Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Re: Standardised Incidence Rate


From   john dark <johndark7@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Re: Standardised Incidence Rate
Date   Fri, 4 May 2012 12:32:42 +0200

Many thanks for this Clyde.

John

On 3 May 2012 16:20, Clyde B Schechter <clyde.schechter@einstein.yu.edu> wrote:
> John,
>
> You have now only to work your data into the form needed for the -dstdize- command.  The help file is somewhat elliptical, but the corresponding section of the manual is quite clear and has helpful examples.
>
> Alternatively, you can "roll your own."  First take your file with the incident cases and aggregate them into counts of cases in each stratum defined by 5-year age band, sex, and year, along with the total actual population in those age-band sex strata in that actual year.  -collapse- or -egen- is probably your friend here.
>
> Then calculate the crude incidence rate in each age-band-sex-year stratum by dividing  the total number of cases by that stratum's population.
>
> The next step is to -merge- this with the file containing the reference population's counts in the age-band-sex stratum, if that isn't already in your working data set.
>
> At this point you should have a file containing the following variables, perhaps under different names:
>
> age_band
> sex
> year
> crude_inc_rate
> ref_population_count
>
> Then, for each year, calculate the weighted average of the year's crude incidence rates, using the reference population counts as weights, as follows.
>
> collapse (mean) std_inc_rate = crude_inc_rate [fweight = ref_population_count], by(year)
>
> takes you home leaving a file with each year and its directly standardized incidence rate.
>
> Use your favorite approach to assessing trends over time.
>
> Done.
>
> By the way, this dialog having begun on Statalist, it should stay there, in case other interested people wish to follow along.  So I did not respond to you directly, though you did contact me off-list.
>
> Clyde Schechter
> Albert Einstein College of Medicine
> Bronx, NY, USA
>
> ________________________________________
> From: john dark [johndark7@googlemail.com]
> Sent: Thursday, May 03, 2012 9:16 AM
> To: statalist@hsphsun2.harvard.edu
> Cc: Tim.Evans@wmciu.nhs.uk; Clyde B Schechter
> Subject: Re: st: Re: Standardised Incidence Rate
>
> Dear Tim and Clyde,
>
> Many thanks for your comments and apologies for not making myself more
> clear. I have now clarified the contents of the dataset with my
> supervisor! It in fact includes data on all patients who were
> diagnosed with a rare type of lung cancer (carcinoid) in the East
> Midlands between 2000 and 2011. Each row in this dataset corresponds
> to a unique patient and includes other variables such as age, gender,
> diagnosis date and tumour stage. I have a second dataset including the
> region's population numbers broken down by 5-year age bands and
> gender. I would like to calculate the annual age and
> gender-standardised incidence rate using the direct method and assess
> the trend over time. Your guidance on how I should go about this
> analysis would be greatly appreciated.
>
> Regards
>
> John
>
> On 2 May 2012 10:43, Tim Evans <Tim.Evans@wmciu.nhs.uk> wrote:
>> Clyde,
>>
>> Yes of course you raise some sensible points regarding exactly what type of standardisation John Dark wants to undertake.
>>
>> I can confirm that 850 lung cancer patients in 10 years does not represent a complete census of 10 years worth of data for the West Midlands region. We observe that just under 3,500 cases were diagnosed in 2009 alone.
>>
>> Best wishes
>>
>> Tim
>>
>> Dr T Evans
>> Cancer Registration Information Manager
>>
>> West Midlands Cancer Intelligence Unit
>> Public Health Building
>> University of Birmingham
>> Birmingham
>> B15 2TT
>>
>> Tel: 0121 414 4274
>>
>> If you are planning to send patient identifiable data please send them from your nhs.net account to t.evans2@nhs.net
>>
>> ** SEE OUR UPDATED LOCAL AUTHORITY CANCER PROFILES **
>>    www.wmciu.nhs.uk/La_profiles.html
>>
>> -----Original Message-----
>> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Clyde B Schechter
>> Sent: 01 May 2012 19:56
>> To: statalist@hsphsun2.harvard.edu
>> Subject: Re: st: Re: Standardised Incidence Rate
>>
>> I think John Dark will need to provide more information if anybody is going to be able to help him out.
>>
>> First, you don't say whether you want direct or indirect standardization of the mortality rate.  They are different procedures, using different "ingredients," and requiring different calculations.   Nor do you tell us whether you are standardizing on both age and sex, or just age (or, less commonly, just sex).
>>
>> Next, your problem is solvable only if the 850  observations you have represent a complete census of lung cancer diagnoses in the West Midlands region over that 10 year period. And in any case you will need to know the size of the West Midlands population, disaggregated by age and sex, over those 10 years.
>>
>> You will also need to develop or find a data set describing the (possibly hypothetical) population to which you wish to standardize your incidence rates.
>>
>> Once you have all those ingredients in place, it is really just a matter of some simple calculations--coding it in Stata will be no difficulty at all.  But from what is described in your post, you are nowhere close to getting started yet.
>>
>> Poisson regression will not be part of the solution, in any case.  It is useful for many purposes, but calculating standardized incidence or mortality rates is not one of them.  Perhaps you are thinking of *adjusted* rather than standardized rates?  Poisson regression would be useful for that.
>>
>> Clyde Schechter
>> Albert Einstein College of Medicine
>> Bronx, NY, USA
>>
>>
>>
>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>>
>> _DISCLAIMER:
>> This email and any attachments hereto contains proprietary information, some or all of which may be confidential or legally privileged. It is for the exclusive use of the intended recipient(s) only. If an addressing or transmission error has misdirected this e-mail and you are not the intended recipient(s), please notify the author by replying to this e-mail. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this e-mail or any attachments, as this may be unlawful.
>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index