| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Malpractice Issues in Radiology |
1 Department of Radiology, Rush Medical College, Chicago, IL 60612, and Department of Radiology, Rush North Shore Medical Center, 9600 Gross Point Rd., Skokie, IL 60076
Received September 1, 1999;
accepted after revision September 15, 1999.
Case summaries are based on actual events and lawsuits, although certain
facts have been omitted or modified by the author, who has supplied and
obtained authorization for the reproduction of the radiographic images. All
opinions expressed herein are those of the author and do not necessarily
reflect those of the American Journal of Roentgenology or the
American Roentgen Ray Society.
Introduction
 Top Introduction The CasesMalpractice IssuesDiscussionSummary and Risk ManagementReferences |
|---|
Alliterative: "The repetition of a sound that is usually a consonant in two or more neighboring words or syllables" [1].
|
TopIntroductionThe CasesMalpractice IssuesDiscussionSummary and Risk ManagementReferences |
|---|
|
|
The irregular density noted on the chest radiograph is well seen on CT. It has a stellate appearance but the overall appearance is suggestive of scarring. If any previous studies are available, it would be helpful to obtain them for comparison.
No further action was taken relative to the chest findings, and the patient underwent knee surgery without incident.
Fifteen months later, the patient was referred for follow-up chest radiography because of complaints of cough and "congestion." Another member of the same hospital-based radiology group interpreted the study as disclosing "no change in the appearance of the irregularly shaped scar since the previous examination" (Fig. 1C). Three months later, a follow-up chest radiograph was interpreted by the same radiologist as disclosing, "no change in appearance of right upper lobe scar. No acute abnormality is noted" (Fig. 1D).
|
|
Nine months later, or 27 months after the initial radiographic studies, the patient returned for follow-up radiography because he was experiencing progressive cough and weight loss. At this time, the same radiologist interpreted the radiographs as showing, "right upper lobe spiculated density; neoplastic process is primary consideration. CT is suggested for further evaluation" (Fig. 1E). CT examination performed later in the day revealed a "3.5-cm spiculated mass compatible with neoplasm. This has changed significantly since previous studies" (Fig. 1F).
|
|
Percutaneous needle biopsy of the lesion revealed well-differentiated adenocarcinoma. Patient underwent thoracotomy, followed by radiation and chemotherapy, but died 10 months later.
Case 2
A 55-year-old man was brought to the emergency department of a hospital
because of injuries sustained in an automobile accident. Although findings on
various bone radiographs were found to be normal, the findings on chest
radiographs (Figs. 2A and
2B) were interpreted by the
radiologist as disclosing the following:
|
|
Left upper lobe poorly marginated infiltration containing irregular radiolucencies. Comparison with previous chest radiographs done in this hospital 2 years ago shows that this is a new lesion. CT scan is suggested for further evaluation.
A CT scan obtained later the same day (Fig. 2C) was reported by the radiologist as showing:
|
...lobulated infiltrate containing radiolucencies in the left upper lobe which has the appearance of an inflammatory lesion. However, if the patient does not have pulmonary symptoms, one must consider the possibility of carcinoma.
After examining the patient, the primary care physician determined that the patient was suffering from pneumonia and prescribed antibiotics. Three months later, the patient returned for follow-up chest radiography. At this time, another radiologist in the same hospital-based radiology group interpreted the study as follows: "In comparison with the previous examination, the density in the left upper lobe has decreased slightly. It remains consistent with a resolving inflammatory process" (Fig. 2D).
|
Five months later, the patient again underwent chest radiography. These radiographs were interpreted by the radiologist as showing "little change since previous study. The stable appearance of left upper lobe irregular opacity is again consistent with an inflammatory process rather than neoplasia" (Fig. 2E).
|
Seven months later, or 15 months after the initial chest radiographs, the patient, because of complaints of weight loss and chest discomfort, was referred for chest radiography to another hospital by another primary care physician. Now, chest radiographs were reported as showing "marked increase in left upper lobe infiltration. Carcinoma is now the prime consideration" (Figs. 2F and 2G). CT examination confirmed the presence of a tumor (Fig. 2H). Biopsy revealed adenocarcinoma. The patient underwent surgery, followed by radiation and chemotherapy. However, 14 months later, the patient died.
|
|
|
|
TopIntroductionThe CasesMalpractice IssuesDiscussionSummary and Risk ManagementReferences |
|---|
Case 1
In the first case, the primary care physician testified at deposition that
he was "lulled into complacency" regarding the possibility that
the patient might have lung cancer because the radiology reports were
"indicative only of a scar." The physician figuratively pointed
the finger at the two defendant radiologists charging that "If they had
told me that the X rays looked like cancer, I would have called in a surgeon
immediately." In their depositions, the defendant radiologists stood by
their original interpretations. "The density on the chest X ray and the
CT looked like a scar," said the radiologist who interpreted the initial
studies. He later added, "I suppose I could have been more aggressive in
giving a differential diagnosis, but I didn't think it was
necessary."
The radiologist who interpreted the follow-up radiologic studies also defended his interpretations as being correct, although he acknowledged that he was "probably influenced" by the interpretation rendered by the first radiologist. When questioned whether he would have raised the possibility of neoplasm on the follow-up radiologic studies if there had been no previous studies or reports, the defendant radiologist who interpreted the follow-up images answered in the affirmative.
A radiology expert retained by the plaintiff was critical of both defendant radiologists. "They should have, at the very least, placed carcinoma in a differential diagnosis and then suggested a biopsy," the expert asserted. The plaintiff's expert was adamant that both defendant radiologists had breached the standard of radiologic care.
A radiology expert retained by the defense disagreed with the plaintiff's expert, stating that in his opinion the radiologic interpretations conformed to the standard of care. "While it's true that the radiologists could have been more aggressive by raising the possibility of carcinoma," explained the expert for the defense, "the appearance of the lesion could well have been indicative of a scar." "Furthermore," emphasized the defense expert, "the fact that the lesion did not change over a period of 18 months could lead any reasonable radiologist to conclude that we were dealing with a benign process." The defense expert also raised the possibility that the abnormal density seen on the initial studies did indeed represent a scar and that the carcinoma developed later.
Eventually, the malpractice lawsuit was settled out of court for $800,000, apportioned equally between the radiologists and the primary care physician.
Case 2
It should be no surprise that the testimony in the second case was similar
to that in the first. In this case, however, the primary care physician did
not lay blame on the radiologists. At her deposition, the primary physician
testified that the physical and laboratory findings of the patient were
consistent with pneumonia. She went on to state that the patient did indeed
improve, but several months later the patient developed recurrence of symptoms
that suggested another inflammatory process. It therefore was "no
surprise," said the physician, that follow-up radiographs showed
"similar findings as before." Although the physician acknowledged
that she would have recommended biopsy or surgical consultation or both if the
radiologist had raised the suspicion of malignancy, particularly at the
8-month follow-up studies, nonetheless, the physician said that "the
responsibility was mine."
The defendant radiologists testified that the findings on the various chest radiographs and CT scans were indicative of an inflammatory process and that although neoplasm was a "possibility," it was not a "probability." The defendant radiologists who interpreted the follow-up images admitted that they were perhaps unduly influenced by the appearance and the written interpretations of the radiologic studies that had been obtained previously. When asked whether they would have interpreted the follow-up radiographs differently if there had been no previous studies, the radiologists who interpreted the later images both responded, "Maybe."
A radiology expert for the plaintiff testified that the defendant radiologists were negligent for not having emphasized the possibility of malignancy and recommending biopsy. A radiology expert for the defense agreed with the opinions and supported the actions of the defendant radiologists, claiming that their conduct was well within the standard of radiologic care.
The lawsuit never proceeded to trial. It was settled for $1 million. Forty percent was paid by the defendant radiologists; the balance, by the primary care physician.
|
TopIntroductionThe CasesMalpractice IssuesDiscussionSummary and Risk ManagementReferences |
|---|
In his classic monograph on radiologic errors published 33 years ago, Marcus J. Smith [5] estimated that judgmental error based on faulty reasoning was responsible for 10% of all mistakes made by radiologists. Smith theorized that judgmental errors were caused by radiologists' failure to "think of possibilities" when interpreting radiographs, because of either incomplete knowledge or, more likely, preconceived notions. Among the factors that lead to preconceived notions is the phenomenon for which Smith coined the term "alliterative error," a phenomenon about which this article will focus in detail.
The alliterative error, according to Smith [5], results from the influence that one radiologist exerts on another. Smith's theory is that if one radiologist fails to detect an abnormality or attaches the wrong significance to an abnormality that is easily perceived, the chance that a subsequent radiologist will repeat the same error is increased. Smith explains that alliterative errors occur because radiologists read the reports of previous examinations before looking at the newly obtained radiographs and are therefore more apt to adopt the same opinion as that rendered previously by a colleague (or oneself). Smith attributes this to a tendency and perhaps a need among people in a social or professional group (in this case radiologists) to conform to their peers.
Other researchers have alluded to alliterative errors, although none, to my knowledge, has used this specific term. In their analysis of 182 radiologic errors committed at the University of Iowa, Renfrew et al. [6] attributed certain cognitive errors to faulty reasoning on the part of radiologists due to their being given information that misled them. In an article about pitfalls in the radiologic diagnosis of lung cancer, Woodring [7] discussed why radiologists sometimes misinterpret otherwise well-known radiologic features of a cancer as indicating benign disease. Implying that radiologists can be shackled by preconceived notions, Woodring urged radiologists to have a "conscious awareness" of their tendency to deliberately misinterpret chest radiographs and to be willing to "freely" entertain alternative diagnoses.
Although this article focuses on adverse outcomes that result from reading previous radiology reports before interpreting newly obtained radiographs, almost all the radiology literature emphasizes the benefits of such review. Elmore et al. [8] evaluated the influence of clinical history of patients on radiologists' interpretations of mammograms. These researchers found that overall diagnostic accuracy is not altered by patient history, but recommendations were affected for appropriate further diagnostic workup. An "alerting" history such as breast symptoms or family history of breast cancer increased the number of workups recommended in patients, but a "nonalerting" history led to fewer recommended workups. To further illustrate how radiologists were influenced by what information was made available to them, Elmore et al. showed that when a sham clinical history was given, 40% of the radiologists changed their diagnostic interpretation about whether to recommend a biopsy in the direction of the misleading sham suggestion.
There is no doubt that knowledge of a patient's history and clinical findings improves accuracy of radiologic interpretations. In a study in which radiologists were told which quadrant of the chest to examine for the presence of pulmonary nodules, nodule detection was found to be considerably more accurate in directed search than in free search [9]. Egglin and Feinstein [10] studied the phenomenon of context bias in diagnostic radiology. They found that a diagnostic test is more likely to have a higher sensitivity in a population that includes many patients with severe or advanced disease than it would in a screening population with less severe disease. Aideyan et al. [11], Berbaum [12], and Doubilet and Herman [13], among others, also have shown that knowledge of history improves the detection rate of radiologic abnormalities.
The standard of radiologic care that requires the comparing of new radiographs with those obtained previously is uncontested [14, 15]. The radiologic standard of care that deals with the requirement about reading previous radiology reports before interpreting newly obtained radiographs is not nearly as clear, however. In fact, there is some controversy as to whether old reports even have value. The "American College of Radiology Standard for Communication: Diagnostic Radiology" states [16]: "Comparisons with previous examinations and reports [italics added] when possible are a part of the radiologic consultation and report." Hunter and Boyle [17] studied the usefulness of reading previous radiology reports before interpreting newly obtained radiographs. These researchers found that reading the previous report was useful in 60% of all cases and that in 22-24% of the cases it provided a significant amount of help. Hunter and Boyle concluded that reading old reports "will do much to enhance the quality of radiologic practice."
In their study evaluating the effectiveness of reading previous radiography reports, White et al. [18] came to an opposite conclusion, however. These researchers found that previous reports were judged to be nonuseful in 48% of the cases in which they were used and that in only 1.5% of cases were the reports believed to be more valuable than previous radiographs.
One additional point pertaining to the first case should be mentioned. The radiology expert for the defense speculated that perhaps the lesion noted on the initial radiologic studies was indeed a scar and that the carcinoma diagnosed on subsequent studies had actually developed within the scar. There is scientific basis for such a hypothesis. Lung malignancies, usually of the adenocarcinoma variety, can arise in preexisting scars [19]. According to Naidich et al. [20], one explanation for this phenomenon is that the increased metabolic activity in areas of scarring causes a mutation in the p53 cancer-suppressing gene that leads to production and survival of cells with abnormal tumorigenic DNA. Naidich et al. also point out that the incidence of true scar carcinoma has probably been overestimated in the past and that a causal relationship between a scar and carcinoma is difficult to prove. The defense attorney in this case decided to not even attempt to convince a jury that the carcinoma was not present initially but had instead developed later and opted for settlement. Whether a carcinoma-developed-in-a-scar defense would be successful in future cases similar to this one remains to be determined.
|
TopIntroductionThe CasesMalpractice IssuesDiscussionSummary and Risk ManagementReferences |
|---|
Risk management with regard to alliterative errors can lessen the likelihood of incurring a medical malpractice lawsuit and maximize chances for a successful defense if such a suit is filed. Risk management can also enhance good patient care. The following risk management pointers will help radiologists meet all three of these objectives.
|
TopIntroductionThe CasesMalpractice IssuesDiscussionSummary and Risk ManagementReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Berlin Accuracy of Diagnostic Procedures: Has It Improved Over the Past Five Decades? Am. J. Roentgenol., May 1, 2007; 188(5): 1173 - 1178. [Full Text] [PDF]
|
|
|
|
|
|
L. Berlin Failure to Diagnose Lung Cancer: Anatomy of a Malpractice Trial Am. J. Roentgenol., January 1, 2003; 180(1): 37 - 45. [Full Text] [PDF]
|
|
|
|
|
|
D. Balsam Malpractice: Avoiding Alliteration Am. J. Roentgenol., October 1, 2000; 175 (4): 1191 - 1191. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
