Importance of diagnostic setting in determining mortality in patients with new-onset heart failure: temporal trends in Denmark from 1997 to 2017

Figures

• 

  Figure 1. Flowchart depicting the inclusion and exclusion criteria for the study population in this study. Light grey boxes indicate inclusion, whereas dark grey boxes indicate exclusion.

• 

  Figure 2. Temporal trend in the incidence of new-onset heart failure in Denmark during 1997–2017 for three patient types: inpatients, outpatients, and all patients. The primary y-axis portrays the incidence of heart failure for the three patient types, depicted by three different graphs (red = inpatients, green = outpatients, and blue = all patients). The secondary y-axis reflects the percentage of heart failure cases diagnosed as inpatients (red shaded area) and outpatients (green shaded area).

• 

  Figure 3. (A–F) Temporal trends in mortality after new-onset heart failure for inpatients and outpatients. (A) The temporal trend in the age-standardized 1-year mortality rate of the previously described patient types. (B) The temporal trend in the age-standardized 5-year mortality rate. Both were standardized to the population in Denmark in 2017. (C–F) The 1-year and 5-year absolute risk of death for inpatients and outpatients, respectively, comparing the following time periods: 1997–2001, 2002–06, 2007–11, and 2012–17.

• 

  Figure 4. (A–F) Absolute risks of first heart failure rehospitalization for inpatients and first heart failure hospitalization for outpatients after new-onset heart failure. The graphs portray the absolute risk of heart failure rehospitalization and the absolute risk of heart failure hospitalization within 30 days (left) and 5 years (right) of follow-up. Note that 1-year and 5-year absolute risks were only estimated for patients who survived 31 days without HF hospitalization. (A–C) Absolute risk of HF rehospitalization for inpatients (red). (D–F) Absolute risk for HF hospitalization for outpatients (green).

• 

  Figure 5. Adjusted hazard ratios of 1-year and 5-year mortality and heart failure hospitalization, respectively, for inpatients vs. outpatients (reference). The forest plot shows the hazard ratios of 1-year and 5-year mortality (in red) and heart failure hospitalization (in blue), respectively, for inpatients vs. outpatients—outpatients being the reference group. Note that hazard ratios for 1-year and 5-year heart failure hospitalization were estimated for inpatients/outpatients who were rehospitalized/hospitalized after 30 days of follow-up. They should therefore be considered hazard ratios for 31 days to 1 year and 31 days to 5 years of heart failure hospitalization. Moreover, the P-values for interaction show whether there was an interaction between patient type and time period for 1- and 5-year mortality and heart failure hospitalization. The absolute risk for inpatients and outpatients were estimated as crude risks (event/at risk). Hazard ratios were adjusted for age, sex, and history of comorbidity (ischaemic heart disease, atrial fibrillation, stroke, and diabetes) at the time of new-onset heart failure. The time periods were defined as follows—period 1: 1997–2001; period 2:…

• 

  Figure 6. (A–D) Temporal trends in the percentage of heart failure patients started on renin–angiotensin-system inhibitors, beta blockers, mineralocorticoid receptor antagonists, and loop diuretics 90 days after new-onset heart failure. HF, heart failure; RASi, renin–angiotensin-system inhibitors; MRAs, mineralocorticoid receptor antagonists.