TECHNICAL NOTE REGISTRATION OF OUTSTANDING DEATHS, QUEENSLAND, 2010


INTRODUCTION

1 In November 2010, the Queensland Registrar of Births, Deaths and Marriages advised the ABS of an outstanding deaths registration initiative undertaken by the Registry. This initiative resulted in the November 2010 registration of 374 previously unregistered deaths which occurred between 1992 and 2006 (including a few for which a date of death was unknown). Of these, around three-quarters (284) were deaths of Aboriginal and Torres Strait Islander Australians.

2 Data in this publication are presented on a year of registration basis. Similarly, mortality indicators presented in various Council of Australian Government (COAG) reports, including the National Indigenous Reform Agreement (NIRA), the National Healthcare Agreement (NHA), the Aboriginal and Torres Strait Islander Health Performance Framework and Overcoming Indigenous Disadvantage reports were compiled using deaths data on a year of registration basis. If this practice is followed for reporting data for the 2010 reference year, mortality indicators for Queensland and any aggregates including Queensland will be overstated and prevent meaningful comparisons over time.

3 This Technical Note summarises the effects of these extra death registrations on selected mortality indicators. A method is outlined to adjust death registrations for Queensland for 2010 that aims to minimise the impact on mortality indicators used in various government reports. This is followed by analysis comparing mortality rates by Indigenous status based on deaths data by: (1) year of registration; (2) year of occurrence and; (3) 'adjusted registration'.


DEATHS BY YEAR OF REGISTRATION AND YEAR OF OCCURRENCE

4 The following table presents deaths of Aboriginal and Torres Strait Islanders by year of registration and year of occurrence separately for those deaths registered up to and including 31 December 2010. The number of deaths by year of occurrence in this series are subject to change, as deaths that occurred up to and including 31 December 2010 but which have not yet been registered, are not included. See paragraphs 26 and 27 of the Explanatory Notes for more information.

5 While the interval between the occurrence and registration of deaths is often longer among Aboriginal and Torres Strait Islander Australians than among the non-Indigenous population, there is normally little difference between the number of deaths registered in a given year and the number of deaths that occurred in the same year for Aboriginal and Torres Strait Islander Australians. This is because for each year, the number of deaths that are not registered in the year that they occur are compensated by deaths that occurred in previous years but were subsequently registered. The table below illustrates that if data are routinely reported on a year of occurrence basis, deaths data for the latest year (and the year prior to the latest year) will be underestimated.

Aboriginal and Torres Strait Islander deaths(a), States and territories(b)(c) - 1994 to 2010
	NSW	Vic.	Qld(d)	SA	WA(e)	Tas.	NT	ACT	Aust.(f)
Year of registration
1994	207	50	np	123	377	np	380	10	1 153
1995	224	50	np	121	384	np	387	9	1 182
1996	177	49	258	118	370	np	328	np	1 306
1997	88	93	531	132	351	5	458	4	1 662
1998	462	123	593	127	378	13	415	3	2 114
1999	435	130	529	116	350	11	399	6	1 976
2000	473	108	535	144	407	np	450	np	2 127
2001	481	93	565	125	345	np	429	np	2 072
2002	516	64	590	107	371	20	462	4	2 136
2003	485	82	569	137	338	23	435	9	2 079
2004	490	54	579	131	400	20	449	10	2 136
2005	507	71	519	142	406	28	454	11	2 141
2006	530	111	584	124	443	20	452	14	2 279
2007	601	95	594	138	502	24	461	6	2 421
2008	559	97	562	141	605	24	467	16	2 472
2009	591	106	632	160	444	30	431	10	2 405
2010	622	117	948	147	436	37	447	13	2 767
Year of occurrence(g)
1994	197	52	np	121	380	np	401	10	1 175
1995	234	49	np	117	371	np	370	10	1 180
1996	166	67	350	122	351	np	383	np	1 444
1997	98	106	567	137	382	5	408	3	1 706
1998	465	108	595	129	389	13	414	3	2 116
1999	433	105	540	117	347	12	438	7	1 999
2000	480	115	619	140	384	np	428	np	2 177
2001	487	81	581	122	368	np	430	np	2 100
2002	508	64	613	108	384	22	470	5	2 176
2003	494	83	594	144	323	22	418	12	2 092
2004	481	54	629	134	391	19	437	7	2 154
2005	507	80	541	140	416	32	493	12	2 224
2006	529	117	587	123	466	16	453	12	2 304
2007	599	94	583	138	513	25	461	8	2 421
2008	559	93	552	143	580	23	483	14	2 448
2009	594	105	623	145	419	33	426	12	2 358
2010	593	101	574	142	355	34	340	11	2 150
(a) From 2007 onwards, Indigenous status for deaths registered in Victoria, South Australia, Western Australia, Tasmania, the Northern Territory and the Australian Capital Territory is sourced from both the Death Registration Form and Medical Certificate of Cause of Death.
(b) State or territory of usual residence.
(c) Due to differing levels of recording Indigenous status by the states and territories and over time, care should be taken in interpreting change in the number of deaths. As a result, data for Australia should not be analysed as a time series.
(d) Queensland began to register Aboriginal and Torres Strait Islander deaths as such in 1996. Care should be taken when interpreting Aboriginal and Torres Strait Islander deaths data for Queensland for 2010 (as per this Technical Note).
(e) ABS are currently investigating the volatility of Aboriginal and Torres Strait Islander deaths in WA in recent years. Until this investigation is finalised, ABS advises caution be used when analysing Aboriginal and Torres Strait Islander deaths data for 2007, 2008 and 2009.
(f) Includes Other Territories.
(g) Includes deaths registered up to and including 31 December 2010. The number of deaths presented on a year of occurrence basis are subject to change, as deaths that occurred prior to 31 December 2010 but have not yet been registered are registered in subsequent years. See paragraphs 26 and 27 of the Explanatory Notes for more information.

DEATH REGISTRATIONS BY YEAR OF OCCURRENCE

6 There were 27,289 deaths registered in 2010 of usual residents of Queensland. All occurred since 1991. Of these deaths, 381 occurred between 1992 and 2006, and a majority of these were registered through the initiative to register outstanding deaths. Of the Aboriginal and Torres Strait Islander deaths registered in Queensland in 2010, 30% (289) occurred between 1992 and 2006. By contrast, less than 1% (31) of non-Indigenous deaths registered in 2010 occurred between 1992 and 2006.

Death registrations, Year of occurrence and Indigenous status - Queensland(a) - 2010
	Aboriginal and Torres Strait Islander	Non-Indigenous	Total(b)
Year of occurrence	persons	persons	persons
1991 and earlier	-	-	-
1992-2006(c)	289	31	381
2007	9	3	13
2008	12	5	18
2009	64	1 220	1 360
2010	574	24 389	25 512
Total(d)	948	25 652	27 289
- nil or rounded to zero (including null cells)
(a) Deaths of usual residents of Queensland.
(b) Includes deaths for which Indigenous status was not stated.
(c) A majority of the deaths that occurred between 1992 and 2006 were registered as part of the initative to register outstanding deaths.
(d) Includes deaths for which year of occurrence was not stated.

ADJUSTMENT OF DEATH REGISTRATIONS

7 The ABS has discussed different methods of adjustment of Queensland death registrations data for 2010 with key stakeholders who are involved in analysis and reporting of COAG 'Closing the Gap' mortality indicators. Following analysis of several methods of adjustment, a paper was presented and discussed at a National Indigenous Reform Agreement Performance Indicators Management Group (NIRAPIMG) meeting on 18 October 2011. Following further analysis, a decision was made by the ABS and key stakeholders to use an adjustment method that added together deaths registered in 2010 for usual residents of Queensland which occurred in 2007, 2008, 2009 and 2010. The method is simple to describe, apply and interpret.

8 Deaths data for 2010 for Aboriginal and Torres Strait Islander, non-Indigenous and total populations have been adjusted. The 'adjusted registrations' for Queensland for 2010 constitute 659 Aboriginal and Torres Strait Islander deaths, 25,621 non-Indigenous deaths and 26,908 total population deaths (which includes deaths for which Indigenous status was not stated).


IMPACT ON AGE-SPECIFIC DEATH RATES

9 The following table presents age-specific death rates (ASDRs) based on year of registration, adjusted registration and year of occurrence for Queensland. In almost all cases, ASDRs based on adjusted registrations fall between rates based on year of occurrence and rates based on year of registration.

Age-specific death rates, Indigenous status - Queensland - 2006-2010
	Registration			Adjusted registration(a)			Occurrence
	Aboriginal and Torres Strait Islander	Non-Indigenous	Total(b)	Aboriginal and Torres Strait Islander	Non-Indigenous	Total(b)	Aboriginal and Torres Strait Islander	Non-Indigenous	Total(b)
	rate	rate	rate	rate	rate	rate	rate	rate	rate
0(c)	8.8	4.6	5.1	8.3	4.6	5.0	8.0	4.5	4.9
1-4	2.8	1.1	1.3	2.5	1.1	1.2	2.3	1.1	1.2
0-4	12.1	5.9	6.5	11.2	5.8	6.4	10.8	5.7	6.3
5-9	1.1	0.5	0.6	1.0	0.5	0.6	1.1	0.5	0.6
10-14	0.8	0.5	0.5	0.8	0.5	0.5	0.7	0.5	0.5
15-19	3.8	1.8	2.0	3.2	1.8	1.9	3.2	1.7	1.9
20-24	6.6	2.4	2.7	6.0	2.4	2.7	5.7	2.4	2.6
25-29	8.3	2.7	3.0	7.7	2.7	3.0	7.1	2.7	2.9
30-34	11.1	3.4	3.8	9.7	3.4	3.7	9.2	3.3	3.6
35-39	17.0	4.1	4.8	15.9	4.1	4.7	15.0	4.0	4.6
40-44	24.3	5.6	6.4	22.6	5.6	6.3	21.4	5.5	6.2
45-49	32.6	8.5	9.4	29.6	8.5	9.3	28.6	8.4	9.2
50-54	47.4	13.1	14.2	43.0	13.1	14.1	41.1	12.9	13.8
55-59	70.0	20.0	21.3	64.5	20.0	21.2	61.8	19.6	20.8
60-64	93.9	31.8	33.4	86.8	31.8	33.3	82.1	31.3	32.7
65-69	154.4	54.3	56.6	140.5	54.3	56.5	136.7	53.4	55.5
70-74	244.5	91.2	94.6	218.3	91.2	94.3	211.4	90.1	93.1
75 and over	471.2	333.6	340.6	428.9	333.6	340.3	422.4	329.1	335.7
Total(d)	21.8	30.1	30.5	19.9	30.1	30.4	19.1	29.7	29.9
(a) Deaths by year of registration used for 2006 to 2009, and adjusted registrations for 2010.
(b) Includes deaths for which Indigenous status was not stated.
(c) Infant deaths per 1,000 live births. Births by year of registration are used as the denominator.
(d) Includes deaths for which age at death was not stated.

IMPACT ON INFANT MORTALITY RATES

10 In the following graph, infant mortality rates (IMRs) for Queensland for the period 2008-2010 are highest for deaths on a year of registration basis and lowest for deaths on a year of occurrence basis (due to time lags in registrations of both deaths and births). IMRs based on adjusted registration fall between the IMRs based on year of registration and year of occurrence data. The difference in rates based on year of registration, year of occurrence or adjusted registration data are greater for the Aboriginal and Torres Strait Islander population than for the non-Indigenous or the total population. This is shown by the ratio of adjusted registration IMRs to year of registration IMRs for Aboriginal and Torres Strait Islander Australians of 0.92 compared with 0.99 for non-Indigenous Australians and 0.98 for the total population.

11 The following table presents IMRs for Queensland and aggregated New South Wales, Queensland, South Australia and the Northern Territory for 2008-2010. The patterns of IMRs are similar for aggregated New South Wales, Queensland, South Australia and the Northern Territory data. However, the effect of using adjusted registration data instead of year of registration or year of occurrence data for IMRs is weaker.

Infant mortality rates(a), Indigenous status - 2008-2010
	Aboriginal and Torres Strait Islander	Non-Indigenous	Total(b)
	rate	rate	rate
Queensland
Registration	8.4	4.6	5.1
Adjusted registration(c)	7.7	4.6	5.0
Occurrence	7.0	4.4	4.8
Total(d)
Registration	7.4	4.2	4.5
Adjusted registration(c)	7.1	4.2	4.5
Occurrence	6.6	4.1	4.3
(a) Infant deaths per 1,000 live births. Births by year of registration used as the denominator.
(b) Includes deaths for which Indigenous status was not stated.
(c) Deaths by year of registration used for 2008 and 2009, and adjusted registrations for 2010.
(d) Data for NSW, Qld, SA and NT combined, based on state and territory of usual residence. Victoria, Tasmania and the ACT are excluded due to small numbers of registered Aboriginal and Torres Strait Islander deaths. WA data are excluded due to ongoing investigation into unusual fluctuations. See paragraph 36 of the Explanatory Notes.

IMPACT ON STANDARDISED DEATH RATES

12 The next graph presents standardised death rates (SDRs) using year of registration, year of occurrence and adjusted registration data for 2010. As with IMRs, SDRs are highest for year of registration data, lowest for year of occurrence data (due to registration lags) and SDRs based on adjusted registration data fall between those based on year of registration and those based on year of occurrence data. The difference in rates based on year of registration, year of occurrence or adjusted registration data are greater for the Aboriginal and Torres Strait Islander population than for the non-Indigenous or the total population. This is shown by the ratio of adjusted registration SDRs to year of registration SDRs for Aboriginal and Torres Strait Islander Australians of 0.68 compared with 1.00 for non-Indigenous Australians and 0.99 for the total population.

13 The following table presents SDRs for Queensland and aggregated New South Wales, Queensland, South Australia, Western Australia and the Northern Territory for 2010. In contrast to patterns for IMR, the SDR using adjusted registrations for the Aboriginal and Torres Strait Islander population is higher (11.3) for aggregated New South Wales, Queensland, South Australia, Western Australia and the Northern Territory than for Queensland (11.0). Whereas, in keeping with patterns for IMRs, the SDR for year of registration is lower (12.8) for aggregated New South Wales, Queensland, South Australia, Western Australia and the Northern Territory than for Queensland (16.0). In both geographic aggregates, SDRs for the non-Indigenous population are identical, and very similar for the total population.

Standardised Death Rates(a), Indigenous status - 2010
	Aboriginal and Torres Strait Islander	Non-Indigenous	Total(b)
	rate	rate	rate
Queensland
Registration	16.0	5.8	6.1
Adjusted registration	11.0	5.8	6.0
Occurrence	9.6	5.5	5.7
Total(c)
Registration	12.8	5.8	6.0
Adjusted registration	11.3	5.8	6.0
Occurrence	10.0	5.5	5.7
(a) Deaths per 1,000 standard population. Standardised death rates are calculated using the total persons in the Australian population at 30 June 2001 as the standard population.
(b) Includes deaths for which Indigenous status was not stated.
(c) Data are for NSW, Qld, SA, WA, and NT combined, based on state or territory of usual residence. Victoria, Tasmania and the ACT are excluded due to small numbers of registered Aboriginal and Torres Strait Islander deaths.

14 The next graph shows the effects of the registration of outstanding deaths on standardised death rates (SDRs) for Aboriginal and Torres Strait Islander Australians in Queensland for the period 2006 to 2010. The inclusion of deaths from earlier years in the 2010 registrations produces an implausibly high SDR in 2010. Conversely, the 2010 year of occurrence based SDR is an underestimate due to a lag in registration of deaths in the year of occurrence (that is, deaths which occurred in 2010 but were not registered until 2011 or later years). Importantly, the adjusted registrations data for 2010 lead to an SDR for 2010 which is consistent with year of registration based SDRs for earlier years.

CONCLUSION

15 More than three-quarters of deaths registered as part of the outstanding deaths initiative were deaths of Aboriginal and Torres Strait Islander Australians. This has impacted on mortality rates of Aboriginal and Torres Strait Islanders more than those of the non-Indigenous population. Using 2010 death registrations data for Queensland, and aggregates including Queensland, overstates mortality rates and prevents meaningful comparisons over time. Reporting using year of occurrence data is not an option since the latest year is understated due to registration lags (see paragraphs 26 and 27 of the Explanatory Notes for more information), and earlier years would be subject to annual and potentially ongoing revisions. Standardised death rates based on adjusted death registration data for Queensland for 2010 maintain a level trend with registrations data from earlier years. As a result, adjusted registration data for Queensland for 2010 are recommended for reporting mortality indicators over time.

16 In addition, the registration in 2011 or later years of any substantial number of additional outstanding deaths which occurred in earlier years may require further adjustments to compile mortality indicators for use in meaningful analysis of mortality trends over time.