Dissection
Budget: £250 – £750 GBP
Study hypothesis: That the sensitivity and specificity for diagnosis of spontaneous coronary artery dissection (SCAD) on computed tomography (CTCA) is suboptimal.
Method:
A total of 80 CTCAs ( 40 SCAD CTCA and 40 non-SCAD CTCA) were interpreted by 5 different readers/raters. Readers were then given education and asked to reinterpret the CTCA.
I have attached 11 files. One file contains the grouped data. I have then separated the data for each individual rater/reader. I have further broken the data down into 1) SCAD versus normal and 2) SCAD versus coronary artery disease for each reader.
Primary question: Could readers identify SCAD from non SCAD on the 80 CTCA? Did education improve diagnosis on the second read?
What was the sensitivity , specificity, negative predictive value and positive predicative value on the initial and second read
Q2: We duplicated 10 non- SCAD and 11 SCAD cases for inter and intra observer variability. What was the inter and intra observer variability on the initial and second read? Did it improve with education.
Q3. What was the confidence ( 5 point scale) of readers for making a diagnosis ? how did their confidence correlate with making the correct diagnosis.
Q4: How did their confidence correlate with the quality of the scan,
Q5: How did the making the correct diagnosis correlate with the quality of the study
5b: what was the frequency of artefact type reported?
Q6: How did the readers correct diagnosis correlate with overcall of SCAD.
Q7: Which coronary arteries were the most likely to have been correctly called a SCAD
Q8: Which portion of the vessel (proximal -mid or distal) was most likely to have been correctly called a SCAD
Q9: How did the making the correct diagnosis correlate with the year of CTCA.
Q10: How did the making the correct diagnosis correlate with the scanner used.
Q11: How did the making the correct diagnosis correlate with the kV
Q12: How did the making the correct diagnosis correlate with the heart rate
Q13: How did the making the correct diagnosis correlate with correctly identifying all the SCAD lesions.
Q14: Rater 1, Rater 2 and Rater 3 are consultants. Rater 4 and Rater 5 are non-consultants. Did the level of confidence vary between consultants versus non-consultants?
Q14: Rater 1, Rater 2 and Rater 3 are consultants. Rater 4 and Rater 5 are non-consultants. Did the level sensitivity , specificity, negative predictive value and positive predicative value vary between consultants versus non-consultants
Method:
A total of 80 CTCAs ( 40 SCAD CTCA and 40 non-SCAD CTCA) were interpreted by 5 different readers/raters. Readers were then given education and asked to reinterpret the CTCA.
I have attached 11 files. One file contains the grouped data. I have then separated the data for each individual rater/reader. I have further broken the data down into 1) SCAD versus normal and 2) SCAD versus coronary artery disease for each reader.
Primary question: Could readers identify SCAD from non SCAD on the 80 CTCA? Did education improve diagnosis on the second read?
What was the sensitivity , specificity, negative predictive value and positive predicative value on the initial and second read
Q2: We duplicated 10 non- SCAD and 11 SCAD cases for inter and intra observer variability. What was the inter and intra observer variability on the initial and second read? Did it improve with education.
Q3. What was the confidence ( 5 point scale) of readers for making a diagnosis ? how did their confidence correlate with making the correct diagnosis.
Q4: How did their confidence correlate with the quality of the scan,
Q5: How did the making the correct diagnosis correlate with the quality of the study
5b: what was the frequency of artefact type reported?
Q6: How did the readers correct diagnosis correlate with overcall of SCAD.
Q7: Which coronary arteries were the most likely to have been correctly called a SCAD
Q8: Which portion of the vessel (proximal -mid or distal) was most likely to have been correctly called a SCAD
Q9: How did the making the correct diagnosis correlate with the year of CTCA.
Q10: How did the making the correct diagnosis correlate with the scanner used.
Q11: How did the making the correct diagnosis correlate with the kV
Q12: How did the making the correct diagnosis correlate with the heart rate
Q13: How did the making the correct diagnosis correlate with correctly identifying all the SCAD lesions.
Q14: Rater 1, Rater 2 and Rater 3 are consultants. Rater 4 and Rater 5 are non-consultants. Did the level of confidence vary between consultants versus non-consultants?
Q14: Rater 1, Rater 2 and Rater 3 are consultants. Rater 4 and Rater 5 are non-consultants. Did the level sensitivity , specificity, negative predictive value and positive predicative value vary between consultants versus non-consultants