README
======

Dataset title:
Paediatric suicide data

Creators:
Sebastian Hökby, Emma Eliasson & Anna Lundh 

Responsible organization:
Karolinska Institutet

See the Researchdata.se catalogue entry for this dataset for complete metadata, including the dataset description, recommended citation, contact information, and information about how to obtain the dataset. 

Description: 
Timeseries dataset (years 1980-2023) with Swedish cause-of-death data and population data for children under age 18. Columns contain information about the year, sex, and the annual aggregated suicide count, the count of suicides plus deaths with undetermined intent, and the corresponding population data. Data columns separate children aged 10-14 from the age group 10-17 years. Data is always stratified by gender in a long format (male data in the first 44 rows/years, and female data in the last 44 rows/years). 

License: 
PDM 1.0



Essential information for dataset reuse.
----------------------------------------

Data extractions:
The dataset contains aggregated data extracted from the official Swedish cause of death registry, informed by individual cause-of-death certificates, as well as official population data extracted from Statistics Sweden (SCB). Both sources provide reliable and high-quality data. Combined, they allow suicide rates to be calculated. See sections ‘Data Sources’ and ‘Data quality’ below for further details, including ethical considerations and permissions. 

Closed for peer review:
The dataset will be shared openly (coupled with a unique DOI) after open access publication of the results, i.e., manuscript publication in a scientific journal. Before publication, it will only be made accessible to journal editors/staff and peer reviewers. 

Intended analysis:
The dataset was created with the intention of analysing annual time trends in suicides during the 44-year period: 1980–2023. It only focuses on Swedish children below the age of 18. The national suicide/death count was aggregated by year, but stratified by sex (male vs. female), for two specific age groups (ages 10-14 vs. ages 10-17). All numbers are presented at the national level – never on a local level. Cases were only excluded if they had missing data regarding sex, age, or year of death (missing exact day of death was acceptable). The dataset is comprised only of dead human subjects and is not linked to any specific register. Suicide for ages 0-9 was examined but dropped from the dataset as the number of deaths was too small to be informative to longitudinal statistical analysis. (the annual suicide count was often zero and therefore not informative to the planned analyses). The dataset covers 1030 deaths in total, divided over the 44 years and two sexes (totals 88 rows).

Statistical software:
The aggregated dataset is provided in common CSV-format. No special software or code is required to read the file. The data can be imported and analysed with most common statistical software, such as R, Stata, and SPSS. However, the dataset was created and structured with the intention of analysing it a specific software – the Joinpoint Software, provided by the National Cancer Institute (2025). Although the study has not been completed yet, for the sake of Joinpoint replication, one may already assume that the same software version (v.5.4.0.0; setup file: "JP_Setup_540.exe") should be used to reproduce the findings, as future versions might work differently. Software download is free of charge but requires email registration. 

- The default settings for the current version (v.5.4.0.0) are described in the uploaded document: “V 5.0 - Default Settings Changes — Joinpoint Help System.pdf” (National Cancer Institute, 2025).  
- The methodological aspects of Joinpoint Regression have been described in two regularly cited papers by Kim and colleagues (Kim et al., 2000; Kim et al., 2020). 
- Additional considerations have been discussed by Ingram and colleagues (Ingram et al., 2018). 
- Information about the Joinpoint software, including extensive documentations about the working algorithms, equations, assumptions, and frequently asked questions, can be found at the developer’s website (https://surveillance.cancer.gov/joinpoint/).


Datafile descriptions
----------------------
The "Paediatric suicide data" dataset - Aggregated paediatric suicide timeseries.
Dataset filename: PaediatricSuicideData.csv
The first three columns (A, B, C) of the dataset are applicable to all analyses. The next three columns (D, E, F) are used to analyse the age group 10-14. The last three columns (G, H, I) are used to analyse the age group 10-17.  
•	Cases (“RowCase”) can be sorted from smallest-to-largest value (column A). This properly sorts years from 1980-2023, first by Male (=1) sex, then by Female (=2) sex. 
•	All data is aggregated by Year 1980-2023 (column B). This is the Independent Variable.
•	The data is stratified by Sex, in a long data format (column C). This is a stratification variable (also called ‘By-variable’).
•	The remaining variables are used to calculate different dependent variables. Data for Ages 10-14 vs. Ages 10-17, are provided separately (columns D, E, F vs. columns G, H, I), and for each of the two age groups. Moreover, the dataset contains:
o	Population data: The mid-year population counts (columns D vs. G).
o	Primary cause of death data: The annual suicide counts (columns E vs. H).
o	Secondary cause of death data: The annual counts of ‘Suicides’ and with ‘Deaths with undetermined intent’, have been added together (columns F vs. I) in a way that separates them from suicide counts. 

Data definitions:
•	“Suicide” corresponds to ICD-10 diagnoses: X60-X84. In the 8th and 9th edition of the ICD, the corresponding diagnoses are: E950-E959. Cases of suicide are used for primary analysis purposes.
•	“Death with undermined intent” corresponds to ICD-10 diagnosis: Y10-Y34). In the 8th and 9th edition of the ICD, the corresponding diagnoses are: E980-E989. Cases with undermined intent are included for sensitivity analyses purposes.
•	“MidPop” = is an abbreviation of “Mid-year population count” which is calculated as the average count of two consecutive years (e.g., 2022+2023/2 = Mid-year count for 2023). This calculation applied as the Swedish population was counted on December 31st each year.

Variable list [with variable explanations]
---------------------
A)	Case_Row [Sorts cases from 1-88; serves no other purpose]
B)	Year [Independent variable datapoints, 1980-2023]
C)	Sex_Male_1_Female_2 [Datapoints are coded Male=1 and Female=2]
D)	MidPop_Ages10to14 [Mid-year population count in Ages 10-14]
E)	Count_Suicides_Ages10to14 [Suicide count in Ages 10-14]
F)	Count_SuicidesAndUndetermined_Ages10to14 [Suicides + Undetermined deaths in Ages 10-14]
G)	MidPop_Ages10to17 [Mid-year population count in Ages 10-17]
H)	Count_Suicide_Ages10to17 [Suicide count in Ages 10-17]
I)	Count_SuicidesAndUndetermined_Ages10to17 [Suicides + Undetermined deaths in Ages 10-17]

Reference information
---------------------
The dataset holds the annual suicide counts (with and without deaths with undetermined intent) and corresponding mid-year population counts (see variable list). However, it does not contain calculated crude suicide rates (e.g., number of suicides per 100 000 population). Such rates are possible to calculate within the Joinpoint software, for example to automatically model heteroscedastic standard errors. To instead calculate such suicide rates by hand, simply divide a death count variable over its population variable, and multiply by 100 000. For example, to get the number of suicides per 100 000 capita in age group 10-14, calculate: Column E / Column D * 100 000. For age group 10-17, use column H and G instead. If using Microsoft Excel, this can be specified as: [ =(H/G)*100000 ].  

Data sources
This is an aggregated dataset. It includes data from two other datasets which are not linked. The primary source is the Swedish cause of death registry, reflecting cause-of-death certificates on the individual-level (one death per row). These cause-of-death data were originally provided by the National Board of Social Affairs and Health, requested by the National Centre for Suicide Research and Prevention, NASP Sweden (data delivery on 2024-September-26; Socialstyrelsen Dnr.: 122437/2023). The original file contained high-quality, detailed, nation-wide, individual-level data, restricted to maximum 44 years (1980-2023) and restricted to only contain deaths related to suicide and deaths with undetermined intent. The individual-level data is referred to as the “Original Data file”; it requires special permission to use and will not be shared. The dataset shared dataset is referred to as the “Raw Dataset” because planned analyses only rely on annually aggregated data which allows for independent replication. At the first step, all death counts were extracted for age groups 10-14 and 10-17, respectively. Secondly, the data was aggregated by year but stratified by sex and diagnosis type (suicide vs. undetermined). The resulting annual counts appear in the dataset as columns with 44 rows of male rows/years, followed by 44 female rows/years. The second data source was used to extract and calculate the male and female mid-year population counts for the same age groups. Mid-year counts are derived from end-year population counts (e.g., 1979+1980/2 = Mid-year count for 1980). The population data is public domain, and free access was provided by Statistics Sweden (SCB, 2025). Population data appears in separate column in the dataset, so that suicide rates (e.g., per 100 000 capita) can be modelled with or without it, and if necessary, model suicide rates with heteroscedastic standard errors (e.g., should it turn out that the population sizes for males and females differ a lot).

Data Quality:
The Swedish cause of death registry data is of high quality compared to international standards (Brooke et al., 2017). One purpose of the registry is to enable high-quality research, including time trend analyses. Annual data is published once per year and the present dataset relies on data that was delivered in year 2024, when data for year 2023 was published. The register was started in 1952 (de Munter 2022) but historical data between 1952-1979 has not been requested by researchers at NASP, which limits the data coverage to the 44-year period of 1980-2023. Data starting from 1980 is requested by NASP researchers on an annual basis, and the recorded data is consistent from year to year. Upon delivery, the first and second author independently checked the entire register for completeness and convergence with previous versions, using IBM SPSS and R, respectively.
Although the data is considered very reliable, some biases may still exist (de Munter, 2022). Minor retroactive changes are sometimes made by the National Board of Social Affairs and Health, Sweden (Socialstyrelsen). For example, diseased persons not registered in Sweden, are as a rule added to the national statistics with one year of extra delay. This can also happen if a physician does not timely submit the death certificate to the Swedish Tax Agency or if the cause-of-death certificate is partially incomplete Retroactive changes to the suicide statistics are often negligible, tends to apply only to the most recent data years. A more important source of bias is the doctor’s assessment of the underlying cause of death, hence why deaths with undetermined intent is included in the present dataset, as it enables sensitivity analyses in this regard.  

Parts of this dataset can be checked against the online cause of death database, curated by the National Board of Social Affairs and Health, but is only available in Swedish (URL: https://sdb.socialstyrelsen.se/if_dor/val.aspx). For example, suicide counts in age group 10-14 can be retrieved from this database, starting from year 1997 and onwards (i.e., only deaths coded with ICD-10).

References:
1.	Brooke HL, Talbäck M, Hörnblad J, Johansson LA, Ludvigsson JF, Druid H, Feychting M, Ljung R. The Swedish cause of death register. Eur J Epidemiol. 2017 Sep;32(9):765-773. doi: 10.1007/s10654-017-0316-1. Epub 2017 Oct 5. PMID: 28983736; PMCID: PMC5662659.
2.	SCB, Statistics Sweden. (2025). Database. ”Population size according to region, civil status, age and sex. Years 1968 – 2024”. Internet. Accessed [2025-July-08] from: https://www.statistikdatabasen.scb.se/pxweb/sv/ssd/START__BE__BE0101__BE0101A/BefolkningN
3.	De Munter (2022). Statistical register's production and quality -National Cause of Death Register. Department of Registers and Statistics, National Board of Social Affairs and Health, item no.: 2022-5-7909. Published 2022-February, Sweden. Accessed 2025-10-07, from URL: https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/dokument-webb/ovrigt/production-and-quality-dors.pdf#page=8.47 
4.	Surveillance Research Program, National Cancer Institute (2025). Joinpoint Regression Software, Version 5.4.0 - April 2025. (https://surveillance.cancer.gov/joinpoint/).
5.	Ingram DD, Malec DJ, Makuc DM, Kruszon-Moran D, Gindi RM, Albert M, Beresovsky V, Hamilton BE, Holmes J, Schiller J, Sengupta M. National Center for Health Statistics Guidelines for Analysis of Trends. Vital Health Stat 2. 2018 Apr;(179):1-71. PMID: 29775435.
6.	Kim HJ, Chen HS, Byrne J, Wheeler B, Feuer EJ. Twenty years since Joinpoint 1.0: Two major enhancements, their justification, and impact. Stat Med. 2022 Jul 20;41(16):3102-3130. doi: 10.1002/sim.9407. Epub 2022 May 6. PMID: 35522060. https://pubmed.ncbi.nlm.nih.gov/35522060/  
7.	Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med. 2000 Feb 15;19(3):335-51. doi: 10.1002/(sici)1097-0258(20000215)19:3<335::aid-sim336>3.0.co;2-z. Erratum in: Stat Med 2001 Feb 28;20(4):655. PMID: 10649300. https://pubmed.ncbi.nlm.nih.gov/10649300/ 











