The investigators aim to emulate a target trial to compare the risk of acute decompensated heart failure in users of methylprednisolone compared to prednisolone. The exposure is prescriptions of methylprednisolone or prednisolone tablets and the primary outcome is heart failure hospitalizations within the following 6 months, i.e. acute contacts with secondary or tertiary health care resulting in a primary diagnosis of heart failure. Secondary outcomes include broader heart failure diagnoses and the initiation of loop diuretics. Data preparation includes an initial observation period of 2 years, inclusion criteria such as age, corticosteroid formulation types / dosages and prescription codes, as well as removal of duplicate prescriptions and measures to reduce potential carry-over effects.

1. Individuals aged < 18 years in January 2008 will be excluded
2. Analyses will be restricted to tablets, as prednisolone is only available in this form.
3. To allow for equipotent dose comparisons, we will focus only on 4 mg and 16 mg methylprednisolone and their counterparts in prednisolone at 5 mg and 20 mg, respectively. Patients with other doses will be excluded.
4. To facilitate comparisons of treatments prescribed for similar conditions, we will only include prescriptions with an International Classification of Diseases, 10th Edition (ICD10) or International Classification of Primary Care, 2nd Edition (ICPC2) refund code.
5. Patients with prescriptions for diagnosis codes including palliative care, cancer symptoms, medical complications, and organ transplants will be excluded. This exclusion aims to reduce the inherent heterogeneity within these groups, allowing for a more comparable baseline risk.
6. An initial 2-year observational period, spanning from January 2008 to December 2009, will be used to establish comorbidities. Accordingly, only prescriptions of methylprednisolone or prednisolone dispensed from January 2010, onward, will be eligible for analyses.
7. To minimize the risk of carry-over effects, we will exclude all prescriptions prescribed < 6 months after another prescription for a corticosteroid. Accordingly, the same patient can contribute data on several treatment courses, provided that there was > 6 months since the last prescription was dispensed (See section 7 for justification for why a 6-month interval was chosen).
8. Duplicate prescriptions (same date, drug and dosage) will be kept in the analyses as a single treatment course. Cases of duplicate prescriptions on the same date, but where the drug or dosage is different will be excluded.
9. Baseline of the treatment course will be defined as the date when the prescription of either methylprednisolone or prednisolone was dispensed.

This registry-based study will include data from the nationwide Norwegian Cardio-Rheuma Registry (NCRR) that encompasses the entire Norwegian adult population over a ten-year period from January 2008 to December 2017.

Individual participant data will not be made available to other researchers.