The top seven analytics-driven approaches for reducing diagnostic error and improving patient safety


Diagnostic errors, including missed diagnoses and misdiagnoses, are an increasingly recognized concern within healthcare. In a 2015 New England Journal of Medicine-published report, The National Academy of Medicine (NAM)—formerly known as the Institute of Medicine—highlighted the industry's rising worry over diagnostic errors and their clinical and financial impact.

As the authors of the report note, "diagnostic errors are clinically and financially costlier today than ever before." Consequently, the industry must dedicate more attention and resources to the problem and increase efforts to identify, monitor and reduce errors.

Back to Basics: Defining Diagnostic Error

NAM defines diagnostic error as "the failure to (a) establish an accurate and timely explanation of the patient's health problem(s), or, (b) communicate that explanation to the patient." A diagnosis is considered "wrong" if it's inaccurate; if it is incomplete and either doesn't represent the patient's true condition or doesn't reveal enough detail for optimal treatment; or, if it is determined too late to guide effective treatment decisions.

The Society to Improve Diagnosis in Medicine (SIDM) categorizes the three different types of diagnostic errors as follows:

Missed diagnosis - when diagnostic tests fail to provide an explanation for a patient's complaints, as is common with patients suffering from chronic fatigue or chronic pain.
Wrong diagnosis - when the original diagnosis is found to be incorrect because the true cause is discovered later, e.g., when a patient is initially diagnosed with acid indigestion but later diagnosed with a heart attack.
Delayed diagnosis: when the diagnosis should have been made earlier.

The SIDM also contends that it's a misconception to associate diagnostic errors with sub-optimal care. On the contrary, the SIDM states that "most diagnostic errors are made by conscientious clinicians practicing in first-rate medical organizations." The SIDM stresses that diagnostic errors are more likely the consequence of the complexity of diagnosis and healthcare delivery, as well as basic human mistakes.

The Consequences of Diagnostic Error

Many health systems fail to take measures to reduce diagnostic error because they're unaware of the potential severity of errors. However, recent research demonstrates that diagnostic errors occur frequently and can have a tremendous impact on patient safety and the cost of care.

Patient Safety Consequences

Diagnostic errors account for 17 percent of preventable deaths in hospital patients, according to a Harvard Medical Practice study. In addition, 9 percent of patients experience a major diagnostic error that is undetected in their lifetime, according to a systematic review of autopsy studies over four decades.

Diagnostic errors are linked to thousands of deaths a year. One multicenter review of 669 cases from 22 institutions revealed that 28 percent of reported diagnostic errors were life-threatening or resulted in patient death or permanent disability. The same analysis of malpractice claims found that diagnostic errors led to death more often than other allegation groups and were the leading cause of claims-associated death and disability.

Communication failures also contribute to diagnostic errors and impact patient outcomes. For a diagnosis to be useful to a patient, the provider much explain it in a way that helps the patient make appropriate treatment decisions or lifestyle choices. Providers must clearly communicate why a particular test is critical, why it must be done in a timely fashion, and what the results indicate about the patient's health.

Financial Consequences

Diagnostic error can be expensive, especially when considering the costs associated with permanent disabilities, unnecessary testing, lost productivity, and increased insurance payouts. A 1986–2010 analysis of malpractice claims, for example, assigned an inflation-adjusted cost of $38.8 billion (or an average of $386,849 per claim) for diagnosis-related claim payments. Over 28 percent of all the malpractice claims were tied to diagnostic errors, as were 35 percent of all payments.

Where to Start: The Top Seven Analytics-Driven Approaches for Reducing Diagnostic Error

One essential strategy for reducing diagnostic error is to utilize advanced healthcare analytics. Healthcare analytics provide a source of truth about which diagnostic procedures are effective in which circumstances, and which procedures are less likely to improve outcomes. By reviewing healthcare data, organizations can quickly glean insights about where to begin making changes that reduce diagnostic error.

When working to minimize diagnostic error and improve patient safety, health systems should consider the following seven analytics-driven approaches:

Approach #1: Use Key Process Analysis (KPA) to Target Improvement Areas

Use an 80/20 Pareto analysis to target improvement areas, such as cost-driving clinical segments and varietal care processes. Combine both clinical and financial data to highlight the best opportunities for improvement and for cost reduction, and to guide the development of applications to support improvement initiatives. The analysis should reveal areas with the greatest cost and volume, and show the degree of variation within subpopulations and within select clinical programs.

Approach #2: Always Consider Delayed Diagnosis

Keeping in mind that the cause of delayed treatment is often delayed diagnosis, organizations should consider time-sensitive conditions that often fail to meet time targets for intervention, such as tissue plasminogen activator for myocardial infarction or stroke and the 3-hour bundle for sepsis.

Approach #3: Diagnose Earlier Using Data

By using analytics, health systems can investigate why certain conditions are routinely overlooked until complications ensue, such as pre-diabetes in patients who aren't diagnosed until they present with neuropathy, hypertension or other issues. An examination of data should help organizations determine potential opportunities for earlier diagnosis, particularly when earlier diagnosis improves patient outcomes.

Approach #4: Use the Choosing Wisely Initiative as a Guide

The Choosing Wisely Initiative promotes patient-physician conversations about unnecessary medical tests and procedures, and delayed diagnosis. The initiative also promotes testing that is evidence-based, safe, and necessary. Using Choosing Wisely recommendations as a guide, health systems can leverage analytics to answer several key questions:

• Are we doing too many unnecessary tests, such as routine X-rays/CTs/MRIs for low back pain?
• Do patients always meet indications for certain imaging or lab tests?
• Are we taking too much time getting to the right tests?
• Are we regularly mistaking symptoms for the wrong condition, such as heartburn instead of a heart attack, and then employing dramatically different treatment?
• How do the above factors influence a delayed diagnosis or wrong diagnosis?

Approach #5: Understand Patient Populations Using Data

The use of analytics helps to understand characteristics of a patient population, including the social and economic factors influencing individuals' health. In the absence of specific guidelines or regulations, providers should implement local disease screening standards that balance out the costs, potential benefits, and the potential for, and impact of, false-positives. For example, universal chlamydia screening may have high value and utility in high-prevalence areas or with high-risk patients, but it may also lead to a high false-positive rate in low-prevalence areas.

Approach #6: Collaborate with Improvement Teams

Ask members of different improvement teams, such as cardiovascular or respiratory, to identify their greatest diagnostic challenges. Use analytics to understand the size and impact of delayed, mistaken, or missed diagnoses, then investigate possible causes and quantify the impact of improvement.

Diagnostic tests are often duplicated when a provider is unaware a patient has been tested previously, or when the provider is unable to access test results. Improved communication can reduce wasteful tests and speed diagnosis. The solution could be as simple as establishing a standard practice of asking patients if they had been seen previously for the same problem, and if so, when and where.

Approach #7: Include Patients and Their Families

Health systems must engage not only patients, but also their family members. Family members offer additional insight into a patient's life and medical condition, as well as the patient's social determinants of health. Better insights lead to more engaged patients who comply with treatment recommendations, improved diagnostic accuracy and timeliness, and ultimately better outcomes.

Reducing Diagnostic Error: A Success Story

Sepsis is a costly, life-threatening disease that is responsible for approximately 200,000 deaths a year in the United States. One large medical center decided to leverage advanced healthcare analytics to tackle the significant clinical and financial consequences of sepsis. The health system deployed a scalable analytics platform that integrated clinical, financial, operational, and other data sources to track how various interventions impacted sepsis rates. They also formed a cross-functional team that used the sepsis analytics platform to define cohorts and recommend best practices based on data-driven sepsis diagnosis and outcomes information.

The health system targeted provider compliance with the 3-hour sepsis bundle and used advanced healthcare analytics to track the rate at which nurses complied with the guidelines. Their goal was full compliance with the assessment and 3-hour sepsis bundle program.

The health system developed a prototype electronic algorithm for automating sepsis assessment. Since implementing the technology, they've tracked an 80 percent relative increase in compliance and a 41 percent absolute increase in nurses' compliance with completing assessments. In less than 12 months, the health system significantly improved outcomes and reduced costs by $1.3 million. They also realized a 22 percent decrease in mortality rates.

Reducing Diagnostic Error Is a Healthcare Imperative

Reducing diagnostic error is a healthcare imperative that improves patient safety and outcomes, reduces costs and protects resources. With accurate and timely diagnosis, healthcare systems can help patients get the care they need when they need it, and avoid wasting money and resources on ineffective treatments. The adoption of advanced healthcare analytics helps providers make appropriate, data-driven diagnoses, eliminates diagnostic error and improves clinical outcomes.

By Tracy Vayo, Vice President of Knowledge Services, Health Catalyst

The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.


Featured Whitepapers

Featured Webinars