Skip to main content

Table 1 Validity of breast cancer flags in individual datasets compared with the Cancer Registry, July 2004-December 2008

From: Ascertaining invasive breast cancer cases; the validity of administrative and self-reported data sources in Australia

Data set

Breast cancer flags

Positive predictive value

Sensitivity

Specificity

45 and Up baseline survey1

Self-reported diagnosis of breast cancer 2 within a year of birth-year reported

49.8%

84.7%

99.6%

Admitted Patient Data Collection

Diagnosis3 of invasive breast cancer

85.9%

86.1%

99.8%

Lumpectomy

52.0%

60.7%

99.2%

Mastectomy

70.8%

32.6%

99.8%

Lumpectomy OR mastectomy

56.4%

84.4%

99.1%

Lumpectomy AND diagnosis of invasive breast cancer

89.0%

59.1%

99.9%

Mastectomy AND diagnosis of invasive breast cancer

85.4%

31.8%

99.1%

(Lumpectomy or mastectomy) AND diagnosis of invasive breast cancer

87.7%

82.3%

99.8%

(Lumpectomy or mastectomy) OR diagnosis of invasive breast cancer

56.5%

88.2%

99.0%

Mastectomy OR diagnosis of invasive breast cancer

79.7%

87.6%

99.7%

Lumpectomy OR diagnosis of invasive breast cancer

58.2%

87.6%

99.1%

Medicare Benefits Schedule

Breast radiotherapy

72.8%

57.6%

99.7%

Pharmaceutical Benefits Scheme

Dispensed medicine for breast cancer 4

45.5%

65.4%

98.9%

  1. 1: Comparison to Cancer Registry was restricted to July 2004-December 2005.
  2. 2: Reported cancer with birth-year occurring in or overlapping with the period July 2004-December 2005.
  3. 3: Primary diagnosis field.
  4. 4: Tamoxifen, toremifene, anastrozole, exemestane, letrozole, goserelin, trastuzumab, lapatinib, and medroxyprogesterone 500 mg.