Aims: In patients with heart failure (HF), we aimed to assess (i) the time trends in N-terminal pro-B-type natriuretic peptide (NT-proBNP) testing; (ii) patient characteristics associated with NT-proBNP testing; (iii) distribution of NT-proBNP levels, focusing on the subgroups with (WHFE) vs. without (NWHFE) a worsening HF event, defined as an HF hospitalization; and (iv) changes of NT-proBNP levels over time.Methods and results: NT-proBNP testing and levels were investigated in HF patients enrolled in the Swedish Heart Failure Registry (SwedeHF) linked with the Stockholm CREAtinine Measurements project from January 2011 to December 2018. Index date was the first registration in SwedeHF. Patterns of change in NT-proBNP levels before (in the previous 6 +/- 3 months) and after (in the following 6 +/- 3 months) the index date were categorized as follows: (i) <3000 ng/L at both measurements = stable low; (ii) <3000 ng/L at the first measurement and >= 3000 ng/L at the second measurement = increased; (iii) >= 3000 ng/L at the first measurement and <3000 ng/L at the second measurement = decreased; and (iv) >= 3000 ng/L at both measurements = stable high. Univariable and multivariable logistic regression models, expressed as odds ratios (ORs) and 95% confidence intervals (95% CIs), were performed to assess the associations between (i) clinical characteristics and NT-proBNP testing and (ii) changes in NT-proBNP from 6 months prior to the index date and the index date and a WHFE. Consistency analyses were performed in HF with reduced ejection fraction (HFrEF) alone. A total of 4424 HF patients were included (median age 74 years, women 34%, HFrEF 53%), 33% with a WHFE. NT-proBNP testing increased over time, up to 55% in 2018, and was almost two-fold as frequent, and time to testing was less than half, in patients with WHFE vs. NWHFE. Independent predictors of testing were WHFE, higher heart rate, diuretic use, and preserved ejection fraction. Median NT-proBNP was 3070 ng/L (Q1-Q3: 1220-7395), approximately three-fold higher in WHFE vs. NWHFE. Compared with stable low NT-proBNP levels, increased (OR 4.27, 95% CI 2.47-7.37) and stable high levels (OR 2.48, 95% CI 1.58-3.88) were independently associated with a higher risk of WHFE. Results were consistent in the HFrEF population.Conclusions: NT-proBNP testing increased over time but still was only performed in half of the patients. Testing was associated with a WHFE, with features of more severe HF and for differential diagnosis purposes. Increased and stable high levels were associated with a WHFE. Overall, our data highlight the potential benefits of carrying further implementation of NT-proBNP testing in clinical practice.