Table 2.
Secondary outcomes of the MYCHOICEa study, including measurement tool and statistical analyses.
| Outcome | Description | Measurement | Statistical analyses |
| QoLb parameters | QoL is expected to be negatively correlated with symptom severity. When symptoms decrease, the QoL is expected to improve. | UFS-QoLc questionnaire and EQ-5D-5Ld questionnaire. | The time course of the change in health-related quality of life after treatment will be analyzed using longitudinal covariance analysis similar to the analysis of the change in SSSe. |
| Adverse events and complications | The nature of adverse events and complications of the 2 treatment arms are expected to differ. | Adverse events will be classified according to the classification of surgical complications [32]. | Adverse events are analyzed using descriptive statistics. Adverse events per treatment group and treatment will be presented with their occurrence rate. |
| Length of hospital stay | Reduced hospital stay is beneficial in terms of health care costs and is also considered as a great advantage by patients. | Length of hospital stay will be collected from the patient hospital file. | The average length of the hospital stay will be reported as mean (SD) or median (interquartile range). |
| Periprocedural and postprocedural pain | Pain perception may influence treatment experience and therefore satisfaction with the treatment. | Periprocedural and postprocedural pain will be measured on a numerical rating scale from 0 to 10 in the 3 months’ follow-up questionnaire. Pain complaints at 3, 6, 12, and 24 months after treatment will be registered by the amount and duration of pain killers used. | The numerical rating score is considered to be a semicontinuous measure (range 0-10: higher score represents more pain). Pain experienced will be reported as mean (SD) or median interquartile range. |
| Patient-reported satisfaction with treatment and treatment preference (PREMf) | Because a randomized controlled trial will be performed, satisfaction with treatment might be affected by not being allocated to preferred treatment. Furthermore, we expect women declining participation because of the randomization aspect of the trial. | The PREM consists of a concise set of statements about the experience of the patients with the treatment and whether they would recommend the treatment to a friend. In addition, the preference will be registered for a particular treatment of potential participants before randomization. Women who decline participation in the study will also be asked if they are willing to disclose their reasons for declining. | The PREM score is scored on a 5-point Likert scale: higher score represents better experience. Whether there is a difference in PREM outcome between the 2 treatment arms is determined by linear regression analysis. |
| (Co)medication | Women (still) experiencing symptoms after treatment may take or may start to take medication to relieve these symptoms. Medication might also be used as contraceptive and to mask possible fibroid-related symptoms at the same time. | Data on any prescribed or over-the-counter medication taken to reduce fibroid symptoms as reported by the patient will be collected via the questionnaires at baseline, 3, 6, 12, and 24 months after treatment. | The number of women taking medication to suppress fibroid-related symptoms is measured at baseline and 3, 6, 12, and 24 months after treatment. The absolute numbers and percentage of women taking comedication per group per time point will be presented. |
| Reintervention rate and time to reintervention | A reintervention is defined as an additional intervention due to persisting or recurring symptoms of the treated fibroid or due to complications of the initial fibroid treatment. | Occurrence and type of reintervention are collected via both electronic patient file and questionnaires at 3-, 6-, 12-, and 24-month follow-up. | The reintervention rate at the follow-up time point of 24 months after MR-HIFUg is presented as percentage reinterventions with its 95% CI. Reintervention rate will be presented per treatment arm but also per treatment. To investigate whether the time to reintervention differs between the 2 treatment arms, Cox proportional hazards analysis will be used. |
| Onset of menopause after uterus-saving treatments | Uterine fibroid symptoms diminish after menopause along with fibroid-related symptoms. Because this may affect symptom reduction, QoL, and the possible need for a reintervention, the menopausal state of the participants will be determined. | Onset of menopause is defined as 1 year without menstrual bleeding and measured by a questionnaire at the 24-month follow-up. | The absolute numbers and percentage of postmenopausal women per group and treatment will be presented. |
| Reproductive outcomes after uterus-saving treatments | Only women with an active wish to conceive within 1 year after treatment will be excluded from this study. Thus, some women may get pregnant after the uterus-saving treatment. | Reproductive outcomes will be collected of all women that underwent a uterus-saving treatment by a questionnaire at 24-month follow-up. | Reproductive outcomes will be presented per uterus-saving group using absolute numbers and percentage. |
| NPVh and fibroid shrinkage after MR-HIFU treatment | Technical success of MR-HIFU is commonly presented as NPV percentage directly after treatment or fibroid shrinkage 6 months after treatment, as determined on an MRIi scan. | NPV% will be measured on an MRI scan performed directly after treatment. Fibroid shrinkage will be measured by comparing volume measured on 6 months’ follow-up MRI scan with pretreatment volume. | We will investigate whether technical success (NPV% [NPV/initial volume of the fibroid] or fibroid shrinkage) is associated with (long-term) effectiveness using regression analysis. |
| Other study parameters | Several patient characteristics are collected from the medical record of the patients and from the baseline questionnaire such as age, amount and size of fibroids, location of fibroid, position of uterus, duration of treatment, ethnicity, parity, height, weight, relevant medical, and medical history. | Data analysis will be stratified by center to check for differences in results between centers. | In case necessary, multilevel analysis will be used to correct for differences between centers. Multivariate analysis will be performed for symptom reduction, QoL improvement, and reintervention correcting for comedication or menopause as possible confounder. In addition, we will investigate whether certain baseline characteristics such as age, BMI, number of uterine fibroids, and target fibroid size are associated with symptom reduction and reintervention using linear, logistic, and Cox regression analysis. |
aMYCHOICE: MYoma treatment Comparison study: High-intensity image–guided fOcused ultrasound versus standard (minimally) Invasive fibroid care—a (Cost-)Effectiveness analysis.
bQoL: quality of life.
cUFS-QoL: Uterine Fibroid Symptom and Quality of Life questionnaire.
dEQ-5D-5L: 5-level version of the EuroQoL questionnaire.
eSSS: symptom severity score.
fPREM: patient-reported experience measurement.
gMR-HIFU: magnetic resonance image–guided high-intensity focused ultrasound.
hNPV: nonperfused volume.
iMRI: magnetic resonance imaging.