The pandemic has created room for technological progress in the medical field. COVID-19 strains on frontline workers and existing systems used to treat, support and service patients are spurring the development of new technologies to support and meet the demands of complex jobs. I put on
Technologies like AI chatbots and ChatGPT can handle tasks like responding to insurance claims, freeing up doctors to focus on patient-centric work. The potential of AI makes it tempting to believe that at some point it will be able to complete tasks without the need for human effort or input.
But even with recent advances in AI capabilities, artificial intelligence will never eliminate the need for pharmacists, said Sam Underleg, CEO of pharmacy information technology company DocStation.
Healthcare IT News spoke to Anderegg to find out exactly why he believes this to be the case, and what the future holds for AI and pharmacy.
Q. You advocate how technology can enhance healthcare, but you warn healthcare professionals against putting too much emphasis on artificial intelligence. What kind of balance are you trying to strike here?
A. Artificial intelligence has the incredible potential to transform nearly every imaginable industry. Today, we are witnessing the power of consumer tools like ChatGPT to digest massive amounts of data on the internet to answer complex questions and draft helpful summaries, articles, emails and itineraries. It makes our daily life easier.
We are also seeing AI play a role in businesses, such as creating virtual meeting minutes or suggesting targeted advertisements based on interactions with social media posts and content.
The Silicon Valley ethos of “moving fast and breaking things” is perfect for building a marketing or e-commerce company. But breaking something in a medical context can have detrimental consequences, including misdiagnosis, causing medication errors, and most seriously, endangering the patient’s life.
Done wrong, it can harm patients, have adverse consequences, and increase costs. The great thing is when it works. There are so many manual and labor intensive administrative tasks in the healthcare process that we are unable to maximize our time as a clinician doing and doing what we can.
Just as AI makes life easier for us at home and at work, it also reduces or hopefully eliminates the simple non-clinical tasks that are now part of our daily routines, helping us It may allow you to spend more time on your work. What we love and have been trained to do is take care of our patients. The ultimate goal is to leverage AI to enhance patient care and improve the patient and provider experience.
However, we must be careful, cautious and responsible in how we leverage AI relative to other industries to reduce the risk and potential for harm.
Q. You warn healthcare executives not to think AI will replace pharmacists. Why can’t AI achieve such feats?
A. There are several traditional pharmacy functions that artificial intelligence may replace. What people often overlook is the rapidly evolving role of the pharmacist in the healthcare team.
Early in this century, most vaccines were provided in primary care clinics. More than 20 years later, in the midst of a pandemic, more than three-quarters of people are now vaccinated in pharmacies.
Healthcare is rapidly moving towards convenient and accessible retail healthcare facilities, and pharmacies are at the forefront of this transformation. In response, the pharmacy practice model is evolving to be more service-oriented.
Pharmacists now occupy a mid-level provider role within the health care team and are ideally positioned to solve the health care access problem, especially in rural and underserved areas. That said, the patient, like her PCP, values the human touch the pharmacist provides.
AI may be able to provide information, but it cannot provide the same level of understanding and problem-solving as a human pharmacist. Using this frame of reference, we can think of AI impacting the role of pharmacists in the same way that we think of AI impacting the role of primary care providers.
AI has the potential to enhance some aspects of the care delivery process, but patients still need to communicate with, educate, and guide healthcare providers through their care journey. Pharmacists are often the most accessible and frequently visited providers in the health care team.
Patients and the healthcare industry at large will continue to rely on pharmacists to serve as frontline healthcare providers and trusted advocates for patients.
Q. So what can AI do to help pharmacists do their job better and more efficiently?
A. Like other healthcare providers, artificial intelligence has the potential to make pharmacists’ jobs easier and better by automating administrative tasks and enhancing patient care processes.
One potential application is to make administrative tasks and functions often provided by support staff faster and easier. During the pandemic, many pharmacies have deployed software to schedule appointments for vaccinations and other services.
There are now several AI-assisted chat services that can assist patients with non-clinical functions such as scheduling appointments, initiating medication refills, and contacting patients for follow-up.
As pharmacies evolve from drug dispensing to service delivery, there are many applications that AI can potentially improve when it comes to the medical billing and revenue cycle processes. Manual claims performed by our support staff may result in errors and refund denials or delays.
Automation helps eliminate errors by flagging incomplete or inaccurate information before a claim is submitted. This reduces the number of denied claims, makes the claims process more efficient, and maximizes potential revenue.
Leveraging these tools to perform administrative tasks is generally low risk, but can help facilitate a highly inefficient and cumbersome process without the risk of causing error or harm to patients. increase.
Pharmacies are also in the business of working with payers on value-based arrangements to improve clinical outcomes.
While some software vendors and tools rely on basic logic (such as if/then decision trees) to identify at-risk patients, AI has the potential to predict which patients will soon be at risk. For example, if a patient is regularly topped up with medication, and in consecutive months begins to be behind schedule in receiving medication, the AI will identify this drug dissatisfaction trend, summarize the information, and use it for evaluation and patient assistance. can be presented to the pharmacist at Guaranteed.
Such applications have the potential to prevent adverse events, improve patient outcomes, and save money for the healthcare system.
Just as AI reads vast amounts of information and creates vacation itineraries, it summarizes medical information such as diagnoses, allergies, laboratory values, treatment history, progress notes, automates the recording of medical appointments, and potentially It can also be used to diagnose and identify symptoms. Recommend possible treatments.
A relatively benign example might be to recommend the most affordable drug option given the drug formulary used by a particular patient’s pharmacy/medical insurance.
Leveraging AI for clinical use cases is a major advance, but there can be information that may lead to risks, is not associated with, or considered by AI.
For this reason, pharmacists and other health care providers still need to evaluate recommendations for safety, efficacy, and suitability from a holistic approach before implementing recommendations.
These examples just scratch the surface of how AI could make a pharmacist’s job and daily responsibilities easier, more efficient, and more fulfilling in the short term.
Q. What are some legal and ethical considerations that pharmacists struggle with for AI to manage?
A. The first legal and ethical considerations pharmacists and other healthcare professionals should be aware of regarding potential biases associated with the use of artificial intelligence in healthcare settings. AI algorithms are trained using historical data, but the data can be inherently biased because underserved people lack access to healthcare.
Without adequate access to healthcare, underserved populations may be underdiagnosed or not receive the same treatment compared to other populations with the same healthcare risk profile. . For pharmacists and other providers, it is important to understand how AI algorithms are developed and trained, and how quality assurance is performed to understand potential biases.
It is also very important for providers to understand the risk of error in clinical applications of AI. AI has the potential to make clinical recommendations, but algorithms may not have access to the critical information needed to make “informed” decisions about treatment plans.
One serious negative effect is the heavy or blind reliance of healthcare professionals on AI to make treatment decisions. Our ethical responsibility as a provider is to do no harm. It is important to understand that when pharmacists rely on AI to enhance clinical decision-making, they can do harm.
Another consideration that may be less obvious is patient consent. The use of AI, especially in medical settings, is still very experimental. Considering it in context facilitates comparisons to the field of clinical trials.
Just as patients understand the potential risks and benefits of participating in clinical trials involving new therapies, patients and their healthcare providers understand the risks of using AI as part of their healthcare experience. And you need to understand the benefits. The technology industry is known to be a generally unregulated sector based on the speed of innovation relative to the policy-making process (think cryptocurrency regulation in the financial industry).
To “do no harm” is part of our responsibility to adopt the same informed consent practices in AI that we use in other areas of the healthcare industry.
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email the author: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.
