Skip to main content

Guest blog: Data Saves Lives

By Wayne Gault, Lead Officer, Alcohol & Drugs Partnership, NHS Grampian

Although the promise of affordable personalised medicine from massive ‘omics’ datasets (Dash, Shakyawar et al. 2019) has yet to materialise (D’Adamo, Widdop et al. 2021), there are examples where the proposition “data saves lives” can be shown to be true:

  • Preventing iatrogenesis to improve patient outcomes (Haraden and Leitch 2011).
  • Proving the causal link between smoking and lung cancer (Hammond 1954).
  • Proving the causal link between adverse experiences during childhood and multiple risk factors for death in adults (Felitti, Anda et al. 1998).
  • Informing the COVID-19 response by guiding lockdown (Pachetti, Marini et al. 2020); identifying shielding candidates (Sperrin and Mcmillan 2020); revealing dexamethasone as a treatment for those on invasive mechanical ventilation (RECOVERY Collaborative Group 2021); and rapidly assuring vaccines for safety (Remmel 2021).
Why then did the phrase, “data saves lives” begin to emerge from 2013? This author posits that it was to repair trust lost due to a catalogue of badly implemented UK government initiatives, one of which, ironically, was designed to make data more available to improve NHS quality and research (Health and Social Care Information Centre 2014).

In 2013, the UK government was under great pressure to overhaul NHS data systems following several scandals including an infamous Mid-Staffordshire hospital where between 2005 and 2009 up to 1,200 people died due to systemic failings (Francis 2013). In 2007, the Healthcare Commission became concerned after Hospital Standardised Mortality Ratio data revealed unusually high death rates there. Several inquiries were held, including two chaired by Robert Frances QC, which recommended that NHS data systems should be dramatically improved (Francis 2013).

The first published UK reference to “data saves lives” was an article analysing Frances’ recommendations, entitled “Better Data Saves Lives” (Alderson 2013). These recommendations quickly led to the creation of “care.Data” in 2013 (Health and Social Care Information Centre 2014). This was to extract patient data from English GP practices into a centralised database to be made available in an anonymised form to academics and others. Public awareness of the proposed transfer of data and opt-out options was low and this led to a variety of criticisms, including of inadequate confidentiality safeguarding (Mathers, Sullivan et al. 2017). Recommendations by Fiona Caldicott, the National Data Guardian, led to care.Data being abandoned in July 2016 (Boseley 2016).

The choice of Atos to run care.Data compounded matters. Atos was implicated in the deaths of 590 people between 2010 and 2013 (Barr, Taylor-Robinson et al. 2016) and 2,380 between December 2011 and February 2014 (Department for Work and Pensions 2015) after having been assessed by Atos as fit to work as part of the Department of Work and Pensions’ Work Capability Assessment (WCA) scheme. Atos were relieved of the WCA contract shortly after they won the care.Data contract (Siddique 2014).

Following these trust-sapping debacles, several organisations launched “Data Saves Lives” campaigns between 2014 and 2020:

  • The Farr Institute, Connected Health Cities and the Health eResearch Centre (HeRC) launched their #DataSavesLives campaign in 2014 (Farr Institute 2015, Northern Health Science Alliance 2016, Susuzlu 2018).
  • Understanding Patient Data was created in 2016 “to support conversations with the public, patients and healthcare professionals about uses of health and care data” (Understanding Patient Data 2016, Understanding Patient Data 2019).
  • HeRC published a #DataSavesLives animation in 2018 (Creative Connection 2018).
  • The European Patients’ Forum and European Institute for Innovation through Health Data launched their “Data Saves Lives” initiative in 2019 (European Health Data & Evidence Network 2019).
  • The Open Data Institute Leeds, Beautiful Information and NHS Digital launched “Open Data Saves Lives in 2020” (OpenDataSavesLives 2020).
Nevertheless, public trust continues to decline with 71% saying they don’t trust government to maintain their privacy (IPSOS MORI 2019). This might be rational given claims that unfettered government access to health data:

  • Is a threat to civil liberties (Rubinstein, Nojeim et al. 2014).
  • Is open to abuse because of insecure public sector databases (Anderson 2009).
  • Threatens privacy because anonymisation can be reversed (Rocher, Hendrickx et al. 2019).
  • Can lead to prejudiced automated decisions due to biased data (Noor 2020).
Other reasons for distrust include:

  • NHS IT security vulnerabilities - dramatically revealed during the ‘WannaCry’ ransomware attack in 2017 (Arthur 2017).
  • Many public and commercial actors alleged to have breached health data protection standards (medConfidential 2021).
  • The sale of UK patient data without patient knowledge or consent (Helm 2019), possibly leading to inflated life insurance premiums (Donnelly 2014).
  • The current UK government’s “move fast and break things” approach to public policy (Anonymous 2020).
  • America’s alleged wish to gain access to UK health records following Brexit (Helm 2019).
  • Government repeating care.Data era mistakes: the UK government recently instructed NHS Digital to extract English GP data to a centralised pseudonymised database (Murgia 2021). The British Medical Association and Royal College of General Practitioners have again spoken out about limited public engagement and limited opportunities for patients to opt-out (Marshall and Vautrey 2021).
Perceptions of risk and benefit are inversely related (Alhakami and Slovic 1994) meaning that trust is also undermined where the public don’t perceive public benefits from sharing their health data (Castell and Evans 2016, Hopkins, Kinsella et al. 2021):

  • It’s unclear yet whether sharing data will lead to discoveries that actually improve diagnosis (Agrawal and Prabakaran 2020), or improve treatments (Wadvalla 2021), or improve outcomes (Blastland 2018).
  • Institutions encouraging the public to share health data but not reciprocate creates information asymmetries that erode trust. For example, researchers don’t readily share their academic data (Longo and Drazen 2016), contributing to the reproducibility crisis increasingly evident in clinical research (Havenaar 2018).
  • Patients still can’t readily compare treatments or performance of services, despite the principle of informed consent and Francis’ recommendations almost 10 years ago (Francis 2013, Cordina and Greenberg 2020).

Description
Credit: Jane Gilbert


“Data saves lives” has some truth but whether public willingness to share their most intimate of information will continue will depend on whether the public trust that their health data will be used transparently for the public good and protected from misuse. Without such guarantees, the one-sided “data saves lives” proposition could at best lead to scepticism and at worst lead to a modern day “tragedy of the commons” (Hardin 1968).


References

  • @msjanebond007 [J. Gilbert]. (2021). "Data Saves Lives." Twitter. https://twitter.com/msjanebond007/status/1402350240511016964.Agrawal, R. and S. Prabakaran (2020). "Big data in digital healthcare: lessons learnt and recommendations for general practice." Heredity 124(4): 525-534.
  • Alderson, C. (2013). "Better data saves lives." Health Service Journal(3/10/2013).
  • Alhakami, A. S. and P. Slovic (1994). "A Psychological Study of the Inverse Relationship Between Perceived Risk and Perceived Benefit." Risk analysis 14(6): 1085-1096.
  • Anderson, R. (2009). Database state : a report commissioned by the Joseph Rowntree Reform Trust Ltd. York, Joseph Rowntree Reform Trust Ltd.
  • Anonymous (2020). "A Misguided Counter-revolution: Remaking the British State." The Economist(19 Nov 2020).
  • Arthur, C. (2017). The ransomware attack is all about the insufficient funding of the NHS. The Guardian.
  • Barr, B., D. Taylor-Robinson, D. Stuckler, R. Loopstra, A. Reeves and M. Whitehead (2016). "‘First, do no harm’: are disability assessments associated with adverse trends in mental health? A longitudinal ecological study." Journal of Epidemiology and Community Health 70(4): 339-345.
  • Blastland, M. (2018). The dark side of early diagnosis. Prospect.
  • Boseley, S. (2016). NHS to scrap single database of patients' medical details. The Guardian.
  • Castell, S. and H. Evans (2016). The One-Way Mirror: Public attitudes to commercial access to health data. I. MORI, Ipsos MORI.
  • Cordina, J. and S. Greenberg (2020). Consumer decision making in healthcare: The role of information transparency. McKinsey Insights, McKinsey & Company.
  • Creative Connection (2018). DataSavesLives. C. Connection", Vimeo.
  • D’Adamo, G. L., J. T. Widdop and E. M. Giles (2021). "The future is now? Clinical and translational aspects of “Omics” technologies." Immunology & Cell Biology 99(2): 168-176.
  • Dash, S., S. K. Shakyawar, M. Sharma and S. Kaushik (2019). "Big data in healthcare: management, analysis and future prospects." Journal of Big Data 6(1).
  • Department for Work and Pensions (2015). Mortality Statistics: Employment and Support Allowance, Incapacity Benefit or Severe Disablement Allowance London: page 8.
  • Donnelly, L. (2014). Hospital records of all NHS patients sold to insurers. The Telegraph, Telegraph Group Ltd.
  • European Health Data & Evidence Network. (2019, 12/11/2019). "‘Data Saves Lives’ unveiled at EPF Congress." Retrieved 13/6/2021, 2021, from https://www.ehden.eu/data-saves-lives-unveiled-at-epf-congress/.
  • Farr Institute. (2015). "Data Saves Lives Story." Retrieved 13/6/2021, from https://social.shorthand.com/HeRC_Farr/uywx8Axhif/the-datasaveslives-story.
  • Felitti, V. J., R. F. Anda, D. Nordenberg, D. F. Williamson, A. M. Spitz, V. Edwards, M. P. Koss and J. S. Marks (1998). "Relationship of Childhood Abuse and Household Dysfunction to Many of the Leading Causes of Death in Adults." American Journal of Preventive Medicine 14(4): 245-258.
  • Francis, R. (2013). Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry. London.
  • Hammond, E. C. (1954). "The Relationship Between Human Smoking Habits and Death Rates." Journal of the American Medical Association 155(15): 1316.
  • Haraden, C. and J. Leitch (2011). "Scotland's successful national approach to improving patient safety in acute care." Health affairs (Project Hope) 30(4): 755-763.
  • Hardin, G. (1968). "The Tragedy of the Commons." Science 162(3859): 1243-1248.
  • Havenaar, M. (2018). "Is medical research facing a replication crisis?" https://www.castoredc.com/blog/replication-crisis-medical-research/ Accessed 13/6/2021 2021.
  • Health and Social Care Information Centre (2014). Health and Social Care Information Centre Annual Report and Accounts 2013/14.
  • Helm, T. (2019). Patient data from GP surgeries sold to US companies. The Guardian.
  • Hopkins, H., S. Kinsella, G. Evans and S. Reid (2021). Putting Good into Practice: A public dialogue on making public benefit assessments when using health and care data. London, Hopkins Van Mil.
  • IPSOS MORI (2019). Trust: The Truth?
  • Longo, D. L. and J. M. Drazen (2016). "Data Sharing." New England Journal of Medicine 374(3): 276-277.
  • Marshall, M. and R. Vautrey. (2021, 27/5/2021). " General Practice Data for Planning and Research - communications with the public." Retrieved 13/6/2021, 2021, from https://www.rcgp.org.uk/-/media/Files/News/rcgp-bma-letter-gpdpr.ashx?la=en.
  • Mathers, N., R. Sullivan, A. Dhillon, I. Rafi and A. Bell (2017). "The use of NHS patient data: report by the National Data Guardian for Health and Care." British Journal of General Practice 67(655): 56-57.
  • medConfidential. (2021). "NHS Digital Data Release Register - reformatted." Retrieved 13/6/2021, 2021, from https://www.theysolditanyway.com/.
  • Murgia, M. (2021). England’s NHS plans to share patient records with third parties. Financial Times. London.Noor, P. (2020). "Can we trust AI not to further embed racial bias and prejudice?" BMJ 368: m363.
  • Northern Health Science Alliance. (2016, 8/9/2016). "Health North: Connected Health Cities launch." Groundbreaking £20 Million Scheme to Help Improve Health In Northern Cities Retrieved 13/6/2021, 2021, from https://www.thenhsa.co.uk/2016/09/health-north-connected-health-cities-launch/.
  • OpenDataSavesLives. (2020, 26/3/2020). "#OpenDataSavesLives COVID-19 Response." Retrieved 13/6/2021, 2021, from https://opendatasaveslives.org/news/covid-19-response/.
  • Pachetti, M., B. Marini, F. Giudici, F. Benedetti, S. Angeletti, M. Ciccozzi, C. Masciovecchio, R. Ippodrino and D. Zella (2020). "Impact of lockdown on Covid-19 case fatality rate and viral mutations spread in 7 countries in Europe and North America." Journal of Translational Medicine 18(1).
  • RECOVERY Collaborative Group (2021). "Dexamethasone in Hospitalized Patients with Covid-19." New England Journal of Medicine 384(8): 693-704.
  • Remmel, A. (2021). "COVID vaccines and safety: what the research says." Nature 590(7847): 538-540.
  • Rocher, L., J. M. Hendrickx and Y.-A. De Montjoye (2019). "Estimating the success of re-identifications in incomplete datasets using generative models." Nature Communications 10(1).
  • Rubinstein, I. S., G. T. Nojeim and R. D. Lee (2014). "Systematic government access to personal data: a comparative analysis." International Data Privacy Law 4(2): 96-119.
  • Siddique, H. (2014). Atos quits £500m work capability assessment contract early. The Guardian.
  • Sperrin, M. and B. Mcmillan (2020). "Prediction models for covid-19 outcomes." BMJ 371: m3777.
  • Susuzlu, G. (2018). "#DataSavesLives." Retrieved 13/6/2021, 2021, from https://datasaveslives.eu/datasaveslives.
  • Understanding Patient Data. (2016). "Understanding Patient Data: About us." Retrieved 13/6/2021, from https://understandingpatientdata.org.uk/about-us.
  • Understanding Patient Data (2019). Understanding Patient Data Annual Report 2017/18, Understanding Patient Data.
  • Wadvalla, B.-A. (2021). "Covid-19: Ivermectin’s politicisation is a warning sign for doctors." BMJ 373: n747.