[R-sig-Epi] Epidemiology Task View

[Bernardo Rangel Tura]

On Fri, 2008-11-21 at 11:14 +0000, Neil Shephard wrote:
> Thanks to all who have responded both on the list and directly.
> 
> It seems that an Epidemiology Task View would be useful and a number
> of suggestions have already been made.  Incorporating and expanding on
> this I now have the following categories with suggested packages...
> 
> Case-Control : Epi, epiR epibasic, epicalc, epitools
> Cohort : Epi epiR, epibasic epicalc epitools
> Incidence/Mortality :
> Demography : demogR
> 
> Additional Sections (Bendix Carstensen & Bernardo Rangel Tura)
> 
> Multistate Models : msm, MSsim() function in Epi  (BC)
> Survival : Refer to Survival Task View (BC)
> Design of Studies : clinfun, PwrGSD , crossdes, Design (BRT)
> Diagnosis : DiagnosisMed (BRT)
> Clinical Trials : clinfun (BRT)
> Epidemiologic Surveillance (BRT)
> Meta-analysis : rmeta (BRT)
> GIS/geographical mapping : r-spatial (http://r-spatial.sourceforge.net/) (NS)
> 
> Additional Packages (Bernardo Rangel Tura) not sure which sections these fit in
> rhosp: Side effect risks in hospital : simulation and estimation
> catmap: Case-control And Tdt Meta-Analysis Package  (is this used in
> epidemiology?  I'd have used in statistical genetics/genetic
> epidemiology, and its already in the Genetics Task View  but does it
> have a wider application I'm unaware of?)

Well... I think meta-analysis of case-control is very useful in specific
fields of epidemiology ...

> pARccs: Estimation of partial attributable risks (PAR) from case-control data
> GroupSeq: Performing computations related to group sequential designs
> BayHaz: R Functions for Bayesian Hazard Rate Estimation
> 
> 
> One difference from the Genetics Task View which generally describes
> which packages to use for a given area is that many of the
> Epidemiological packages have funcitons that cover several of the
> suggested sections (i.e. Epi/epiR/epibasic/epicalc/epitools span
> several sections).  Thus I feel that it would be useful to mention
> which functions within a given package are useful for a particular
> task.
> 
> Clearly there are more packages than I'm familiar with, but I can read
> the package descriptions to find out whats what and which functions
> are available for each area.
> 
> I'll have a go at knocking up a draft document over the coming weeks
> and will post a copy to the list when I've got something worth
> distributing.  No time frame I'm afraid (yet).  Any more input is of
> course more than welcome.

when draft is ready I can review de text....
> 
> Thanks for the encouragement,
> 
> Neil

Thanks for your effort

-- 
Bernardo Rangel Tura, M.D,MPH,Ph.D
National Institute of Cardiology
Brazil