Upcoming webinar: Thursday 13 August, 5:30 pm - 6:30 pm AEST. The webinar will be available via Zoom, and will incl… https://t.co/3b5aVq1R03
ASHM Report Back
Clinical posts from members and guests of the Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine (ASHM) from various international medical and scientific conferences on HIV, AIDS, viral hepatitis, and sexual health.
Day 1: Biostatistics 101
This whirlwind run with two enthusiastic and knowledgeable presenters reinforced that implementation of Evidence Based Medicine means clinicians need a basic understanding of how the data is being collected, sifted and analysed to provide treatment recommendations. I’ve just completed (with some difficulty!) an introductory biostatistics university course and was looking forward to consolidating the knowledge.
First, Professor Matthew Law gave a rundown of the basics of statistical inference. His Key Points:
• 5% of all studies with a significant finding have occurred by chance – thinking of all the studies in all the journals in all the world, this is a sobering perspective!
• Look at confidence intervals to get an idea of how precise this estimate is – narrower is better, but crossing the ‘no difference’ value negates significant p-values.
• Failure to reject a null hypothesis doesn’t mean the null hypothesis is true.
• One tailed tests suggest mathematical jiggery-pokery and should be approached with caution!
Next, an explanation of what the different types of models and tests all mean, by A/Prof Kathy Petoumenos. She ran through the differences in variable types, model types, and how to interpret reported calculations such as relative risk, odds ratios and hazard ratios.
Her (very reassuring) Key Point: Ask a friendly statistician!
In this era of information galore, KPIs and rapidly evolving evidence base, we as clinicians should consider basic statistics knowledge for ourselves mandatory, and biostatisticians part of our multidisciplinary team.