The Central Statistics Office (CSO).

Vital Statistics Section

Central Statistics Office,
Skehard Road,
Mahon,
Cork
T12X00E

Data on causes of death (CoD) provide information on mortality patterns and form a major element of public health information.

CoD data refer to the underlying cause which - according to the World Health Organisation (WHO) - is "the disease or injury which initiated the train of morbid events leading directly to death, or the circumstances of the accident or violence which produced the fatal injury".

CoD data are derived from death certificates. The information provided in the medical certificate of cause of death is mapped to the International Statistical Classification of Diseases and Related Health Problems (ICD).

Eurostat's CoD statistics build on standards set out by the World Health Organisation (WHO) in the International Statistical Classification of Diseases and Related Health Problems (ICD).

The regional breakdown is based on the Nomenclature of Territorial Units for Statistics (NUTS 2).

Classification and updates applied by years

Public Health.

Concepts and definitions are described in the Commission regulation (EU) No 328/2011 in articles 2 and 3.

From 2007 onwards, the population concept of usual residence has been used, i.e. all persons usually resident and present in the State on census night, plus absent persons who are usually resident in Ireland, but are temporarily away from home and outside the State on Census night. All persons are classified according to the region of their usual residence.

Stillbirth weighing 500 grams or more or at gestational age of 24 weeks or more.

The statistical units are the deceased persons and the stillborns, respectively.

All deaths that occur in the State including those visiting the country at time of death but are residing outside of the State. - Irish residents that die abroad are not included in the population.

Neonates of non-resident mothers are included in the published data.

Non-residents that die in the State are included in the mortality statistics.

Irish residents dying abroad are not included in the statistics. Only deaths that occurred in the State are included in the population.

The Irish State (26 counties). 

Northern and Western (Border and West), Southern (Mid-West, South-East and South-West) and Eastern and Midland (Dublin, Mid-East and Midland).

Ireland is providing  COD data to Eurostat in line with EU regulation for the years 2011 (causes of death (EU) No 328/2011) onwards.  National publications adopted ICD-10 from 2007 et seq.

Not applicable.

The unit is number.

Vital Statstics Annual Report 2021.

The Central Statistics Office (CSO) prepares releases and publications for the Minister for Social Protection and Employment Affairs in accordance with the provisions of Section 2 of the Vital Statistics and Births, Deaths and Marriages Registration Act 1952 and the Government Order(SI 831 of 2007).   Supplementary information collected on Form 104 in respect of external causes of death are collected under the Statistics Act 1993.

CoD data was submitted to Eurostat on the basis of a gentleman's agreement established in the framework Eurostat's Working Group on "Public Health Statistics" until data with reference year 2010.

A Regulation on Community statistics on public health and health and safety at work (EC) No 1338/2008 was signed by the European Parliament and the Council on 16 December 2008. This Regulation is the framework of the data collection on the domain.

Within the context of this framework Regulation, a Regulation on Community statistics on public health and health and safety at work, as regards statistics on causes of death (EU) No 328/2011 was signed by the European Parliament and the Council on 5 April 2011. 

CoD data according to this regulation is submitted to Eurostat since reference year 2011.

Researchers can apply to access mortality microdata by completing a Research Microdata Application Form and other relevant forms.  Very strict procedures/criteria are in place to protect the confidentiality/security of data.  Only applicants that meet this strict criteria are conferred with Officer of Statistics status.  No microdata can be accessed from outside of the State.  All researcher outputs must be approved by the relevant Statistician before being released to the researcher.

Ireland has provided data to the World Health Otganisation (WHO) for some years previously.

The CSO suppresses any sensitive data to ensure that there is no data published whereby an individual can be identified in the data.  As a general rule location, gender and individual ages are not published together in a single table. Data is aggregated and for example  if there are single units reported than the underlying cause of death is included in a range and is not a specific ucod code or age is reported within an age range or the gender isn't identifiable.  The same applies to information disseminated in response to queries.

With regard to micro-date, strict criteria/protocols are in place.  Researchers must become Officers of Statistics before access to microdata is granted. Furthermore, mortality data cannot be merged to any other dataset.  All outputs must be approved by the relevenat Statistician before outputs are released to researchers.

Care is taken to ensure that the data disseminated by the CSO is not explosive where age, location and gender coupled with the core variable of interest doesn't allow the identiity of the deceased when releasing sensitive cause of death data.  For example in annual report where the number of deaths reported are less than 3 by general location and sex then the underlying cause is reported as a range e.g. R00-R99 or V01-Y89

Age-group, location, sex and specific cause of death code are never disseminated together in one table.  For example if there is a single unit reported in 15-24 age group, sex is male and cause of death range is A00-B99 and no location is provided, then the identity of the deceased is suppressed.

For stillbirth and neonatal figures, no breakdown by parity is displayed to ensure confidentiality.

When dealing with specific queries, the one age-group is aggregated with another age-group if necessary in order to ensure primary and secondary confidentiality.

The CSO disseminated the underlying cause of death data for 2021 based on the date the death occurred (final data) on the 31st October 2023 (i.e. T+22 months).  Provisional mortality data is published on a quarterly basis.  The most recent quarterly data i.e. Quarter 2 2023 was published on the 30th November 2023 and is based on the date of registration of the death.

The situation in Ireland is that the next-of-kin has 3 months from the date of death in which to register the death.  Also, all unnatural deaths are referred to the Coroners Office and some cases can take approximately 2 years to finalise and for the death to be subsequently registered by that office.

See link to the CSO calendar:  https://www.cso.ie/en/csolatestnews/releasecalendar/

The release calendar can be accessed via the CSO website www.cso.ie.

Then going to the Release Calendar - see link below:

https://www.cso.ie/en/csolatestnews/releasecalendar/

All releases and publications are scheduled on the CSO Release Calendar at least six months in advance.  Information is disseminated to all customers/users at the same time.  A Press Release, selecting key facts contained in the Release/Publication, are presented in the Press Release.  Other forms of social media are used e.g. twitter to flag the relevant Release/Publication.  Key users/organisations are also provided with details of the Release/Publication as the same time as it is disseminated to all.

All of the vital statistics reports are downloaded by the user in electronic format.

Quarterly vital statistics releases are disseminated on a quarterly basis and a summary of the four quarters is disseminated at the end of May each year.  The data is generally published within five months from the end of the relevant quater.  The registered data is regarded as provisional data.

The Vital Statistics Annual Report is disseminated within 22 months from the end of the relevant period, for example the report for 2020 was published on the 28/10/2022 while the 2021 report was published on 31/10/2023. The annual data is regarded as final data.

Life Tables, each cycle using three years of mortality data around the Census Year. is published every five years.  The last analysis was calculated based on the mortality (final) data for the years 2015, 2016 and 2017 and used census 2016 data. The most recent iteration of Life Tables was published in June 2020.

A press release was issued which each of the following releases/publications:

Summary of the vital statics quarterly releases 2021

https://www.cso.ie/en/releasesandpublications/ep/p-vsys/vitalstatisticsyearlysummary2021/

https://www.cso.ie/en/releasesandpublications/ep/p-vsys/vitalstatisticsyearlysummary2022/

Irish Life Tables 2015-2017

https://www.cso.ie/en/csolatestnews/pressreleases/2020pressreleases/pressstatementirishlifetables2015-2017/

Quarter 2 2023

https://www.cso.ie/en/releasesandpublications/ep/p-vs/vitalstatisticssecondquarter2023/

Vital Statistics Yearly Summary 2021 - CSO - Central Statistics Office

Vital Statistics Annual Report 2021

https://www.cso.ie/en/releasesandpublications/ep/p-vsar/vitalstatisticsannualreport2021/

 All releases and publications can be downloaded in electronic format from the CSO website - see link below:

 https://www.cso.ie/en/statistics/birthsdeathsandmarriages/

Not Available

The mortality data can be accessed on line using the CSO Statbank.  See link below:

https://www.cso.ie/en/databases/

Some social media used.

Not available.

For broad overview of the methodology please see the background notes in the link https://www.cso.ie/en/releasesandpublications/ep/p-vsar/vitalstatisticsannualreport2021/backgroundnotes/

Terms used in relation to stillbirths and infant mortality are defined in the report (chapter 4 for stillbirths and infant mortality).

A quality report is also available on the CSO website, see link below:

https://www.cso.ie/en/media/csoie/methods/birthsdeathsandmarriages/Quality_Report_-_Deaths.pdf

100%

Quality Report on mortality statistics on the CSO website.

The causes of death data are based on a regulation, which defines scope, definitions of variables and characteristics of the data.

Cause of Death coding is very complicated and has a subjective element, so errors can also occur here. However, the editing process will identify where the deceased has been assigned an underlying cause of death inappropriate to their age or sex. 

There is ongoing training by senior mortality coders.  Coders aim to keep abreast of all the major updates by the WHO.  The IRIS software version 5.6 includes the WHO major and minor updates in the automatic coding.   Our senior coders contribute regularly to the international mortality forum and are continuously networking and building solid relationships with COD experts in other EU member states and further afield e.g. Canada and Australia.

Difficult cases are discussed by the coders to achieve consensus on arriving at the most suitable cause of death code.

The Quality of the data is very good. The CSO uses IRIS software for cause of death coding - there is around 55%-'58% automatic  coding done in IRIS each week.   The balance iof the coding including all external causes of death are coded manually by the Mortality Coders in the CSO.  The Statistician has several edits run in SAS to highlight partiular situations e.g. women that have been coded to testicular cancer.  There are also many quality edits built in to our Data Management System which will highlight inconsistencies in the cause of death coding.  The CSO also has the benefits of an expert Nosologist on site.

The quality of the mortality coding is dependent on very good quality medical certification on the death certificate. 

Mortality statistics are fundamental to medical and social demographic analysis. They are also one of the most reliable and important statistics.  All mortality data is sourced from the General Registration Office (GRO), any obvious error or incomplete data is queried and corrected.  

There were 2 consultations with representatives of agencies/organisations engaged under the umbrella the Suicide Liaison Group to discuss how best to present data on late registered deaths due to intentional self-harm.  Late registered deaths from intentional se;f-harm are often registered around 2 years from the date of death as such cases must be referred to the Coroners Office for further investigation, Inquest and subsequent registration of the death.  Tables have been updated on the CSO Statbank for users. 

Mortality statistics are fundamental to demographic analysis. They are also one of the most reliable and important statistics for health researchers. The CSO is involved in various liaison groups and user groups to ensure that customer needs are met where possible.

For example, the Vital Statistics section is involved in a suicide liaison group to develop further the detail that can be provided to researachers, academics, the general public etc. There a many customers that sign up as Officers of Statistics to interrogate the mortality data - we run special outputs for certain key medical researchers e.g. National Cancer Registry Ireland. 

There are numerous researchers engaged in a variety of medical research projects that are conferred with Officer of Statistics status (i.e respiratory, circulatory, neoplasms etc.). 

The Vital Statistics section provide data to numerous health researchers, that are conferred with the 'Officer of Statistics' status, many of which are involved in longitudinal studies.  The section responds to numerous requests for data, from students, interest groups, researchers, academics and the general public.  The feedback, in general, is very positive. For example we published new tables on what type of institution the person dies in e.g. hospice, maternity unit, nursing home, clinical setting, psychiatric hospital etc.  this was widely welcomed especially by the Irish Hospice Foundation as they can use these statistics for planning etc.

The mortality data received from the General Registration Office (GRO) is comprehensive and of good quality and any queries raised by the Vital Statistics Office is provided where possible. Details on Nationality and Country of birth is improving over time, generally,where this information is absent, there is enough information provided to make a definitive decision in respect of both these variables. Otherwise, 'unknown' is returned.

1. For mandatory variables:

 All mandatory variables provided e.g. year of death by occurrence, sex, cause(es) of death  etc -  100%

2. For voluntary variables:

Circa 30% of records are referred to the Coroners Office (i.e. deaths that occur in Nursing Homes, Unnatural deaths etc) and it is that Office that registers the death.   There are often gaps in information in some variables namely Nationality and Country of Birth and these variables are completed by Vital Statistics Section in the CSO bssed on other information that is provided on the electronic death certificate.  

3. For additional variables:

• External CoD

 For external causes of death included in data for 2021 - narrative provided in each death record - 100%

• Place of occurrence for external CoD

 Place of occurrence (fourth digit of COD code is provided in each UCOD)

• Activity for external CoD

 Activity for external COD is not collected.

Overall, the quality of the data received from the General Registration Office (GRO) is of very high quality.  Any amendments to the data are received subsequently from the GRO. Furthermore, edits are a part of the CSO data capture system which will highlight any inconsistencies in the data.  Further edits are done in SAS to check for inconsistencies.

Overall, the quality of the data is very good.

Not applicable. Data collection is from administrative sources.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Publication is generally T+22 after the reference period. 

Not applicable.

Not applicable.

From data collection with reference year 2011 onwards, Eurostat asks for the submission of final data at national and regional level and related metadata for the year N at N+24 months, according to the Implementing Regulation (EC) No. 328/2011, Article 4.

The size of the geographical regions are not the same - Dublin is the largest region - there are no obvious issues arising when comparing regions - rates are calculated using the population of relevant regions.

The same variables are collected in respect of the deceased irrespective of what region the deceased resided.  

The same process of collecting, coding and the transmission of data is uniform nationally and doesn't differ by region.

Not applicable.

Note that due to the fact that 2011 data is the first data collection with a legal basis (and few changes in the requested variables and breakdowns), the data between 1994-2010 and starting from 2011 are not always comparable (in part due to the different groupings of causes of deaths). Moreover time series for data on stillbirths starts in 2011 and no information on previous data is available.

Mortality data is available for Ireland since 1864, although the rules for assigning an underlying cause of death has changed over time.  The mortality coders adhere to the implementation of the WHO ICD-10 cause of death classification taking into account the major updates by the WHO.

The CSO implemented ICD-10 from 2007 onwards.  Iris software has been used since 1st January 2018 to code the underlying cause of death. This was a break from the American MICAR/ACME software that was previously used.  There is approx. 50-55% automatic coding using this software and the balance of the coding is done manually.  The use of Iris software by the other EU member states serves to make cause of death statistics comporable over time and accross countries.

Prior to the provision of cause of death data under a legal framework, data was provided from 1994 based on a gentlemans agreement.

Not known.

Not applicable, only annual data are available.

Not applicable.

Data coherent across all tables.  Coherence adhered to, based on the most current data available.

The cost and burden of the data collection is reduced by using validation and dissemination IT tools.

Vital statistics data is collected by means of administrative data and as such there are no survey costs are incurred.

Not applicable. Data is final for the reference year 2021.

Data is final for mortality data for 2021 - as this data is based on occurrence, the year the person died rather then the year the death was registered.

Provisional data based on registered data is publised on a quarterly basis and final data is disemminated in the annual report on Vital Statistics based on date of occurrence.

Not applicable.

For births and deaths the source data is the General Registration Office (GRO), Roscommon town, Co Roscommon. Data is encrypted, and received electronically each Thursday. This data is then captured in the CSO Data Management System (DMS).

For births and deaths the source data is the General Registration Office (GRO), Roscommon town, Co Roscommon. Data is encrypted, and received electronically each Thursday. This data is then captured in the CSO Data Management System (DMS).

Data is transmitted electronically each Thursday from the General Registration Office based on the date of registeration of the death and this data is published based on date of registeration for each quarter( T+5) and this data is regarded as provisional data.  Each year the CSO publishes deaths data based on the year the death occurred in T+22 from the end of the relevant reference period.  This data is regarded as final data. 

Data is transmitted from the General Registeration Office, in encrypted format, electronically each Thursday of every week.  The General Registration Office also sends a weekly Reconciliation sheet separately to the CSO and the number of deaths downloaded each week from the GRO should equal the number of deaths uploaded by the CSO in their Data Capture System.  If there is a discrepancy between the two figures then the GRO are contacted immediately to resolve any issue.

a) Stillbirths

Routine death certificate format received - interpreted to the WHO stillbirths standard

b) Neonates

Routine death certificate format received - UC coding forced to Chapter 16 - ICD-10 where possible.

Edits are run each week in our data capture system and any inconsistencies are checked and corrected as necessary.  The statistician carries out other consistency and code checks in the SAS environment and any inconsistancies are checked and corrected as is necessary.

escription of coding procedure (central level, distributed among other bodies, etc.):

The mortality data is received electronically from the General Registration Office (GRO) every Thursday.  Once the number of weekly deaths received are confirmed as correct, the data is uploaded in our Data Capture System.  The CSO uses IRIS software to assign the underlying cause of death code in line with the WHO ICD-10 classification rules.  There is approximately 56% of coding finalised in IRIS automatically.  the balance of the codes which includes external causes of death codes (which are always coded manually) are coded manually by the Mortality Coders in the CSO.

Description of the procedures to detect errors (i.e.errors such as potential inconsistency in the death certificate or error due to mistake when filling the deaths certificates):

1. There are edits included in the CSO Data Capture System (DMS) which isolates certain ICD-10 codes and these are reviewed by the Mortality Coder and the Head Mortality Coder.

2.  Where the narrative of the causes of death are complex then these cases are referred to our Lead Coder for further investigation and coding and also these examples are a source of training for our less experienced Coders.

3.  The Statistician runs other edits in the SAS environment to isolate any other inconsistancies in the coding and these records are reviewd and recoded asis necessary by the Mortality Coders.

Description of the measures taken in order to solve detected errors:

If the Lead Coder detects an error in the IRIS Software then he contacts the IRIS Institute for feedback/correction.

If there is a need for further training of the Mortality Coders then this is provided.

An additional edit is written either in the DMS or generally in the SAS environment.

Coding performed by a certifier:

Coding is not performed by Certifier

Estimation of the percentage of autopsy from which information is available for coding:

Not Availabale.

Description of double coding exercises and rate of codification errors for underlying cause of death:

 Not Available.

ICD codes for the underlying cause (% of the Total)

Unknown country/region (%) for residents and non-residents who died in the country

Data collected by the General Registratrar's Office via localised registration offices.

Not applicable.

Not applicable.

Not applicable.

Not applicable.