Are Socioeconomic Factors, Sex and Country of Birth Associated With Decisions to Withhold or Withdraw Intensive Care in Swedish Hospitals?
Are Socioeconomic Factors, Sex and Country of Birth Associated With Decisions to Withhold or Withdraw Intensive Care in Swedish Hospitals?
Medical and socioeconomic data are extracted from the Swedish Intensive Care Registry (SIR), the Swedish National Patient Registry and Statistics Sweden for all adult patients admitted to in Swedish intensive care units between 2014-01-01 and 2020-12-31 with a diagnosis of sepsis and/or acute respiratory distress syndrome (ARDS) and/or coronavirus 19 (COVID-19) infection, and registered in SIR. The impact of demographic and socioeconomic factors on decisions to withhold or withdraw intensive care, and on mortality, are studied and statistically adjusted for level of chronic comorbidity and severity of acute illness.
Patient identification and data acquisition With permission from the ethical board of Sweden, all registered intensive care episodes with a diagnosis of sepsis, acute respiratory distress syndrome (ARDS) or coronavirus 19 (COVID-19) and admission between 2014-01-01 and 2020-12-31 are identified in the Swedish Intensive Care Registry (SIR). Basic demographic data (age, sex) and intensive care data from SIR are combined with comorbidity data from the National Patient Registry of the Board of Health and Welfare (Socialstyrelsen), and data on country of birth, household size, education and economy from Statistics Sweden (SCB) to produce a pseudonymized study data set. For patients with multiple intensive care episodes, the first one is used.
Analysis Descriptive data on the frequency of decisions to withhold or withdraw intensive care, and on demographic and socioeconomic factors, morbidity and mortality in patients with or without such decisions, are produced. Logistic regression is used to assess the association between demographic and socioeconomic factors and limitations in intensive care, adjusting for acute and chronic morbidity and accounting for multicenter data. Secondarily, association between the same factors and mortality is studied.
Inclusion Criteria:
Exclusion Criteria: