Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States

Abstract

Background:

Endometriosis symptoms have multifaceted manifestations, and there are few approved nonsurgical treatment options. Gonadotropin-releasing hormone (GnRH) agonists/antagonists for endometriosis vary on efficacy, safety profile, and out-of-pocket (OOP) cost, among other features.

Objectives:

This study quantified the importance that women with endometriosis in the United States (US) placed on pain and non-pain features that differ among these medications.

Design:

English-speaking, 18 to 50 years (premenopausal) US women with healthcare coverage for the past 3 years, who self-reported a laparoscopy-confirmed endometriosis diagnosis, were recruited via healthcare research panels to complete a cross-sectional survey.

Methods:

A discrete choice experiment (DCE) with a surgery opt-out option assessed preferences for eight GnRH agonist/antagonist attributes (reducing different types of pain, treatment administration, impact on daily activities, etc.). Best-worst scaling (BWS) assessed preferences for 11 non-pain medication attributes (dosage flexibility, short treatment onset, reversible side effects, etc.). Relative importance (RI) was estimated for each attribute.

Results:

Overall, 300 women were included (mean age 33.0 years; 76.7% White). Across DCE choice tasks, GnRH agonist/antagonist was chosen over surgery 46.7% of the time. Non-menstrual pelvic pain relief (RI = 23.1%), reducing monthly OOP cost (RI = 22.1%), and relief of dyspareunia (RI = 21.4%) and dysmenorrhea (RI = 12.9%) were most important for GnRH treatment choice. Among non-pain attributes in the BWS, short onset of treatment effect (RI = 13.1%), long-term safety (RI = 12.9%), and reducing fatigue (RI = 11.2%) were most important to women when choosing a pharmacologic endometriosis treatment.

Conclusion:

Relieving the three types of endometriosis pain and reducing cost are the most important considerations for women when selecting GnRH agonist/antagonist treatment. Women with endometriosis strongly prefer a medication that can be safely taken for longer periods of time, takes effect within a few menstrual cycles, and can reduce endometriosis-related fatigue. Findings can inform discussions between patients and healthcare providers to better align endometriosis treatment decision-making with patients’ individual needs and preferences.

Keywords

patient preference discrete choice experiment best-worst scaling endometriosis pelvic pain shared decision-making

Introduction

Endometriosis is a chronic gynecological disorder, defined by the presence of abnormally or ectopically located endometrial tissue, affecting women in their reproductive years.¹ Because surgery is generally required for a definitive diagnosis and large population-based studies are missing, the true prevalence of endometriosis is unknown; globally, the prevalence of endometriosis among women of reproductive age was estimated to range from 0.7% to 8.6% in the general population, according to a prior systematic review.² Endometriosis symptomology is heterogeneous, and endometriosis symptoms have similar presentations to other diseases, which often results in long delays between symptom onset and diagnosis.³ To facilitate earlier endometriosis case identification and diagnosis, it is recommended that all women who present with pelvic pain be further evaluated, as the majority of women with endometriosis present for care with some type of pelvic pain.⁴

The symptoms most commonly reported by women with endometriosis include dysmenorrhea, non-menstrual pelvic pain, and dyspareunia, which may lead to long-term sequelae.¹ A wide array of other symptoms often accompany pelvic pain, such as atypical uterine bleeding (e.g. premenstrual spotting and heavy menstrual bleeding), bowel symptoms (e.g. diarrhea and constipation), abdominal bloating, urinary difficulties (e.g. urinary urgency and painful urination), and significant fatigue.⁴ Endometriosis is also associated with comorbid health conditions, including depression and anxiety.⁵ Moreover, endometriosis pain is associated with substantial worsening of health-related quality of life (HRQoL) and considerable work and household productivity loss, with the negative impact on HRQoL and productivity increasing with a number of symptoms experienced and with symptom severity.^6,7 Results from a recent systematic review highlighted the considerable impact of endometriosis on various aspects of HRQoL, including physical health, emotional well-being, and daily activities, with chronic pain, fatigue, and comorbid mental health conditions often limiting work productivity and intimate relationships.⁸

Because there is no cure for endometriosis, the primary goal of treatment is relieving the heterogeneous compilation of symptoms. Pharmacological treatment options for endometriosis primarily include hormonal therapies; while hormonal contraceptives are used off-label to treat endometriosis, progestins, danazol, and gonadotropin-releasing hormone (GnRH) agonists/antagonists are indicated for endometriosis.⁹ Medical treatments, such as GnRH agonists/antagonists, can alleviate symptoms and improve functioning but may be associated with side effects that affect adherence, whereas surgical options offer relief by removing lesions but carry risks of recurrence and repeated procedures.¹⁰ The aim of surgical options, such as hysterectomy with or without oophorectomy, laparoscopy, or laparotomy, is to remove existing endometriosis lesions.⁹ However, endometriosis lesions frequently recur following surgical treatment, necessitating repeated surgeries.¹¹ While nonsteroidal anti-inflammatory drugs (NSAIDs), opioids, and other analgesics do not treat endometriosis, these medications are recommended to manage episodes of acute endometriosis pain.⁹

Notably, up to a third of patients may not respond or may be intolerant to hormonal contraceptives, which are often used as first-line treatment, according to a systematic literature review.¹² GnRH antagonists, which have recently been approved for use in endometriosis, provide pain relief and treat endometriosis by regulating the levels of sex steroid hormones associated with the growth of endometriosis lesions.¹³ Currently available GnRH agonists and antagonists differ in key attributes, including efficacy, side-effect profile, dosing schedule, and mode of administration.¹³ Given these variations, an appreciation of women’s willingness to trade-off between the potential benefits, risks, and other attributes associated with GnRH agonists/antagonists will be important for shared decision-making between patients and healthcare providers, which is recommended as part of the process of providing obstetric/gynecological care.¹⁴

The current study sought to evaluate how women with endometriosis assess various attributes that differentiate among currently available GnRH agonists/antagonists by eliciting their willingness to trade-off between GnRH agonist/antagonist treatment attributes and between pharmacologic treatment with a GnRH agonist/antagonist and surgery. In addition, this study examined the importance that women with endometriosis place on a broader set of non-pain attributes (eg, short treatment onset and reducing fatigue) associated with GnRH agonists/antagonists. A better understanding of how women with endometriosis view the variety of factors associated with treatment will raise awareness among healthcare providers about what is important to their patients and will support joint informed decision-making between patients and healthcare providers when selecting endometriosis treatments.

Materials and methods

The STROBE guidelines for cross-sectional studies were consulted when preparing the manuscript.¹⁵

Study design

A cross-sectional online quantitative survey, comprising a discrete choice experiment (DCE) and a best-worst scaling (BWS) exercise, was conducted among US English-reading women with endometriosis who had insurance coverage for the past 3 years. The DCE and BWS exercise provided different, yet complementary, information about patients’ preferences; the DCE provided information primarily regarding the importance of relieving different types of endometriosis pain and side effects of GnRH treatment, while the BWS provided information on the importance of non-pain attributes of pharmacological treatment for endometriosis.

A DCE was implemented to assess trade-offs between desirable and undesirable features of GnRH agonists/antagonists; in addition, the DCE assessed the extent to which a GnRH agonist/antagonist option would be preferred over surgery, which was presented in the DCE choice tasks as an alternative to pharmacologic treatment with a GnRH agonist/antagonist. A DCE is a method for quantifying the trade-offs patients are willing to make among different attributes of potential treatments by presenting patients with a series of choice tasks.¹⁶ The fundamental principle underlying the DCE is that the utility patients derive from a treatment is a function of its attributes. By asking respondents to choose between hypothetical treatment options with varying levels of these attributes, researchers can statistically infer the relative importance (RI) that patients place on each attribute and the trade-offs patients are willing to make between attributes.

An object case BWS exercise was used to quantify the RI of a larger set of non-pain pharmacologic treatment attributes. Specifically, in an object case BWS exercise, a subset of attributes is presented to respondents in each of a series of choice tasks. Respondents are asked to choose the attribute in the subset they view as best and the one they view as worst in each choice task, with this process repeating until a comprehensive ranking of the full set of attributes is obtained.¹⁷ The data for this study were collected from 30 November 2021 to 13 July 2022.

Survey instrument

For the current DCE study, hypothetical pharmacologic treatment profiles consisted of combinations of attributes, each with varying levels representing the range observed or expected to be observed in clinical trials or observational/real-world studies of GnRH agonists/antagonists. GnRH agonist/antagonist attributes were identified through the published literature. Specifically, GnRH agonist/antagonist treatments can vary in administration mode and frequency,¹⁸ as well as impact on each of the three most common pain symptoms (dysmenorrhea, non-menstrual pelvic pain, and/or dyspareunia), which can interfere with individuals’ daily activities;¹ GnRH agonist/antagonist treatments also differ in safety profile, especially the risk of loss in bone mineral density, which can lead to fractures and hot flashes.¹⁹ Lastly, out-of-pocket (OOP) costs may vary among treatment options, potentially affecting individuals’ ability to access treatment. Accordingly, the final DCE included efficacy, safety, administration, and OOP cost attributes (see Table 1 for attributes and attribute levels).

Table 1.

Discrete choice experiment attributes and levels.

Attribute(not shown)	Item stem	Levels
Dysmenorrhea	Worst pelvic pain during a period = X (0, no pain, to 10, pain as bad as you can imagine)	0 3 7 10
Non-menstrual pelvic pain	Worst pelvic pain when not having a period = X (0, no pain, to 10, pain as bad as you can imagine)	0 3 7 10
Dyspareunia	Worst pelvic pain during vaginal sexual intercourse = X (0, no pain, to 10, pain as bad as you can imagine)	0 3 7 10
Impact on daily activities	___ Limited in daily activities, such as work, school, social events, jobs around the house, or physical activities	Never Sometimes Often
Out-of-pocket cost	Total out-of-pocket cost for medication, including all doses, is ______	$5 per month $75 per month $200 per month $350 per month
Increased risk of bone fracture later in life	X out of 100 additional women (X%) will fracture a bone later in life	0 1 2
Treatment administration	How the medication is taken	One tablet taken once a day One tablet taken twice a day Injection administered by a healthcare provider every 1-3 months
Hot flashes	X out of 100 women (X%) have hot flashes	10 44 77

In the DCE, each participant was presented with a series of seven choice tasks, each including two hypothetical GnRH agonist/antagonist treatment profiles with varying levels of each attribute; for each choice task, the levels shown with each attribute were randomized. As endometriosis treatment guidelines include surgery as an alternative to medical management,⁹ each DCE choice task included a surgery opt-out option, which enabled study participants to select surgery instead of either hypothetical GnRH agonist/antagonist option. Including an option to choose an alternative to the hypothetical treatments is recommended when selecting something other than the hypothetical treatments in the DCE is a medically plausible scenario, and the opt-out alternative closely resembles a real-world option.²⁰ The surgery opt-out option was described in the DCE choice tasks as “any surgery to remove endometriosis tissue in the pelvis, around the uterus, or to remove the uterus and/or the ovaries”. An example DCE choice task is shown in Supplemental Figure 1.

The BWS exercise in this study assessed women’s priorities among 11 key non-pain attributes of pharmacologic treatment, which were informed by the available literature, including dosage flexibility, short treatment onset, reversibility of side effects, reductions in fatigue, depression, and headache, impact on libido, impact on sleep, ability to maintain employment, duration of treatment, and additional need for contraceptive use. In each question in the BWS exercise, women were presented with a set of 4 non-pain treatment attributes and asked to indicate which was most important and which was least important to them. The combination of non-pain attributes in each BWS question was determined by an experimental design, and each respondent answered 11 BWS questions. The brief description for each attribute included in the BWS exercise is shown in Table 2, and an example BWS choice task is shown in Supplemental Figure 2.

Table 2.

Best-worst scaling attributes.

Label	Attribute
Short treatment onset	When starting a treatment, you will begin to feel the treatment’s effects within the first few menstrual cycles.
Safe long-term treatment	You can safely take the treatment for a long period of time.
No libido change	Your interest in sex will not be affected.
Able to maintain employment	Your ability to get or maintain a job will not be affected.
Reducing depression	You will be less depressed.
Dosage flexibility	Your doctor offers different options for the dose strength and how often you take it, as appropriate to your needs.
Avoiding the need for additional contraceptives	You are not required to take additional contraceptives along with the treatment.
Side effects are reversible	Any side effects that you may experience are resolved quickly after treatment stops.
Reducing fatigue	You will be less fatigued or tired.
Reducing headache	You will have fewer headaches or migraines.
Avoiding sleep disruption	Your sleep will not be affected.

In addition to the DCE and BWS exercises, the survey included several questions to characterize the study sample. Specifically, these questions assessed sociodemographic and clinical characteristics, comorbidities, and endometriosis treatments used (ever and currently used). In addition, the survey included the 8-Item Patient Health Questionnaire (PHQ-8), which is a validated tool to assess depressive symptoms. The PHQ-8 asks respondents to report the frequency with which 8 depressive symptoms were experienced in the past 2 weeks; items are rated on a numeric rating scale, with response options ranging from 0 (not at all) to 3 (nearly every day).²¹ Total scores, which can range from 0 to 24, are computed by summing the responses across the eight items. Higher scores reflect greater depressive symptom severity, with scores of ⩾10 indicating a positive screen for depression.

Cognitive pretest interviews were conducted with 12 women who met all study eligibility criteria (described below) via telephone using a ‘think-aloud’ approach to qualitatively assess and refine the final survey instrument before its implementation. The cognitive pretest interviews also provided verification that the survey questions were clear, easy to understand, and interpreted as intended. Cognitive interview participants were not invited to complete the quantitative study.

Participants

For the quantitative survey and cognitive pretest interviews, English-speaking, premenopausal, female US residents aged 18–50 years, who had health insurance coverage during the previous 3 years, and who self-reported both having a diagnosis of endometriosis and that it was confirmed via laparoscopy, were recruited via a healthcare research panel. Women who self-reported surgical treatment of endometriosis or other female-specific/gynecological surgery in the past 3 months, or hysterectomy or oophorectomy at any time in the past, were excluded from the study, as having a hysterectomy, oophorectomy, or recent endometriosis-related or other gynecological surgery may potentially impact women’s treatment choice, pain, and pain management. Those participants who had a cancer diagnosis or cancer treatment in the past 10 years or who had ever been diagnosed with osteoporosis, bone disease, or uterine fibroids were excluded from the study. These women were excluded because hormone-dependent cancers, osteoporosis, and bone disease are contraindicated with GnRH agonists/antagonists; in addition, cancer may impact pain and pain management, which could influence individuals’ treatment choices. Those with uterine fibroids were excluded due to the similarities of this condition with endometriosis in terms of treatment options and symptoms, such as pelvic pain. Women who were pregnant, currently enrolled in a clinical trial, or who were unable or unwilling to provide informed consent were also excluded from participation.

The formula used to determine minimum sample sizes for aggregate level full-profile DCE modeling is nta/c >500, where n is the number of respondents, t is the number of choice tasks, a is the number of alternatives (profiles) per task, and c is equal to the largest number of levels (e.g., a side effect could have three levels, <1%, 10%, and 25% risk) for any one attribute.²² The study aimed to include 300 participants answering seven tasks as part of the experimental design, each with two alternatives per task and a maximum number of levels of four, as well as an opt-out. It was assumed that 15% of the 300 respondents chose the surgery opt-out option, leading to a calculated value of 1050 for the formula, which is much larger than the recommended value of 500.

With respect to the BWS, the minimum sample size required was computed using binomial sampling as an approximation. To be more consistent with the precision estimates around the DCE preference weights, we approximated a minimum sample size based on a ±4% margin of error, which requires an attribute to be seen 500 times. This yielded an overall minimum sample size of 125;²³ hence, the target sample size of 300 participants for the DCE was also sufficient for the BWS.

Data analysis

The DCE data were analyzed consistent with recommendations of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Good Research Practices for Conjoint Analysis Task Force.^20,24,25 Hierarchical Bayes models were used to estimate a preference weight for each attribute level for each respondent.²⁵ Mean preference weights were computed for each attribute level for the full sample. The change in utility (i.e., perceived value or desirability), associated with a change in the levels of each attribute, is represented by the difference between the preference weights for any two levels of an attribute. Larger differences indicate that the change between those two levels is more influential on the overall choice. Attribute RI was computed by dividing the range of each attribute (preference weight of the most preferred level minus preference weight of the least preferred level) by the sum of the ranges of all attributes and multiplying by 100. Attribute RI estimates, which sum to 100% across attributes, are ratio scaled; for example, an attribute with an RI of 30% is twice as important to endometriosis treatment choice as an attribute with an RI of 15%. Women who selected the surgery opt-out option on every DCE choice task were excluded from the computation of attribute-level preference weights and attribute RIs because there was no variation in their responses, which precluded the ability to estimate these parameters. The frequency and percentage of women who selected the surgery opt-out option on all DCE choice tasks are reported.

The BWS data were also analyzed using hierarchical Bayes estimation. Specifically, a coefficient (β), assumed to be from a multivariate normal distribution, was estimated for each item in the BWS exercise for each respondent, and then a modified logit model was used to determine the probability of each possible combination of best and worst ratings. The estimated coefficients were transformed with the logit model into selection probabilities and then standardized to sum to 100% across attributes. The standardized selection probabilities are RI estimates; that is, they represent the RI women in the sample place on each attribute or outcome included in the BWS. Further details on this analytic approach have been described elsewhere.²⁶

Descriptive analysis of the remaining data included means, medians, standard deviations (SDs), first and third quartiles (Q1 and Q3), minimums, and maximums for continuous variables and frequencies and percentages for categorical data. Women were compared, based on their patterns of surgery opt-out selection on the DCE choice tasks (always vs. sometimes vs. never chose surgery opt-out), on sociodemographic and clinical characteristics, comorbidities, and treatment use variables using one-way analysis of variance tests (ANOVAs) and chi-square tests for continuous and categorical data, respectively. For the aforementioned comparisons, p-values <0.05, two-tailed, were considered statistically significant.

Analyses of the DCE and BWS data were performed using Sawtooth Lighthouse Studio v9.13.1.²⁷ All other data analyses were performed using IBM SPSS v28.0.²⁸

Results

Participant characteristics

Overall, 300 eligible women who completed the survey were included in the analyses. Demographic characteristics are shown in Table 3, and clinical characteristics, comorbidities, and treatments used are shown in Table 4 for the overall sample of women with endometriosis. Most women were White (76.7%), with 10.3% identifying as Black/African American. More than half were employed (59.0%), and 42.7% had a college degree or higher. Women had a mean (±SD) age of 33.0 ± 7.6 years, and they had been diagnosed with endometriosis for a median of 9.0 years. Based on a list of options presented to respondents, the ever-experienced endometriosis/menstrual symptoms reported most often were abdominal pain (96.3%), back pain (87.0%), and fatigue (86.0%); 78.3% and 61.3% reported pelvic pain and dyspareunia, respectively. The most frequently reported comorbidities were anxiety (62.3%) and depression (54.3%). The mean PHQ-8 score was 10.2 ± 6.3, with 49.3% screening positive for depression on the PHQ-8.

Table 3.

Participant demographic characteristics.

Variables	Total SampleN=300
Age (years)
Mean ± SD	33.0 ± 7.6
Median (Q1, Q3)	32.0 (27.0, 38.0)
Min-Max	19.0-50.0
Race/Ethnicity, n (%) ^a
White/Caucasian	230 (76.7)
African American/Black	31 (10.3)
Hispanic/Latina	26 (8.7)
Other race/ethnicity^b	21 (7.0)
Education, n (%)
College degree or higher	128 (42.7)
Employment, n (%)
Employed (full-time or part-time)	177 (59.0)
Health insurance, n (%) ^a
Individual/family insurance plans	218 (72.7)
Medicaid	82 (27.3)
Medicare	19 (6.3)
TRICARE	7 (2.3)
VA/CHAMPUS	2 (0.7)

Q1: first quartile; Q3: third quartile; SD: standard deviation; TRICARE: military healthcare program formerly known as CHAMPUS; VA/CHAMPUS: Veterans Administration/Civilian Health and Medical Program of the Uniformed Services.

Women could select ⩾1 response option.

This race/ethnicity category included any responses of American Indian/Alaska native, Asian, other race/ethnicity, and/or native Hawaiian/other Pacific Islander.

Table 4.

Participant clinical characteristics, comorbidities, and treatments used.

Variables	Total SampleN=300
Comorbidities, n (%) ^a,b
Anxiety	187 (62.3)
Depression	163 (54.3)
Migraine	118 (39.3)
Lower back pain	97 (32.3)
Asthma	89 (29.7)
Infertility	77 (25.7)
Polycystic ovary syndrome	65 (21.7)
Irritable bowel syndrome	63 (21.0)
Eczema	46 (15.3)
Hypertension	45 (15.0)
Fibromyalgia	34 (11.3)
Depression (PHQ-8 score)
Mean ± SD	10.2 ± 6.3
Median (Q1, Q3)	9.0 (5.0, 14.0)
Min–Max	0.0-24.0
Depression screen (PHQ-8), n (%)
Positive screen (PHQ-8 score ⩾10)	148 (49.3)
Endometriosis/menstrual symptoms ever experienced, n(%) ^a
Abdominal pain	289 (96.3)
Back pain	261 (87.0)
Fatigue	258 (86.0)
Bloating	246 (82.0)
Headache/migraine	236 (78.7)
Pelvic pain	235 (78.3)
Feeling unwell, faint, or nauseous during period	227 (75.7)
Heavy bleeding during period	227 (75.7)
Gastrointestinal distress	201 (67.0)
Bleeding/spotting between periods	200 (66.7)
Dyspareunia	184 (61.3)
Painful bowel movements	173 (57.7)
Painful urination	99 (33.0)
Other endometriosis/menstrual symptoms	26 (8.7)
Time since endometriosis diagnosis (years) ^c
Mean ± SD	11.0 ± 7.4
Median (Q1, Q3)	9.0 (4.5, 15.2)
Min–Max	0.8-34.5
Therapies ever used for endometriosis/menstrual pain, n (%) ^a
OTC pain medications	279 (93.0)
Contraceptives (including oral birth control pill or IUD)	255 (85.0)
Rx medications (including Rx pain medications or antidepressants)	241 (80.3)
Surgery (any type)^d	190 (63.3)
Alternative therapies	138 (46.0)
OTC muscle relaxers	127 (42.3)
Prescribed physical therapy	47 (15.7)
Never taken/received any treatments	3 (1.0)
Rx medications currently using for endometriosis/menstrual pain (specific treatments), n (%) ^a
Oral contraceptive pill	96 (32.0)
Another Rx medication not listed	96 (32.0)
Opioid-based Rx pain medications	65 (21.7)
Antidepressant	51 (17.0)
Other Rx pain medications	50 (16.7)
Elagolix	32 (10.7)
Hormonal IUD	30 (10.0)
Rx muscle relaxer	30 (10.0)
Progestin	13 (4.3)
Leuprorelin	6 (2.0)
Another GnRH agonist	5 (1.7)
Uses ⩾2 of the Rx medications listed	141 (47.0)
Not currently taking/using Rx medications	24 (8.0)
Rx medications currently using for endometriosis/menstrual pain (treatment categories), n (%) ^a
Any hormonal therapy^e	152 (50.7)
Any other Rx medication^f	129 (43.0)
Any Rx pain medication^g	96 (32.0)
Not currently taking/using Rx medications	24 (8.0)

GnRH: gonadotropin-releasing hormone; IUD: intrauterine device; OTC: over-the-counter; PHQ-8: 8-Item Patient Health Questionnaire; Q1: first quartile; Q3: third quartile; Rx: prescription; SD: standard deviation.

Women could select ⩾1 response option.

Only those comorbidities reported by ⩾10% of women are shown.

N = 2 women were excluded from the calculation of time since diagnosis due to extreme outlier values on this variable.

Reflects the percentage of participants who self-reported receiving “endometrial ablation or excision” and/or “other female-specific/gynecological surgery” for endometriosis.

This category includes any use of oral contraceptive pill, elagolix, hormonal IUD, progestin, leuprorelin, another GnRH agonist, and/or danazol.

This category includes any use of antidepressants, Rx muscle relaxers, and/or another Rx medication not listed.

This category includes any use of opioid-based Rx pain medications and/or any other Rx pain medications.

Of the pharmacologic treatments ever used to manage their endometriosis symptoms and menstrual pain, the most commonly reported were over-the-counter (OTC) pain medications (93.0%) and oral contraceptives (85.0%) (Table 4). Regarding current prescription treatments, hormonal therapies were the most frequently used for the treatment of endometriosis symptoms (50.7%). In addition, 21.7% reported using opioid-based prescription pain medications. Overall, 47.0% of respondents had ⩾2 prescription treatments for endometriosis.

Treatment preferences

DCE results

As 300 participants were each presented with seven DCE choice tasks, a total of 2100 choice tasks were completed. Across the seven DCE choice tasks, each showed 2 options of different GnRH agonist/antagonist profiles, as well as a surgery opt-out option (any surgery to remove endometrial tissue). Of the 2100 total choice tasks, participants chose a GnRH agonist/antagonist profile 46.7% of the time; the surgery opt-out option was selected 53.3% of the time. Among the 300 participants, 227 (75.6%) chose a GnRH agonist/antagonist profile in at least 1, but not in all 7, DCE choice tasks; 38 participants (12.7%) chose a GnRH agonist/antagonist profile in every DCE choice task. The remaining participants (n = 35, 11.7%) selected the surgery opt-out option in every DCE choice task.

Demographic characteristics are shown in Table 5, with clinical and prior treatment characteristics of participants who never, sometimes, and always selected the surgery opt-out option shown in Table 6. Across these three groups, women who always selected the surgery opt-out option were most often White (85.7%), commonly had Medicaid coverage (54.3%), and least often had a college degree (28.6%), relative to those who sometimes (80.6%, 25.1%, and 41.9%, respectively) or never (44.7%, 15.8%, and 60.5%, respectively) selected the surgery opt-out (all, p < 0.05) (Table 5). The three groups were generally similar with regard to comorbidities, except that those who always selected the surgery opt-out option more frequently reported having a diagnosis of depression (74.3%) than those who sometimes or never selected the surgery opt-out (53.7% and 39.5%, respectively, p = 0.011; Table 6). More women who always selected the opt-out option (80.0%) also reported that they had surgery in the past to treat their endometriosis pain than those who sometimes (63.0%) or never (50.0%) selected the surgery opt-out (p = 0.029).

Table 5.

Participant demographic characteristics by surgery opt-out selection.

Variables	Never Chose Surgery Opt-OutN = 38	Sometimes Chose Surgery Opt-OutN = 227	Always Chose Surgery Opt-OutN = 35	p-value
Age (years)
Mean ± SD	34.0 ± 8.2	32.9 ± 7.6	32.1 ± 7.4	0.564
Median (Q1, Q3)	34.5 (29.0, 39.0)	32.0 (27.0, 38.0)	31.0 (28.0, 36.0)
Min–Max	19.0–49.0	19.0–49.0	19.0–50.0
Race/ethnicity, n(%) ^a
White/Caucasian	17 (44.7)	183 (80.6)	30 (85.7)	<0.001
African American/Black	6 (15.8)	22 (9.7)	3 (8.6)	0.487
Hispanic/Latina	7 (18.4)	15 (6.6)	4 (11.4)	0.047
Other race/ethnicity^b	6 (15.8)	13 (5.7)	2 (5.7)	0.076
Education, n(%)
College degree or higher	23 (60.5)	95 (41.9)	10 (28.6)	0.020
Employment, n(%)
Employed (full-time or part-time)	18 (47.4)	138 (60.8)	21 (60.0)	0.295
Health insurance, n(%) ^a
Individual/family insurance plans	30 (78.9)	167 (73.6)	21 (60.0)	0.159
Medicaid	6 (15.8)	57 (25.1)	19 (54.3)	<0.001
Medicare	7 (18.4)	8 (3.5)	4 (11.4)	<0.001
TRICARE	1 (2.6)	6 (2.6)	0 (0.0)	0.623
VA/CHAMPUS	1 (2.6)	1 (0.4)	0 (0.0)	0.269
Not sure/no health insurance/prefer not to answer	0 (0.0)	0 (0.0)	0 (0.0)	N/A

N/A: not applicable; Q1: first quartile; Q3: third quartile; SD: standard deviation; TRICARE: military healthcare program formerly known as CHAMPUS; VA/CHAMPUS: Veterans Administration/Civilian Health and Medical Program of the Uniformed Services.

Women could select ⩾1 response option.

This race/ethnicity category included any responses of American Indian/Alaska native, Asian, other race/ethnicity, or native Hawaiian/other Pacific Islander.

Table 6.

Participant clinical characteristics, comorbidities, and treatments used by surgery opt-out selection.

Variables	Never Chose Surgery Opt-OutN = 38	Sometimes Chose Surgery Opt-OutN = 227	Always Chose Surgery Opt-OutN = 35	p-value
Comorbidities, n(%) ^a,b
Anxiety	21 (55.3)	138 (60.8)	28 (80.0)	0.058
Depression	15 (39.5)	122 (53.7)	26 (74.3)	0.011
Migraine	14 (36.8)	90 (39.6)	14 (40.0)	0.944
Lower back pain	7 (18.4)	75 (33.0)	15 (42.9)	0.075
Asthma	10 (26.3)	68 (30.0)	11 (31.4)	0.876
Infertility	11 (28.9)	58 (25.6)	8 (22.9)	0.835
Polycystic ovary syndrome	4 (10.5)	51 (22.5)	10 (28.6)	0.146
Irritable bowel syndrome	3 (7.9)	52 (22.9)	8 (22.9)	0.105
Eczema	7 (18.4)	32 (14.1)	7 (20.0)	0.567
Hypertension	4 (10.5)	37 (16.3)	4 (11.4)	0.536
Fibromyalgia	3 (7.9)	25 (11.0)	6 (17.1)	0.439
Endometriosis/menstrual symptoms ever experienced, n(%) ^a
Abdominal pain	33 (86.8)	222 (97.8)	34 (97.1)	0.004
Back pain	30 (78.9)	201 (88.5)	30 (85.7)	0.258
Fatigue	29 (76.3)	197 (86.8)	32 (91.4)	0.140
Bloating	28 (73.7)	187 (82.4)	31 (88.6)	0.243
Headache/migraine	30 (78.9)	179 (78.9)	27 (77.1)	0.973
Pelvic pain	21 (55.3)	180 (79.3)	34 (97.1)	<0.001
Feeling unwell, faint, or nauseous during period	24 (63.2)	173 (76.2)	30 (85.7)	0.075
Heavy bleeding during period	18 (47.4)	178 (78.4)	31 (88.6)	<0.001
Gastrointestinal distress	19 (50.0)	153 (67.4)	29 (82.9)	0.011
Bleeding/spotting between periods	18 (47.4)	156 (68.7)	26 (74.3)	0.021
Dyspareunia	12 (31.6)	143 (63.0)	29 (82.9)	<0.001
Painful bowel movements	11 (28.9)	136 (59.9)	26 (74.3)	<0.001
Painful urination	8 (21.1)	71 (31.3)	20 (57.1)	0.002
Other endometriosis/menstrual symptoms	4 (10.5)	15 (6.6)	7 (20.0)	0.029
Time since endometriosis diagnosis (years) ^c
Mean ± SD	11.2 ± 7.0	10.9 ± 7.3	11.1 ± 8.3	0.972
Median (Q1, Q3)	9.1 (5.9, 18.0)	9.3 (4.6, 15.0)	10.1 (3.9, 17.6)
Min–Max	1.8–26.0	0.8–34.5	1.3–32.5
Therapies ever used for endometriosis/menstrual pain, n (%) ^a
OTC pain medications	32 (84.2)	216 (95.2)	31 (88.6)	0.028
Contraceptives (oral birth control pill or IUD)	34 (89.5)	195 (85.9)	26 (74.3)	0.143
Rx medications	30 (78.9)	184 (81.1)	27 (77.1)	0.841
Surgery (any type)^d	19 (50.0)	143 (63.0)	28 (80.0)	0.029
Alternative therapies	19 (50.0)	102 (44.9)	17 (48.6)	0.802
OTC muscle relaxers	15 (39.5)	96 (42.3)	16 (45.7)	0.864
Prescribed physical therapy	2 (5.3)	39 (17.2)	6 (17.1)	0.168
Never taken/received any treatments	0 (0.0)	2 (0.9)	1 (2.9)	0.441
Rx medications currently used for endometriosis/menstrual pain (specific treatments), n (%) ^a
Oral contraceptive pill	15 (39.5)	75 (33.0)	6 (17.1)	0.098
Another Rx medication not listed	13 (34.2)	75 (33.0)	8 (22.9)	0.462
Opioid-based Rx pain medications	11 (28.9)	49 (21.6)	5 (14.3)	0.315
Antidepressant	3 (7.9)	39 (17.2)	9 (25.7)	0.127
Other Rx pain medications	7 (18.4)	37 (16.3)	6 (17.1)	0.946
Elagolix	5 (13.2)	26 (11.5)	1 (2.9)	0.268
Hormonal IUD	2 (5.3)	20 (8.8)	8 (22.9)	0.021
Rx muscle relaxer	2 (5.3)	24 (10.6)	4 (11.4)	0.574
Progestin	1 (2.6)	11 (4.8)	1 (2.9)	0.743
Leuprorelin	0 (0.0)	5 (2.2)	1 (2.9)	0.621
Another GnRH agonist	1 (2.6)	4 (1.8)	0 (0.0)	0.663
Danazol	0 (0.0)	0 (0.0)	0 (0.0)	N/A
Uses ⩾2 of the Rx medications listed	19 (50.0)	107 (47.1)	15 (42.9)	0.827
Not currently taking/using Rx medications	2 (5.3)	19 (8.4)	3 (8.6)	0.801
Rx medications currently used for endometriosis/menstrual pain (treatment categories), n (%) ^a
Any hormonal therapy^e	20 (52.6)	116 (51.1)	16 (45.7)	0.811
Any other Rx medication^f	15 (39.5)	98 (43.2)	16 (45.7)	0.860
Any Rx pain medication^g	14 (36.8)	73 (32.2)	9 (25.7)	0.592
Not currently taking/using Rx medications	2 (5.3)	19 (8.4)	3 (8.6)	0.801

GnRH: gonadotropin-releasing hormone; IUD: intrauterine device; N/A: not applicable; OTC: over-the-counter; Q1: first quartile; Q3: third quartile; Rx: prescription; SD: standard deviation.

Women could select ⩾1 response option.

Only those comorbidities reported by ⩾10% of women in the total sample of N = 300 are shown.

N = 2 women were excluded from the calculation of time since diagnosis due to extreme outlier values on this variable.

Reflects the percentage of participants who self-reported receiving “endometrial ablation or excision” and/or “other female-specific/gynecological surgery” for endometriosis.

This category includes any use of oral contraceptive pill, elagolix, hormonal IUD, progestin, leuprorelin, another GnRH agonist, and/or danazol.

This category includes any use of antidepressants, Rx muscle relaxers, and/or another Rx medication not listed.

This category includes any use of opioid-based Rx pain medications and/or any other Rx pain medications.

The mean attribute-level preference weights for the DCE choice tasks are shown in Supplemental Figure 3. The RI estimates from the DCE indicate that reductions in the three types of endometriosis-related pain were most important when selecting a GnRH agonist/antagonist treatment (Figure 1). Specifically, improvement in non-menstrual pelvic pain (RI = 23.1%) was the most influential attribute in GnRH agonist/antagonist treatment selection, overall; improvements in dyspareunia (RI = 21.4%) and dysmenorrhea (RI = 12.9%) also strongly influenced GnRH agonist/antagonist treatment selection. These reductions in pain were significantly more important to GnRH agonist/antagonist treatment selection than reducing the impact of endometriosis on daily activities (6.1%), a reduction in the risk of hot flashes (RI = 5.9%), a change in treatment administration from injection once every 1–3 months to once-daily oral administration (RI = 4.4%), and lower risk of bone fracture later in life (RI = 4.1%) (all, p < 0.05). A reduction in monthly OOP cost was also a very important consideration for women with endometriosis (RI = 22.1%).

Figure 1.

Attribute relative importance estimates from the discrete-choice experiment.

BWS results

As shown in Figure 2, when choosing a pharmacologic endometriosis treatment, the non-pain attributes that were most important to women included: short treatment onset (RI = 13.1%), safe, long-term treatment option (RI = 12.9%), and reduction in fatigue (RI = 11.2%). The non-pain attributes that were least important to women were: dosage flexibility (RI = 7.8%), avoiding sleep disruption (RI = 6.5%), and avoiding the need for additional contraceptives (RI = 4.3%). Reversible side effects (RI = 9.3%), ability to maintain employment (RI = 9.1%), reducing headache (RI = 8.7%), reducing depression (RI = 8.6%), and avoiding libido change (RI = 8.5%) were more important to pharmacological treatment choice than dosage flexibility, avoiding sleep disruption, and avoiding the need for additional contraceptives, but less important than short treatment onset, safe long-term treatment, and reducing fatigue.

Figure 2.

Attribute relative importance estimates from the best-worst scaling exercise (non-pain attributes).

Discussion

The current study assessed the treatment preferences of US women with endometriosis and provided unique insights into the importance of various attributes associated with GnRH agonist/antagonist treatment from the patient’s perspective. Overall, findings suggest that relieving the 3 most common types of endometriosis-related pain (non-menstrual pelvic pain, dysmenorrhea, and dyspareunia) is highly influential in pharmacologic treatment selection. In addition to endometriosis pain relief, several non-pain attributes, namely short treatment onset, being a safe long-term treatment option, and fatigue reduction emerged as influential to medication choice.

Among the overall sample of women with endometriosis, surgery was selected over a GnRH agonist/antagonist 53.3% of the time across DCE choice tasks. However, endometriosis lesions have been shown to recur in up to 67% of cases after having any type of laparoscopic surgery,²⁹ although recurrence rates vary depending upon post-surgery follow-up time; approximately 15% experienced persistent pain and 3% to 5% developed new endometriosis symptoms or worsening pain following standard hysterectomy, which may subsequently lead to repeat surgeries.¹¹ It is likewise possible that women may have chosen surgical treatment due to mistrust of medical management. For instance, in a prior meta-synthesis of 14 qualitative studies, women reported that healthcare providers often trivialized patients’ endometriosis symptoms, dismissing their pain as just being a normal part of the menstrual cycle, and made little to no effort to identify the underlying cause of their symptoms.³⁰ In addition, some women with endometriosis may have been more willing to accept surgery due to the inadequacy of prior pharmacological treatments they have used to alleviate their symptoms, as approximately 3-in-10 patients do not respond to first-line treatment with hormonal contraceptives.¹² This is further supported, given that one-quarter of those women who always selected the surgery opt-out option reported currently using prescription pain medications, of which over one-third were concomitantly using hormonal therapy (data not shown), suggesting that many of these women experience acute episodes of endometriosis pain even while on hormonal treatment. Nevertheless, women in this study, overall, differed in the frequency with which they chose surgery over a GnRH agonist/antagonist option, as well as in their preferences for pharmacologic treatment attributes. Hence, results suggest there may be opportunities for healthcare providers and patients to have an informed discussion of endometriosis symptoms, as well as the benefits and risks of available alternatives to surgery, including newer GnRH agonists/antagonists.

To date, evidence on patient preferences for endometriosis treatments is limited. In contrast to the current study, in which non-menstrual pelvic pain was the most important attribute for medication choice, a DCE study conducted in the United Kingdom, which included mode of administration, ability to conceive, worst level of pelvic pain, dysmenorrhea, and dyspareunia, need for pain medication (days per month), fracture risk, and OOP cost attributes, reported that dysmenorrhea was the most inﬂuential pain type,³¹ although variations between that study and the current one may reflect differences in the patient population included. Specifically, Germain and colleagues reported that approximately 8-in-10 study participants had severe menstrual pain; hence, it is reasonable to anticipate that relieving dysmenorrhea pain would be of foremost importance to participants in that study. In another DCE study conducted in the US, women placed the greatest importance on the risk of hot ﬂashes, followed by improvements in dyspareunia, non-menstrual pelvic pain, and dysmenorrhea,³² whereas the risk of hot flashes was less important to treatment choice in the current study. The difference in the results of these studies may be due to the fact that hot flashes were described in the DCE conducted by Poulos and colleagues as being moderate to severe, whereas the current study described hot flashes in terms of the percentage of women who will have them, with severity being left to the subjective interpretation of the participant. Participants in the current study may have placed less importance on that attribute if they assumed that the hot flashes would be mild. The findings from the current study suggest that relieving the 3 most common types of endometriosis-related pain was highly influential in medication selection, and these findings are generally aligned with prior research. However, unlike previous research, our study was also able to provide a better understanding of the importance of non-pain attributes by demonstrating that women placed great importance on having safe, long-term treatment, feeling treatment effects within a few menstrual cycles, and reducing fatigue.

Results showed that OOP costs were an important driver of treatment preferences. OOP costs may be of particular relevance to treatment accessibility for lower-income women with endometriosis, as the options available to them would be constrained due to personal budget limitations. As such, some patients may receive suboptimal treatment for their endometriosis because they are unable to afford more effective options. Given that over a quarter of study participants had Medicaid, affordability may potentially impact access to treatment for a sizable minority of women with endometriosis. To mitigate the substantial psychosocial and economic burden associated with endometriosis, public health policymakers should focus on making effective treatments (i.e., pharmacological or surgical) accessible to all patients, with the specific treatment selected based on what is medically optimal for the individual patient.

The current study also identified differences in participant characteristics between groups based on opt-out selection frequency. For instance, those who always selected the surgery opt-out option in the DCE were 2.2 to 3.4 times more likely to have Medicaid than those who never or sometimes selected the opt-out option. It is possible that many participants with Medicaid believe that a one-time surgery would be more affordable than using pharmacological treatment, even though surgery comes with a risk of recurrence and a potential need for additional procedures.¹⁰ Results further showed that those who always selected the surgery opt-out also had a higher rate of depression than the other two groups. As noted in a prior systematic review, depressive symptoms can increase pain intensity and duration.³³ If endometriosis pain is exacerbated by depressive symptoms, and those with depression perceive that surgery will be more likely than pharmacological treatment to effectively alleviate their pain, this may account for the more frequent selection of the surgery opt-out option among these participants. Accordingly, these findings may present potential opportunities for patient education.

Current guidelines recommend that the choice of either hormonal therapy or surgery to treat endometriosis involves a discussion between patients and healthcare professionals about the individual patient’s preferences, as well as considerations regarding the side effects, efficacy, costs, and availability of the particular treatment.⁹ Furthermore, shared decision-making is guideline-recommended as part of routine obstetric/gynecological care to enable patients to make informed decisions about their health with the support of their healthcare provider.¹⁴ Shared decision-making will help to ensure women with endometriosis clearly understand the benefits, risks, costs, and potential outcomes associated with the different treatment options available to them.³⁴ A candid dialogue to address both clinical safety, efficacy, and cost burden with each patient is critical to the management of endometriosis. The current study’s findings provide a fuller and more nuanced depiction of the various factors, including pain and non-pain attributes, that underlie the treatment choices of women with endometriosis. As such, shared decision-making that includes a discussion of both non-pain and pain attributes may help to optimize treatment selection for women with endometriosis.

Limitations

The DCE cannot accommodate all factors that could potentially influence treatment choice or real-world treatment decisions. Accordingly, stated preferences may not perfectly correspond with the actual treatment choices of women with endometriosis. In addition, the description of the surgery opt-out option in the DCE choice tasks was broad, potentially encompassing procedures of varying degrees of invasiveness. Participants were not asked about all the possible types of surgical procedures they received to treat endometriosis, whether any surgery received was curative, or about the severity of involvement, which may potentially influence the severity of their symptoms and their perceptions of the surgery opt-out option in the DCE. As such, it is possible that estimated treatment preferences would differ from those observed in the current study if a specific surgical procedure had been described. Moreover, in the BWS exercise, a specific timeframe was not provided for the long-term treatment attribute, leaving the interpretation of this attribute up to the individual participant, which could limit generalizability to the extent that there may have been wide variation in the time period envisioned by participants. Furthermore, participants were not asked to provide their rationale for selecting the surgery opt-out option on the DCE, thus, future research will be needed to discern the reasons why some women with endometriosis may be more motivated to choose surgery over pharmacological treatment. When entering the survey, participants may have had preexisting perceptions about oral contraceptives, treatment type, and administration, and other factors relevant to treatment choice, which were not controlled for in the estimation of preference weights; likewise, it is also possible that some participants may have been more knowledgeable about the potential benefits and risks of available endometriosis treatments. The profiles presented were not fixed to correspond with specific GnRH agonist/antagonist treatments. Data on clinical characteristics, treatment, and diagnosed endometriosis and comorbid health conditions were self-reported and could not be medically confirmed for accuracy. Participants were not asked to report the dosage of any medications used, and it is possible that dosage might influence participants’ perceptions of endometriosis treatments, including efficacy, side effects, and administration. Furthermore, participants were not asked to report on whether imaging, such as ultrasound or magnetic resonance imaging, was also used to confirm their endometriosis diagnosis.

The study sample may not be fully representative of the overall population of women with endometriosis in the US, which could limit generalizability. For example, recruitment via a healthcare research panel may have led to the inclusion of more educated participants. Notably, the racial distribution of participants in the current study was similar to that reported for US women with diagnosed endometriosis in a recent analysis of National Survey of Family Growth data from the period 2011 to 2019;³⁵ in the current study 76.7% and 10.3% self-identified as White and African American, respectively, with 71.8% and 8.6% reported by Al-Lami and colleagues for these racial groups. However, a prior systematic review and meta-analysis found that African American women are less likely to be diagnosed with endometriosis than White women;³⁶ as such, studies that only include women with diagnosed endometriosis, such as the current research, may underrepresent African Americans and overestimate Whites, thereby reducing the ability to generalize results to the broader population of US women with endometriosis. The current study also excluded women with endometriosis who had prior hysterectomy or oophorectomy, recent endometriosis-related or other gynecological surgery, cancer treatment or diagnosis, and uterine fibroids diagnosis. Given the aforementioned surgeries and health conditions are not contraindicated with GnRH agonists/antagonists (except for hormone-dependent types of cancer), these women could potentially be eligible to receive this treatment; as such, it is unknown whether the findings reflect the preferences and treatment choices of some subpopulations of women with endometriosis. Nevertheless, the characteristics of our study sample were generally similar in terms of age, US geographic distribution, and endometriosis and menstrual symptoms, to those of prior epidemiological studies on the population of US women with endometriosis.^37,38

Conclusion

In the patient population who participated in this study, US women aged 18 to 50 years with endometriosis, relieving non-menstrual pelvic pain, followed closely by reducing OOP costs, were most influential to the choice of GnRH agonist/antagonist treatment. Improvement in dyspareunia, which was nearly as important as relieving non-menstrual pelvic pain, and dysmenorrhea relief, were likewise strong drivers of GnRH agonist/antagonist treatment selection among women with endometriosis. Beyond the desire for endometriosis pain relief and lower OOP costs, women preferred a medication that has a rapid onset, is safe to take long-term, and can reduce fatigue. Findings suggest that discussions between patients and their healthcare providers can encompass a broad array of factors, in addition to pain, to ensure endometriosis treatment selection more closely aligns with women’s preferences and best meets their needs.

Supplemental Material

sj-docx-1-whe-10.1177_17455057251331700 – Supplemental material for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States

Supplemental material, sj-docx-1-whe-10.1177_17455057251331700 for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States by Martine C. Maculaitis, Ruth Kim, Elke Hunsche, Kathleen M. Beusterien, Paul Cislo, Nicole T. Ansani and Brett Hauber in Women’s Health

Supplemental Material

sj-docx-2-whe-10.1177_17455057251331700 – Supplemental material for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States

Supplemental material, sj-docx-2-whe-10.1177_17455057251331700 for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States by Martine C. Maculaitis, Ruth Kim, Elke Hunsche, Kathleen M. Beusterien, Paul Cislo, Nicole T. Ansani and Brett Hauber in Women’s Health

Supplemental Material

sj-tiff-3-whe-10.1177_17455057251331700 – Supplemental material for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States

Supplemental material, sj-tiff-3-whe-10.1177_17455057251331700 for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States by Martine C. Maculaitis, Ruth Kim, Elke Hunsche, Kathleen M. Beusterien, Paul Cislo, Nicole T. Ansani and Brett Hauber in Women’s Health

Supplemental Material

sj-tiff-4-whe-10.1177_17455057251331700 – Supplemental material for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States

Supplemental material, sj-tiff-4-whe-10.1177_17455057251331700 for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States by Martine C. Maculaitis, Ruth Kim, Elke Hunsche, Kathleen M. Beusterien, Paul Cislo, Nicole T. Ansani and Brett Hauber in Women’s Health

Supplemental Material

sj-tiff-5-whe-10.1177_17455057251331700 – Supplemental material for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States

Supplemental material, sj-tiff-5-whe-10.1177_17455057251331700 for Attribute preferences associated with gonadotropin-releasing hormone agonists/antagonists among women with endometriosis in the United States by Martine C. Maculaitis, Ruth Kim, Elke Hunsche, Kathleen M. Beusterien, Paul Cislo, Nicole T. Ansani and Brett Hauber in Women’s Health

Footnotes

Acknowledgements

The authors acknowledge Oliver Will (employee of Oracle Life Sciences) for contributions to study design and statistical analysis and Lewis Kopenhafer (employee of Oracle Life Sciences) for contributions to study design and data acquisition. A portion of the data reported in the manuscript was included in posters presented at the following conferences: International Society for Pharmacoeconomics and Outcomes Research European Congress (ISPOR-Europe; held from November 6–9, 2022, in Vienna, Austria; URL: https://www.valueinhealthjournal.com/article/S1098-3015(22)04429-1/fulltext) and European Society of Human Reproduction and Embryology Annual Meeting (ESHRE; held from July 3–6, 2022, in Milan, Italy; URL: ).

ORCID iDs

Martine C. Maculaitis

Elke Hunsche

Brett Hauber

Statements and Declarations

Supplemental material

Supplemental material for this article is available online.

References

Alimi

Iwanaga

Loukas

, et al. The clinical anatomy of endometriosis: a review. Cureus 2018; 10(9): e3361.

Ghiasi

Kulkarni

Missmer

SA.

Is endometriosis more common and more severe than it was 30 years ago?

J Minim Invasive Gynecol 2020; 27(2): 452–461.

Agarwal

Chapron

Giudice

, et al. Clinical diagnosis of endometriosis: a call to action. Am J Obstet Gynecol 2019; 220(4): 354.e1–354.e12.

Fraser

IS.

Recognising, understanding and managing endometriosis. J Hum Reprod Sci 2008; 1(2): 56–64.

Van Barneveld

Manders

van Osch

FHM

, et al. Depression, anxiety, and correlating factors in endometriosis: a systematic review and meta-analysis. J Womens Health 2022; 31(2): 219–230.

Soliman

Coyne

Zaiser

, et al. The burden of endometriosis symptoms on health-related quality of life in women in the United States: a cross-sectional study. J Psychosom Obstet Gynaecol 2017; 38(4): 238–248.

Soliman

Coyne

Gries

, et al. The effect of endometriosis symptoms on absenteeism and presenteeism in the workplace and at home. J Manag Care Spec Pharm 2017; 23(7): 745–754.

Maulenkul

Kuandyk

Makhadiyeva

, et al. Understanding the impact of endometriosis on women's life: an integrative review of systematic reviews. BMC Womens Health 2024; 24(1): 524.

Becker

Bokor

Heikinheimo

, et al. ESHRE guideline: endometriosis. Hum Reprod Open 2022; 2022(2): hoac009.

10.

D’Alterio

Saponara

Agus

, et al. Medical and surgical interventions to improve the quality of life for endometriosis patients: a systematic review. Gynecol Surg 2021; 18(1): 13.

11.

Vercellini

Barbara

Abbiati

, et al. Repetitive surgery for recurrent symptomatic endometriosis: what to do?. Eur J Obstet Gynecol Reprod Biol 2009; 146(1): 15–21.

12.

Donnez

Dolmans

MM.

Endometriosis and medical therapy: from progestogens to progesterone resistance to GnRH antagonists: a review. J Clin Med 2021; 10(5): 1085.

13.

Saleh

Taylor

HS.

Clinical applications of gonadotropin-releasing hormone analogues: a broad impact on reproductive medicine. F&S Reports 2023; 4(2 Suppl): 83–87.

14.

Informed consent and shared decision making in obstetrics and gynecology: ACOG Committee Opinion, Number 819. Obstet Gynecol 2021; 137(2): e34–e41.

15.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies [published correction appears in Ann Intern Med 2008 Jan 15; 148(2): 168]. Ann Intern Med 2007; 147(8): 573–577.

16.

Ali

Ronaldson

Ordinal preference elicitation methods in health economics and health services research: using discrete choice experiments and ranking methods. Br Med Bull 2012; 103(1): 21–44.

17.

Louviere

Lings

Islam

, et al. An introduction to the application of (case 1) best–worst scaling in marketing research. Int J Res Market 2013; 30(3): 292–303.

18.

Taylor

RN.

Role of GnRH analogues in endometriosis management. Contemporary Ob/Gyn, https://www.contemporaryobgyn.net/view/role-of-gnrh-analogues-in-endometriosis-management (2022, accessed 13 July 2023).

19.

Rzewuska

Żybowska

Sajkiewicz

, et al. Gonadotropin-releasing hormone antagonists—a new hope in endometriosis treatment? J Clin Med 2023; 12(3): 1008.

20.

Bridges

Hauber

Marshall

, et al. Conjoint analysis applications in health–a checklist: a report of the ISPOR Good Research Practices for Conjoint Analysis Task Force. Value Health 2011; 14(4): 403–413.

21.

Kroenke

Strine

Spitzer

, et al. The PHQ-8 as a measure of current depression in the general population. J Affect Disord 2009; 114(1–3): 163–173.

22.

Orme

Getting started with conjoint analysis: strategies for product design and pricing research (4th ed). Research Publishers LLC, 2019.

23.

Orme

Chrzan

Becoming an expert in conjoint analysis, Provo, UT, Sawtooth Software, Inc., 2017.

24.

Bridges

JFP

de Bekker-Grob

Hauber

, et al. A roadmap for increasing the usefulness and impact of patient-preference studies in decision making in health: a Good Practices Report of an ISPOR Task Force. Value Health 2023; 26(2): 153–162.

25.

Hauber

González

Groothuis-Oudshoorn

, et al. Statistical methods for the analysis of discrete choice experiments: a report of the ISPOR Conjoint Analysis Good Research Practices Task Force. Value Health 2016; 19(4): 300–315.

26.

Mühlbacher

Kaczynski

Zweifel

, et al. Experimental measurement of preferences in health and healthcare using best-worst scaling: an overview. Health Econ Rev 2016; 6(1): 2.

27.

Sawtooth Software. The MaxDiff System technical paper, version 9. Report, Sawtooth Software, Inc., Provo, UT, 2020.

28.

IBM Corp. IBM SPSS Statistics for Windows, Version 28.0. IBM Corp, 2021.

29.

Selçuk

Bozdağ

Recurrence of endometriosis; risk factors, mechanisms and biomarkers; review of the literature. J Turk Ger Gynecol Assoc 2013; 14(2): 98–103.

30.

Pettersson

Berterö

CM.

How Women with endometriosis experience health care encounters. Womens Health Rep 2020; 1(1): 529–542.

31.

Germain

Aballea

Andersson

, et al. Patients’ preferences related to treatments for endometriosis-related pain: results from a discrete choice experiment in the UK. Value Health 2017; 20(9): PA523.

32.

Poulos

Soliman

Renz

, et al. Patient preferences for endometriosis pain treatments in the United States. Value Health 2019; 22(6): 728–738.

33.

Bonilla-Jaime

Sánchez-Salcedo

Estevez-Cabrera

, et al. Depression and pain: use of antidepressants. Curr Neuropharmacol 2022; 20(2): 384–402.

34.

Vercellini

Introduction: management of endometriosis: moving toward a problem-oriented and patient-centered approach. Fertil Steril 2015; 104(4): 761–763.

35.

Al-Lami

Taha

Jalloul

, et al. Women with endometriosis in the United States: National Survey of Family Growth, 2011–2019. J Endometr Uterine Disord 2024; 8: 100081.

36.

Bougie

Yap

Sikora

Flaxman

Singh

Influence of race/ethnicity on prevalence and presentation of endometriosis: a systematic review and meta-analysis. BJOG 2019; 126(9): 1104–1115.

37.

Fourquet

Sinaii

Stratton

, et al. Characteristics of women with endometriosis from the USA and Puerto Rico. J Endometr Pelvic Pain Disord 2015; 7(4): 129–135.

38.

Robert

Caraballo-Rivera

Isola

, et al. Demographics and hospital outcomes in American women with endometriosis and psychiatric comorbidities. Cureus 2020; 12(8): e9935.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

0.40 MB

0.10 MB

0.17 MB

0.22 MB