This is a very early, rough start at making a django app that does cool stuff with the prescriptions dataset.
It was written quickly with no real idea of a point, so there is a high chance it's broken or wrong in some way.
Throwing it up here in the hope it turns in to something cool one day.
:)
You will be able to see awesomer & awesomer examples at http://prescriptions.openhealthcare.org.uk
http://prescriptions.openhealthcare.org.uk/examples/group/statins/
http://prescriptions.openhealthcare.org.uk/api/drug/
=> All drugs in the prescribing list
http://prescriptions.openhealthcare.org.uk/api/drug/?name=Rosuvastatin
=> [{name:'Rosuvastatin'...}]
http://prescriptions.openhealthcare.org.uk/api/drug/?name=Rosuvastatin Calcium,Fluvastatin Sodium"
(Must be exact names as returned by the drug list here)
=> [{name:'Rosuvastatin'...},{name: 'Fluvastatin'...}]
http://prescriptions.openhealthcare.org.uk/api/drug/habits/?name=Fluvastatin
=> [{"code": "0212000M0", "name": "Fluvastatin Sodium", "habit": [{"total": 13668, "period": 201202}, ...]]
/api/drug/habits/
=> Don't do this, it will aggregate the entire multiple gig dataset, not particularly efficiently, and return all historical aggregate data for all drugs.
http://prescriptions.openhealthcare.org.uk/api/practice/
=> All practice data
General Postgis dance... General download data dance...
$ pip install -r requirements.txt
$ python manage.py syncdb
$ python manage.py migrate practices
$ python manage.py migrate
$ cat your_practices_file.csv | python manage.py import_practices
$ python manage.py import_prescriptions your_csv_file.csv
$ ./manage.py load_ccgs -f ../data/ccg_data_file.csv
$ ./manage.py bind_practices
Note - your database user must be a superuser to import prescriptions.
The Postcode data can be obtained from Ordnance Survey
http://www.ordnancesurvey.co.uk/oswebsite/products/code-point-open/index.html
The Django management command to import postcodes is
python manage.py mapit_import_postcodes --coord-field-lon 3 --coord-field-lat 4 /tmp/postcodes.csv
...where we assume that postcodes.csv is our combined csv of all of the codepoint stuff.