Part II:  Using the information on disease based illness scripts, analyze the following diagnostic error case.

Mr. John Smith is a 30 year old man with HIV infection for two years and a history of only occasional use of his antiretroviral medications. He presented to the emergency department with complaints of shortness of breath and dry cough for two weeks, He noted that the symptoms were worse when he was lying down and were relieved when he sat up.  He denied fever. A chest xray showed bilateral interstitial infiltrates without hilar adenopathy.  The cardiac silhouette was enlarged, which was attributed to the fact that it was a portable chest xray.  The admitting team documented bilateral rales on pulmonary exam and made a presumptive diagnosis of pneumonia due to pneumoncystis jiroveci.  His oxygen saturation was low so his initial treatment included supplemental oxygen, trimethoprim-sulfamethoxazole, and steroids.  On day 2 of his hospitalization, a medical student chart note documented an S3 gallop. The resident note commented that S3 gallops are sometimes normal in young people.  The CD4 count documented by the lab was 450. He could not comply with sputum induction and a bronchoalveolar lavage was scheduled for later in the week.   On day 3 of the hospitalization, Mr. Jones noted swelling in his legs and worsening of his shortness of breath.  The team felt that these findings were caused by excessive intravenous fluids and discontinued his supplemental fluids.  On day 4 of his hospitalization, Mr. Jones had a cardiopulmonary arrest, was resuscitated and transferred to the ICU, where his cardiac ejection fraction was found to be 15%.  His bronchoalveolar lavage was negative for pneumocystis jiroveci.  

---

On day one, the admitting team estimated that pneumonia due to pneumocystis jiroveci was a Tier I diagnosis in this patient: Very Likely.   What was their patient illness script (epemiology, time course, and syndrome statement) to support their diagnosis?

0 points: No submission
1 point: Only one of the three illness script components included (epidemiology, time course, or syndrome statement)
2 points: Only two of the three illness script components included (epidemiology, time course, or syndrome statement)
3 points: All three illness script components included (epidemiology, time course, and syndrome statement) 
4 points: All three illness script components included (epidemiology, time course, and syndrome statement) AND one is consistent with the answers provided
5 points: All three illness script components included (epidemiology, time course, and syndrome statement) AND two or three are consistent with the answers provided

What data were they missing on day 1 that would help the accuracy of their diagnosis?

0 points: No submission
1 point: Data identified that is not correct and not consistent with the answers provided
2 points: Data identified that is consistent with the answers provided

What data did they have but ignore on day 1?

0 points: No submission
1 point: Data identified that is not correct and not consistent with the answers provided
2 points: Data identified that is consistent with the answers provided

On day 2 of Mr. Smith's hospitalization, a medical student claimed he heard an S3 and the resident explained that S3s can be seen in healthy individuals, which is true. The resident’s belief that the S3 in this patient was a false positive finding rather than a sign that the diagnosis was wrong is an example of what type of bias?

0 points: No submission
1 point: Incorrect type of bias selected
2 points: Correct type of bias selected

The sensitivity of the lab finding of a CD4 count < 200 for pneumocystis pneumonia is 99%.  Obtaining a CD4 count result of 450 for this patient should change the team’s belief that the likelihood of pneumocystis from VERY LIKELY to:

0 points: No submission
1 point: Incorrect answer
2 points: Correct answer

In HIV infected patients, LDH is sometimes used as a diagnostic test for pneumocystis pneumonia.  The sensitivity and specificity of an elevated LDH in HIV infected patients is 100% and 47%, respectively.  Given this information, should you order an LDH as part of your management of this patient?  Why or why not?

0 points: No submission
1 point: Incorrect decision, or correct decision without explanation
2 points: Correct decision and explanation consistent with answers provided

Given the sensitivity and specificity of LDH in this clinical situation, which of the following statements is true?

a.     An elevated LDH would definitively confirm the diagnosis of pneumocystis pneumonia.

b.    A normal LDH would definitively exclude the diagnosis of pneumocystis pneumonia.

c.      A normal LDH would not add to the diagnosis of this patient’s condition

0 points: Incorrect answer selected or no submission
1 point: Correct answer selected

What reasons might exist for the failure of the team to change or reevaluate their diagnosis when the results of the CD4 count became available?

0 points: No submission
1 point: Reasons are not consistent with the answers provided
2 points: Reasons are consistent with the answers provided