In fast-paced environments with high stakes, emergency physicians have to quickly make the best decisions possible with imperfect or incomplete information. Under the circumstances, this information asymmetry often forces physicians to rely on their heuristics to diagnose and treat their patients. While this approach consistently saves lives, this reliance on heuristics/implicit bias—we now know—can be problematic. The asymmetry of information in these settings can allow implicit bias to creep in—even for well-meaning care teams that may not consciously subscribe to those biases—and ultimately affect patient care.
The good news is technology can play a role here. Implementing processes and solutions that surface insights into care team workflows upfront can help reduce the need for relying on surface-level information and disarm unconscious biases for better patient outcomes.
Step one: identifying what we’re missing—and why that matters
Care teams are trained to ask, “what am I missing?” when diagnosing a patient in an Emergency situation. However, with limited factual information to rely on, outcomes data shows that race, ethnicity and gender—often apparent in the patient standing in front of you—can impact next steps rather than other factors that might be more pressing. This is where technology can fill in the gap. If we can surface key information that is not actually visible, we are relying on fact rather than unconscious bias instinct to fill in the missing pieces.
Underlying social health factors—like, zip code, food security, access to housing, income, etc.—contribute to a patient’s overall health and are as important as physical symptoms when diagnosing a patient. When we use race/ethnicity to proxy for these underlying factors, we are doing a disservice to patients and activating bias that can often be inappropriate, leading to outcomes that are far from ideal. This proxy is what may lead a provider to deny pain medication to a sickle cell patient in chronic pain based on an implicit bias that may tie certain ethnicities to drug seeking behavior. Ideally, factors like the history of a pertinent illness (in this case sickle cell) are surfaced to a provider in the same form—and with the same importance—as all the other pieces of information that help them become familiar with and diagnose a patient.
While surfacing more data would help question implicit biases, it is not a silver bullet. As articles such as this one from the NYTimes, reveal underlying medical protocols may contain more inherent bias than we know. If we are committed to overhauling gender and racial bias in treatment, the “what am I missing?” must extend beyond the individual doctor to the industry and underlying protocols that care teams have been taught for generations.
Step two: learning how to recognize and address hidden determinants to health
Not all patients will be forthcoming in disclosing other, more sensitive, situations like homelessness, food insecurity, substance use disorder, or dangerous living conditions. Understanding that by disclosing this information, patients and their families could face consequences (like losing parental rights or even incarceration), the burden of finding these hidden contributors falls on the care teams that help these individuals.
For example, if a 34-year-old middle-class mother of three presents at the ED complaining of lower back pain, there may be no immediate cause for alarm. But if a deeper dive into the woman’s prescription history reveals that she has presented at four other EDs complaining of lower back pain in the past 90 days, the solution changes. Rather than prescribing a pain medication, the physician can work with a case manager to connect her with community resources that help her address the root of her complaint and potential misuse of substances.
Homelessness is another hidden contributor to health that can put a strain on both the patients and the healthcare system that serves them. As these transient patients move between hospitals, it may be difficult to tell upfront that anything is wrong. In an attempt to better identify and care for homeless populations, California passed a bill that requires ED physicians to dig deeper with patients with suspected homelessness and work to identify an appropriate post-discharge location and needed community behavioral health resources. Using existing technology to help flag, track, and coordinate care for these patients—even as they travel—has become critical to not only meeting the requirements outlined in this bill, but improving health outcomes for the homeless population by delivering key insights into provider workflows.
Step three: working together to make things better
An analysis from the Kaiser Family Foundation estimates that disparities produce $93 billion in excess medical care costs and social health factors‚—including education, segregation, poverty, etc.— account for over one-third of total deaths in the United States in a year. Taking on these disparities alone can feel overwhelming, but in the age of technology, we can work together to help source information about each patient as an overall human being—not just their physical symptoms, their DOB and their SSN.
By utilizing health IT that aggregates and automatically flags the most important pieces of clinically actionable patient information, providers can combine what tangible physical symptoms they observe with insights into social health conditions they can’t necessarily see.
A tool (like the Collective platform) can quickly identify, and in many cases predict, if a patient might be struggling with homelessness, housing insecurity or a substance use or behavioral health disorder and alert care teams of this information in real-time, making it easier to connect patients to a better care setting or community resources. It can also surface information about other interactions a patient has had with the healthcare system. By integrating a more complete set of data points into existing workflows, making informed decisions—even in a fast-paced ED—becomes easier, allowing providers to focus more of their time delivering quality care to patients while accommodating (and sometimes helping) their social health, too.
For true change to happen, all the different stakeholders in healthcare—providers, technology, and community partners—need to fully acknowledge and start working together to close information gaps. The sooner we can come to the table together and create an open dialogue about biases and information gaps, the more likely we are to actually make changes that move us in the right direction.
Related Reading: Understanding and Addressing Physician Suicide
Mandira Singh is the chief operating officer of Collective Medical.