Sample Characteristics
Of 1448 scheduled laparoscopies (since August 2021), 875 women were interested in the study, of which 699 completed the baseline survey (T0). Due to some women not having an endometriosis diagnosis, the decision against laparoscopic surgery and other dropout reasons, the final sample consisted of 393 participants (see Fig. 1).
Characteristics of the investigated sample and the distributions of the six predictors are presented in Table 1. Participants were 18 to 51 years old (M = 29.74, SD = 6.52). About one-sixth had a migrant background, i.e. women whose one or both parents were not born in Germany. Almost every tenth woman was incapable of working. In the present study, patients showed four times higher pain disability scores (M = 30.77, SD = 13.28 vs. M = 6.9, SD = 11.1, N = 1,368 (37)) and three times higher somatic impairment scores (M = 15.61, SD = 5.07 vs. M = 6.31, SD = 4.06, N = 4,815 (38)) than the general population. Pain catastrophising scores were M = 18.33 on average (SD = 7.81). Furthermore, participants showed higher state depression (M = 2.52, SD = 1.40) and state anxiety scores (M = 2.36, SD = 1.64) than healthy subjects (7, 39). Almost half of the women showed signs of major depression (40.2%) and/or an anxiety disorder (39.9%). Overall, more than every fourth woman simultaneously showed signs of major depression and a general anxiety disorder (27.0%). The duration of symptoms ranged from 0 to 486 months (M = 84.85, SD = 83.97), respectively M = 7.1 (SD = 7.0) years. Every fifth woman stated prior experiences with laparoscopy (19.8%). Table 1 presents more information on the characteristics of the sample under analysis and descriptive statistics.
Treatment expectations concerning the laparoscopy of participants were, in general, rather positive. However, TEX-Q mean scores were lower than in the validation sample (M = 6.70, SD = 1.24 vs. M = 7.34, SD = 1.20, N = 251 (24)) and distributed slightly left-skewed (skewness: -0.23). The positive dimension of the GEEE (i.e. expected improvement) was rated with higher values (M = 7.08, SD = 2.04) and distributed left-skewed (skewness: -0.88). Negative dimensions of the GEEE (i.e. expected worsening and expected complaints) were rated with lower values (M = 1.12 and 3.83, SD = 1.68 and 2.33) and distributed right-skewed (skewness: 1.84 and 0.31). Distributions of all measurement instruments and items assessing treatment expectations are presented in Table 2.
Table 1
Sample characteristics at baseline (T0).
|
Women with endometriosis (N = 392)
|
n
|
%
|
M
|
SD
|
Observed range
|
Potential range
|
Gender (female vs. non-binary)
|
388
|
98.7
|
|
|
|
|
Age (in years)
|
|
|
29.74
|
6.51
|
18–51
|
18–99
|
Migrant background
|
61
|
15.5
|
|
|
|
|
Unable to work
|
36
|
9.2
|
|
|
|
|
Pain disability (PDI)
|
|
|
30.77
|
13.28
|
0–69
|
0–70
|
Pain catastrophisation (CSQ)
|
|
|
18.33
|
7.81
|
0–36
|
0–36
|
Somatic symptom impairment (PHQ-15)
|
|
|
15.61
|
5.07
|
3–27
|
0–30
|
State depression (PHQ-2)
|
|
|
2.52
|
1.40
|
0–6
|
0–6
|
High likelihood of major depression (PHQ-2 ≥ 3)
|
158
|
40.2
|
|
|
3–6
|
3–6
|
State anxiety (GAD-2)
|
|
|
2.36
|
1.64
|
0–6
|
0–6
|
High likelihood of general anxiety disorder (GAD-2 ≥ 3)
|
157
|
39.9
|
|
|
3–6
|
3–6
|
High likelihood of major depresson and general anxiety disorder (PHQ-2 + GAD-2 ≥ 3)
|
106
|
27.0
|
|
|
|
|
Prior experiences with laparoscopy
|
78
|
19.8
|
|
|
|
|
Duration of symptoms (in months)
|
|
|
84.85
|
83.97
|
0-486
|
≥ 0
|
Deep infiltrating endometriosis
|
78
|
19.8
|
|
|
|
|
rASRM (stage of endometriosis)
|
|
|
|
|
|
|
minimal
|
250
|
63.6
|
|
|
|
|
mild
|
88
|
22.4
|
|
|
|
|
moderate
|
38
|
9.7
|
|
|
|
|
severe
|
7
|
1.8
|
|
|
|
|
Recommendation of further hormone therapy
|
279
|
71.0
|
|
|
|
|
Note. CSQ = catastrophisation subscale of the Coping Strategies Questionnaire; GAD-2 = Generalized Anxiety Disorder Scale; PDI = Pain Disability Index; PHQ-2 = Patient Health Questionnaire for Depression; PHQ-15 = Patient Health Questionnaire; rASRM = revised American Society for Reproductive Medicine Score. |
Table 2
Treatment expectation scales.
Subscale
|
M
|
SD
|
Observed range
|
Skewness (SE)
|
Kurtosis (SE)
|
Percentiles
|
25
|
50
|
75
|
TEX-Q total score
|
6.70
|
1.24
|
2.4–10
|
-0.226 (.12)
|
.226 (.25)
|
5.87
|
6.73
|
7.53
|
Treatment benefit
|
6.77
|
2.15
|
1–10
|
-0.745 (.12)
|
0.544
|
6.00
|
7.33
|
8.33
|
Positive impact
|
6.36
|
2.56
|
0–10
|
-0.725 (.12)
|
0.035
|
5.00
|
7.00
|
8.33
|
Adverse events
|
2.83
|
1.96
|
0–9
|
0.349 (.12)
|
-0.262
|
2.33
|
3.33
|
5.00
|
Negative impact
|
2.19
|
2.40
|
0–10
|
0.849 (.12)
|
-0.016
|
0.50
|
2.00
|
4.00
|
Process
|
6.55
|
2.33
|
0–10
|
-0.359 (.12)
|
-0.053
|
5.50
|
7.00
|
8.00
|
Behavioural control
|
6.03
|
2.60
|
0–10
|
-0.277 (.12)
|
-0.445
|
4.50
|
6.00
|
7.50
|
GEEEexpect_improvement
|
7.08
|
2.04
|
0–10
|
-0.879 (.12)
|
0.843 (.25)
|
6.00
|
7.00
|
8.00
|
GEEEexpect_worsening
|
1.12
|
1.68
|
0–9
|
1.838 (.12)
|
3.385 (.25)
|
0.00
|
0.00
|
2.00
|
GEEEexpect_complaints
|
3.83
|
2.33
|
0–10
|
0.312 (.12)
|
-0.516 (.25)
|
2.00
|
4.00
|
5.00
|
Note. The mean TEX-Q total score was computed after inverting TEX-Q_7 to TEX-Q_11. Higher scores indicate more positive treatment expectations. GEEEexpect = Generic Rating Scale for Treatment Expectations; TEX-Q = Treatment Expectation Questionnaire. |
Bivariate Associations
A correlation matrix of bivariate associations between the outcome measures and predictors is presented in Table 3. All bivariate associations between individual items measuring treatment expectations are presented in Supplementary Table 1.
Women with higher scores of anxiety reported significantly lower overall treatment expectations measured by the TEX-Q total score, r = − .149, p = .003. Treatment expectations regarding treatment benefit were not significantly associated with any predictors. Women with higher scores of pain disability, pain catastrophising, and somatic symptom impairment reported significantly higher positive impact expectations, r = .230, p < .005, r = .204, p < .005, and r = .175, p < .005. Women with higher scores of pain disability, pain catastrophizing, somatic symptom impairment, depression, and anxiety reported significantly higher treatment expectations regarding adverse events (i.e. negative expectations regarding treatment), r = .190, p < .00), r = .152, p < .005, r = .165, p < .005, r = .148, p < .005, and r = .184, p < .005. Women with higher scores of pain disability and somatic symptom impairment reported significantly higher treatment expectations regarding negative impact (i.e. negative expectations regarding treatment), r = .161, p < .005 and r = .134, p = .008. Women with higher scores of anxiety and a longer duration of symptoms reported significantly lower treatment process expectations (i.e. a less straightforward process), r = − .152, p < .005 and r = − .107, p = .034. Women with higher scores of somatic symptom impairment reported significantly higher treatment expectations regarding behavioural control (i.e. being able to influence treatment success), r = .141, p = .005.
Treatment improvement expectations measured with the G-EEE were not significantly associated with any predictors. Women with higher scores of pain catastrophizing reported significantly higher treatment worsening expectations (i.e. negative expectations regarding treatment), r = .116, p = .021. Women with higher scores of pain disability, somatic symptom impairment, and depression reported significantly higher treatment complaint expectations (i.e. negative expectations regarding treatment), r = .171, p < .005, r = .144, p < .005, r = .107, p = .035. All significant bivariate associations were small.
Prior experiences with laparoscopy were not significantly associated with any of the outcome variables. Regarding medical information, neither the rASRM-score nor having deeply infiltrated endometriosis were significantly associated with any of the outcome variables.
Prediction of treatment expectations
Seven of the ten multiple linear regression analyses found a significant regression equation. There was a significant model fit for total treatment expectations (TEX-Q_total: F(8,384) = 2.72, p = .006, R2 = .034), positive impact (TEX-Q_positiveimpact: F(8,384) = 5.48, p < .001, R2 = .084), adverse events (TEX-Q_adverseevents: F(8,384) = 3.56, p < .001, R2 = .046), negative impact (TEX-Q_negativeimpact: F(8,384) = 1.96, p = .048, R2 = .020), process (TEX-Q_process: F(8,384) = 2.34, p = .019, R2 = .027), behavioural control (TEX-Q_behaviouralcontrol: F(8,384) = 2.21, p = .026, R2 = .024), and impairment (F(8,384) = 2.73, p = .006, R2 = .034). The expected treatment benefit (TEX-Q_treatmentbenefit), improvement (GEEEexpect_improvement), and worsening (GEEEexpect_worsening) had no significant model fit.
Women with prior experiences expected significantly fewer complaints due to treatment (GEEEexpect_complaints: B = − .102, p = .045) than women without prior experiences with endometriosis laparoscopy. Women with a higher disability due to pain had significantly higher treatment expectations regarding positive impact (B = 1.37, p = .027), expected significantly more adverse events due to treatment (B = .132, p = .036), had significantly higher treatment expectations regarding negative impact (B = .165, p = .010), and expected more complaints due to the treatment (GEEEexpect_complaints: B = .143, p = .025). Patients with a higher level of pain catastrophising had significantly higher total treatment expectations (B = .131, p = .044) and expected higher positive impact due to the treatment (B = .197, p = .002). Participants with a higher level of physical impairment expected significantly higher behavioural control (B = .188, p = .002). Women with a higher level of state depression did not significantly differentiate in terms of their treatment expectations from those who had lower levels of state depression. Women with a higher level of state anxiety had significantly lower total treatment expectations (B = − .190, p = .005), lower expectations regarding positive impact due to the treatment (B = − .155, p = .019), and lower expectations regarding the treatment process (B = − .192, p = .005). Women with a higher level of state anxiety expected significantly more adverse events due to treatment (B = .141, p = .036). Patients of older age reported significantly higher total treatment expectations (B = .119, p = .029) and expected significantly higher positive impact due to the treatment (B = .118, p = .025). Participants with a longer duration of endometriosis-related symptoms had significantly lower total treatment expectations (B = − .119, p = .023), and had significantly lower expectations regarding the treatment process (B = − .140, p = .008). All significant and non-significant regression equations and model coefficients can be found in Tables 4 and 5.
Table 4
Multiple linear regression analyses of the total TEX-Q score and subscales.
|
TEX-Q total score
|
TEX-Q treatment benefit
|
TEX-Q Positive impact
|
TEX-Q Adverse events
|
TEX-Q Negative impact
|
TEX-Q Process
|
TEX-Q Behavioural control
|
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
(Constant)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Treatment Experience
|
− .002
|
.964
|
− .002
|
.962
|
− .043
|
.384
|
− .079
|
.118
|
.017
|
.734
|
− .009
|
.863
|
− .014
|
.777
|
Pain disability (PDI)
|
− .030
|
.634
|
− .029
|
.654
|
.137
|
.027
|
.132
|
.036
|
.165
|
.010
|
− .009
|
.890
|
.042
|
.510
|
Pain catastrophising (CSQ)
|
.131
|
.044
|
.074
|
.259
|
.197
|
.002
|
.005
|
.941
|
− .083
|
.206
|
.073
|
.263
|
.015
|
.823
|
Somatic symptom impairment (PHQ-15)
|
.073
|
.231
|
.059
|
.341
|
.101
|
.089
|
.064
|
.293
|
.088
|
.153
|
.060
|
.324
|
.188
|
.002
|
State depression (PHQ-2)
|
− .040
|
.572
|
− .076
|
.288
|
− .025
|
.717
|
− .026
|
.706
|
− .071
|
.318
|
− .020
|
.780
|
− .117
|
.098
|
State anxiety (GAD-2)
|
− .190
|
.005
|
− .073
|
.289
|
− .155
|
.019
|
.141
|
.036
|
.082
|
.229
|
− .192
|
.005
|
− .044
|
.521
|
Age (in years)
|
.119
|
.029
|
.117
|
.034
|
.118
|
.025
|
− .025
|
.642
|
− .010
|
.849
|
.054
|
.324
|
.079
|
.146
|
Duration of symptoms (in months)
|
− .119
|
.023
|
− .127
|
.07
|
− .094
|
.066
|
.024
|
.645
|
.053
|
.315
|
− .140
|
.008
|
.003
|
.956
|
Modell fit
Adjusted R2
SE of the Estimate
F Change
df1/ df2
Sig. Change in F
|
.034
1.23
2.72
8 / 384
.006
|
.012
1.84
1.59
8 / 384
.126
|
.084
2.19
5.48
8 / 384
.001
|
.046
1.85
3.56
8 / 384
.001
|
.020
2.29
1.98
8 / 384
.048
|
.027
1.82
2.34
8 / 384
.019
|
.024
2.35
2.21
8 / 384
.026
|
|
Note. Significant p-values are in bold. CSQ = Sum score of the catastrophisation subscale of the Coping Strategies Questionnaire; GAD-2 = Sum score of the Generalized Anxiety Disorder Scale; PDI = Sum score of the Pain Disability Index; PHQ-2 = Sum score of the Patient Health Questionnaire for Depression; PHQ-15 = Sum score of the Patient Health Questionnaire; TEX-Q = Treatment Expectations Questionnaire.
Table 5
Multiple linear regression analyses of the three GEEEexpect items.
|
GEEEexpect_improvement
|
GEEEexpect_worsening
|
GEEEexpect_complaints
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
Stand. Beta
|
p-value
|
(Constant)
|
|
|
|
|
|
|
Treatment Experience
|
.029
|
.577
|
.059
|
.251
|
− .102
|
.045
|
Pain disability (PDI)
|
− .005
|
.936
|
.024
|
.705
|
.143
|
.025
|
Pain catastrophising (CSQ)
|
.100
|
.127
|
.076
|
.250
|
− .057
|
.377
|
Somatic symptom impairment (PHQ-15)
|
.102
|
.099
|
.029
|
.638
|
.088
|
.148
|
State depression (PHQ-2)
|
− .061
|
.391
|
.007
|
.918
|
.039
|
.578
|
State anxiety (GAD-2)
|
− .082
|
.233
|
.012
|
.867
|
.009
|
.898
|
Age (in years)
|
.088
|
.109
|
− .007
|
.899
|
.050
|
.354
|
Duration of symptoms (in months)
|
− .106
|
.045
|
.054
|
.313
|
.059
|
.260
|
Modell fit
Adjusted R2
SE of the Estimate
F Change
df1/ df2
Sig. Change in F
|
.
0.13
2.03
1.63
8 / 384
.114
|
.001
1.67
1.06
8 / 384
.388
|
.034
2.29
2.73
8 / 384
.006
|
Note. Significant p-values are in bold. CSQ = Sum score of the catastrophisation subscale of the Coping Strategies Questionnaire; GAD-2 = Sum score of the Generalized Anxiety Disorder Scale; GEEEexpect = Generic Rating Scale for Treatment Expectations; PDI = Sum score of the Pain Disability Index; PHQ-2 = Sum score of the Patient Health Questionnaire for Depression; PHQ-15 = Sum score of the Patient Health Questionnaire.