Digital Assistants/Chatbots: What Should Patients Expect?
Digital assistants like Siri , Alexa, Cortana and Google Assistant are becoming commonplace in our houses, devices and apps. Soon these same assistants will begin offering healthcare related services. Here’s a look at what to expect for the next 2 years.
Soon these digital assistants may also order our medications or schedule dental cleanings. For example, Microsoft and others now offer AI-platforms for developers to create bot services. Possibly, creating digital medical assistant services. Imagine asking:
“Siri: Find me a Heart Surgeon!”
But are patients ready to accept medical advice from an algorithm?
Currently, healthcare digital assistants are chatbots. Their user interface is primarily textual. Equipped with natural language processing, machine learning, sentiment analysis and interfaces to converse, learn and interact with humans.
We expect the healthcare industry to offer Artificial Intelligence (AI)-equipped chatbots. To perform routine and repetitive tasks, such as, helping employees navigate company healthcare plans. Here is how a chatbot conversation might begin:
Chatbot: Hi Jeff, My name is Kirk. How can I help you today?
Jeff: What are my health insurance options?
Chatbot: Okay. Lets get started.
Chatbot: Jeff I will need to get some information from you first.
Healthcare costs are measured differently than commercial costs. If chatbots simplify or improve patients’ experience they are cost effective. For example, chatbots will require more expensive human redundancy. Not all users will be tech savvy and mistakes will be made. As a result, patients should expect to interact with AI-equipped applications and devices to include:
- Billing-AI automates the medical coding process and improves accuracy and reduces fraud.
- Claims Management – AI conducts lifecycle reviews to detect anomalies.
- Drug Interactions- AI predicts side effects and offer alternatives.
- History-AI aggregates medical records, scans images and transcribes conversations.
- Nutrition- AI monitors a person’s diet and recommends changes.
- Monitoring-AI monitors patient’s vital sings, their prescriptions and diagnoses conditions.
- Schedule –Al schedules appointments.
Applications that are currently available include:
Florence — A digital healthcare assistant that is accessible on Facebook Messenger KIK and Skype. To track a person’s health, prescribed medications and schedule appointments.
Your. MD — A digital healthcare assistant that is accessible on Android, Facebook Messenger, KIK, iOS, Slack and Telegram. To act as a personal health directory (medical providers, pharmacies, clinics) and to diagnose symptoms then schedule appointments.
Safedrugbot — A digital healthcare assistant accessible on the Telegram app messenger service. To assist medical personnel about certain prescribed drugs and their potential side effects during pregnancy and nursing.
Government is interested.
People organize their lives around their mobility. Governments are attempting to combine this mobility with chatbots. To better manage finite resources. The Canadian Province of British Columbia and the United Kingdom’s National Health Service (NHS) are deploying Babylon (Telus Health) triage chatbot for symptom recognition and video consults. (https://www.babylonhealth.com).
Before we can expect healthcare chatbots and similar AI-based technologies to become as common as their household equivalents, the healthcare context poses a number of specific challenges:
AI software interacts with many databases, systems and interfaces. The probability for exchanging patients PII could violate Health Information Portability and Accountability Act (HIPPA) rules.
This technological innovation lacks in-depth healthcare experiences. Chatbots require platform API integration, hosting and monitoring. What happens if integration, hosting or certain software fails?
Patients should expect extra- verification steps (human intervention) to remove mistakes or even patient complacency.
The healthcare profession generates Big Data from multiple sources. Determining the best data (scope of work) strategy is vital.
Patients should expect to use chabots for specific tasks (fraud detection) with more capability added (e.g. claims management) overtime.
AI connects to many data sources that could implicate PII. Thus, AI-chatbot platforms may become a center of gravity for cyber attack.
Patients should expect online user training to help guard them against hacks, manipulations and other threats.
Patients should expect the following:
Patient-chatbot interactions will be limited to patient education, engagement and routine tasks.
Home-based digital assistants will gradually offer more personalized medical services.
What Do You Think?
I am the author of this article and it expresses my own opinions. I have no vested interest in any of the products, firms or institutions mentioned in this post. Nor does the Analyst Syndicate. This is not a sponsored post.