Robots, AI, and cognitive analysis
How technology is disrupting hospitals around the globe
Authors : Dr. Terri Cooper & Dr. Stephanie Allen
A view from Deloitte Center for Health Solutions
When we look back at hospitals 20, or even 10 years from now, we might wonder how we accomplished so much without robots, artificial intelligence (AI), avatars, or cognitive analysis. We will likely marvel at how hospital clinicians were able to make complex decisions about our care while also completing an unending list of administrative tasks. We might ask, how did our bodies heal in those drab patient rooms, under bright and buzzing incandescent lights, surrounded by machines and intercoms that beeped, bonged, and blared around the clock?
Whether in an industrialized or developing country, and regardless of the health care system’s structure, hospital leaders, their employees, and their patients struggle with many of the same issues. Cost containment might be at the top of the list among hospitals globally. In the US, for example, health care makes up about 18 percent of the gross domestic product (GDP).1 Although government-run health care systems in Australia, Canada, and the United Kingdom spend far less – about 10 percent of GDP – health care won’t be affordable if costs continue to rise along the same trajectory. Many emerging cognitive technologies promise to bend this cost curve and transform the journey of care significantly.
A lack of connectivity is another issue commonly shared by public and private health care systems worldwide. While the use of electronic health records (EHRs) in the US has grown since the enactment of the Affordable Care Act (ACA), a lack of integration can make it difficult for hospitals and other providers to easily share patient information, such as test results and diagnoses. While Australia, Canada, and the UK are considered to have more integrated health care systems by some, they haven’t yet implemented universal EHRs, and their hospitals continue to rely on the transmission of paper based-records where electronic records don’t exist.
Requiring patients to repeat tests and complete identical paperwork not only can create a negative hospital experience for them, it often piles unnecessary work onto nurses, clinicians, and other hospital employees. But we can be optimistic that such scenarios can be reduced or even eliminated as hospitals take advantage of emerging technologies and seek new and better ways to engage with consumers. In a new report, the Deloitte Center for Health Solutions explores the digital hospital of the future.
A growing number of inpatient health care services are already being pushed to the home or to outpatient ambulatory facilities, which helps reduce patient volume. But patients with complex conditions or advanced illnesses will likely continue to need acute inpatient services.
Technology and AI will help enhance decision making
Clinical workforce shortages are a significant issue among hospitals in the US and across Europe. Nurses and other clinicians tend to spend too much time completing administrative tasks and chasing down test results. Nurses devote as much as a quarter of their time on non-patient care.2 We know that when people are overloaded, mistakes can happen.
Cognitive technologies have the potential to not only relieve some of the administrative burden, but also to support and augment the clinical workforce. Some hospitals are already using robots to transport linens, meals, specimens, and to collect diagnostic results – tasks typically performed by clinical staff. Case in point: The South Glasgow University Hospital in Scotland has 26 robots that move medical equipment, linens, food, and waste. These robots have their own underground tunnel (through which they transport supplies) and a dedicated elevator.
In the US, drug dispensary automation has helped improve patient safety by eliminating some decision making, but it isn’t used as much in other parts of the world.3 Hospitals in some countries still keep medications in drawers, which can lead to mis-doses or overdoses because fluids and injectable drugs tend to look similar.
While robotic process automation (RPA) is offloading some duties from the shoulders of caregivers, combining it with AI can take this a giant step further by helping doctors and nurses make appropriate care decisions much more quickly. Let’s use cancer as an example. There are numerous types of cancer, endless variables, and a wide range of treatment options. An oncologist often has to rely on personal experience, the experience of colleagues, and maybe the hospital’s database when deciding the best course of treatment. This can be a time-consuming process…and time is everything for a cancer patient. AI can sift through mountains of data, including studies related to a patient’s tumor, and offer the oncologist well-researched treatment options. Automating this process can give the doctor and patient more time to discuss the diagnosis (based on AI research and the doctor’s own experience) and treatment plan.
Expect a patient-friendly atmosphere and more cognitive workers
Physically, hospitals of the future will likely look different than they do today. They will likely include attractive visitor lounges, dayrooms, and views of natural or green surroundings, such as healing gardens that help reduce patient anxiety and expedite healing. Adjustable ambient lighting will likely be designed to improve the patient experience, particularly in terms of both mood and the experience of pain. These facilities may also be quieter to reduce stress and to help ensure patients get needed rest. Noiseless alarms on medical devices, soundscapes, noiseless paging, and health acoustic engineering will help minimize ambient noise levels.
Patient rooms may look different, too. The walls might be prepopulated with pictures of the patient’s family members, or photos from a recent trip. Along with such “picture walls,” a patient could create a customized music playlist, make video calls with friends and family, or access the Internet. The bathrooms could have integrated sensors to monitor any unusual activities, such as falls, which would trigger a call for immediate help.
The hospital workforce of the future will likely also be different, and could include the following:
- Cognitive workers: Some hospitals have started to supplement their back-office staff with “cognitive workers” that can complete routine tasks. These virtual employees are easy to train and can regularly be updated with new skills by outside vendors. Moreover, cognitive workers don’t take vacations, don’t have emotions, and don’t get distracted. Human workers of the future will likely need different skills and will take on roles that do not yet exist. The hospital workforce of the future could be made up of augmented teams of humans and cognitive workers.
- Avatar counselors: In Australia, avatars are being used in some hospitals to provide health counseling to patients. One virtual caregiver was used in a closed trial of diabetes management. Avatars also can help patients deal with depression and anxiety issues. These cognitive counselors can understand natural language patterns, including sarcasm, and can interpret facial expressions, such as sadness, agitation, and even dishonesty. Clinical trials indicate these avatars provide a consistent and quality interaction, and deliver better diagnosis and outcomes, particularly in relation to mental health and sexual health counseling where the threat of human judgment can sometimes be impeded by the full disclosure needed.
- Discharge planning robots: Practitioners from the health care and social services industry groups from several Deloitte member firms are building an RPA and AI discharge planning “bot” that initiates and coordinates concurrent activities including the discharge summary, prescriptions, transportation, referrals to home support services, and the scheduling of follow-up visits for specialists or occupational therapy. Existing discharge management can be susceptible to problems. An incomplete process, for example, could extend a patient’s length of stay by one or two days.
- Health care companions: Pillo, an AI health care companion that can assist patients in the home, uses the latest in voice and facial recognition technologies to hear, see, and understand natural language. This enables it to adapt the functionalities to serve consumer-specific needs. It can answer health and wellness questions, connect directly with health care professionals, and securely manage vitamins and medication – storing, dispensing, and even ordering refills when people need them.
Rapidly-evolving technologies and growing consumerism, along with demographic and economic changes, are already having a dramatic impact on hospitals. Several megatrends could have significant implications on how hospitals of the future are staffed, sized, and designed. Around the world, hospital executives will likely want the vision for their hospitals to address these megatrends and plan for investments in people, processes, and premises. Ten or 20 years from now, we will likely wonder how we ever got along without cognitive analysis, AI, and other emerging technologies.