Dear fellow STATA users,
Having made my initial attempt at implementing the DISTRATE function, I am reaching out to other STATA users in hopes of gaining insight around troubleshooting a seeming misstep in my methods, or a mis-match to my intended aims.
Specifically, I am attempting to generate age-adjusted rates and CIs on the gamma distribution for each of numerous classifications in leading cause of death as defined by the CDC National Center for Health Statistics; using the US 2000 Standard proportional distribution for weighting. As I am aiming to generate multiple classification-specific rates simultaneously, I am collapsing data across several variables including: age group, geographic area, and cause of death classification.
At first pass using DISTRATE, the rate point estimates I am getting from STATA differ notably from those I calculate by direct standardization methods using a spreadsheet [In each case these STATA estimates are greater]. They differ as well from those published by the state health jurisdiction here in WA USA [I have confirmed that my case counts are correct]. I have ensured that my weights are sorted appropriately, and cross-checked aggregate counts in my collapsed dataset with tabulations from the root dataset.
I am seeking guidance as to how I might go about trouble-shooting/ correcting my implementation of this program. As perhaps a clue, I have noted that in my listed output of findings there is included a row with missing or "0" values for point estimates and gamma CIs.
Any assistance would be very much appreciated. And my thanks to Kim and Enzo for making this utility available.
Sincerely,
Shawn
Shawn McBrien, MPH
Epidemiologist
Thurston County
Public Health and Social Services
412 Lilly Road NE
Olympia WA 98506-5132
ph: (360) 786-5581, ext 6950
*
* 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/
