Getting caught up in a patient’s bad mood may set up a lose-lose proposition.
Physicians could use simple mind-games to safeguard against negative emotions
In two separate experiments, primary care residents proved to be less accurate when diagnosing patients who demonstrated disruptive behavior compared with patients showing more neutral behavior, researchers reported.
Average diagnostic accuracy scores were 10% lower when first- and second-year internal medicine residents evaluated clinical vignettes portraying disruptive patients compared with behavior-neutral patients, according to Silvia Mamede, MD, PhD, of Erasmus Medical Center in Rotterdam, the Netherlands, and colleagues.
This 10% reduction in diagnostic accuracy persisted across a second experiment involving more experienced, third-year family medicine residents who were asked to perform similar evaluations under time pressure, the same authors reported. Both studies were published in BMJ Quality & Safety.
"The findings show a significant decrease in physician diagnostic accuracy when combined across scenarios and scaled so that '0.00' denotes a faulty diagnoses and '1.00' denotes an accurate diagnoses," noted Donald A. Redelmeier, MD, and Edward E. Etchells, MD, of Sunnybrook Health Sciences Centre in Toronto in an accompanying editorial.
They added that "physicians need safeguards against the negative emotions that may undermine diagnostic accuracy."
Redelmeier and Etchells suggested physicians could use simple mind-games to safeguard against negative emotions that may undermine diagnostic accuracy, such as self-reflection, imaging the patient as easy instead of difficult, employing diagnostic checklists or computer programs, or following up with patients.
"Real people, however, cannot always control their temper when suffering or in pain ... we believe these logically coherent patient strategies are unlikely to be popular or effective despite the implication ... that etiquette can foster better diagnostic accuracy," they wrote.
In the first experiment, 74 PGY-1 and PGY-2 internal medicine residents at Erasmus Medical Center were randomized to diagnose eight clinical vignettes. All of the vignettes had the exact same clinical condition, however, each scenario represented a different patient who demonstrated either neutral or disruptive behavior in the following ways:
âº Frequent demander
âº Questioned doctor's competence
âº Ignored doctor's advice
âº Low expectations of doctor's support
âº Utterly helpless
âº Threatening to the doctor
âº Accused the doctor of discrimination
The average diagnostic accuracy scores, which ranged from 0-1, were lower for the difficult patients compared with the neutral patients (0.41 vs 0.51, respectively, P<.01). And residents were more likely to recall clinical findings (32.5% versus 29.8%, P<.001), and less likely to recall patient behavior, in neutral scenarios compared with difficult patients (17.9% versus 25.5%, P<.001).
Study limitations included the fact that the authors looked at three potential underlying mechanisms of the decrease in diagnostic performance in difficult patients, and not any cognitive explanations for this phenomenon, which may have other mechanisms. Also, the results may have differed with more experienced physicians.
However, the authors noted that while it's possible more experienced doctors develop the ability to prevent a negative response, "having dealt with many difficult patients might make negative attitudes towards them stronger and, consequently, more easily activated in the mind."
In the second study, 63 family practice residents at the same facility who were in their last quarter of their third year were asked to perform a two-part diagnosis for six clinical vignettes while being timed.
The vignettes were a combination of neutral and difficult patients with the same behaviors as those patients from the first experiment. But in this experiment, the vignettes also included one of the following neutral or complex clinical conditions: community-acquired pneumonia, pulmonary embolism, meningoencephalitis, hyperthyroidism, appendicitis, and acute alcoholic pancreatitis.
With a large digital clock in the room, a reminder for expedience, and a directive to record start and finish time of each of the six diagnoses, the residents began the first part of the experiment.
For the second part, the residents were asked to deliberately reflect upon the cases for a period of time, and then return to the vignettes to double-check the initial diagnosis, and either provide a list of justifications for the diagnosis or any other possible explanations for the symptoms. If the original diagnosis was incorrect, the residents were asked to list an alternative diagnosis, and provide analysis for the new diagnosis.
Average diagnostic accuracy scores were lower for difficult patients compared with neutral patients (0.54 versus 0.64, P=0.017). However, regardless of patient attitudes, diagnostic accuracy was higher for simple cases compared with complex cases (0.88 vs 0.94). During the reflection period, diagnostic scores improved regardless of case complexity or patient behavior (0.60 vs 0.68, P=0.002).
The time it took to reach a diagnosis did not differ greatly between vignettes for either study.
Redelmeier and Etchells suggested the study was subject to limitations inherent to scenario experiments, such as surrogate endpoint bias, confounding bias, and selection bias.
Study limitations included the fact that the vignette design provided all of the necessary information to make a diagnosis, a circumstance not reflective of clinical practice. It was also possible that the residents perceived the disruptive behavior as a symptom relevant to the diagnosis.
"I think these studies offer reasonably compelling evidence that medicine is more than collecting and interpreting clinical data," commented Sean Tackett, MD, of Johns Hopkins University, said in an email to MedPage Today. "Good medical care relies on physicians and patients being on the same page, especially in primary care."
"Physicians need to recognize when they are with 'difficult patients' so they can seek help from the healthcare team to avoid making diagnostic errors," wrote Tackett, who was not involved in the study.
The editorialists pointed out that the study authors did not "test corrective procedures, thereby highlighting opportunities for future research."
Mamede and co-authors disclosed no relevant relationships with industry.
The editorial was supported by a Canada Research Chair in Medical Decision Sciences and the Canadian Institutes of Health Research.
Redelmeier and Etchells disclosed no relevant relationships with industry.
Reviewed byRobert Jasmer, MD Associate Clinical Professor of Medicine, University of California, San Francisco and Dorothy Caputo, MA, BSN, RN, Nurse Planner.