Formulate hypotheses for different study models
This article was originally published here
J Korean Med Sci. Dec 27, 2021; 36 (50): e338. doi: 10.3346 / jkms.2021.36.e338.
Generating a verifiable working hypothesis is the first step towards conducting original research. Such research can prove or disprove the proposed hypothesis. Case reports, case series, online surveys and other observational studies, clinical trials and narrative reviews help generate hypotheses. Observational and interventional studies make it possible to test hypotheses. A good hypothesis is usually based on previous evidence-based reports. Hypotheses without evidence-based justification and a priori ideas are not welcomed by the scientific community. Original research to test a hypothesis must be carefully planned to ensure appropriate methodology and adequate statistical power. While assumptions can challenge conventional thinking and can be controversial, they shouldn’t be destructive. A hypothesis must be tested by ethically sound experiments with significant ethical and clinical implications. The 2019 coronavirus disease pandemic has brought to light many hypotheses, some of which have been proven (e.g. the effectiveness of corticosteroids in people with hypoxia) while others have been refuted (e.g. the ineffectiveness of hydroxychloroquine and ivermectin).