We wish to monitor the effects of long term hypnotic (night sedation) medication on the elderly in residential care. As over 62% have dementia / cognitive impairment an observational tool is required.
It strikes me that observation tools are troubled by observer bias. If nurses are primed to monitor physical effects of hypnotics, then they will tend to see what they expect. They will also tend not to see what they don't expect.
The next problem is one of expectation: what is the expected rate of the things you observe in the absence of hypnotics? This requires observing a control group. Then, you face another problem: that patients on hypnotics are in many ways different to patients not on hypnotics. Some of the things you observe may have been indications for the prescription of hypnotics rather than the consequence of their use.
In short, this is a particularly magnificent can of worms, and, I agree, really really important as a research topic. It's an example of W Lambert Gardiner's law:
Rigour = 1 – relevance
I fully agree with you Professor Conroy, which is where the major problem for this study lies. Despite the fact that i in Ireland in 2002, the Benzodiazepine Committee of the Department of Health recommending that these drugs should not be prescribed for longer than 2–4 weeks, depending on their indication, with slightly longer treatment periods permitted for anxiety relief in comparison to insomnia, I've sourced a 2013 study evidencing that the prevelance of use hadn't changed from around 33%. Our study shows 36% usage. So = what is going wrog in our healthcae system that we haven't madee any improvement despite the miriad of resarch showing that long term use has little or no effect except for a 'minority/' of cases. and in fact is detrimental. I would very much appreciate meetingg up on this as ther has to be some sort of rationle to why it's still continueing 13 years after our own Dept of Health making their reccomendations.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics
- Survey Design
- Questionnaire