Recently I gave a talk at the Revolutionizing Healthcare with IT Conference in Mumbai around ROI of Health IT. Here’s the gist!
Before I delve any deeper, let’s understand what ROI is.
ROI can mean different things to different people. To nurses and infection control teams, ROI means ‘Risk of Infection’. To most of us burdened by home loans, car loans and education loans, ROI means ‘Rate of Interest’. To the CEO who makes gut-wrenching investments and wants to make money back, ROI means ‘Return on Investment’. As for me, the emergency physician in me wants to take ROI at its face value, but the clinical informaticist in me thinks of ROI as the ‘Radius of Information’.
Let me illustrate what this means. Think about the “radius of information” as a measure of the circle of knowledge, i.e. the super-set of all that we know in medicine (Fig 1). From all this knowledge, there are things that you know and things that you don’t know i.e., you know what you know and you know what you don’t know. However, with an explosion in the rate of medical information growth coupled with the slow adoption of research findings into clinical practice, more often than not, physicians don’t know what they don’t know. This is what is known as knowledge variability. No two doctors, no two nurses have the same amount of knowledge, and that is at the core of knowledge-based medical errors.
Figure 1The other fundamental cause of medical errors is operational variability. Operational variability arises, for example, when a physician’s handwriting results in a prescription error. Similarly, there are many categories of operational lapses that result in medical errors.
Reducing operational variability is usually the mainstay of most quality programs and health IT is often deployed to yield returns from operational optimization. But, the question remains, what about knowledge-based errors, where care providers at best fail to provide high-value care and, at worst, cause preventable medical errors and deaths due their knowledge gaps? Knowledge variability is hardly ever addressed by health IT.
Knowledge variability stems from the “Knowledge Dilemma” (Fig 2) posed by two determinants of medical knowledge – Diffusion Time and Doubling Time. On one hand, it takes, on an average, 17 years for research to move into day-to-day clinical practice and on the other hand, medical knowledge will soon double every 73 days!
Reducing operational variability is usually the mainstay of most quality programs and health IT is often deployed to yield returns from operational optimization. But, the question remains, what about knowledge-based errors, where care providers at best fail to provide high-value care and, at worst, cause preventable medical errors and deaths due their knowledge gaps? Knowledge variability is hardly ever addressed by health IT.
Knowledge variability stems from the “Knowledge Dilemma” (Fig 2) posed by two determinants of medical knowledge – Diffusion Time and Doubling Time. On one hand, it takes, on an average, 17 years for research to move into day-to-day clinical practice and on the other hand, medical knowledge will soon double every 73 days!
Figure 2
Now, with this disastrous collision of realities, is the practice of medicine no longer “humanly” possible? How can we avoid variability in patient care while keeping current? And how can we optimize ROI of health IT?
In order to truly address new challenges appearing as our entire healthcare delivery model evolves, technology needs to be evidence-adaptive i.e., to have a knowledge-base that continually reflects current evidence so that it can bring the right information to physicians at the right time, at the point-of-care. Clinical Decision Support Systems (CDSS) are the most pertinent answer to the vast and destructive problem of variability in care delivery. Evidence-adaptive CDSS from authoritative sources have proven to reduce spending on unnecessary tests and procedures as well as avoid costly adverse events (and in many systems, malpractice litigation claims). Although at present evidence-adaptive platforms require human intervention, we are now beginning to see the inclusion of artificial neural networks (deep learning), Bayesian networks, reinforcement learning, and other artificial intelligence techniques for synthesizing evidence relevant to patient data in real-time, with potentially unprecedented insights for clinicians.
Intelligence Augmentation (IA) with evidence-adaptive CDSS, where technology amplifies the decision-making capabilities of humans, links healthcare providers to vast amounts of patient data in the Electronic Health Record (EHR) with relevant clinical knowledge, in real-time, at the point-of-care. This can dramatically increase clinicians’ ROI – radius of information, thereby improving clinical outcomes and consequently increasing the healthcare organization’s ROI – return on investment. Indeed, a longer ROI for a higher ROI!
End note: Wondering if you should build evidence-adaptive CDSS in-house or opt for a third-party solution? In my latest whitepaper, I weigh the pros and cons of each option. You can read more about it: http://bit.ly/2AXR6h9
In order to truly address new challenges appearing as our entire healthcare delivery model evolves, technology needs to be evidence-adaptive i.e., to have a knowledge-base that continually reflects current evidence so that it can bring the right information to physicians at the right time, at the point-of-care. Clinical Decision Support Systems (CDSS) are the most pertinent answer to the vast and destructive problem of variability in care delivery. Evidence-adaptive CDSS from authoritative sources have proven to reduce spending on unnecessary tests and procedures as well as avoid costly adverse events (and in many systems, malpractice litigation claims). Although at present evidence-adaptive platforms require human intervention, we are now beginning to see the inclusion of artificial neural networks (deep learning), Bayesian networks, reinforcement learning, and other artificial intelligence techniques for synthesizing evidence relevant to patient data in real-time, with potentially unprecedented insights for clinicians.
Intelligence Augmentation (IA) with evidence-adaptive CDSS, where technology amplifies the decision-making capabilities of humans, links healthcare providers to vast amounts of patient data in the Electronic Health Record (EHR) with relevant clinical knowledge, in real-time, at the point-of-care. This can dramatically increase clinicians’ ROI – radius of information, thereby improving clinical outcomes and consequently increasing the healthcare organization’s ROI – return on investment. Indeed, a longer ROI for a higher ROI!
End note: Wondering if you should build evidence-adaptive CDSS in-house or opt for a third-party solution? In my latest whitepaper, I weigh the pros and cons of each option. You can read more about it: http://bit.ly/2AXR6h9
Author
Dr. Ujjwal Rao
Dr. Ujjwal Rao is Senior Clinical Specialist in Integrated Decision Support Solutions, and is based in New Delhi, India. He provides strategic counsel to health providers on designing world-class clinical decision support systems with Elsevier’s comprehensive suite of current and evidence-based information solutions that can improve the quality and efficient delivery of healthcare.
An experienced emergency physician, executive, clinical informaticist and technology evangelist, Dr. Rao has a decade of experience serving in trust and corporate hospitals in various roles ranging from clinical administration, hospital operations to quality & accreditation. In his former positions, Dr. Rao led EHR implementations for large hospital groups and designed bespoke healthcare analytic solutions to raise profitability.
His passion to see transformation through technology led him to volunteer as a quality consultant with the United Nations. He also currently serves as an Assessor on the Panel of the Quality Council of India for the National Healthcare Accreditation Standards body, NABH.
Dr. Rao obtained his degree in Medicine and then specialized in Hospital and Health Systems Management, Medical Law and Ethics before completing his PhD in Quality and Medical Informatics.
