Ojective: to test the feasibilty of writing and conducting a web-based, robust diagnostic study on dying in a home palliative care setting, usually not doing experimental research.
Methods: Two structured meetings of 15 nurses and 6 physicians from four home palliative care groups of Veneto Region (Italy) and discussions via email in August through October 2003.
Results: after the first meeting, the key elements of internal and external validity of good quantitative research were outlined. Following, a draft protocol was written by one of the authors (FB), e-mailed for discussion to the rest of the group and presented in the second meeting for final version to be implemented. The chosen study design was an independent, blind comparison of some aspects of hystory and physical examination with reference standard (death certificate) for the diagnosis of dying, applied in a series of home terminally ill patients.Sample size was based on pre-test probability. Sensibility, specificity and likelyhood ratios were research measurements. A 95% confidence interval and a 4% precision were considered. For test results not to influence the reference standard, researchers (nurses) are different from caregivers. Nurses collect data with automatic entry via Internet. About six hundred patients will be recruited during the two years study, following our inclusion/exclusion criteria. The study will be conducted in four cities of Veneto, starting from January 2004.
Conclusions: Writing and beginning to conduct a collaborative experimental study on bedside diagnosis is feasible in a home palliative care setting. The task is made easier by using the Internet.