# Development of an early prediction model for endometriosis risk: the simplified adolescent factors for endometriosis (SAFE) score

## TL;DR

Researchers developed a risk prediction model called SAFE score to help identify endometriosis risk early in young women.

## Contribution

The SAFE score is a novel, evidence-based tool for early endometriosis risk prediction using adolescent factors.

## Key findings

- The SAFE score achieved AUCs of 0.79–0.81 in training and internal test samples.
- A SAFE score cut-point of ≥2 showed high specificity and negative predictive value.
- The model included six stable risk factors like pelvic pain, menstrual disorders, and family history.

## Abstract

Endometriosis affects 11% of women of reproductive age. Diagnosis is often delayed by 6–8 years due to nonspecific symptoms and lack of early detection tools. This study aimed to develop an early risk prediction model for endometriosis to support referral decisions in primary care.

Data were from an endometriosis-focused sub-study of the Australian Longitudinal Study on Women's Health. Data were split into a training (75% of women born 1989–95, n = 4005), internal test (remaining 25%, n = 1335), and external test (women born 1973–78, n = 4077) samples. Stable risk factors were identified using a bootstrapping procedure with multivariable logistic regression. A logistic regression model based on these factors was used to develop the Simplified Adolescent Factors for Endometriosis (SAFE) score, calculated as a count of risk factors. Models were evaluated in all three samples.

Six stable risk factors were identified; three factors related to severity of general pelvic pain, two with menstrual disorders, and one with a family history of endometriosis. For the final logistic regression model, the areas under the curves (AUCs) were 0.81, 0.78, and 0.72 for the training, internal, and external test samples, respectively. Using the SAFE score (values 0–6), the corresponding AUCs were 0.79 (95% CI: 0.77–0.82), 0.79 (95% CI: 0.75–0.83), and 0.71 (95% CI: 0.69–0.73). Using a cut-point of ≥2 for the SAFE score, the corresponding sensitivity was 59.6, 64.2, and 46.7, specificity was 83.4, 82.6, and 85.7, and negative predictive value was 94.6, 94.4, and 89.7.

The SAFE score is an evidence-based tool to guide endometriosis referrals in primary care. It shows strong statistical performance and warrants clinical evaluation.

The ALSWH study is funded by the Australian Government Department of Health, Disability, and Ageing. The GELLES study is supported by the Medical Research Future Fund (MRFF) (MRFF1199785). GDM and GWM are Australian National Health and Medical Research Council (NHMRC) Fellows (APP2009577; GNT1177194). SM is a National Endometriosis Clinical and Scientific Trials Network Fellow.