This study aimed to calculate costs and health-related quality of life of women with endometriosis-associated symptoms treated in referral centres. A prospective, multi-centre, questionnaire-based survey measured costs and quality of life in ambulatory care and in 12 tertiary care centres in 10 countries. The study enrolled women with a diagnosis of endometriosis and with at least one centre-specific contact related to endometriosis-associated symptoms in 2008. The main outcome measures were health care costs, costs of productivity loss, total costs and quality-adjusted life years. Predictors of costs were identified using regression analysis. Data analysis of 909 women demonstrated that the average annual total cost per woman was €9579 (95% confidence interval €8559–€10 599). Costs of productivity loss of €6298 per woman were double the health care costs of €3113 per woman. Health care costs were mainly due to surgery (29%), monitoring tests (19%) and hospitalization (18%) and physician visits (16%). Endometriosis-associated symptoms generated 0.809 quality-adjusted life years per woman. Decreased quality of life was the most important predictor of direct health care and total costs. Costs were greater with increasing severity of endometriosis, presence of pelvic pain, presence of infertility and a higher number of years since diagnosis. Our study invited women to report resource use based on endometriosis-associated symptoms only, rather than drawing on a control population of women without endometriosis. Our study showed that the economic burden associated with endometriosis treated in referral centres is high and is similar to other chronic diseases (diabetes, Crohn's disease, rheumatoid arthritis). It arises predominantly from productivity loss, and is predicted by decreased quality of life.

A variety of methods have been developed to demultiplex pooled samples in a single cell RNA sequencing (scRNA-seq) experiment which either require hashtag barcodes or sample genotypes prior to pooling. We introduce scSplit which utilizes genetic differences inferred from scRNA-seq data alone to demultiplex pooled samples. scSplit also enables mapping clusters to original samples. Using simulated, merged, and pooled multi-individual datasets, we show that scSplit prediction is highly concordant with demuxlet predictions and is highly consistent with the known truth in cell-hashing dataset. scSplit is ideally suited to samples without external genotype information and is available at: https://github.com/jon-xu/scSplit