Patient Safety Day – thoughts from a clinician

By Piers Manser, Senior Clinical Safety Officer

The concept of Patient Safety has been around for over 2000 years since it was first attributed to Hippocrates and the Hippocratic Oath. This oath was still adopted as recently as last century by the medical profession across numerous countries, used as a statement of intent that emphasised the safety of the patient as paramount. It has however been replaced in more recent times with more modern ethical codes that reflect the expected conduct, behaviour and obligations for clinical staff.

For instance, in the UK, the General Medical Council (GMC) and the Nursing and Midwifery Council (NMC) both have codes of conduct that registered members are expected to adhere to and that prioritise patient safety. The principle of Primum non nocere (First do no harm) is cited as one of the foundations of patient safety and is related to the Hippocratic oath. Often, the actual term is inaccurately ascribed to the Greeks when the language is Latin and was probably first used as late as the 1800s. 

So, is the practicality of Primum non nocere appropriate or even relevant in current times?

In the treatment of patients, as clinicians, we are often facing a challenge of risk vs reward. Should I prescribe a medicine with potential, or even inevitable, side effects if the diagnosis is not 100% definite? Should I consider invasive treatment for someone with a poor prognosis? It is always a difficult decision and one that needs to be addressed holistically and also needs participation from the patient, in not just the acceptance of the risk, but the understanding of the outcome. To first do no harm therefore becomes a difficult proposition in many clinical scenarios.

What does all this have to do with the modern evolution of digital health?

It is widely accepted that we are at a pivotal moment in modern computing, where data science can transform areas of understanding in the realms of health through machine learning and provide discovery through big data. There has been a realisation over the last year that AI (Artificial Intelligence) provides an exciting future but could also be a threat to current safeguards around how technology can be applied. In healthcare, one of the challenges we face is how we can utilise this to further improve patient safety.

Well, how about being able to balance risk vs reward based on, not just your own personal clinical experience, but having the ability to model outcomes based on whole population level information?

Hippocrates and Asclepius understood the need for evidence-based research, although they were often limited by anecdotal evidence. We are in the beneficial position of being able to map the health of vast numbers of people that experience similar symptoms or disease profiles.

Understanding the complexities of an individual's medical history and how it relates to other patients with similar comorbidities and on similar medications unlocks opportunities to improve the care we can offer. This provides alternative treatment plans and different perspectives on managing cohorts of patients more directly as their need arises rather than, for example, at the point they attend a twice-yearly clinic appointment. Then, placing those patients on a virtual ward for crisis prevention using simple, easy to deploy solutions, such as the Docobo remote monitoring solution. This type of intervention itself provides safety benefits such as reducing the risk of falls, cross-infection or medication errors that can be associated with inpatient admission.

Or how about identifying patients that may not be on the care teams' radar, ensuring that early intervention and appropriate diagnostic procedures are done before an emergency develops?

One such example of this is the case finding tools that Graphnet have delivered to Integrated Care Boards, and clinical teams across England as part of its Population Health Solution. Through the "Gaps in Care" programme it is possible to identify patients from a host of previously unconnected data from across the health and social care systems, to enable us to zero in on the patients in need and make sure they are provided for.

Transforming care provision further, we can deliver these tools directly to the first-line clinical teams through access on a ubiquitous platform not constrained by a single location or local system. Shared Records such as Graphnet's CareCentric provide that platform. It can be utilised across the multi-disciplinary teams working at local and regional levels.

Access to information (use within the correct boundaries of role-based access) is a dramatic safety improvement. This was highlighted in the Caldecott Report (1997) and subsequent review (2013); "Good information sharing is essential for providing safe and effective care."

Through harnessing the information available to us we can make healthcare safer and give clinicians the confidence to improve their practice.

When the physicians of Ancient Greece were first documenting the requirements for their newly established medical community in the Hippocratic Corpus, they knew that key to looking after their patients was "the importance of understanding the patient’s health, independence of mind, and the need for harmony between the individual, social and natural environment" (Kleisiaris 2014). We are now at a point where we have even more insight into this than ever before. As clinicians it is our duty of care to use it wherever appropriate and through new technology, we can make the NHS a safer environment for patients and create a culture where ‘first do no harm’ is far more achievable.