Evaluation of Heart Failure Treatment Guided by N-terminal Pro B-type Natriuretic Peptide (NTproBNP) vs Clinical Symptoms and Signs Alone

The non-parametric scale is constructed from 3 variables, modified after Cleland. Each patient receives a rank score from 1 to 246 (246-number of patients in the study). The lowest score receive patients who die (due to CV event), next patients still alive at end-of-study with the worst composite score, the best alive patients with 0 days in hospital and the largest improvement in the KCCQ (self-administered by patient symptom score, where the higher score reflect better health status). Scores will be summarized using non-parametric calculations. The mean of non-parametric scores is presented

Number of deaths

Each overnight stay is counted as one day. The lower the better

Changes from baseline in the symptom score subset (question 3, 5, 7 and 9) of KCCQ (swelling, fatigue, shortness of breath, shortness of breath night time). KCCQ is a self-administered by patient symptom score, where higher score reflect better health status. Scale scores are transformed to a 0 to 100 range by subtracting the lowest possible scale score, dividing by the range of the scale and multiplying by 100. This mean that the KCCQ scale is from 0 to 100 with the higher value showing a better health status.

The 95% confidential interval (CI) is given as measure of dispersion

Change range -100 to 100. The higher the better.

Each titration step in prescribed medication is counted as one step, either up or down. One step up indicates an increase of dose in prescribed medication and one step down indicates a decrease of dose in prescribed medication. The sum of steps is given as a score. Score is given for each arm as a total number of titration steps for all patients in arm.

Number of patients discontinued due to adverse events'