Its Drivers and Consequences for Work and WorkersRead the full report
A joint report with the UC Berkeley Labor Center, written by Adam Seth Litwin, associate professor of industrial and labor relations at the ILR School at Cornell University.
Unprecedented challenges confront the U.S. health care sector. The spread of COVID-19 has exposed and amplified many of the underlying weaknesses of the U.S. system, including a lack of widespread access to care, inefficiencies, and runaway costs. The U.S. health care industry also lags behind other domestic industries and other countries’ health care systems in terms of technology adoption.
There exists an untapped potential for new technology to propel changes in health care delivery; COVID-19 may speed up some of these changes, such as the use of telemedicine and autonomous robots in hospitals. While some new technologies may be able to improve outcomes both for patients and for health care workers, this is not a foregone conclusion. The consequences of technological change in health care will depend on the choices policymakers and industry stakeholders make in this current moment of crisis and in the future.
This report examines the drivers of technological change in the U.S. health care industry and explores how technologies may be used in response to the challenges facing the industry over the next five to 10 years. We also assess how technological change in health care may affect health care workers, who represent 12% of total employment in the United States—around 18 million workers. As recent events have emphasized, workers throughout the health care industry—whether janitors, nursing assistants, registered nurses, or doctors—are essential to the functioning of our society and economy. Women and people of color make up a greater share of workers in health care than in the economy at large, and many of the sector’s front-line workers have not completed a college degree.
We asked the following three questions in our research:
Our research involved interviews with hospital and home health agency administrators, union representatives, health care IT experts and consultants, and technology developers. We also attended health care conferences and trade shows, and analyzed government-collected labor market data. We conducted 32 interviews overall, either in person or via web or telephone, between April 2018 and June 2019.
Our interviews showed there are four objectives guiding the health care industry in the United States; these establish the conditions and motivations for technological change:
Ideally, technology can facilitate providers’ ability to offer care to more people. Subsumed within this objective is the goal of using technology to reduce the unit cost of care delivery.
Consolidating care allows providers to serve more patients with a broader range of services while helping to reduce costs through economies of scale and scope. It may necessitate increased reliance on new technologies for managing patient flow and coordinating care delivery.
A shift toward “value-based care” (away from fee-for-service care) has given providers a financial incentive to keep patients healthy and better manage their chronic conditions. This calls for technologies that help monitor and nudge patients and facilitate regular communication with their providers.
People are living longer, increasing the prevalence of conditions that will require long-term care. Providers will turn to technology that responds to the increasing demand for long-term care, in particular, home care.
Our research identified three types of emerging technologies most aligned to the health care sector’s guiding objectives:
In simplest terms, this category includes any smartphone or internet-connected computer. However, digital communications technologies have a broad range of applications, including in the home care setting and in the virtual provision of patient care, e.g., telehealth, telemedicine, and telehospitals. Digital technologies have aided the transition from paper-based to electronic health records and allowed for richer, more data-dependent ways of leveraging interconnected health records.
While humanoid, caregiving machines largely remain the province of fiction, a simpler form of service robot already traverses hospital hallways. These robots accept external commands from users and can maneuver and operate on their own by taking in, processing, and reacting to information absorbed through sensors. They can pick up soiled sheets and dirty dishes, and they can deliver meals and medications, among many other tasks.
The use of artificial intelligence (AI) in health care has only just begun. AI differs from other technologies in its ability to “teach” itself: rather than following predetermined, detailed directions provided by programmers, one particular form of AI—machine learning—allows technology to develop its own rules and responses once “trained” by existing data. AI can essentially supercharge existing digital technologies, including those allowing for virtual care delivery.
The consequences of these technologies for work and workers depend on how they are put to use towards addressing the objectives guiding the health care industry. For example, regulation has promoted the use of electronic visit verification (EVV), which monitors direct care workers through a smartphone. This technology has facilitated documentation but it has also increased micro-management of workers. However, similar technology could be used by the same workers in a very different manner, potentially empowering them to serve as a patient’s point person for the entire team of providers contributing to their care plan.
For example, service robots and other AI applications could be used simply to trim the workforce and justify more limited activities and pay for workers. Or, they could be used in such a way that they free up time for these workers to focus on other activities, in particular those that depend on skills at which humans excel compared to robots, such as empathy and communication. The ways in which new technologies are deployed hinges on the choices that we make about the future direction of the industry.
Increasing access to health care while improving the quality of that care and containing its costs are common goals across the sector. How to get there remains a subject for debate. The federal government has an outsized stake as the funder of Medicare and Medicaid, giving it leverage and buying power. Workers enjoy stable jobs and long-term careers, especially women, people of color, and those with little or no formal education beyond high school. And as stated previously, unions and their members are well-represented in the health care sector; they have played a key role in preserving and improving wages and working conditions.
New technologies, if thoughtfully deployed also may improve the performance of the health care industry. To that end, our research identified three specific choice points regarding new technologies and how they are deployed:
Provider organizations will only invest in a particular technology if they think it makes sense financially. Hastening the shift toward value-based care (from fee-for-service care) will likely accelerate the adoption and diffusion of quality-enhancing health care technologies. Likewise, Medicare and Medicaid account for a substantial share of our national health care bill. If policymakers think telehealth will improve access while containing costs, then reimbursement rules can be tweaked to ensure virtual visits are adequately reimbursed. This has happened, albeit on a temporary basis, in response to the COVID-19 pandemic.
Careful experimentation will be necessary to assess the effectiveness and quality of care delivered virtually. In addition to adjusting reimbursement rules, this may require adapting state-based medical licensing and harmonizing scope-of-practice rules to allow responsibility and accountability across a wider range of health care professionals. Policymakers and managers should assess the effectiveness of technologies intended to empower front-line workers with better health care information and patient data, to enhance the role of frontline workers in patient care.
By default, most employers focus on the amount and variety the technology under consideration can do, then give it as many tasks as it can manage, leaving remaining tasks for workers. Using a work-centered approach to new technology begins instead by asking, “What are people—RNs, direct care workers, etc.—really good at, and how might technology best exploit these strengths?” This approach respects human dignity as well as the constraints imposed by economics and technology: human labor will long remain part of the health care delivery process, will be in shorter supply, and cannot be altogether supplanted by new technology. Addressing worker shortages will require that workers be paid more and that they take on new roles, and it will require investments in upskilling.
Given the challenges facing the U.S. health care industry and the opportunities that new technologies present, the choices we make now can lead us in one of two directions. Policymakers and industry leaders can choose a high-road path, in which the benefits of technological change are shared among patients, providers, and health care workers alike. Or, they can continue along the default trajectory, in which technology is deployed primarily to increase returns for atomized actors, and to reduce staffing and increase micromanagement of workers. Taking the high road will require coordinated efforts to improve industry outcomes that involve a voice for all stakeholders—in particular, health care workers.
Our findings suggest that technological advances in health care can be used to help build the high road if they are deployed in specific ways: toward the fulfillment of value-based care; under the auspices of policymakers and managers open to experimentation; and via the adoption of a work-centered approach.
Along the high road, adapting payment rules and embracing experimentation would increase options for the application of telehealth technologies, which could create new avenues for patient engagement and new career opportunities. Likewise, digital communications in the form of EVV hardware and software would be left behind in favor of more empowering uses for this same technology. Under an augmented home health model, an aide would take on the role of care coordinator for their client—using the smartphone not simply for clocking in and clocking out, but for connecting themselves to the rest of the care team. Furthermore, he or she would be trained (and compensated) to leverage their proximity to the client, serving as a two-way information conduit and front-line care coordinator.
Where do semi-autonomous service robots figure along the high road? We expect hospitals to continue using these technologies, though we see big differences in how they are deployed under the two scenarios. Along the default path, employers use robots to relieve themselves of labor obligations and to de-skill workers’ jobs. Under the work-centered approach that characterizes the high road, employers instead consider how robots could assume some of the less enjoyable, lower valued-added tasks for which workers have long been responsible, freeing those workers (such as dietary clerks or orderlies) to enhance their roles and to provide compassionate care as only humans can. It would have the additional benefit of bolstering patient perceptions of genuine empathy, which could also be a boon to hospitals’ performance metrics.
Artificial intelligence permeates many existing technologies, including autonomous robots and chatbots—with more applications coming online daily. However, potential applications of AI and machine learning are seemingly boundless—and at this stage, largely speculative. It is up to us to imagine applications that would fit within a high-road vision for the future of the health care industry. For instance, in the future, AI such as clinical decision support (CDS) could equip the next generation of caregivers to fill a new, highly trained and well-compensated role interacting with and examining patients while interfacing with a standardized but self-evolving diagnostic and treatment system powered by AI and machine learning (ML). In its initial incarnation, the machine would sit physically in the exam room alongside the practitioner and the patient. Later on, the machine could instead be used by teleproviders delivering care remotely. Aside from supporting efforts to optimize for cost, access, and quality of care, this type of AI/ML deployment could generate job opportunities in an entirely new category of highly-trained and well-compensated medical professional.
Our nation’s health care sector has a history of underperformance in the areas of access, quality, and cost. The COVID-19 pandemic has further exposed the frailty and ineffectiveness of the system, and pointed to the need to leverage technology toward more efficient use of the health care workforce. Technology can play an important role in moving the nation toward the health care high road, particularly if we are thoughtful in how and to what ends it is deployed. Our research suggests greater use of technology in a work-centered approach could not only improve industry performance for patients and providers, but could also improve job quality and career prospects for health care workers. We submit that getting there will require a bold change of direction.
That said, early indications are that telehealth and semi-autonomous robots have both played key roles in the system’s and policymakers’ response to the pandemic. The use of telehealth, in particular, was actually facilitated by a direct but thus far temporary policy pronouncement by the Trump administration to commit to Medicare and Medicaid reimbursement for such services. It remains to be seen how permanent pandemic-related sectoral changes will become.
The unique manner in which the United States delivers and finances health care seems to provide a guarantee that present market forces will not beget solutions to leverage the use of new technologies for improvements in industry performance and worker wellbeing. Without careful, coordinated decision-making, technological choices are likely to undermine workers’ job quality and their ability to exercise their voice at work. The effect of this will be to limit the possible ways in which technology could be used to improve outcomes for patients, providers, frontline workers, taxpayers, and society at large.
This report is part of a larger, multi-industry project generously supported by the Ford Foundation, the W.K. Kellogg Foundation and the Open Society Foundations. This project received additional support from the SEIU California State Council and a Cornell University ILR School “Technology and the Future of Work” theme project grant.