Istat, the Italian national institute of statistics

Population Health, Healthcare and Social Assistance Division

viale Liegi 13, 00198 Roma, ITALY

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".

Causes of death are classified by the 86 causes of the "European shortlist" of causes of death. This shortlist is based on the International Statistical Classification of Diseases and Related Health Problems (ICD).
COD data are derived from death certificates. The medical certification of death is an obligation in Italy. The information provided in the medical certificate of cause of death is coded into ICD codes according to the rules specified in the ICD.
Data are broken down by sex, 5-year age groups, cause of death and regional level (NUTS level 2), residence and occurrence. 
Annual national data are provided in absolute number. 

The Vital statistics system on causes of death is the main source for the evaluation of the health status of the population, and for the health programs and resources allocation. Data on causes of all deaths occurring in Italy during a calendar year are collected by the death certificates Istat/D.4 and D.4 bis. The physician must fill the health section of the certificate (part A) and the civil status officer of the appurtenant municipality must fill the demographic section of the certificate (part B).

The main topics observed are: Mortality.
Reporting units: Municipal register offices.
Statistical units: Deaths. Mortality statistics include all deaths occuring in Italy during a calendar year.

Public Health, Causes of death.

1. National definition used for usual residency

Usual residency refers to the Italian legal residency registered at Civil Registry and it is the place where a person normally spends the daily period of rest, regardless of temporary absences.

2. Stillbirths

a) National definition used for stillbirths

In Italy there is a definition applied in the survey on spontaneous abortion: stillbirths are foetal deaths with a gestional age of more than 180 days.
Nevertheless, the official information source used to report stillbirths, that is Birth Delivery Certificate (Decree of Ministry of Health 349/2001), collects data about stillbirths irrespective of the duration of pregnancy. This information source collects, in addition, all the characteristics used to classify and group stillbirths (gestational age, weight, crown-heel) according to the Commission Regulation (EU) No 328/2011.

b) What are the characteristics that you collect (gestational age, weight, crown-heel)?

Gestational age, weight, crown-heel.

The statistical unit is the deceased person. The reporting unit is a physician for health section and the municipal register officer for socio-demographic section.

Statistical population: all deaths and stillbirths occurring in Italy, distinguishing residents and non-residents.

1. Neonates: Are neonatals of non-resident mothers considered residents?

No, they are not residents as the residence of newborns is the residence of the mother.

2. Coverage

a) Do you include non-residents in your national statistics?

Yes, if they die in our country. About  stillbirths, our national statistics include events occured in Italy from non-resident mothers.

b) Do you include residents dying abroad in your national statistics? If yes, how do you record the cause of death? 

No,our national statistics refer only to deaths occurred in the country. 

The statistical data on causes of death refer to Italy.

Region of occurrence and residence (Nuts2) are indicated.

Italian data are available from 1994 onwards. For 2004 and 2005 only demographic information are available.
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 could not be 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.

Not applicable.

In all EU countries, the medical certification of death is an obligation. Most countries already use WHO's international standard model for all but perinatal deaths (0 to 1 week). For perinatal deaths, WHO recommends a specific form which is less frequently applied. The objective of the medical certificate of cause of death is to allow the certifier to enter as clearly and completely as possible the causes of death, i.e. describing the sequence of diseases and conditions leading to the death, mentioning other contributing conditions etc. In most countries, the medical certificates of cause of death are forwarded to COD statistics offices for centralised coding. COD statistics also require information on sex, age, place of residence etc. of the deceased. Depending on the country, this information is either collected through the death certificate or taken from other sources.
For calculating crude and standardised death rates, the annual average population available in Eurostat's demography database is used.
However, national legal requirements as well as national practices concerning the registration of residents dying abroad and domestic deaths of non-residents are far from being harmonised across European countries. Therefore, information about residents dying abroad might not be included in all countries, while deaths of non-residents are mandatory information.

For each death a death certificate is filled in by a physician. He must indicate "all diseases, morbid or traumatic conditions that led or contributed to death, and the circumstances of the accident or violence that provoked these traumatisms" in the health section of certificate, hereby also reporting other relevant information linked to the death. Once the physician has filled in the health section, the form is sent to the pertinent municipality. The registrar must complete the certificate by indicating the socio-demographic data of the deceased. In particular, other than the dates of birth and death this information concerns the place of birth and residence, marital status, educational level, profession, branch of economic activity, citizenship and individual code of the deceased. If death has occurred during the first year of life, the form is slightly different and, as regards the demographic section, the requested information mainly refers to the condition of the parents. Having been filled out in duplicate, the form follows two different paths: one copy is sent to the ASL (where the death occurred), whereas the other copy, before being sent to Istat, is sent to the Prefectures and Regional Offices.

Annual.

1. Certification

Description of the certification process and the validation process

In Italy, the death form reflects the structure of the WHO certificate. Part 1 contains four line, the upper one is for the underlying cause of death and the lines below are for the complications and final cause. There is a special item for external cause description to be filled out in case of deaths due to injury and poisoning. Another item is for reporting deaths due to transport accidents.

Certification is paper based, the certificate is filled out by the patient’s attending physician. When this is not available, the cause of death is reported by the physician who ascertained the death. The completion is mandatory for all deaths occurred in Italy and should be provided to the municipality within 24 hours from the ascertainment of the death. Successively a copy of the certificate is sent to the local health unit and one to the Italian National Institute of Statistics.

How are the certification errors assessed?
To evaluate the certification errors we use an indicator based on the ratio between the number of ill-defined deaths and the total deaths. Moreover we analyze it by the geographic and time variations. 

Is training in certification obligatory in your country?
No, the training certificaton is not obligatory in Italy. It is carried out occasionally. 

 NA= not available. In Italy a standard query procedure is not implemented and training in certification is not carried out.

In Italy, data are collected by a paper death certificate. The same certificate for males and females is used.

5. Stillbirths and Neonates: Do you have a different death certificate or do you code these data differently from other deaths? If yes, how?

a) Stillbirths

Yes, about  stillbirths, the source of data is Birth Delivery Certificate for which the Ministry of Health is responsible (Decree of Ministry of Health 349/2001). This information source collects data about births, stillbirths and congenital anomalies at birth, irrespective of the duration of pregnancy.

b) Neonates

Yes, Istat has a different death certificate for neonatal and infant deaths (all deaths under 1 year of age).

1. Coding

Description of how the coding is performed and what procedures are used to detect coding errors (i.e.errors such as potential inconsistency in the death certificate or error due to mistake when filling the deaths certificates).

Causes of death are coded by an automated coding system (ACS), that requires different steps and the reject of the ACS are manually coded. The data processing is managed by means of the software CODSANII that allows to run the different steps of coding. In detail the procedure requires the following steps: data recognition is performed by the software ACTR. The output of this procedure is inputted to Micar and the rejects are manually coded. For some of them the manual coder corrects the multiple cause codes while for some others the UC is selected (all external causes are handled in this second way). Records with completed multiple cause are inputted to ACME which selects the UC. The reject are manually coded by attributing an UC.

About 75% of all certificates are completely automatically coded.

Coding errors were systematically assessed after a major training carried out in 2013 for new coders in which for a sample of about 4.000 certificates the coding of each trainee was compared to the gold standard provided by expert coders.
Description of the outcome of the error detection procedures in place

The study revealed a concordance with the gold standard of about 90%. Higher rates were observed for malignant neoplasm while poorer rates were obtained for viral hepatitis and sequela codes.
Is coding carried out at a central level?

Yes, the coding is carried out at central level in the National Institute of Statistics.
Estimation of the percentage of autopsy from which information is available for coding
Not applicable, because in Italy the autopsy results are not sistematically collected.
Is coding performed by a certifier?
No, the coding is performed by expert nosologists.
Are double coding exercises performed  (if yes, please indicate the rate of codification errors for uderlying cause of death/if not give an estimate of the rate of codification errors)?
See above ("Description of the outcome... ").

2. Validation process

What are the criteria (external source, local registers, …) used for the validation of the coverage?

We use external sources such as the data of demographic events of civil registry.
Regarding stillbirths, as external source we use Hospital Discharges Information for the validation of the coverage of deliveries.

The Italian statistics on causes of death are produced by a centralized system of coding and data processing.

Causes of death are coded by an automated coding system (ACS), that requires alphanumerical registration of the health section of the death certificate. Only causes rejected by this system are manually coded. Since reference year 2003 also data on multiple causes are coded.

An editing and imputation procedure is applied to check and correct the measurement errors and item non-responses. The procedure is based on both deterministic (health and socio-demographic data) and probabilistic approach (socio-demographic data), to input incorrect and missing data. Much of the probabilistic procedure is based on the Nearest-neighbour Imputation Methodology (NIM, developed by Statistics Canada).

Not applicable.

Since the 90s Istat adopted a systematic approach to ensure quality in both statistical information and service to the community: http://www.istat.it/en/about-istat/quality

From 2011, the causes of death data are based on a regulation, which defines scope, definitions of variables and characteristics of the data.

During 2015 an audit procedure to assess the system of statistics on causes of death has been performed.

The overall assessment was positive, the aims of system are clear and well defined, and they are shown on a wiki page. The audit showed that there are not issues related to the coverage and the list of reference of the survey. Furthermore any attempt to reduce the statistical burden was performed. Regarding the IT, the validated data are systematically stored on an Oracle DB, and they are adequately protected following the standards of the Institute. The data released on the web are accompanied by links to SIQual, the Istat official quality documentation system.

Data on causes of death provide information on mortality patterns and form a major element of public health information.
The relevance of an instrument has to be assessed in the light of the needs of its users. As for the COD data collection the main users are the following:
· Institutional users like DG SANTE or other Commission services;
· Statistical users in Eurostat or in Member States National Statistical Institutes to feed publications; and
· End users – including the media - interested in health statistics in the EU.

Main users are public national and local government, public research institutes and private researchers.

Data are mainly used to support the decisions about public policies on health and to study the mortality (trend, risk factors, etc).

Sometimes users ask for more detailed data (e.g. nuts3 level, microdata), but generally these requests cannot be met for confidentiality reasons.

The views and opinions of the users are not regularly collected but during the meeting for the arrangement of the National Statistical Program (PSN) the Committee of Statistical Information Users (CUIS) is consulted to highlight the possible weaknesses of data. In general terms the users resulted satisfied.

All data requested are sent and disseminated on Eurostat's website.

1. mandatory variables:
   All variables required by the Commission regulation n°328/2011 are transmitted to the European Commission.

1. voluntary variables:
   for stillbirths

The following variables are available:

Region of occurrence – nuts 2.

Country of residence of the mother.

The following variables have been studied and tested and they will be available from 2014 onwards:

Sex.

Distinction between the first or the second group of stillbirth

Age of the mother.

As regards the following other variables, they are not transmitted because they have a significant degree of incompleteness:

Underlying cause of death (the variable is completed on average in 24% of cases. This is caused by the delay in the availability of the autopsy that is rarely available in time for the compilation of the Birth Delivery Certificate. In the Italian National Statistical Programme, a joint project has been activated between the Ministry of Health and the National Institute of Statistics aimed at improving the statistics on stillbirths)

Parity (some measures have been implemented to improve the quality of this variable).

for neonatal deaths

We are not able to transmit the following variables:

Parity (because the high degree of incompleteness, but we are trying to improve the quality of this information).

Region of residence of the mother - nuts2 (not available only for non-residents who died in Italy).

for other deaths

we are not able to transmit:

Region of residence - nuts2 (not available only for non-residents who died in Italy).

In general, regarding the region of residence - nuts 2, the main difficulty concerns the collection of this information for non residents who died in Italy. Actually, for deaths of residents in Italy, the region of residence – nuts 2 is available.

The accuracy of statistical outputs in the general statistical sense is the degree of closeness of estimates to the true values. Statistics can be different from the true values because of random variability and/or bias (for example, in case of administrative data the lack of information for a subpopulation).

As the data collection is total population based, the sampling error is not applicable. Many efforts are spent to detect and to obtain the missing models by the non-respondent municipalities, reaching a coverage level of about 100%. An editing and imputation procedure is applied to data to check and correct the measurement errors and item non-responses. Significant investments (tools to support encoding, training coders, etc.) are continously made to improve the quality of causes of death coding.

Not applicable.

Not applicable.

see 6.1

The frame for unit Municipal register offices is List of Italian Municipalities managed by Istat (National Statistical Institute of Italy).
There is not a coverage-error because the List of Italian Municipalities is always upgrated.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

There is no systematic revision of previous year data. Data are occasionally revised, e.g. if the "NUTS" changes.

There is no systematic revision of previous year data. Data are occasionally revised, e.g. if the "NUTS" changes.

Optional field

From data collection with reference year 2011 onwards the submission of final data for the year N is at N+24 months, as requested by the European Regulation.

Optional field

Optional field

From data collection with reference year 2011 onwards, Eurostat asks for the submission of final data for the year N at N+24 months.

From data collection with reference year 2011 onwards, Istat mets the deadline of the Regulation.

Optional field

Currently the definition adopted in Italy distinguishes miscarriages and stillbirths according to whether the expulsion of the fetus occurred before or after 180 days of gestation.

Miscarriage events are collected by Istat, while stillbirth events are collected by the Ministry of Health through the Birth Delivery Certificate.

The harmonization with respect to fetal deaths European definitions requires a careful assessment of the impacts on the Italian legislation about the safeguard of working mothers, the civil register and mortuary police state, as well as on the activities of health personnel. The project study included in the National Statistical Programme, designed to improve the statistics on stillbirths, aims to assess the suitability and feasibility of the definition change.

The coherence of two or more statistical outputs refers to the degree to which the statistical processes, by which they were generated, used the same concepts and harmonised methods.

The comparability of the data across different countries is limited by the fact that the revision of classification used to collect information on underlying causes of death may be different. Italy adopts the tenth revision of the ICD since 2003 reference year. The recommended WHO's updates are applied as indicated in the table of point 3.2.

Not applicable.

Due to the fact that in 2003 the tenth revision of the ICD has been adopted the data between 1994-2002 and starting from 2003 could not be always comparable. Moreover time series for data on stillbirths starts in 2011 and no information on previous data is available.

Optional field

Two different data sources for collecting mortality data in Italy are carried out by the National Institute of Statistics (Istat):

1)       “Deleted to death from population registers” (Demographic Statistics or DS)

2)       “Causes of death” (CoD). 

The DS survey refers to cancellation due to death from the Population register of all residents in Italy, including those died abroad. CoD statistics instead refer to all deaths occurred in Italy (for both residents and non residents).

The discrepancies found between DS and CoD data are primarily ascribed to the different data source: acquisition of an administrative archive cancellation for the DS data versus a data collection of death certificates completed by physicians for the CoD data and to the different population that these two different data sources belongs to, actually the residents dying outside the country are collected only by DS.

Is there any missing information (uncomplete geographical coverage, missing certificates, …)? If yes, please indicate also the %.

Yes, there may be missing certificates. We can estimate an uncomplete geographical coverage rate of about 1% of municipalities (but only very small municipalities). Regarding stillbirths, geographical coverage is complete and missing events are of negligible order.

Optional field

Not applicable.

Italian data on causes of death are internally consistent.

At Eurostat level, news releases on-line.

At National level it is not applicable.

Causes of death data are disseminated in many different publicatons "multi-source".

Among the most relevant there are:

"Italy in figures"   (http://www.istat.it/en/archive/166226)

"Italian Statistical Yearbook" (http://www.istat.it/it/archivio/134686)

"BES. Benessere Equo Sostenibile" (http://www.istat.it/it/files/2015/12/Rapporto_BES_2015.pdf)

Please consult free data on-line: http://ec.europa.eu/eurostat/data/database

I.Stat: http://dati.istat.it/

Health For All: http://www.istat.it/it/archivio/14562

According to the national laws, techniques have been adopted to protect the confidentiality of the microdata.

Given the sensitiveness of data, the microdata access is limited to the projects included in the National Statistical Program (PSN) approved by law.

Only for data of the reference year 2012 a release was disseminated : http://www.istat.it/it/archivio/140871

A relevant publication is also the following: "La mortalità dei bambini ieri e oggi: l’Italia post-unitaria a confronto con i Paesi in via di sviluppo" http://www.istat.it/it/archivio/40505

Documents for COD are available in CIRCABC, Causes of Death section.

The official quality documentation on causes of death statistics is available on the Istat official quality documentation system (SIQual) : http://siqual.istat.it/SIQual/visualizza.do?id=5000131

Optional field

Optional field

  Information not available

Main costs concern the recording and coding phases. Nevertheless increasing use of IT tools allowed to reduce costs. 

Optional field

Regulation (EC) No 223/2009 on European statistics (recital 24 and Article 20(4)) of 11 March 2009 (OJ L 87, p. 164), stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society.

Information on Istat confidentiality policy is available on the Istat website: http://www.istat.it/en/tools/methods-and-it-tools/analyse

All age groups showing a total mortality of less than 4 cases are considered as confidential. Therefore, any 'confidential' agegroup is grouped with another one to have higher numbers. In practice, this problem mainly occurs for young ages so, either the ages from 0 to 14 years old, or the ages from 0 to 14 and 15 to 24 years old are grouped. The age groups considered as confidential show then the value ':' and the age group '0-14y' (and '15-24y' if needed) shows the sum of all ages before 15 years old (or between 15 and 24 years old).

None.