Skip to main content

Table 4 Results from mixed multinomial logits (diagnostic tests)

From: The effect of cost-sharing design characteristics on use of health care recommended by the treating physician; a discrete choice experiment

Health Service: Diagnostic tests
Group: Forgoing health care Utilizing health care
Weighting: Unweighted Inverse Probability Weighting Unweighted Inverse Probability Weighting
Coefficient (SE) Coefficient (SE) Coefficient (SE) Coefficient (SE)
Variable:
 Type of payments
  -Copayments 0.664 (0.116)** 1.041 (0.457)* 0.773 (0.057)** 1.032 (0.234)**
  -Coinsurances (reference) − 0.664 (0.116)** − 1.041 (0.457)* − 0.773 (0.057)** − 1.032 (0.234)**
 Rate of payments
  -Rate1 a 1.123 (0.102)** 1.758 (0.759)* 0.643 (0.033)** 0.841 (0.106)**
  -Rate2 a 0.522 (0.089)** 0.202 (0.350) 0.424 (0.030)** 0.334 (0.092)**
  -Rate3 a (reference) −1.646 (0.133)** −1.959 (0.760)* − 1.067 (0.042)** −1.174 (0.150)**
 Annual caps on cost-sharing payment
  -Half cap −0.005 (0.055) 0.035 (0.207) −0.003 (0.022) 0.039 (0.036)
  -Full cap (reference) 0.005 (0.055) −0.035 (0.207) 0.003 (0.022) −0.039 (0.036)
  -Half cap * Copayments −0.050 (0.052) −0.266 0.240) − 0.057 (0.019)** −0.084 (0.063)
  -Half cap * Rate1 a 0.037 (0.081) 0.227 (0.347) −0.056 (0.033) −0.103 (0.078)
  -Half cap * Rate2 a − 0.147 (0.090) −0.884 (0.363)* − 0.019 (0.038) − 0.053 (0.121)
  -Half cap * Afterwards billed by health insurer 0.017 (0.057) 0.226 (0.231) −0.079 (0.020)** −0.221 (0.079)**
 Moment of payment
  -Afterwards billed by health insurer 0.414 (0.086)** 1.581 (0.831) 3.416 (0.108)** 2.660 (0.390)**
  -Directly at point of care (reference) −0.414 (0.086)** −1.581 (0.831) −3.416 (0.108)** −2.660 (0.390)**
 Alternative specific constant
  -Opt-out option 3.752 (0.146)** 4.503 (1.060)** 0.556 (0.033)** 0.319 (0.247)
 SD of random parameters
  Copayments 2.702 (0.143)** 2.599 (0.689)** 3.236 (0.092)** 3.480 (0.451)**
  Rate1 a −1.486 (0.128)** −2.132 (0.804)** −0.907 (0.048)** 0.974 (0.161)**
  Rate2 a 1.412 (0.130)** 2.338 (1.035)* 0.827 (0.047)** 0.789 (0.172)**
  Afterwards billed by health insurer 1.908 (0.113)** 3.829 (1.098)** 4.690 (0.132)** 4.521 (0.591)**
Model
 Number of individuals 738 738 5510 5510
 Number of observations 13,284 13,284 99,180 99,180
 Log-likelihood − 3187.780 − 3180.645 −20,580.500 −16,919.590
  1. Model: mixed multinomial logits modelling eq. 1. Random parameters assumed a normal distribution and estimated based on 5000 Halton draws
  2. SE Standard Error, SD Standard Deviation
  3. a labels Diagnostic tests: rate1 (€40/15%), rate2 (€50/20%), rate3 (€80/30%)
  4. * = p-value≤.05, ** = p-value≤.01
Back to article page

BMC Health Services Research

ISSN: 1472-6963

Contact us