Skip to main content

Table 2 Linkage stages in the PIAC project

From: Empirical aspects of record linkage across multiple data sets using statistical linkage keys: the experience of the PIAC cohort study

Stage

Dataset 1

Dataset 2

Minimum match rate (from one-step deterministic linkage on SLK-581)

Final match rate (from stepwise deterministic linkage)

   

% dataset 1

% dataset 2

% dataset 1

% dataset 2

1

Residential care 2002-06(a)

Community care packages 2002-06(a)

9.4

44.0

10.2

47.7

2

Deaths July 03-Dec 06(a)

RCCP 2002-06(a) (from stage 1)

32.6

36.9

36.2

41.0

3

PIAC cohort (ACAP 2003-04) (b)

RCCP 2002-06 (a)

64.4

16.3

72.6

18.4

4

Deaths July 03-Dec 06 not linked to RCCP (a)

ACAP 2003-04 not linked to RCCP(b)

3.0

31.4

3.3

34.7

5

PIAC cohort(b)

Aged care assessments 2004-05(b)

29.4

21.8

30.9

22.9

6a

PIAC cohort(b)

Home and Community Care 2002-03(b)

41.6

7.1

46.3

7.9

6b

PIAC cohort(b)

Home and Community Care 2003-04(b)

53.4

8.3

58.5

9.1

6c

PIAC cohort(b)

Home and Community Care 2004-05(b)

35.9

5.3

39.5

6.0

6d

PIAC cohort(b)

Home and Community Care 2005-06(b)

22.8

3.4

26.2

3.9

7

PIAC cohort(b)

Veterans' Home Care January 2001 - March 2008(a)

11.4

7.3

12.2

7.8

  1. (a) Clients identified through administrative processes, and de-duplication.
  2. (b) Clients identified by SLK-581 within broad region (1st digit of postcode)
  3. Note: For the datasets with administratively-derived person identifiers, data for all years in the study were linked at the same time. For data sets with clients identified by SLK-581 data were linked for each year separately to allow for variation over time in reported SLK-581.