Table 3

Examples of gold standard criteria for adult breast cancer, adult acute myocardial infarction, child pneumonia, and neonatal birth asphyxia

Adult breast cancer

Level 1

One of the following:

• Operative specimen with histological confirmation

• Biopsy/fine needle aspiration cytology

Level 2A

Both of the following:

• Mammography diagnosis

• Imaging evidence of metastases in bone, lung, etc. based on CT scan/MRI/X-rays

Level 2B

Patient under treatment from a recognized cancer hospital or cancer unit for breast cancer in cases where the basis for the initial diagnosis is no longer available.

Adult acute myocardial infarction

Level 1

Evidence of acute MI within three months preceding death based upon one or more of the following:

• Cardiac perfusion scan

• ECG changes

• Documented history of CABG or PTCA or stenting

• Coronary angiography

• Enzyme changes (any troponin elevation or CK-MB isoenzyme elevation >2 times the upper limit of normal) in the context of myocardial ischemia

Level 2A

Clinical evidence of the following:

• Sudden death within six hours of the onset of characteristic shock and chest pain when the case has been witnessed by a physician

Child pneumonia

Level 1

Chest X-ray showing primary end-point consolidation, pleural effusion or other consolidation/infiltration, plus two or more of the following:

• Respiratory rate >70/minute

• Severe lower chest indrawing

• Abnormal breath sounds (i.e., grunting, decreased breath sounds, crepitations)

• Rectal temperature >38°C or <36°C

• Oral or axillary temperature >37.5°C or <35.5°C

Neonatal birth asphyxia

Level 1

Each of the following:

• Failure both to breathe spontaneously and to cry at birth

• No major congenital abnormality

• Not a stillbirth (one or more signs of life at birth like pulse or movement)

Plus one of the following in the 24 hours after birth:

• Not feeding

• Hypotonia

• Seizures

• Needed and failed resuscitation at birth

Level 1 is the most stringent criteria, while level 2A or 2B were also collected for some causes.

Murray et al. Population Health Metrics 2011 9:27   doi:10.1186/1478-7954-9-27

Open Data