Page 161 - Strategies for non-invasive managementof high-grade cervical intraepithelial neoplasia - prognostic biomarkers and immunotherapy Margot Maria Koeneman

Table 3. Results of the multinomial model
ASC Imiquimod
Treatment success Premature birth
Subfertility
Vaginal pruritus and/or pain Vaginal discharge
Flu-like symptoms ASC LLETZ
Pregnancy wish*subfertility Pregnancy wish*premature birth Pregnancy wish* LLETZ LL=-738.33
Number of observations=894† Number of individuals=100
Coefficient 0.060 0.084*** -0.126*** -0.109*** -0.031*** -0.013* -0.029** 4.25***
95%CI
-0.17 – 0.29 0.06 –0.10 -0.19 – -0.06 -0.15 – -0.06 -0.05 – -0.01 -0.03 – -0.001 -0.05 – -0.005 3.33 – 5.16
Coefficient 95%CI 0.14 -0.11 – 0.40 0.077*** 0.06 – 0.10
-0.05 -0.14 – 0.03
-0.006 -0.07 – 0.05
-0.024** -0.04 – -0.003 -0.009 -0.03 – 0.007 -0.014 -0.04 – 0.011 6.37*** 5.09 – 7.66
-0.178*** -0.246 – -0.111 8 -0.103** -0.193 – -0.013
-0.863*** -1.16 – -0.56
LL=-612.69
Number of observations=778‡ Number of individuals=87
Patient preferences in treatment of high-grade CIN
of premature birth in subsequent pregnancy,” and the fixed alternative LLETZ treatment are significantly different depending on a woman’s plans for a future pregnancy. The LLETZ treatment provides a significantly higher benefit for women without a pregnancy wish compared to women with a pregnancy wish. The model 2, which includes the interactions, significantly improves the model fit compared to model 1 (likelihood ratio test χ2 (251) ≥ χ2 (7.81); p < 0.05).
Attributes
Main effect model (1)
Model with interactions (2)
Likelihood ratio chi square test: model 2 vs model 1 χ2 (251)≥ χ2 (7,81);
p-value<0.05
***: p-value< 0.01;**p-value<0.05; *p-value<0.1
† 100 subjects, 9 questions per subject, six questions were not completed: (100*9)-6 = 894 observations
‡ 13 subjects were excluded because of uncertainty about future pregnancy wish, five questions were not completed: (900-(13*9)-5) = 778 observations
ASC = alternative specific constant
LLETZ = large loop excision of the transformation zone
LL = log likelihood
DCE results: Subgroup analysis of women with and without the desire for a future pregnancy
Table 4 presents the results for women with a future pregnancy wish and without a pregnancy wish. Women with a future pregnancy wish show a significant preference for a higher chance of treatment success and a lower risk of subfertility and premature birth. The other attributes do not play a role in their decision for imiquimod treatment. They also assign a positive utility (3.35) to the LLETZ treatment, although this value is significantly lower compared to women who do not desire a future pregnancy (5.76). For women who are planning a future pregnancy, imiquimod must have a treatment success of at least around 72% (0.08 * 72%) with low levels for subfertility
159