Skip to main content

Table 3 Laboratory data and PDA diagnosis compared

From: Quality of EHR data extractions for studies of preterm birth in a tertiary care center: guidelines for obtaining reliable data

 

1. Manually abstracted database, # of subjects

2. EHR extract-ion, # of subjects

3. Discrepancy (% and # of subjects) between the databases a

4. Manually abstracted database errors

5. EHR-extracted data errors

6. Median discrepancy

7. Discrepancy range

1st WBC count b

1257

1437

3.2 % (40)

2.5 %(32)

0.6 % (8) c***

0.75 k/mm3

0.01–109 k/mm3

1st Hemoglobin

1333

1460

11.9 % (158)

5.8 % (77)

8.3 % (110) d *

1.4 g/dl

0.1–25.9 g/dl

Peak total bilirubin

1565

1336

11.4 % (152)

6.9 % (92)

5.1 % (68) e *

1.45 mg/dl

0.1–15.2 mg/dl

Peak direct bilirubin

681

674

4.9 % (33)

4.5 % (30)

0.9 % (6) f ****

0.5 mg/dl

0.1–16.4 mg/dl

PDA

512

414

12.8 % (53)

12.8 %(53)

8.2 % (34) g ***

NA

NA

  1. Laboratory data and PDA parameters compared in the paper. The denominator for the percentage is the smaller of the corresponding values in the first two columns
  2. *P0.05; **P0.01; ***P0.001; ****P0.0001
  3. a - In general, the sum of the error counts in columns 4 and 5 do not add up to the number in column 3, because the error occurred in both manually and electronically extracted data, or the cause was ambiguous
  4. b - Re-calculated discrepancies after adjusting for the inappropriate Hispanic category in the race column
  5. c - Difference statistically significant, p = 1.3 × 10−4 by Chi-square test
  6. d - Difference statistically significant, p = 0.012 by Chi-square test
  7. e - Difference statistically significant, p = 0.05 by Chi-square test
  8. f - Difference statistically significant, p = 4.9 × 10−5 by Chi-square test
  9. g - Difference statistically significant, p = 0.001 by Chi-square test