| Literature DB >> 34761281 |
Dirk Stengel1, Wolf Mutschler2, Luzi Dubs3, Stephan Kirschner4, Tobias Renkawitz5.
Abstract
Informative, participatory clinical decision-making needs to combine both skills and expertise as well as current scientific evidence. The flood of digital information makes it difficult in everyday clinical practice to keep up to date with the latest publications. This article provides assistance for coping with this problem. A basic understanding of prior and posterior probabilities as well as systematic error (bias) makes it easier to weigh up the benefits and risks, e.g. of a (surgical) intervention compared to a nonsurgical treatment. Randomized controlled trials (RCT, with all modern modifications) deliver undistorted results but in orthopedic and trauma surgery can lead to a heavily selected nonrepresentative sample and the results must be confirmed or refuted by further, independent RCTs. Large-scale observational data (e.g. from registries) can be modelled in a quasi-experimental manner and accompany RCTs in health technology assessment.Entities:
Keywords: Bias; Confidence intervals; Probability; Randomized controlled trial; Scientific misconduct
Mesh:
Year: 2021 PMID: 34761281 PMCID: PMC8579904 DOI: 10.1007/s00113-021-01101-8
Source DB: PubMed Journal: Unfallchirurg ISSN: 0177-5537 Impact factor: 1.000



| Szenario 1: ideale Randomisierung | Szenario 2: zufällige Ungleichverteilung von Risikofaktoren in beiden Interventionsgruppen | Szenario 3: überzufällige Verteilung von Risikofaktoren in eine Interventionsgruppe | ||||||
|---|---|---|---|---|---|---|---|---|
| Variable | Platte | Nagel | Variable | Platte | Nagel | Variable | Platte | Nagel |
| 51 | 52 | 51 | 52 | 51 | 52 | |||
| Geschlecht | Geschlecht | Geschlecht | ||||||
| Weiblich | 21 (41 %) | 20 (38 %) | Weiblich | 15 (29 %) | 22 (42 %) | Weiblich | 15 (29 %) | 22 (42 %) |
| Männlich | 29 (59 %) | 30 (62 %) | Männlich | 30 (58 %) | Männlich | 30 (58 %) | ||
| Alter (Jahre) | 45 (± 9) | 47 (± 10) | Alter (Jahre) | 42 (± 8) | Alter (Jahre) | 42 (± 8) | ||
| BMI (kg/m2) | 25 (± 5) | 24 (± 6) | BMI (kg/m2) | 23 (± 7) | BMI (kg/m2) | 23 (± 7) | ||
| Raucher | 10 (20 %) | 9 (17 %) | Raucher | 5 (10 %) | Raucher | 5 (10 %) | ||
| Diabetes | 4 (8 %) | 7 (15 %) | Diabetes | 3 (6 %) | Diabetes | 3 (6 %) | ||