UKPDS Outcomes Model
When the years since pre-existing event is greater than the age at diagnosis, the spreadsheet gives an error 'Years since existing <event> must not be greater than age'.
This is an incorrect restriction in the software. The work-around is to instead enter the age as that is the maximum which will be allowed. For IHD and Blindness the model will not be affected, however for other events it will have a slight impact on the calculation of Diabetes Death in the absence of any other events..
The UKPDS Risk Engine and the UKPDS Outcomes Model have both been developed using UKPDS data, but have distinct roles.
The UKPDS Risk Engine is a Type 2 diabetes-specific Risk Calculator designed to estimate the likelihood that a person with recently diagnosed Type 2 diabetes, who has not had a CVD event, will have one in the next defined period (usually 10 years). This calculator has been designed predominately for clinical use to help patients and health care providers decide on the most appropriate management for individuals who develop Type 2 diabetes. UKPDS Risk Engine outputs are the estimated risks and 95% CIs for a first CHD or stroke event within a given period.
The UKPDS Outcomes Model is also Type 2 diabetes-specific but requires considerably more input data and is a simulation model designed to assess the total burden of disease over an extrapolated lifetime for populations. The UKPDS Outcomes Model uses a wider variety of input data, including knowledge of previous events for individuals, and can take account of updated risk factor data over time where available. The UKPDS Outcomes Model outputs are estimated Life Expectancy and Quality Adjusted Life Expectancy.
This was a bug that is corrected in version 1.2.
Version 1.2 includes a standalone implementation of the software that has no limit to the number of subjects that it can process.
Version 1.2 allows alphanumeric values
Windows : Windows 2000 or later (2000/XP/Vista etc)
Macintosh : Intel or PowerPC Macintosh, Mac OS X 10.2 or later
Linux : x86-based Linux distributions with GTK+ 2.8 (or higher), glibc-2.4, libstdc++.so.6
This is a typographical error. Cell U5 on the Input sheet should read CHF, not CHD.
If risk factor values at diagnosis are not available it is still possible to use the Outcomes Model by taking one of two approaches. Firstly, estimates of missing values can be entered to enable the Outcome Model to run e.g. based on current values that have been adjusted for any secular trends observed. If a significant proportion of risk factor values at diagnosis are missing the user should consider making a range of assumptions in a sensitivity analysis, in order to test the degree to which the outcomes of interest vary with different assumptions. Secondly, the user can supply predictions of future time paths for these risk factors. Again it would be good practice to examine the degree to which results varied across different assumptions.