Future trends in clinical decision support systems (CDSS)

Vivek Patkar, MBBS, MS, MRCS, FEBS Fellow of European Board of Breast Surgery

Chief Medical Officer at Deontics

Future trends in clinical decision support systems (CDSS)

21 September 2022 | 9min

Quick Takes

  • Clinical Decision Support Systems (CDSSs) are revolutionizing patient and chronic disease management in four areas: therapeutic pathways, precision medicine, diagnostic imaging, and personalized healthcare

  • Several trends where CDSSs can be deployed include data-driven patient management, learning healthcare systems, remote patient monitoring, sensor-based systems, and value-based healthcare

  • While the technology behind these systems is powerful, other human and organizational factors must be considered for it to be successfully implemented

Clinical decision support systems (CSDSSs) are being used to revolutionize healthcare in various ways including patient management, chronic disease management, and even in the shift toward value-based health care (VBHC)

To dive deeper and learn more about the upcoming trends for CDSSs and how to gain the most value out of these systems, we sat down with Vivek Patkar, Chief Medical Officer at Deontics.

Use cases for CDSS and the ways they are revolutionizing patient care delivery

HT: What are the current main use cases for clinical decision support systems (CDSSs) and in which ways are they revolutionizing the delivery of patient care? 

Vivek Patkar: The clinical decision support system market will be $USD 2.2 billion in the next three to five years from $USD 1.3 billion today.1 New AI-based medical decision support systems are driving this. 

There are four areas of current use cases. Let’s look at how they’re going to revolutionize the delivery of patient care in the coming years.   

1. Therapeutic pathways: Standardized pathways for chronic disease management are becoming more common practice in most healthcare organizations where CDSSs are playing major roles. 

Chronic diseases include common diseases, like respiratory disease, or cardiac diseases, and those requiring more specialist care such as cancer. Chronic diseases must be managed throughout the entire patient journey from prevention to diagnosis, diagnosis to survivorship, to palliative care in some cases. 

2. Precision medicine: Rapidly advancing technologies of high-throughput sciences such as genome sequencing, and pharmacogenomics are changing the healthcare industry both in terms of our understanding of diseases and the availability of targeted therapeutics. 

Targeted therapies allow physicians to minimize unwanted harmful effects by precisely targeting diseased cells based on an individual’s genomic makeup. However, keeping up with rapidly expanding knowledge in the era of precision medicine is becoming a humanly impossible task.

Pharmacogenomics can impact the medication choice for a particular patient, from painkillers to cancer medications. For example, the US Food and Drug Administration (FDA) lists over 100 cancer drugs that have pharmacogenomic associations.2-3  Medical decision support systems can help general practitioners or doctors in hospitals to individualize the dosages of these drugs based on individual genetic make-up, thus preventing many common side effects of these drugs.  

All of this information is difficult to put into practice for an unaided physician without clinical decision support technology. Especially when you are managing patients on multiple medications for many different conditions. 

3. Diagnostic imaging: In diagnostic fields, artificial intelligence (AI)-based support systems are advancing image processing. Whether it is radiology or pathology, these clever AI systems can use many different algorithms to help clinicians and specialists potentially make quicker and safer decisions. 

The AI-based systems assist pathologists and radiologists by pointing out abnormal lesions or areas in the image which could be overlooked or could take a longer time to identify for an unaided clinician.  

4. Personalized healthcare: Understanding patient preferences, values, and individual circumstances, like occupation, plays a key role in the shared decision-making process. Combining clinical evidence from randomized clinical trials with patient values and preferences provides more personalized healthcare, which may help result in better patient satisfaction and improved compliance with the treatments. 

Personalized healthcare goes beyond precision medicine to consider patient needs

Vivek Patkar: Sometimes precision medicine and personalized healthcare are often used interchangeably, but personalized medicine is much more than just precision medicine. Precision medicine takes into account an individual’s molecular and genomic profile to select targeted treatments to help avoid unnecessary side effects, while personalized medicine takes into account personal preferences, goals, and values in optimizing treatments. Both are complementary and not alternatives. 

Personalized healthcare goes beyond simple medication prescriptions. It is not just about therapeutics, but about considering personal circumstances and preferences in decision-making. One size does not fit all and two patients with similar disease conditions may need different therapeutic options that suit their personal circumstances best.  

For example, a younger prostate cancer patient may like to avoid treatment that could adversely impact sexual health and the quality of life, even if it is the most potent treatment available. He may choose instead the treatment that does not impact these areas as much, yet comes at a cost of a slight reduction in length of survival time. He may weigh things differently from a person with the same cancer but who is sexually less active. 

Upcoming CDSS trends to look out for in the next 5 to 10 years

HT: Where do you see clinical decision support systems (CDSSs) playing a major role in the next 5 to 10 years?

Vivek Patkar: There are several upcoming trends where CDSS can be deployed in the next 5 to 10 years. 

1. Data-driven patient management. The nineties witnessed a huge shift towards evidence-based medicine (EBM), which was more about randomized clinical trials (RCT) and applying the results of RCTs to develop guidelines and protocols to standardize treatments. It will remain a part of the medical decision-making process, however, unprecedented access to real-world medical data (big data), and advancements in machine learning and AI has opened doors to a new paradigm of data-driven medicine. 

RCTs represent a very small patient population. For instance, it is estimated that less than 5% of adult cancer patients enroll in cancer clinical trials.4 Most of the patients are treated or managed outside RCTs. We can now unlock the insights from this real-world data using the power of analytics and machine learning AI. 

2. Learning healthcare systems. As the name suggests, a learning health care system uses real-world data to learn from clinical practice variations. The system audits its own recommendations and the actual decisions made by clinicians and then analyzes the outcomes to refine its recommendations. These learning systems can fast-track the clinical practice guideline development process.

3. Remote or virtual monitoring. The aging population with multiple comorbidities is a huge burden on healthcare systems worldwide and hospitals are stretched to the limit. There is an increasing trend in managing and monitoring less complex patients away from the hospitals in their own homes/care homes, by means of virtual care. This frees up hospital beds for more complex patients that really need close monitoring.

4. Sensor technology. Advances in sensor technology have made the remote/virtual monitoring of patients a clinical reality. Routinely used devices such as our mobile phones are equipped with sensors that can detect and process clinical information such as vitals, activity, sleep duration, and so on. 

Due to the rising demand for medical sensors, partly driven by consumers and their increasing health awareness, the global medical sensor market is expected to grow from $USD 15.23 billion in 2020 to $USD 32.15 billion by the end of 2028.5  The CDSSs integrated with devices will be critical to making use of the data from sensors and drive recommendations to help patients and carers in making the right decisions at the right time, taking away some of the burdens from specialists in hospitals. 

5. Value-based healthcare. Business as usual is no longer financially sustainable for the healthcare system with an aging population and growing costs of therapeutics. CDSSs can help design and implement VBHC pathways to optimize costs while helping to achieve the outcomes that matter most to patients.   

How to successfully implement a CDSS in your organization

HT: In your experience, what must be in place for healthcare organizations to gain the most value from a clinical decision support systems (CDSSs)? 

Vivek Patkar:   Healthcare organizations will gain the most value from a CDSS if they have the following in place:

1 The human aspect: change management. Many times while implementing innovation such as CDSS, we focus on the technology ignoring the human aspect of the system at our own peril. Change management is equally important if you want successful CDSS adoption to generate the most value possible out of it.
2 Appropriate interoperability standards. Enforcing interoperability standards mitigates one of the key challenges for CDSS, which is getting information from different systems. CDSS is only as good as the data or information that goes into the system. 
3 Vision about overall systems architecture. Having a unified vision about how different technology systems and platforms within the organization will fit together to provide a seamless experience to all the users is essential to avoid fragmented deployment where each IT system is a silo.
4 Stakeholder buying. All the key stakeholders involved in patient care should feel a sense of ownership. If they are alienated from the process of deployment of new technology, then adoption is challenging. 
5 Clinical governance structures. These should be in place for continuous monitoring and timely updates of any CDS system.

Validating CDSS in healthcare organizations

HT: How would healthcare organizations go about validating clinical decision support systems (CDSSs) to ensure it’s beneficial to their needs?

Vivek Patkar: Different organizations have different needs depending upon the population they serve, their resources, the way they are structured, technology infrastructure, etc. It is important from an organization’s perspective that whichever CDSS solution they are using, it is benefiting them. Here are a few steps organizations can follow to do this. 

  1. Identify and measure your current practices and identify gaps. Until organizations understand what is going wrong or where there is room for improvement, it will be difficult to measure the benefit or success.
  2. Set performance metrics and goals. The best way to do this is to analyze your organization’s data to enable measuring baseline outcomes and setting performance metrics and goals. The clinical impact, operational impact, and financial impact of any CDS solution must be considered. 
  3. Monitor the CDSS. Ensure the organization is meeting its goals in all the three areas (clinical, operational, and financial) mentioned above. For example, a CDS solution may improve diagnostic efficacy but unless subsequent management processes are also streamlined, it will clog the system downstream. This would negate any benefits of quicker diagnosis. So always evaluate clinical, operational, and financial impact. 

Recommendations for executives looking to integrate a CDSS into their organization

HTWhat main advice would you give to healthcare executives who are thinking about integrating clinical decision support systems (CDSSs) into their organization? 

Vivek Patkar: My recommendations would be as follows:

  1. Have an organization-wide digital strategy and vision. There are many different moving parts when you are deploying any IT system. So, having an agreed organization-wide strategy, which all stakeholders have bought into makes things easy. Stakeholders should include not just executive management but also the end users, managers, IT technical staff, support staff, and patient representatives. 
  2. When you are looking for a clinical decision support platform, preferably it should be electronic medical record (EMR) or electronic health record (EHR) agnostic in order to avoid getting locked into a specific EMR/EHR platform.
  3. Think about how you can learn from the data generated by the CDS system to make your organization a learning healthcare system.

Vivek Patkar, MBBS, MS, MRCS, FEBS Fellow of European Board of Breast Surgery is Chief Medical Officer and one of the founders of Deontics Ltd, a medtech company in the field of AI and Clinical Decision Support. Vivek was originally trained as surgical oncologist at TATA Memorial hospital, the largest tertiary cancer hospital and research centre in India, after graduating in medicine from the University of Bombay in 1995. Vivek has developed and piloted a number of clinical decision support applications for cancer as well as other indications and led a team of researchers at the University College London medical school (Royal Free Hospital) which worked on improving the quality of decisions made in cancer multi-disciplinary team meetings using clinical decision-support. Most recently Vivek and collaborators at Guys & Kings received an NHSx and NIHR AI in health and care award in 2020.


  1. Businesswire.com. (2021). Article available from https://www.businesswire.com/news/home/20210617005730/en/Global-Clinical-Decision-Support-System-Market-2021-to-2026—Players-Include-Cerner-McKesson-and-Meditech-Among-Others—ResearchAndMarkets.com [Accessed September 2022]
  2. FDA. (2022). Report available from https://www.fda.gov/medical-devices/precision-medicine/table-pharmacogenetic-associations [Accessed September 2022]
  3. Edelman E. (2022). Article available from https://www.jax.org/news-and-insights/jax-blog/2022/february/when-is-pharmacogenomic-testing-useful-in-cancer-care [Accessed September 2022]
  4. Unger et al. (2016). Am Soc Clin Oncol Educ Book, 35. 185-198
  5. Transparaceny Market Research. (2022). Report available from https://www.transparencymarketresearch.com/medical-sensors-market.html [Accessed September 2022]