It's 2018, and Alexa has become a common household name. The iPhone has turned 10 and roughly two-thirds of adults worldwide own a smartphone. About 84 percent of the global population lives in an area where mobile broadband access is available. At the same time, genetic testing has become more affordable and accessible to consumers, with direct-to-consumer genetic testing pioneer, 23andMe, now boasting five million users in over 50 countries.
As the interest in and demand for genetic testing has increased, the demand for genetic counsellors – experts trained to help patients understand and make use of genetic information – is increasing. There are currently over 4800 certified genetic counsellors in North America. Clear Genetics, a healthcare technology company based in San Francisco, has collaborated with genetic counsellors working in a variety of patient-care settings to develop a chatbot named GIA (Genetic Information Assistant). GIA is a clinical-grade chatbot who assists patients pursuing genetic counselling, risk assessment and testing.
For those of you wondering what a bot, or chatbot, is – these are software-based simulated conversation tools. Chatbots use artificial intelligence and natural language processing to answer simple questions, increase and maintain consumer engagement, promote services, and provide convenient, easy access between consumers and service providers. Bots often handle repetitive questions from users, such as, 'Is my flight on time?' In the field of genetics, they can respond to questions like, 'Should my children also be tested?'
One American health entity using chatbot technology is Geisinger, a non-profit integrated health system serving over three million residents in Pennsylvania and southern New Jersey. Geisinger and Clear Genetics have collaborated to develop chatbots for communication with patients enrolled in the MyCode Community Health Initiative. MyCode is a large research biobank returning genetic test results for genes known to be associated with an increased risk for treatable and preventable heritable heart diseases and cancers.
Over 210,000 Geisinger patients are enrolled in MyCode, with at least 2 percent expected to have a disease-causing genetic variant for which there are potential treatment options. With so many patients enrolled, automation of some repetitive, scriptable communications would free up staff members and genetic counsellors for tasks that require human interaction.
To recruit patients into the MyCode study, Geisinger and Clear Genetics developed a consent chatbot that walks patients through components of the consent form, allowing them to opt to receive more or less detail on key topics (goals, benefits, risks, and so on). The tool is compliant with the Health Insurance Portability and Accountability Act 1996, and can be used on any mobile device. It is designed to record the decision (consent, considering, decline) in the patient's electronic health record. Geisinger patients who tested the consent chatbot in focus groups reacted positively, favoring the more informative and customisable chatbot experience to an in-person approach.
In addition to the consent chatbot, a patient 'follow-up' chatbot was developed to remind MyCode participants of suggested actions after they got their results. Via the follow-up chatbot, MyCode participants could indicate whether they had received a result packet, gathered their family history data, met with a provider to review results, and shared results with relatives. Patients could also request a genetic counselling visit and type in questions.
To help patients communicate their genetic test results with at-risk family members, a family sharing tool was developed as an easy mechanism to electronically share results and inform family members of the importance of genetic counselling and testing. The tool allows the patient to send their relatives a link to a 'cascade' genetic counselling chatbot by text message, email or Facebook messenger. The cascade chatbot describes the patient's result, associated disease risks and recommended management.
The patient's at-risk relative can request a genetic counselling visit via the chatbot and ask questions. Geisinger patients who have interacted with the family sharing tool and cascade chatbot report that they have enjoyed the ability to preview what their relative will see and welcome a tool that 'has the detailed information' and 'answers that wouldn't be on the tip of the tongue' when talking with family members about genetic test results.
Beyond the consent, follow up and cascade use cases described here, a chatbot like Clear Genetics' GIA, can free up genetic counsellors by automating many other routine tasks, such as delivering negative results and providing standard pre-test education.
Sceptics may lament that 'the bots are taking over'. But the goal is not to replace genetic counsellors with chatbots. Rather, the goal is to delegate important tasks that are repetitive, time-consuming and scriptable to a chatbot, like GIA, to leave genetic counsellors more time for more valuable and highly skilled patient care. More counsellor time will lead to greater accessibility of genetic counselling services. To that we enthusiastically say, bring on the bots.