Safe and Effective Communication to Prevent Diagnostic Errors
Diagnostic errors (i.e., diagnoses that are delayed, wrong, or missed) are increasingly recognized as a patient safety concern in ambulatory care (Singh & Graber, 2010). A recent report from the American Medical Association (AMA) Center for Patient Safety (Lorincz et al., 2011) highlighted the importance of diagnostic error and the critical need for future research on this topic. Although multiple factors influence the diagnostic process, communication is a central theme (Singh, Thomas, Khan, & Petersen, 2007; Singh et al., 2010b). From the initial patient-provider encounter to confirmation of a diagnosis through diagnostic testing, procedures, or subspecialty referrals, good communication is essential to timely and accurate diagnosis. Accordingly, communication breakdowns are emerging as a leading preventable cause of diagnostic errors and are the focus of our recent work. In this article, I discuss some of the early lessons and challenges in this work and share some recent developments in the field of diagnostic errors.
Most ambulatory malpractice claims data suggest that diagnostic errors are the largest category among U.S. malpractice claims (Singh & Weingart, 2009). Outpatient diagnostic errors may not necessarily involve only rare diseases or unusual disease presentations, (Singh et al., 2009b) but also relatively common conditions such as cancer, ischemic heart disease, and infection (Singh et al., 2009a; Singh et al., 2010a; Singh, 2013b). Many such errors involve communication breakdowns, which are at times complex and difficult to define. It is not surprising that those breakdowns occur—ambulatory care involves several settings of care and is longitudinal in nature, making it increasingly chaotic for information processing (Sarkar et al., 2012).
Communication and the Diagnostic Process: A Hard Nut to Crack
Communication challenges are virtually a given in ambulatory care settings, where barriers include time and workload pressures on busy clinicians, the sheer volume of both verbal and electronic communication among providers, and several patient factors that affect information transfer (Singh & Weingart, 2009). Identifying the point(s) at which critical communication breakdowns occur is a first step in understanding the origins of error.
It is important to recognize that, in healthcare settings, communication is often intended not only to transmit information but also to elicit some response from the recipient. For instance, when providers receive notification of abnormal test results, they might order follow-up diagnostic tests, notify patients, or refer to subspecialists. Thus, the desired outcomes of communication can be viewed in steps: message transmission (sending accurate, complete, and unambiguous information); message reception (perceiving the information accurately and taking appropriate next steps); and message acknowledgment (providing feedback that the message has been received and/or acted upon) (Singh, Naik, Rao R, & Petersen, 2008). Pinpointing the weakest links in those steps can help prioritize interventions.
Using a process-based conceptual framework to advance patient safety in this complex area is also important. Our work has shown that errors in the diagnostic process span five interactive dimensions, each of which is closely related to one or more aspects of communication (Singh, 2013b).
- Patient-provider encounter: Problems with history, physical exam, or ordering diagnostic tests for further work-up.
- Diagnostic tests: Problems with ordered tests either not performed or performed/interpreted incorrectly.
- Follow-up and tracking: Problems with follow-up of abnormal diagnostic test results or scheduling of follow-up visits.
- Referrals: Lack of appropriate actions on requested consultation or communication breakdown between consultant and referring provider.
- Patient factors: Delay in seeking care or non-adherence to appointments or advice.
Our recent work on a study of diagnostic errors in primary care has shown that process breakdowns related to the patient-provider encounters were involved in 80% of the errors (Singh, 2013b).
Integrated electronic health records (EHRs) readily address certain problems that are endemic to paper-based record systems, such as illegible handwriting, misplaced documents, and distance barriers between providers. However, the EHR must resolve communication problems that might contribute to errors in any of the five interactive dimensions. Meanwhile, researchers must be vigilant for problems that are uncovered or introduced by new technologies. For instance, clinical decision support (CDS) interventions in the EHR can enhance communication by prompting important questions or actions during the diagnostic work-up. However, to avoid overlooking important information, we must also ensure that CDS interventions are manageable and fit into providers’ workflow (Singh, 2013a).
Similarly, the EHR eliminates the need for a physical “paper trail” and replaces it with centralized, easily accessible referral requests. However, our work suggests that there are several remaining vulnerabilities that must be addressed to prevent patients from being lost to follow-up (Singh et al., 2011; Singh & Weingart, 2009). Breakdowns in communication can lead to patient “no-shows” for appointments, tests, or procedures, and subsequent delays in diagnosis (Singh et al., 2011).
Multidisciplinary interventions are needed to ensure that communication processes reflect our best knowledge of human cognition, human-computer interface design, and patient and provider behavior in complex care settings (Sittig & Singh, 2010). Incorporating this knowledge into systems, policies, and procedures can enhance communication while avoiding the pitfalls of simply introducing more technology to solve the problem (Sittig & Singh, 2010). Our ongoing work, for instance, applies a multifaceted approach to improving EHR-based communication, with the ultimate goal of reducing missed and delayed diagnoses (Singh et al., 2012b).
Finally, advances in provider-patient communication are also needed. Personal health records and secure messaging to improve patient-provider communication are a few such innovations that merit further study. Bringing patients into the communication loop is a potentially powerful but underdeveloped strategy to help ensure the quality and safety of care in the outpatient setting (Singh et al., 2012a).
The field of diagnostic error has now begun to gain attention from several key stakeholders. For the past six years, the AHRQ-sponsored Diagnostic Error in Medicine (DEM) conference has enabled continued development and collaboration in the nascent field of diagnostic error. The DEM conferences have offered an important resource and satisfy an otherwise unmet need to bring together researchers, practitioners, and noted speakers and experts to focus on issues that lie at the intersection of diagnostic error and fields such as clinical informatics, cognitive psychology, and human factors. The conference has begun to create an impact in the areas of research, education, and clinical care.
One of the most significant developments of the DEM conferences is the formation of the Society to Improve Diagnosis in Medicine (SIDM; www.improvediagnosis.org), a new, independent organization dedicated to research, education, and awareness about diagnostic error in medicine. SIDM was conceived at the 2010 DEM conference and subsequently organized as an independent entity to spearhead the diagnostic error movement. One of SIDM’s overarching goals is to define interventions that reduce diagnostic error. SIDM members are now actively engaged in building relationships with diverse key stakeholders, including patients.
In sum, communication breakdowns are preventable and often involved in diagnostic errors. Research in this area is promoting a better understanding of errors, identifying specific areas of vulnerability, and fostering the creation of novel intervention strategies to improve the quality of diagnosis.
Hardeep Singh is chief of the Health Policy, Quality and Informatics Program at the Houston VA Health Services Research and Development Center of Excellence at the Michael E. DeBakey Veterans Affairs Medical Center, and assistant professor at Baylor College of Medicine. He may be contacted at firstname.lastname@example.org.
Note: This article is updated and adapted from “Safe and Effective Communication to Prevent Diagnostic Errors,” by Hardeep Singh, MD, MPH. Veterans Affairs Health Services Research and Development Forum: Research Highlights, August 2010.