Table 2
Results from the discrete choice experiment
| Treatment Attribute/Level | Country/N | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
BE | NO | GE | SE | NL | UK | HU | SP | FR | IT | PT | RO | All countries | |
| 33 | 41 | 44 | 24 | 63 | 40 | 71 | 63 | 40 | 59 | 39 | 42 | 559 | |
| (a) Results from the random parameters logit model (β coefficients per attribute level reflecting physician's preferences for the respective attribute) | |||||||||||||
| Efficacy | |||||||||||||
| Moderate response | Reference level (efficacy) | ||||||||||||
| Good response | 2.08*** | 2.93*** | 2.93*** | 2.81*** | 2.40*** | 1.92*** | 2.31*** | 3.33*** | 2.49*** | 2.41*** | 3.40*** | 4.96*** | 2.30*** |
| Remission | 3.55*** | 4.32*** | 5.44*** | 5.05*** | 4.40*** | 3.47*** | 4.34*** | 5.35*** | 3.90*** | 4.66*** | 5.07*** | 8.86*** | 3.91*** |
| Safety | |||||||||||||
| Very rare | Reference level (safety) | ||||||||||||
| Rare | −0.37 | −0.30 | −0.65** | −0.82** | 0.14 | 0.00 | −0.41** | −0.16 | −0.61*** | −0.89*** | −1.14*** | −1.15** | −0.41*** |
| Uncommon | −0.97*** | −1.04*** | −1.21*** | −1.28*** | 0.05 | −0.19 | −1.13*** | −1.13*** | −1.78*** | −1.75*** | −2.49*** | −3.31*** | −1.04*** |
| Patient’s preference | |||||||||||||
| Patient favours treatment | 0.17 | 0.24 | 0.57** | 0.47 | 0.29* | 0.44** | 0.62** | 0.71*** | 0.53** | 0.49*** | 0.00 | 0.04 | 0.40*** |
| Patient is neutral | Reference level (patient’s preference) | ||||||||||||
| Patient disfavours treatment | −0.84*** | −1.27*** | −1.28*** | −1.75*** | −1.45*** | −1.22*** | −0.87*** | −1.44*** | −1.40*** | −0.97*** | −1.69*** | −1.12** | −1.03*** |
| Cost-effectiveness (10 000 EUR/QALY) | −0.15*** | −0.21*** | −0.03*** | −0.19*** | −0.23*** | −0.31*** | −0.22*** | −0.21*** | −0.05 | −0.14 | −0.18** | −0.11 | −0.15*** |
| Overall medication costs (1000 EUR/year) | −0.12*** | −0.19*** | −0.19*** | −0.15*** | −0.18*** | −0.14*** | −0.09*** | −0.16*** | −0.09*** | −0.12*** | −0.19*** | −0.23*** | −0.12*** |
| Constant | 0.14 | −0.15 | −0.40** | −0.30 | −0.26** | −0.32** | −0.15 | 0.02 | −0.16 | −0.31** | −0.24 | −0.57 | −0.15*** |
| McFadden pseudo R-squared (R2). | 0.61 | 0.65 | 0.66 | 0.67 | 0.46 | 0.63 | 0.63 | 0.65 | 0.58 | 0.63 | 0.64 | 0.63 | 0.39 |
| (b) Contribution of each of the treatment attributes to the overall preference for a treatment choice (mean per country, in per cent of total explained variance) | |||||||||||||
| Country clusters | Country cluster 1 | Country cluster 2 | |||||||||||
| Efficacy Improvement DAS28 | 44% | 40% | 42% | 43% | 44% | 39% | 46% | 45% | 43% | 45% | 39% | 52% | 44% |
| Safety Probability of a serious AE | 10% | 10% | 9% | 11% | 1% | 2% | 12% | 10% | 20% | 18% | 19% | 19% | 12% |
| Patient′s preference | 14% | 14% | 14% | 19% | 17% | 19% | 16% | 18% | 21% | 14% | 14% | 7% | 16% |
| Cost-effectiveness ICER, QALY/year | 9% | 12% | 14% | 10% | 14% | 21% | 14% | 10% | 3% | 8% | 8% | 4% | 10% |
| Medication costs per year, in local currency | 23% | 24% | 19% | 17% | 24% | 19% | 12% | 17% | 13% | 15% | 20% | 18% | 18% |
| Model fit, R2 | 0.61 | 0.65 | 0.66 | 0.67 | 0.46 | 0.63 | 0.63 | 0.65 | 0.58 | 0.63 | 0.64 | 0.63 | 0.39 |
*significant at 10%, **significant at 5%, ***Significant at 1%.
All parameters were included as random parameters and assumed to be normally distributed. The estimation was conducted using 1000 Halton draws.
AE, adverse event; BE, Belgium; DAS28, disease activity score in 28 joints; EUR, Euro; FR, France; GE, Germany; HU, Hungary; ICER, incremental cost-effectiveness ratio; IT, Italy; N, number of responses per country and overall; n.a., not applicable; NL, the Netherlands; NO, Norway; PT, Portugal; QALY, quality-adjusted life year; R2, McFadden pseudo R-squared; RO, Romania; SE, Sweden; SP, Spain.