Australian National Diabetes Information Audit & Benchmarking (ANDIAB) 2011 Final Report

1.5 Data Verification and Validation

Describes the system used for changing any incorrect data through validation reports

Page last updated: December 2011

As in previous years every effort was made to ensure data completeness and correctness, with specific ‘validation reports’ generated for each site [Table 3].

Thus, attempts were made to provide sites with the opportunity to improve their data with the generation of these specific reports containing lists of missing or potentially invalid data, as well as possible duplicate individual entries. These were forwarded to the sites through the NADC Secretariat and returned to the central site once reviewed. In 2011, 31 sites returned Validation Reports. All additional or corrected data items were entered / corrected respectively, in the pooled database, prior to final data analysis. For duplicate data, where duplicates were identified, these were reviewed and the first entered record retained, supplemented by any additional data in the second record that was missing in the original. The second entered record was then deleted.

Table 3 - Specific Reports

[a] Missing Vital Data Fields :
A list was generated of all instances of missing data in any of the following fields :

Sex; Date of Birth; Year of Diagnosis; Diabetes Type; Diabetes Management; Height; Weight; Initial Visit; Fasting Status if Lipids.
[b] Potentially Invalid Data Values :
A list was generated to check potential data inaccuracies as follows :

[1] Male = Pregnant OR Female <15 or > 50 = Pregnant;
[2] Type 1 and Treatment []] = Diet or Tablets;
[3] GDM and ] = Tablets;
[4] Year of Diagnosis or Year Started Insulin < Date of Birth;
[5] On Insulin Since Year < Date of Diagnosis;
[6] Systolic BP is < Diastolic BP OR Systolic or Diastolic is missing;
[7] Systolic BP < 70 or > 200 OR Diastolic BP < 40 or > 130 mmHg;
[8] Age >14 and Height < 1.3 Metres or > 2.0 Metres;
[9] Age >14 and Weight < 40 Kilograms or > 150 Kilograms;
[10] BMI > 50 OR Age >14 and BMI <15;
[11] Type 1 and On Insulin Since Year ≥ 3 years after Date of Diagnosis;
[12] Age < 20 and On anti-hypertensive therapy;
[13] Age < 2 [check Date of Birth is correct];
[14] Female with erectile dysfunction;
[15] Stated to be On Ezetrol AND On Vytorin.

Whilst some of these may be possible (eg 7, 8, 9, 12), they most probably represent incorrect data recording or interpretation by the computer system.

[c] Possible Duplicates : (double individual registration / data entry), based on Sex and Date of Birth match – (or on an ID match with significantly different data).

[d] Potentially Incorrect Haemoglobin A1c, Lipid, and Creatinine Values : HbA1c <3.5, Cholesterol <2.0, HDL Cholesterol >2.5, Triglyceride <0.5, Creatinine <50 or >1000.

[e] Visual Acuity - Blindness & Creatinine – Renal Failure Checks : The reported results and status (respectively) were cross-checked to question likely or possible errors (eg V/As <6/60 recorded as Blind or High Creatinine/not ESRD).

Document download

This publication is available as a downloadable document.

Australian National Diabetes Information Audit & Benchmarking (ANDIAB) 2011 Final Report(PDF 559 KB)