 to −0.05) −0.06 (−0.07 to −0.04)
Choice elasticity (in %)
 Information 12.3 (11.0 to 13.6) 10.6 (8.3 to 12.9) 15.0 (12.9 to 17.0)
 Situation 16.9 (15.4 to 18.3) 16.1 (13.6 to 18.6) 21.9 (19.3 to 24.5)
 Living well 15.8 (14.5 to 17.1) 14.6 (12.5 to 16.7) 19.4 (17.2 to 21.6)
 Communication 3.8 (2.7 to 4.8) 4.2 (2.6 to 5.8) 5.4 (3.5 to 7.2)
 Cost −1.1 (−1.2 to −0.9) −1.0 (−1.3 to −0.8)
Model statistics
 Individuals 517 200 206
 Observations 12 408 4800 4944
 Parameters 5 5 4
 Log likelihood −12 004 −4739 −4542
 BIC 24 054 9520 9117

*Coefficient for increase by £1 in cost of service

BIC, Bayesian Information Criterion.