New Zealand Census-Mortality Study WebTable Results

Welcome to the New Zealand Census-Mortality Study WebTable

Please Note : Any result that is also in the newly released Tracking Disparities report has been checked that the results are correct (especially Income and Total Ethnicity results). However please treat all other results as provisional at this stage (especially Labour Force Status and Occupational Class 1991 results). Please email June Atkinson if you find any strange results or files or things that are not working correctly.

Overview

The New Zealand Census Mortality Study (NZCMS) is a record linkage study of census and mortality records. The study was initially funded by the Health Research Council of New Zealand, and is now funded by the Ministry of Health as a joint project between the University of Otago and the Ministry of Health. The NZCMS is conducted in collaboration with Statistics New Zealand.

The most recent NZCMS is made up of five cohorts. The NZCMS has linked mortality records for the three years following each census back to 1981. A follow up period of three-years was selected, as the linkage of census data with mortality data is dependent on residential stability. Limiting the follow up period also allows for timely reporting on mortality inequalities.

The information in this WebTable allows policy and research communities to analyse inequalities in standardised mortality rates and relative rates (rate ratios and rate differences). Standardised mortality rates are calculated for males and females in the 1-74 year age group as well as age groups within this range. Rates be analysed for all ethnic groups as well as specific ethnic groups (including Maori, Pacific, Asian and Non-Maori/Non-Pacific/Non-Asian) and across a number of exposure variables including various indicators of socio-economic position. The NZCMS WebTable also permits analysis of specific causes of death.

The information is freely available and all tables and figures are downloadable. NZCMS WebTable should be acknowledged as the source when WebTable information is being quoted (refer to citation format below). Please contact the School of Medicine and Health Sciences if information from this WebTable makes up more than 10% of your publication.

We have attempted to report as many results as possible. Please note that a number of results have wide confidence intervals, caution should be applied when using these. Results have not been made available where there is insufficient data (i.e. too few deaths).

The WebTable is also available on CD-ROM. Please contact the School of Medicine if you prefer this method of access.

The following background information is provided to assist you in using the NZCMS Web Table. It is advised you read this before accessing results.

• Statistics New Zealand Security Statement
• Variables Available for Analysis
• Cause of Death
• Exposure Variables
• All Exposures
• Measures of socio-economic position
• Measures of Ethnicity
• Ethnic Groups
• Age Groups
• Initial Restrictions on the Data for Analysis
• Criteria for Reporting
• Notes on the Methods
• Contact Information
• Citation Format

A list of NZCMS publications can be accessed from the NZCMS Publications page

^ Top of page

---

Statistics New Zealand Security Statement

The New Zealand Census – Mortality Study is a study of the relationship between social factors and mortality in New Zealand, based on the integration of anonymised population census data from Statistics New Zealand and mortality data from the New Zealand Health Information Service.

The project was approved by Statistics New Zealand as a Data Laboratory project under the Microdata Access Protocols in 1997. The data sets created by the integration process are covered by the Statistics Act and can be used for statistical purposes only. Only approved researchers who have signed Statistics New Zealand’s declaration of secrecy can access the integrated data in the Data Laboratory. For further information about confidentiality matters in regard to this study, please contact Statistics New Zealand.

(The full security statement is available at http://www.wnmeds.ac.nz/academic/dph/research/HIRP/nzcms/nzcmssnz.html).

^ Top of page

---

Variables Available for Analysis

Cause of Death

Cause of Death	ICD-9	ICD-10
All-Cause	All Codes	All Codes
Cardiovascular disease	410-414, 393-409, 415-459	I20-I25, I05-I15, I26-I99
IHD	410-414	I20-I25
Stroke	430-438	I60-I69
Cancer	140-209	C00-C97
Lung	162	C33-C34
Colorectal	153-154	C18-C21
Breast	174	C50
Prostate	185	C61
Non-Lung	140-161, 163-209
Stomach	151
Pancreas	157
Melanoma	172
Brain/Nervous System	191
Respiratory	470-478, 490-519
Chronic Lung Disease	490-493, 495-496
COPD	490-492, 495-496
Asthma	493
Diabetes	250	E10-E14
Infectious Diseases	001-139, 320-323, 390-392, 460-466, 480-487, 590, 595, 614-616, 680-686, 711, 771
Unintentional injury	800-949	V01-X59
Road traffic crash	810-825	V01-V99
Other	800-809, 826-949
Suicide	950-959, 980-989	X60-X84
External Injuries	800-999

Exposure Variables

All Exposures

1. All Exposures
   The user may chose to examine overall mortality rates opposed to mortality across a socio-economic or ethnicity variable.




Measures of socio-economic position

1. Household equivalent income

The revised Jensen index (Blakely 2002; Jensen 1988) has been used to equivalise household incomes in this study. Incomes have also been adjusted for inflation using the Consumer Price Index (base year 1996). Please refer to the Decades of Disparity II publication for details.
Three levels:
1. Low Income (less than $26,109)
2. Medium Income (between $26,109 and $43,016)
3. High Income (over $43,106). (Reference group)


2. Education

Highest personal education qualification obtained at time of Census.
Three levels:
1. No Qualifications
2. School Qualifications
3. Post-School Qualifications. (Reference group)
Analyses restricted to those 25 years old and over.


3. New Zealand Index of Deprivation

The New Zealand Index of Deprivation (a measure of neighbourhood deprivation) has been assigned using the place of usual residence. The deprivation index is provided for the most recent three cohorts only.
Five levels:
1. Deprivation levels 1 and 2 (Reference group)
2. Deprivation levels 3 and 4
3. Deprivation levels 5 and 6
4. Deprivation levels 7 and 8
5. Deprivation levels 9 and 10.


4. Labour Force Status

Labour force status at time of Census.
Three levels:
1. Employed (Reference group)
2. Unemployed (but seeking work)
3. Non Labour Force (not employed and not seeking work).
Analyses restricted to those 25 years old and over.


5. Occupational Class

Occupational class has been assigned to people employed at the time of the Census using the New Zealand Standard Classifications of Occupation 1968 (NZSCO68) Elley-Irving.
Four levels:
1. Occupational Classes 1 and 2 (Reference group)
2. Occupational Class 3
3. Occupational Class 4
4. Occupational Classes 5 and 6.
Analyses restricted to those 25 years old and over.


6. Car Access

Access to a car in the household.
Three levels:
1. 0 cars available
2. 1 car available
3. 2 or more cars available. (Reference group)


7. Marital Status

Marital Status at time of Census.
Three levels:
1. Never Married
2. Currently Married
(Reference group)
3. Separated, Divorced, or Widowed.




Measures of Ethnicity

The WebTable supports five alternative classifications of ethnicity but the preferred measure is Total Ethnicity.

The treatment of non-response to the Census ethnicity question differs from previous NZCMS analyses. Where individuals have not specified their ethnic affiliation they have been excluded from the NZCMS ethnicity analyses. Previously non-respondents were included in the European/Other group.


1. Total Ethnicity

Total ethnicity is the new statistical standard for reporting ethnicity (Statistics New Zealand 2005). The new standard rejects the notion of prioritising one ethnicity over others for people with multiple identities in favour of a ‘total response’ concept.
Four levels:
1. Maori
2. Pacific
3. Asian
4. Non-Maori non-Pacific non-Asian (also known as European/Other). (Reference group)


2. Maori/Non-Maori

This classification uses the Total ethnicity standard (i.e. anyone who chose Maori as one of their ethnic affiliations is classified as Maori, the remainder are Non-Maori).
Two levels:
1. Maori
2. Non-Maori. (Reference group)


3. Pacific/Non-Pacific

This classification uses the Total ethnicity standard (i.e. anyone who chose Pacific as one of their ethnic affiliations is classified as Pacific, the remainder are Non-Pacific).
Two levels:
1. Pacific
2. Non-Pacific. (Reference group)


4. Asian/Non-Asian

This classification uses the Total ethnicity standard (i.e. anyone who chose Asian as one of their ethnic affiliations is classified as Asian, the remainder are Non-Asian).
Two levels:
1. Asian
2. Non-Asian. (Reference group)


5. Prioritised Ethnicity

For historical reasons, prioritised ethnicity is also available on the WebTable. Using this method Maori has priority coding (i.e. anyone who chose Maori as one of the ethnic groups to which they belong is classified as Maori), of the remainder, anyone who chose any Pacific group is classified as Pacific, of the remainder anyone who chose any Asian group is classified as Asian. The remainder are classified as non-Maori non-Pacific non-Asian.
Four levels:
1. Maori
2. Pacific
3. Asian
4. Non-Maori non-Pacific non-Asian (also known as European/Other). (Reference group)

Ethnic Groups

A single ethnic group can be examined (opposed to selecting ethnicity as an exposure variable thus performing analysis across ethnic groups). Single ethnic groups are defined using the new statistical standard of Total Ethnicity (Statistics New Zealand 2005). The new standard rejects the notion of prioritising one ethnicity over others for people with multiple identities in favour of a ‘total response’ concept.

The treatment of non-response to the Census ethnicity question differs from previous NZCMS analyses. Where individuals have not specified their ethnic affiliation they have been excluded from the NZCMS ethnicity analyses. Previously non-respondents were included in the European/Other group.
Five Groupings:
1. All Ethnicities
2. Total Maori
3. Total Pacific
4. Total Asian
5. Non-Maori non-Pacific non-Asian (also known as European/Other).



Age Groups

Standardised mortality rates are calculated (using the WHO World Population as the reference) for males and females in the 1-74 year age group as well as age groups within this range.
Nine Groupings:
1. 1-74 year olds
2. 25-74 year olds
3. 1-14 year olds
4. 15-24 year olds
5. 25-44 year olds
6. 45-64 year olds
7. 65-74 year olds
8. 25-59 year olds
9. 60-74 year olds.

Note that not all analyses are relevant for all age groups. For example, education, labour force status, occupational class, and marital status analyses are not relevant for 1-14 year olds or 15-24 year olds.

Sex

All analyses are performed separately for males and females. The exceptions being causes of death prostate cancer and breast cancer that are relevant to males only and females only respectively.

Cohorts

Five cohorts have been analysed: 1981-84, 1986-89, 1991-94, 1996-99 and 2001-04.



^ Top of page

---

Initial Restrictions on the Data for Analysis

Mortality rates reported relate to deaths at ages 1-74 only. Infant mortality rates are not reported as the NZCMS consists of closed cohorts (i.e., no newborns enter the study after census night). At the other end of the age spectrum, the first four NZCMS cohorts did not include people aged 75 years and older on census night. Previously people aged 74 on census night were followed up for mortality till age 77; however in this update we cutoff ‘cleanly’ at the 75th birthday.

^ Top of page

---

Criteria for Reporting

We have endeavoured to report as many valid results as possible. Criteria have been applied as follows:

Tables and graphs have been made available where there is sufficient valid data. ‘Sufficient’ has been defined as having a low number of cells in a table where the standardised mortality rate has been derived from less than ten deaths. The exact number of allowable cells with less than ten deaths is dependent on the analysis variable selected and the related number of levels. The number of allowable cells with a low death count is never more than four when results are reported for both males and females. The symbol ‘.’ is used to signify an invalid result in a table, while in the graphs bars are missing.

Results have been reported for both males and females where there is sufficient valid data. Where only male or only female results are valid one sex has been reported.

There are a few instances where mortality rates for a particular level of analysis variable have been omitted while remaining levels have been reported. This typically occurs when the variable being analysed is total ethnicity or prioritised ethnicity and the level omitted is Asian.

Education, labour force status, occupational class and marital status results have only been reported for those 25 years and over.

Breast cancer has only been reported for females. (The few cases of breast cancer for males have been omitted from these analyses.) Similarly, prostate cancer has only been reported for males.

We have attempted to report as many results as possible. Please note that a number of results have wide confidence intervals, caution should be applied when using these.

^ Top of page

---

Notes on the Methods

The analyses shown have all been adjusted for linkage bias using the age and ethnicity weighting (see NZCMS publications for full explanation).

All the data has been age standardised using the WHO World Population as the reference.

Results for 'All Ethnicities' combined have been age and ethnicity standardised. Results by ethnicity have been age standardised only.

Standardised Mortality Rates (SR) have been calculated for each level of analysis variable.

Standardised Rate Ratios (SRR) have been calculated against the relevant reference level.

Standardised Rate Differences (SRD) have been calculated against the relevant reference level.

For each of SR, SRR or SRD 95% confidence intervals have been calculated.

Where an analysis variable other than ‘All Exposures’ has been selected the Rich Text Format (RTF) files consist of three pages of tables and graphs, one page each for SR, SRR and SRD results. Where ‘All Exposures’ has been selected as the analysis variable the RTF file is one page reporting SR only.

The RTF graphs can be selected and put into a presentation package (e.g. MS PowerPoint). If it is necessary to increase the size of the image, we recommend keeping the aspect ratio constant to avoid distortion of the image.

^ Top of page

---

Contact Information

Any queries, comments or suggested enhancements to these web tables can be emailed to June Atkinson or Tony Blakely, School of Medicine and Health Sciences, or Martin Tobias, Ministry of Health or written to:

June Atkinson
Department of Public Health
School of Medicine & Health Sciences
University of Otago, Wellington
PO Box 7343
Wellington South
New Zealand


Tel 64 4 385 5541 (ext 6085)
Fax 64 4 389 5319
Email June.Atkinson@otago.ac.nz


^ Top of page

---

Citation Format

The following is a suggested format for citing our WebTable system:

Atkinson J., Bastiampillai N. New Zealand Census-Mortality Study WebTable. Department of Public Health, School of Medicine and Health Sciences, University of Otago, Wellington. http://www.otago.ac.nz/NZCMSWebTable. [Date Accessed]

^ Top of page

^ Top of page