Surgery is a fundamental component of modern healthcare, yet postoperative outcomes vary substantially between patients and may be influenced by factors extending beyond the surgical procedure itself. The overall aim of this thesis was to provide a comprehensive audit of the Swedish surgical population and investigate factors associated with postoperative outcomes, with particular emphasis on major surgery and long-term mortality.

Four observational studies were conducted; three used large-scale registry data and one a prospective multicentre cohort. The adult surgical population in Sweden was characterised using the Epidemiology of Perioperative Care in Sweden (EPeCS) database (Study I). Subsequent analyses investigated perioperative risk factors (Study II) and socioeconomic status (SES) (Study IV) after acute hip fracture surgery, and preoperative anaemia in major noncardiac surgery (Study III).

Study I included 1 319 048 individuals undergoing surgery between 2015 and 2020; 59.4% underwent major surgery, and crude 30-day and 1-year all-cause mortality were 1.3% and 5.1%, respectively. Educational attainment and disposable household income were associated with several perioperative care processes and with postoperative mortality. For 1-year mortality, the lowest income quartile was associated with an aOR of 1.25 (95% CI 1.20–1.30) compared with the highest quartile after adjustment for patient, procedural, perioperative, and healthcare characteristics. In the multicentre cohort of 1060 patients undergoing major noncardiac surgery (Study III), lower preoperative haemoglobin was associated with adverse long-term outcomes, whereas perioperative RBC transfusion did not appear to mediate this relationship. In the hip fracture cohorts, educational attainment was identified as the SES factor most strongly associated with long-term mortality, with an adjusted difference in 2-year mortality of 6.3 percentage points between the lowest and highest educational groups and a corresponding attributable fraction of 14% (Study IV). Anaesthetic technique, surgical waiting time, and after-hours surgery were not independently associated with postoperative mortality after comprehensive risk adjustment (Study II). An association with hospital type observed in Study II was no longer evident after additional adjustment for socioeconomic factors in Study IV.

In conclusion, this thesis characterises an unselected national surgical population and demonstrates low postoperative mortality in Sweden in an international context. The findings challenge previously reported associations between several surgery-related factors and postoperative outcomes after hip surgery, highlighting the importance of comprehensive risk adjustment. Preoperative anaemia remains an important marker of long-term postoperative risk, an association that does not appear to be mediated by perioperative RBC transfusion. Finally, socioeconomic status was associated with several perioperative care processes and with postoperative mortality, highlighting the importance of considering socioeconomic circumstances when evaluating perioperative care and long-term outcomes.