Industry
Retail
Omnichannel CX and Intelligent Innovation – The Future of Retail
HIGHLIGHTS
- The current process of transcribing prescription instructions onto medicine labels by pharmacy retailers using sig codes can lead to errors, especially during busy periods.
- Artificial intelligence (AI) can enhance sig code automation rates, and the technology can be quickly customized to meet the needs of pharmacies.
- Automating the sig process with AI will not only reduce the number of quality and safety alerts for patients and pharmacists but will also save the time spent on ensuring process efficiencies for pharmacists and technicians.
On this page
Streamlining prescription labelling through sig automation
Pharmacists must be extremely careful while filing out prescriptions, particularly for patients with chronic illnesses or multiple medications.
At the same time, it can be challenging for patients to correctly interpret the dosage time of medications, frequency, and amounts prescribed by their doctors. The method that pharmacy retailers currently use to transcribe prescription instructions onto medicine labels can lead to errors, especially during busy periods. The current rule-based system that automatically processes prescription instructions is not 100% accurate, meaning the rest of the instructions still need to be handled and vetted manually by pharmacists.
This challenge is prompting pharmacy retailers to explore the use of artificial intelligence (AI) to improve the efficiency and accuracy of prescription labelling.
Electronic prescriptions (eRx) have revolutionized the healthcare and pharmacy retail sectors by improving efficiency, accuracy, and accessibility. However, despite these advancements, a significant challenge remains in accurately identifying sig codes, derived from the Latin word ‘signatura’, meaning directions for use. Prescribers provide instructions for administering each medication using sig codes, which combine natural language and medical mnemonics. These codes convey essential information, including medication dosages, measurements, timing or frequency, and routes of administration to pharmacists, pharmacy technicians, and patients. This ensures patients understand how and when to use their medications. However, the current process requires significant manual effort, which can be improved through a sig-code automation system.
The impact of sig code errors on patient safety
The issues surrounding sig codes can lead to significant errors in prescribing medication and increased non-compliance among patients.
These challenges can also result in substantial financial costs for healthcare providers. Incorrect or unclear sig codes can lead to patients taking the wrong dosage or improperly consuming their medications, potentially causing adverse drug events.
In the absence of proper labelling, patients may misunderstand how to take their medications, leading to poor adherence and suboptimal treatment outcomes. This situation also increases the workload for pharmacists and healthcare providers, who must spend additional time clarifying unclear prescriptions, which can delay medication dispensing and disrupt the workflow. Additionally, misidentified or incomplete sig codes require follow-up communications between prescribers and pharmacies, causing delays in medication availability for patients.
Moreover, inaccurate sig data can expose healthcare providers and retail pharmacies to legal risks if errors result in patient harm. The additional resources required to manage and rectify these sig code-related errors also contribute to increased operational costs for healthcare organizations.
Limitations of sig automation in prescription systems
The existing prescription systems rely on a rule-based engine to automate the interpretation of sig code messages, usually from electronic prescriptions.
Automation, currently, is only about 40%, which means more than half of the incoming prescriptions require manual intervention (through the triage queue) from pharmacists to identify the correct sig values. Pharmacists must interpret prescription messages, print medication labels, and provide patients with instructions on the administration of medication.
In addition, there are limitations to the existing set of crosswalked sigs, which are standardized mappings of different sig codes used to ensure consistency across prescriptions. The sig codes are essentially maintained manually, and the use of AI in analyzing sig patterns across various drug classifications is minimal. The manual process is time consuming for pharmacists and technicians and also prone to errors, posing risks to patient safety.
Consequently, the low level of sig automation results in significant manual labor to interpret and fix sig errors. Further, the introduction of new drugs and Rx directions takes time to integrate into the automation rules. Major changes or upgrades to the existing system will mean a longer time to market, ultimately impacting patient service.
AI for seamless sig automation
The key to sig automation is to accurately parse prescriber instructions as standardized sig codes or content.
This information must then be presented to the end user in a clear, detailed format that is easy to understand. An AI-backed software to interpret sig codes can allow new changes to be incorporated seamlessly and also pave the way for faster time to market.
AI can enhance sig code automation rates, and the technology can be quickly customized to meet the needs of pharmacies. The four key pillars of leveraging AI for sig automation are natural language processing (NLP), machine learning and predictive analytics, rule-based systems, and decision support systems.
Implementing AI-driven systems in large pharmacies with enterprise pharmacy management systems will seamlessly integrate the new software and speed up the overall sig code transformation process.
AI algorithms can analyze prescriptions and automatically extract sig instructions accurately, eliminating the need for manual interpretation by pharmacists.
This can be achieved by implementing NLP techniques, which allow AI to accurately interpret diverse sig formats, with in-built logic for customizable electronic prescriptions.
To avoid complicating the system architecture with a separate solution, pharmacy retailers must integrate the AI software into the existing workflow of the prescription dispensing system. The AI software will automatically check for patterns in sig data, cross-referencing specific drugs, conditions, and prescriber information. This integration will enable automatic conversion of uncrosswalked sigs and flag potential issues based on the analyzed data.
This approach will reduce the number of safety alerts caused by sig code errors and shorten the time taken to address them through AI-based pre-population of relevant data. Pharmacy retailers must adopt the following measures to enhance the existing system:
- Standardize sig instructions: Pharmacies will need to modify the algorithms to standardize sig instructions to ensure consistency and uniformity in medication usage directions.
- Enhance AI accuracy: Training the AI model using a diverse dataset of prescriptions will improve its accuracy in recognizing and interpreting sig instructions. This can be achieved by leveraging the existing proprietary model itself. Additionally, maintaining an AI-based self-learning rule set and probabilistic model for newly introduced drugs and new iterations on Rx directions will reduce manual intervention.
- Adopt feedback-driven AI learning: Retailers should implement a feedback mechanism where the corrections of pharmacists and technicians to sig data are automatically fed into the AI model, allowing the system to learn and improve over time.
- Ensure continuous AI model updates: Companies will have to regularly update the AI model using new prescription data, regulatory data, sig codes, condition data, and so on, and integrate advancements in NLP and AI technologies.
The road to good health
Automating the sig process using AI can significantly reduce the number of quality and safety alerts for patients and pharmacists.
Additionally, it will save valuable time by improving process efficiencies for pharmacists and technicians. For drugmakers, this translates to quicker realization of business value, reduced time to market, and optimized inventory management, all of which contribute to an improved patient experience.
Moreover, this automation would allow more electronic prescriptions to go directly into the workflow without requiring manual intervention as pharmacists will not have to manually interpret sig instructions.
AI has the potential to improve sig automation rates by up to 15% within six months, resulting in a highly streamlined and automated workflow. Additionally, pharmacy retailers can secure a higher return on investment (RoI) by adopting this new system while ensuring seamless integration of AI software. Automation will facilitate faster prescription processing, which, in turn, will free up more time for pharmacists and enhance customer satisfaction. Embracing these advancements will help pharmacy retailers transform pharmacy operations and significantly enhance patient care and safety.
