*THIN short course will be back from 11-15 November 2013!*
Use and analysis of electronic health records
The department of Primary Care and Population Health (PCPH), UCL is offering two short courses on use and analysis of electronic health records in November 2013.
- An introduction to primary care databases
- Missing data and new methods for multiple imputation of longitudinal health records
Both courses include a mixture of lectures and computer based practical sessions based on Stata 12.
Course 1: An introduction to Primary care databases - 11-13 November 2013
This course provides an introduction to analyses of primary care databases. The course material is developed around analyses of The Health Improvement Network (THIN), but participants can easily adapt the analyses and data management tools to other data sources. It is a practical course with a mixture of lectures and hands-on practical sessions.
Tutors: Laura Horsfall, Sarah Hardoon, Linda Wijlaars, Shuk-Li Man, Louise Marston, Irene Petersen
- What do primary care database data look like and how can they be used for epidemiological research?
- How do you identify
individuals with specific treatment and/or diseases: creating code lists
- Ways to deal with large datasets using Stata: an introduction to loops, macros and other tools for analyses of large datasets
- Basic analysis of primary care databases
including calculation of incidence rates and regression analysis
Course 2: Missing data and new methods for multiple imputation of longitudinal electronic health records – 14-15 November
This practical course will introduce participants to missing data issues in electronic health records such as primary care databases (THIN and GPRD), but the methods are also applicable to other datasets with longitudinal records. The course will provide a brief introduction to the statistical theory around missing data followed by hands-on practical sessions using the ice Stata command for multiple imputation and subsequent sessions using the newly developed algorithms to deal with multiple imputation of longitudinal records. Please notice some basic knowledge and practical experience with Stata will be an advantage for this course.
Tutors: Cathy Welch, James Carpenter, Jonathan Bartlett, Irene Petersen
Day 1:
Missing data in electronic health records
- Missingness mechanisms
- Limitations of “ad-hoc” methods
Introduction to multiple imputation (MI)
- MI explanation
- Fully conditional specification (FCS)
- Missing At Random (MAR) assumption
- Selecting variables to include in imputation model
- Practical session: Using ice in Stata
Day 2:
Multiple imputation of longitudinal data in electronic health records - Two-fold FCS MI algorithm
- Issues of using MI to impute in longitudinal electronic health records
- A new method for MI in the longitudinal data - two-fold FCS algorithm
- Practical session: using twofold in Stata
General information
Both courses will be held in the Department of Primary Care and Population Health (PCPH), which is situated at the Royal Free Hospital, Hampstead in North London. The practical sessions will be based on participants bringing their own laptop with Stata 11 or 12 installed. However, most hands on sessions will require participants to work in pairs. Please indicate at registration whether you will bring laptop with Stata 11 or 12.
Lunch, tea and coffee will be provided.
After registration, you will receive an email confirming your place on the course(s) and with details about payment.
For more information or any queries related to these courses, contact our course administrator Nadine Soteriou: n.soteriou [at] ucl.ac.uk (opens your email software)
Nearby hotels
Premier Inn Hampstead (£109 single / double per night)
The Premier Inn is a 5 minute walk from the Royal Free Hospital, and 5 minutes from the nearest tube station (Belsize Park).
Hampstead Guesthouse (£65 small single with shared facilities - £110 double)
This guest house is located a 5-10 minute walk from the Royal Free, and 5 mins walk from Hampstead Heath overground station.
Holiday Inn Swiss Cottage (£94 standard room per night)
The Holiday Inn is situated in front of Swiss Cottage tube station (Jubilee line). From there you can take the Overground train (10 minutes), bus 46, 268 or C11 (15 minutes) or walk (20 minutes) to the Royal Free.
La Gaffe (£110 standard room per night)
From La Gaffe it is a 5 minute walk to Hampstead station (Northern line), or 20 minutes to the Royal Free. By bus, the 268 or 46, it is only 10 minutes.
Camden Lock Hotel (£67 single / £79 double room per night)
This hotel is right next to Chalk Farm station (Northern Line) and is a 20 minute walk from the Royal Free, or 10 minutes by bus (the 168 or 24) or tube.
Address
Research Department of Primary Care and
Population Health
Upper Third Floor
UCL Medical School (Royal Free Campus)
Rowland Hill Street
London
NW3 2PF
Reception Telephone: +44 (0)20 7830 2239
Fax: +44 (0)20 7794 1224
Transport
Tube
The nearest Underground station is Belsize Park, on
the Northern Line (Edgware Branch) and situated about 10 minutes walk from the department.
Train
The nearest London Overground train station is Hampstead Heath and situated about 5 minutes walk from the department.
The
nearest mainline train stations are King's Cross St Pancras and Euston.
From there, take the Northern line (Edgware Branch) up to Belsize Park.
Buses
Buses 24, 46, 168, 268 and C11 stop in the vicinity.
Bike
There are bike racks available at the main entrance, and A&E entrance to the Royal Free Hospital. There are also bike racks on Hampstead Heath overground station.
Directions to the department
The Research Department of Primary Care and Population Health is based at the Royal Free campus of UCL.
Click here for a map to the department from the Rowland Hill Street entrance.
Click here for a Royal Free Hospital location map.
Click here for UCL Maps, including the UCL campus, routefinder, public transport guide, and iPhone application.
Location Map
View UCL-PCPH in a larger map
Registration for the 2013 THIN courses will open in July 2013
Page last modified on 05 sep 12 11:00 by Linda Wijlaars
