[Thurs, 23d May] ODI Meetup #7 "Give us our health data"
Denis Parfenov
The theme for ODI Meetup #7 is "Give us our health data"
Building on the health data meetup that took place in March, we will be hearing from some great speakers about how open health data can save lives and drive innovation. We'll also have a panel discussion on open health data to see how we can overcome some of the challenges.
We are incredibly honoured to have E-Patient Dave deBronkart, who gave the seminal "Let Patients Help" Ted talk with the "Gimme my damn data" rap, joining us for a Google Hangout.
This meetup will be facilitated by Niall O'Neill
Guest Speakers and panelists include:
- E-Patient Dave deBronkart / http://epatientdave.com/
- Roisin Doherty (HSE)
- Daragh O'Brien
- Others to be announced
Many thanks to Peter McCanney for a new 'Open Data Ireland' banner!
Please register (https://tito.io/open-data-ireland/meetup-7) and help to spread the word. (Twitter #hashtag: #OpenDataIRL )
Best regards,
Denis Parfenov
Denis Parfenov
Open Data Ireland - "Give us our health data"
Open Data Ireland Meetup - notes for evening
Thursday, 23d May 2013 (18:30)
Registration page https://tito.io/open-data-ireland/meetup-7
Running Order
Introduction - Denis & Niall
e-Patient Dave (~15 mins)
Roisin Doherty (HSE)
Demo of prescribing analytics - Niall
Panel discussion chaired by Niall:
Daragh O'Brien - Information Quality, Info Gov & Data Protection Consultant
Leonora O'Brien - CEO of Pharmapod, an Irish startup providing an incident reporting and learning platform for pharmacies
Howard Johnson - Clinical Lead of Health Intelligence, Health Intelligence Unit, HSE
Discussion points for panel discussion:
Is the solution to privacy concerns to anonymise data? Can health data be truly anonymous?
Does the good of society outweigh the privacy of the individual?
Allowing patients to opt-out is only effective if patients are aware of how their data might be used, and can trust the system. It must be informed consent. How should we tackle this challenge?
Is health data just too complex to be opened for public analysis? Taking surgical outcomes, is the range of factors influencing outcomes too complex for 'apples to apples' comparison of outcomes? For example, timescale for which outcomes are considered; level of risk of patients pre-surgery.
Quality of data was a key public concern in the Research Council UK public consultation doc due to the risk of poor decision making - but on the other hand, open data can expose data quality issues. What are your views on the balance between data quality and openness?
Does the public really have a role in accountability in healthcare? (e.g. Michael Neary - the system failed to detect his malpractice - shows the need for 'many eyes' on data in order to spot unusual patterns/malpractice?).
Who should steward open health data? How do we address the question of data quality, variation in the quality of data capture across the myriad of systems, the responsibility of effectively anonymising data.
Do patients really want to access their own data? Does this demand exist in Ireland?
In Ben Goldacre's book Bad Pharma, he gives an example of how an academic spotted a link between an anti-diabetic drug, Rosiglitazone, and heart problems - previously unpublicised. He was using data that had been made more publically available than the norm. The drug was suspended by regulators.
What is the role of open data in pharmaceutical testing?
Research
Todd Parks, US CTO:
"Data is fuel for innovation, and developers and entrepreneurs are key players who can turn that fuel into innovations that matter. That’s why one of the many ways the US Government is contributing to the transformation of health care is by unleashing vast amounts of data from the vaults of government, while rigorously protecting privacy."
Quote from an administrator in a GP practice in Ireland, when asked if the patient could have a print out of their own blood test: "I don't think we're allowed to do that"
Ireland - Data concerns over audit system
New national audit system for Ireland to monitor and compare death rates in hospitals - delayed over confidentiality and data protection issues.
Run on behalf of HSE by National Office of Clinical Audit (NOCA) at RSCI.
Under system, hospitals would report surgical death rates.
However, the process and outcomes would be confidential: "He said the audit was intended to operate on a confidential basis, in order to ensure adequate engagement from surgeons in terms of providing data for the audit."
Is this a system designed to protect the surgeon - not the patient.
UK have been publishing data for many years since Sir Bruce Keogh, now National Medical Director of the NHS, convinced his heart surgeon colleagues to start publishing outcomes.
There are now 1,000 fewer deaths a year after surgery - clearly advances in medicine account for some of this change; but Keogh is convinced that "professional transparency" has driven improvement.
Death rates are half that of Germany - which has also benefited from medical advances.
UK - Open Prescription Data
GP prescription data is published, giving the types, volumes and brands of drugs prescribed by GP practice over time.
Analysis on this data by a group of UK data analysts and NHS doctors revealed that 27 million is wasted every month on branded statins.
Branded statins are clinically no different to generic statins - yet GPs continue to prescribe more expensive branded statins over generic statins.
In Ireland, prices for drugs paid for by the Primary Care Reimbursement Service are published - but prescription patterns are not.
Example:
Generic Statin - Simvastatin (KRKA Pharma) Film Coated Tabs. 10 mg. 28 (A) - 3.49
Branded Statin - Atorvastatin (Actavis) Film Coated Tabs. 10 mg. 28 (A) - 11.56
UK - purchasing of individual data by private companies
Private companies in the UK can buy patient data from the NHS via Health and Social Care Information Centre (HSCIC).
Data is either anonymous or identified (by gender, postcode and age).
Cost for 20 years of data for a person (identified) is 8k.
Data is only provided when patient consent has been given.
Secretary of State for Health assures: “The Guidance that we will be giving to GPs, is that, if people ask for data not to be shared, then it wont be” ”If people ask for their information not to go beyond the GP’s surgery, GPs will respect that”. Existing consent will stand.
Links:
NHS - open data commitment
NHS has made three commitments around transparency:
Big data: the NHS Commissioning Board has launched a new initiative, care.data, to transform the availability of data to clinicians, patients and the public. From the summer, we will publish clinical quality measures and survival rates from national clinical audits of the consultants practising in England in 10 surgical specialities including cardiac, vascular and orthopaedic surgery.
User data: this is where the citizen is the source of information. Examples include TripAdvisor-style feedback for collation of customer insight. The NHS is rolling out a new "friends-and-family" test to see whether patients recommend services to loved ones.
My data: this is personal information that empowers the individual. By March 2015, every patient will be entitled to online access to their general practice record.
Notes by Niall O’Neill
Denis Parfenov
It would be great, if - in the spirit of ODI community - you could use your networks to spread the word about tomorrow event
i.e.
Join
Or just click RT button:
- https://twitter.com/opengovpart/status/336811101213822977
- https://twitter.com/epsiplatform/status/336839515933134849
- https://twitter.com/FingalOpenData/status/336413567244988416
Cheers and I hope to see you tomorrow.
Best regards,
Denis
On Monday, May 13, 2013 3:42:50 PM UTC+1, Denis Parfenov wrote: