Mortality in persons with recent primary or secondary care contacts for mental disorders in Finland

Kimmo Suokas, Christian Hakulinen, Reijo Sund, Olli Kampman, Sami Pirkola

2022-09-24

Description: This document contains supplementary information on the study:

Suokas K, Hakulinen C, Sund R, Kampman O, Pirkola S. Mortality in persons with recent primary or secondary care contacts for mental disorders in Finland. World Psychiatry. 2022;21(3):470-1

Background in brief

• Mental disorders are commonly treated in primary care, but excess mortality related to mental disorders is usually studied with secondary care data.

• In Finland, primary care is included in the national health care registers since 2001.

• This nationwide study examined excess mortality in individuals with a one-year history of any mental heatlh treatments in primary or secondary care.

Main findings

Excess mortality

Table 1. Age-standardized mortality rates (ASMRs) and mortality rate ratios (MRRs) in individuals with and without a one-year history of mental health treatments

^aAge-standardized mortality rate per 1 000 person-years. Standardized to the 2013 European Standard Population by 5-year age groups. ^bModel 1: adjusted for Model 1: adjusted for calendar year and age group (5-year intervals). ^cModel 2: adjusted for Model 2: adjusted for calendar year, age group, urbanicity, region, education, living alone, household income, and economic activity. ^dModel 3: adjusted for Model 3: adjusted for calendar year, age group, and Charlson comorbidity index (CCI). ^eMortality in individuals with secondary care treatments compared to individuals without secondary care treatments (primary care ignored).

Excess mortality by mental disorder

Mortality rate ratios in individuals with a one-year history of mental health treatments compared to individuals without such a history, stratified by mental disorder

Figure 1. Primary and secondary care combined

Figure 2. Primary care only

Figure 3. Secondary care only

Note X axis values

Number of deaths and person-years by diagnosis

Number of deaths and person-years at risk by gender and diagnostic categories in individuals with a history of mental health treatments in primary or secondary care

Click on legend to select/un-select one trace, double-click to isolate.

Mental health treatments included both primary and secondary care. Mental disorders are presented based on ICD-10 sub-chapter categories; the group “organic and substance use excluded” presents all mental disorders excluding the ICD-10 sub-chapters F0 “Organic, including symptomatic, mental disorders” and F1 “Mental and behavioural disorders due to psychoactive substance use”. Category F8 “Disorders of psychological development” is not shown due to the low number of deaths (on average less than 10 per year in women). Model 1 was adjusted for Model 1: adjusted for calendar year and age group (5-year intervals). Modle 2 was adjusted for Model 2: adjusted for calendar year, age group, urbanicity, region, education, living alone, household income, and economic activity. Model 3 was adjusted for Model 3: adjusted for calendar year, age group, and Charlson comorbidity index (CCI). Model 4 was adjusted for Model 4: adjusted for calendar year, age group, urbanicity, region, education, living alone, household income, economic activity, and CCI. Error bars represent 95% CIs.

Excess mortality by age

Age-specific mortality rate ratios in individuals with a one-year history of mental health treatments compared to those without

Figure 4. Primary and secondary care combined

Figure 5. Primary care

Figure 6. Secondary care

Mortality rate ratios are shown on a log scale. Model 1 was adjusted for calendar year. Model 2 was adjusted for calendar year, urbanicity, region, education, living alone, household income, and economic activity. Model 3 was adjusted for calendar year, and Charlson Comorbidity Index (CCI). Model 4 was adjusted for calendar year, urbanicity, region, education, living alone, household income, economic activity, and CCI. Error bars represent 95% CIs.

Excess mortality stratified by comorbidity

Mortality rate ratios in individuals with one-, three-, or five-year histories of mental health treatments compared to individuals without such a history, stratified by the presence of physical comorbidities and health care setting

Model adjusted for calendar year, age group, urbanicity, region, education, living alone, household income, and economic activity. Error bars represent 95% CIs.

Sensitivity analysis

Age-standardized mortality rates (ASMRs) and mortality rate ratios (MRRs) in individuals with one-, three-, or five-year histories of mental health treatments compared to individuals without such treatments, primary and secondary care combined

^aAge-standardized mortality rate per 1 000 person-years. Standardized to the 2013 European Standard Population by 5-year age groups. ^bModel 1: adjusted for Model 1: adjusted for calendar year and age group (5-year intervals). ^cModel 2: adjusted for Model 2: adjusted for calendar year, age group, urbanicity, region, education, living alone, household income, and economic activity. ^dModel 3: adjusted for Model 3: adjusted for calendar year, age group, and Charlson comorbidity index (CCI).

Supplementary methods

Assessment of the history of mental health treatments

Information on mental health care was obtained from the Finnish Care Register for Health Care. Primary care has been included since 2011. The register has a good accuracy of mental health diagnoses¹.

The diagnostic system used was International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10). We described specific disorders with the ten-level ICD-10 sub-chapter categories (see Supplementary methods). In addition, individuals with mental disorders excluding organic disorders (F00-F09) and substance use disorders (F10-F19) were recognized. In primary care, ICPC-2 International Classification of Primary Care, instead of ICD-10 is used in some facilities, and ICPC-2 mental health-related diagnoses were converted to corresponding ICD-10 sub-chapter categories.² A description of the method used for handling partly overlapping register data entries is publicly available.³

For every individual in the study, person-time at risk of death was dynamically labeled based on the history of mental health treatments. A history of mental health treatments was defined as having any medical contact with secondary care psychiatric inpatient or outpatient services, or with primary care with a diagnosis of any mental disorder (i.e. ICD-10 Chapter V: Mental and behavioural disorders (F00-F99), or ICPC-2 chapter P: Psychological) within the previous year.

Time spent in psychiatric inpatient care and the following year after the discharge date, and one year after any Secondary care psychiatric outpatient visits, was labeled as follow-up time with a one-year history of secondary mental health care. From exactly one year after the latest discharge, or the latest Secondary care outpatient contact, if still at risk of death, the individual no longer contributed to the Secondary care population and returned to either the population with a one-year history of Primary care, or the reference population with no one-year history of mental health treatments.

One year following any Primary care visit with a diagnosis of mental disorder was labeled as follow-up time with a one-year history of primary mental health care. Exactly one year after the latest Primary care contact, if still at risk of death and with no new Secondary care contacts, the individual returned to contributing to the reference population with no one-year history of mental health treatments. If a contact to Secondary care occurred during the follow-up time with a one-year history of Primary care, the individual’s follow-up time was relabeled to Secondary care based on the date of the Secondary care contact. If a Primary care contact emerged less than one year after the latest Secondary care contact, the follow-up time was relabeled to Primary care one year after the latest Secondary care contact, until one year had passed since the latest Primary care contact.

ICD-10 sub-chapter categories of mental disorders

The International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) has been used in Finland since 1996. The 10 sub-chapter categories of mental disorders include:

Conversion of the ICPC-2 International Classification of Primary Care diagnoses

According to the ICPC-2 manual,² “ICPC is based on a simple bi-axial structure: 17 chapters based on body systems on one axis, each with an alpha code, and seven identical components with rubrics bearing a two-digit numeric code as the second axis”. Conditions listed in ICPC-2, chapter P: “Psychological” were included in this study. In the Finnish version of ICPC-2, the chapter P is translated to the Finnish equivalent of “mental health” not “psychological”·⁴ The Finnish version was used throughout the conversion process.

The conversion codes from ICPC-2 to ICD-10 are provided in the ICPC-2 manual and were adopted in this study. ICPC-2 concepts not represented exactly in the ICD-10 were included as a separate group of their own and are not included in the ICD-10 sub-chapter categories.

Cofactors

The following individual-level data on the last day of each study year from the population registers were included:

• gender: man or woman,
• urbanicity of residence area: a seven level classification,⁵
• region of residence: 19 regions,
• living alone status: living alone or not,
• level of educational attainment: less than upper secondary, upper secondary, or tertiary. A national classification based on the United Nations Educational, Scientific and Cultural Organization’s International Standard Classification of Education 2011,
• economic activity: employed, unemployed, students, pensioners and others outside the labour force, and
• equivalized household net income deciles. Net income was obtained after subtracting taxes and was adjusted for the size of household dwelling unit, using the Organisation for Economic Co-operation and Development–modified equivalence scale. Income measurement with a three-year lag was used to account for potential reverse causation.
• Physical comorbidity was assessed using the Charlson comorbidity index (CCI),⁶ a widely used comorbidity index with a weighted score of 17 comorbid conditions. For every study year and for every individual in the study, the CCI score was calculated using available ICD-10 diagnoses of any actual treatment contacts in the health care registers from the beginning of the previous calendar year. Age was not included in the CCI scores, but adjusted in the main model, instead. CCI scores were categorized by previously used cut-points: none, 1-3, and ≥4.⁷

Statistical analysis

Deaths and person-time at risk were labelled based on the one-year history of mental health treatments. Individual follow-up time was allocated dynamically between primary and secondary care and the reference population based on the actual dates. All variables were treated as time-varying factors. If the exact time of change in covariate status was not known, mid-year was assumed. We aggregated the number of deaths and person-years at risk according to calendar year, treatment history, and the covariates and we calculated age- and stratum-specific mortality rates. The 2013 Revision of the European Standard Population was used for direct age standardization.

Mortality rate ratios (MRRs) were examined using a Poisson regression model with a robust variance estimator. The cells in the aggregated data were taken as the unit of analysis, with the logarithms of the aggregated person-years counts set as an offset variable. Bayesian information criteria were used for model selection.

References

1.

Sund R. Quality of the Finnish Hospital Discharge Register: a systematic review. Scandinavian journal of public health 2012; 40: 505–15.

2.

WONCA International Classification Committee (WICC). ICPC-2-R: International Classification of Primary Care. New York: Oxford University Press Inc., 2005.

3.

Suokas K. Hilmo_identify_episodes (v1.1.0). Epub ahead of print 2021. DOI: 10.5281/zenodo.5381082.

4.

Kvist M, Savolainen T, Suomen Kuntaliitto. ICPC-2 : perusterveydenhuollon kansainvälinen luokitus. Helsinki: Suomen Kuntaliitto, 2010.

5.

Finnish Environment Institute. Environment > Urban-rural classification, https://www.ymparisto.fi/en-US/Living_environment_and_planning/Community_structure/Information_about_the_community_structure/Urbanrural_classification (2014, accessed 18 May 2022).

6.

Bannay A, Chaignot C, Blotière P-O, et al. The Best Use of the Charlson Comorbidity Index With Electronic Health Care Database to Predict Mortality. Medical Care 2016; 54: 188–194.

7.

Erlangsen A, Stenager E, Conwell Y, et al. Association between Neurological Disorders and Death by Suicide in Denmark. JAMA 2020; 323: 444–454.