Clinical Decision Support Systems(CDSS)- A Companion or Contradiction to Physicians ??

We always seek guidance and active support when faced with complex situations, especially in decision-making tasks. Call them Experts or Mentors, they are in huge demand due to the ever-changing landscape of jobs and responsibilities. So, do Physicians and healthcare professionals. For a long, those support systems during Doctors’ hours of dilemma or ambiguity have been conventional knowledge-based support systems like UpToDate or structured, concise handbooks like Washington Manuals; which highlight the guidelines, only for the treating Physician to make the final call about treatment decisions. Though this system of Evidence-Based Practice helped in providing standardised care, what was lacking was the robustness of the decision process. Remember, a Clinician has to arrive at a prescribing decision in a matter of few minutes, with the available patient information, lab results and the limited scope of his knowledge. This style of decision-making is prone to human errors and misjudgements.

With the introduction of EMRs, and with an increasing acceptance of their usage, we now have large and significant patient Data, that can be data mined and analysed to improve patient outcomes by making necessary interventions. These large data sets are beyond human capabilities to analyse and interpret; this is exactly where CDSS will help Clinicians; to provide inputs and suggestions based on the knowledge it gained by analysing large and wider volumes of patient characteristics. The role of CDSS isn’t just limited to Diagnosis; it can assist in prescribing orders and labs, predicting Drug-Drug or other interactions, and sending critical alerts

CDSS has evolved and the latest versions have been tried and tested to be more effective and reliable. CDSS can play a ‘Consulting’ role, becoming the adviser proactively or a ‘Critiquing’ role, responding after the Physician triggers. The widely used models are the ones based on Deep Learning, Artificial Intelligence or Convolute Neural Networks(CNN). These models follow a stepwise ‘flow chart’ process as mentioned in our medical texts. That’s the biggest catch of CDSS; having a decision-making system that follows a guideline-based ‘decision-tree’ approach, rather than ‘black box’ warning methods. To cite an example of the evolution of CDSS; what was once a simple drug dosage and dispensation software, has now transformed into a complete ‘Pharmacy practice’, suggesting dose modifications, dose adjustments, potential Drug-Drug interactions, to even recommending alternate treatment options based on patient criteria and medical condition.

The National Health Authority, INDIA in collaboration with AIIMS, New Delhi has recently developed a digital health technology-based CDSS under the name of CARDIOMETCARE-m to help Primary Care Physicians in managing Non-Communicable diseases. The aim is to provide Evidence-based recommendations to Physicians by integrating CDSS into existing EMRs.

If CDSS is such a vital and groundbreaking tool that enhances a Clincian’s accuracy in decision-making at the click of a button, why is its adoption so infinitesimal, especially in India, and why haven’t Healthcare Organisations absorbed this yet? Let’s explore a few bottlenecks, advantages and disadvantages of CDSS in the context of India.

Bottlenecks:

The biggest Bottleneck is that most Physicians haven’t used it, nor are inclined to adopt it, to even understand its full potential. The inertia to begin, the reluctance to evolve and the confidence and dependence on their expertise are the biggest bottlenecks. Nevertheless, these issues can be addressed by harbouring a collaborative and step-wise approach to implement it in phases and adopt it in batches.

Advantages of Clinical Decision Support Systems – CDSS:

Improves patient safety with accurate suggestions and timely reminders(like in a Diabetic Patient on Insulin Infusion)
Improves adherence to Clinical Guidelines based Management, with Guidelines in place, it’s easy to conduct Clinical research too.
Cost-effectiveness: CDSS can reduce the length of stay, avoid duplication of orders and can even suggest cheaper alternatives to prescribed drugs.
Administrative Ease: Easily pick ICDs(International Classification of Diseases), alert vaccinations and other important reminders. It can facilitate easy coding for Insurance Billing and other administrative tasks
Diagnostic Decision Support Systems(DDSS) for improved diagnosis: AI and Deep Learning models in Medical Diagnosis & Imaging are fast catching ground. We have seen how a Google Project accurately Diagnosed Diabetic Retinopathy by scanning and analysing lakhs of retinal images at par with Ophthalmologists.
Patient Health Records(PHR): With the integration of EMRs and PHRs, CDSS can become a game-changer in Healthcare-Patient communications, transferring all important and critical data between them.

Disadvantages:

Impaired workflows of professionals and reduced patient interactions and interface can become a potential disadvantage
Alert fatigue: Excess, unwanted alerts may create burnout and distrust among the Clinicians
Impact on reasoning and skills: Overtime; overdependence on CDSS may ensue, which could negatively impact human thinking abilities. A tell-tale example is the usage of calculators which led to their over-reliance even for minor mathematical calculations.
A decent level of Computer literacy is expected which is lacking among Clinicians, especially if the algorithms are complex to use
Annual Maintenance of the systems, applications and databases is a costly affair.
Regular upgradation and incorporation of Evidence-based rules and formulas is a tedious and challenging job, it needs major overhauls.
Interoperability issues: This is by far the biggest drawback of CDSS, where the exchange of data between entities and organisations is limited by the legal framework and sensitivities involved. However, most of the commercial EMR vendors have been adopting Fast Healthcare Interoperability Resources(FHIR) which have addressed these core ethical issues to a larger extent. The NHA, under the auspices of ABDM(Ayushman Bharat Digital Mission), has made “interoperability’ an important and mandatory standard for all EMR software accredited by the body. The Government of India has come up with the “Digital Personal Data Protection(DPDP) rules in 2025 that laid down guidelines and emphasized adherence to Data Protection by all operators and stakeholders.

Clinical Decision Support Systems(CDSS) : CDSS is an invaluable tool for Clinicians in Decision-making, and it should rather act like a facilitator and not a dictator. Even with the most validated and accredited software, the Clinician should always use his acumen and experience in drawing conclusions. CDSS systems should be more extensively studied, tested and validated in large models; always comparing the results with expert panel groups. This re-validation and reassessment process should be a continuous ritual to keep pace with the changing guidelines and scenarios. The potential for misuse of these systems by unqualified persons looms large, and the management teams along with regulatory authorities should ensure strict compliance of usage.

“ Human-human interactions produce results with emotion while Human-Machine interactions produce results with accuracy; CDSS should address both, accuracy of a machine with a human touch “!!