The demented elderly man arrived in the emergency department with a note from his nursing home: “cough and shortness of breath worsening over the past 7 days.” He had a low grade fever but did not seem too unwell, and my examination revealed only some crepitations at the base of the right lung. Chest infection, I thought, and the x ray did show some patchy opacification in the right lower lobe. I started some antibiotics and sent him to the ward. A couple of hours later, the medical registrar phoned me and gently informed me that the opacities I had seen were, in fact, the entire lung—collapsed as a result of the huge pneumothorax which ought to have been impossible to miss.
One Saturday night a few weeks later, I found myself the only doctor on an island, home to an Aboriginal community. A 14 year old girl, 32 weeks pregnant, a shy and diffident historian, attended the island's hospital complaining of lower abdominal pain. She too had a low grade fever. I suspected urinary tractinfection, and a very cloudy urine specimen was positive for blood and protein. I sent her away with antibiotics. Fortunately for us both, and very fortunately for her baby, she re-presented soon afterwards, giving me the chance to make the correct diagnosis (labour) in time for a helicopter to take her to someone more qualified than I to manage the footling breech delivery of a premature baby. The urine had been cloudy because it was full of vernix.
I was young, inexperienced, and overconfident. In each case, lacking a clear history, I made things harder for myself by doing an inadequate examination. But the biggest mistake, common to both of these stories, was of leaping to a conclusion early and then seeing what I expected to see. Expectations can fog your vision. It's best to wait until all the evidence is in before attempting a synthesis.
Competing interests: None declared.