Radiology and Errors

In the days of active consumerism , reducing credibility of  medical profession and  informed decision making , it is vital to   understand the  scenario, and act accordingly (instead of the usual  , I  am very good  and do not make mistakes,  while others less knowledgeable  than me,  make all the mistakes  !). An excellent write up submitted by Dr MGK Murthy, Sr Consultant Radiologist

A mistake is  a mistake.(  neither it is mine (make me guilty!)  nor yours (makes me arrogant!), and  it need not be owned.

“I would give great praise to the physician whose mistakes are small, for perfect accuracy is seldom to be seen. “Hippocrates 

“Errors in judgement  must occur in the practice of an art  which consists largely in balancing the probabilities “  Sir William Osler ( considered Father of modern medicine , 1849-1919)

From a practical point of view  once an abnormality has been pointed out  and a lay person can see it , it is not easy to convince that  a Radiologist  who is trained for the job and paid  for seeing it , should be exonerated for missing it.

1. In all branches of medicine , there is an  inevitable element of patient exposure  to problems arising from human error ,and this is increasingly subject of bad publicity , often skewed towards an assumption that perfection is achievable , and any error or discrepancy represents wrong and must be punished!

Unfortunately the public (and hence the   executive &judiciary) frequently expects a medical investigation will produce “the correct answer” all the time .

2. Radiology involves decision making under conditions of uncertainty , and therefore can not  always produce  infallible interpretations of reports . Interpretation of human picture (radiology in essence) is not a  binary process ; the answer is not always  normal or abnormal , cancer or not etc

         Final report is  often influenced by many variables  , not least among them , available clinical/ other information  at the time of reporting .

3. With respect to radiology investigations , the use of the word  “error” is often unsuitable .It is more appropriate to  concentrate on “discrepancies “ between a report and a retrospective review of a film or outcome .

4. “Opinion” may be defined as “ a conclusion  arrived at,  after some  weighing of evidence , but open to debate or suggestion” and hence radiology reports are not expected to be incontrovertible

       Somewhere between  clear cut errors and the inevitable difference of opinion in interpretation of a picture , is an arbitrary division defining the limit of  professional acceptability .

5. Approximately about 1 billion radiological  investigations are carried  out across the globe annually, and  literature quoted ,average error rate of 4% (across all modalities ,  though range is 2-20%  in a large 20 year review study)is considered as acceptable , making an appx 40 million radiological investigations suspect  for their utility.

          A large study at  MGH, Boston ( 2010 ) suggested   after  a double blinded study (of abdominal CT scans)   an inter observer  difference of appx 26%(between two different  similarly trained Radiologists  )  and(ironically! )intra observer (same person reading the same pictures at different times ) rate higher  at 30%. We can safely say hence “we  differ more with ourselves than with others!

          Another  recent  global  study has revised the figure of  “real time”   error rate in day to day  radiological practice averaging at 3-5%.

6. The factors for the discrepancies  are many including  the time of viewing the  film (4 secs is considered as  optimal for Chest x ray viewing and any prolongation (visual dwell )  would lead to higher false +ve or -ves/ system related  parameters including  acquisition parameters / Technique used OR  available data at the time  / work load of the day / mental frame of the Radiologist / Viewing  conditions  etc etc

7. Common experience  in radiology suggests  that many errors are of little or no significance to the patient and some significant errors  remain  undiscovered

8. Perfection =  An  imaginary state or quality  distinguished from the actual by an element of known as excellence , an attribute  of the critic .

        Legal basis for negligence  involves a breach of standard of care , which is usually defined as “care exercised by an average physician  of similar knowledge , skill  and ability under similar conditions”

9. In conclusion, there is an absolutely  unavoidable human factor at work in review of films/ images  ; some abnormalities ( even obvious ones ) would be missed ; the mere fact that  a Radiologist  misses an abnormality, does not constitute malpractice ; and also  not all radiographic misses are excusable ;

              Therefore the focus of attention  should be on  issues   such as Proof of competence , habits of practice, and use of proper   techniques. A few simple steps  would mostly reduce , if not eliminate the so called errors or discrepancies. 

1. Train  technician to  obtain adequate  history / clinical data ,  and  make him /her  the preliminary reader of the images with ref to the   clinical question(you may or may not agree with the  opinion, at least   it can  facilitate your  second look at his region of interest ). I have immensely benefitted from such interactions  (if one is blessed  to have PGs , they suit the role brilliantly)

2. Start your report with the clinical question and why is the investigation ordered

3. Answer the findings in relation to the question ,+vely or -vely in the very  first paragraph  itself

4. Do not use diagnostic terms in the findings (for eg : glioblastoma/ cancer etc). Instead use only anatomical descriptions .

5. Impression/ conclusion should bring  abbreviated  highlights of +ve findings in relation  to the clinical question  and a suitable confirmatory or exclusive investigation (preferably radiological) should be offered (with the words as would help to give legal  leverage  to the clinician/ patient  to act  appropriately)

6. Rest of the study shows …. Should be next conclusion (brief … and not copy paste the Findings ) , with a similar suitable advice of other investigations if necessary.  Conclusion should use diagnostic terms /  reflect  one opinion about disease and not  an echogenic  …. Etc  . one can take the help of , could rep…. in view of …… etc

7. Comparison with old  images should follow ,  expressing various components  including the effect of therapy/ progression of  the disease  findings etc

8. Always proof read your report with the help of your data entry operators( who would even confirm that you have reported the correct patient , i.e. images and report belong to the same individual/ investigation done is  what the clinician ordered  etc)

9. Do not try to  make report as if one has studied the tissue  and not a picture ! Murphys law states  “if something can go wrong , it WILL go wrong)

10. Spend  a few minutes summarizing your days work before departure, rather than  feeling  sorry at a later stage ( a bit  of unpaid  time at work can keep our dopamine/ serotonin/ oxytocin levels(happy guys)  high enough to feel contended)