| | 6 | |
| | 7 | == QRISK 2 Batch Processor |
| | 8 | |
| | 9 | There are several software options for calculating QRISK2 scores. Below is an email from a representative of qrisk.org that provide the software. |
| | 10 | |
| | 11 | Dear Richard |
| | 12 | |
| | 13 | Standard reply |
| | 14 | |
| | 15 | For the last few years, we have published the QRISK2 algorithm (and its annual update) as open source software, which you can find at qrisk.org. The open source is our preferred route for academic studies, as it is transparent, open, and means that the fruit of your labours is open for others to use in turn.. We don't provide support, and we don't include our Postcode to Townsend score deprivation map -- you have to calculate your own Townsend score for each patient -- but you may not need that for your modelling. The open source will make sense if you work with it alongside the original academic papers. Of course, you have to comply with the terms of the open source licence -- they basically ensure that others can use what you've done. |
| | 16 | |
| | 17 | We offer an annual licence for a Windows batch processor. This is easy to use, includes our Postcode to Townsend score deprivation map, and takes an Excel file (.csv) as input, and produces one as its output, detailing any missing value substitutions that it makes on its way. This would certainly be much more straightforward for you, especially if you don't have the software expertise in house. Several academics choose to go this this route too. |
| | 18 | |
| | 19 | We can offer you a year's licence for the commercial version of the QRISK2-2013 batch processor for £500 + VAT. This will include the 2014 update released in April. |
| | 20 | |
| | 21 | There are several advantages to going the open source route, if you can manage it, e.g. you can code up the score in whatever stats package you fancy, and have it at the heart of the model you are running. Several folk have done just that. I'm sure that you understand, however, that we can't offer support to everyone who goes this route -- you are on your own! |
| | 22 | |
| | 23 | Let me know if you want to take us up on the offer of the commercial batch processor, and if so, who to send the invoice to. |
| | 24 | |
| | 25 | Personal reply |
| | 26 | |
| | 27 | If you want to merely get a baseline on your patient data, I'm happy to run one off for you for free. You'd need to send me a file of the patient data (anonymised, of course) in the Windows batch processor format, which I can supply. |
| | 28 | |
| | 29 | You'll only need the SDK if you want to build the score into your database, i.e. for live update of patient data, or for sophisticated what-if scenarios. |
| | 30 | |
| | 31 | The best way to use it would be to build a web service that your database application calls as and when it wants. |
| | 32 | You'll need a Java programmer with 2-5 years experience, depending on how good they are, i.e. one that is comfortable with building and hosting web services. |
| | 33 | We could do that at cost for you of £1206 + VAT for a year's licence, including support, loads of test data, a java batch processor and the upgrade to the 2014 version of the score. |
| | 34 | |
| | 35 | Have you seen qstroke.org? I'm happy to bundle a batch processor for that in for free if you take the QRISK2 one. Ditto with the free baseline score. |
| | 36 | |
| | 37 | All the best |
| | 38 | Stephen |
