Mattias Sjöström
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mattias Sjöström has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Jenny Selander
Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
2 shared publications - Magnus Alderling
Centre for Occupational and Environmental Medicine, Stockholm County Council, Sweden
2 shared publications Marie Lewné
Institute of Environmental Medicine, Karolinska Institutet , Solnavägen 4, 10th floor, SE 113 65 Stockholm , Sweden
2 shared publications- Marie Lewné
Karolinska Institute, Stockholm, Sweden
2 shared publications - Per Gustavsson
Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
2 shared publications - Anneli Julander
IVL Swedish Environmental Research Institute
1 shared publication
Latest Publications
- Open Access
- 5 May 2026
- Published by Elsevier in SSRN Electronic Journal
Occupational Noise Exposure During Pregnancy and Birth Outcomes: A Large-Scale Cohort Study in Nordic Countries
Objectives: Adverse birth outcomes such as low birth weight (LBW), small for gestational age (SGA) and preterm birth (PTB), are major indicators of perinatal health, yet evidence on the effects of occupational noise during pregnancy remains limited.Methods: This multi-country register-based study (n=590,454) examined associations between occupational noise exposure estimated using a job-exposure matrix with daily 8-hour equivalent sound levels (<70, ≥70–<75, ≥75–<80, ≥80 dBA) and adverse birth outcomes in metropolitan areas of Denmark, Finland, Norway, and Sweden.Results: Results across the four countries were inconsistent. Indications of higher odds of LBW, SGA, and PTB, particularly in Norway and Sweden were seen, though these associations attenuated after full adjustment. Meta-analyses of fully adjusted models indicated no pooled associations between occupational noise exposure (≥80 dBA category) and birth weight (estimate = –0.35 g, 95% CI [–14.58, 13.88]), LBW (OR = 1.03, 95% CI [0.88, 1.20]), SGA (OR = 1.07, 95% CI [0.88, 1.29]), or PTB (OR = 1.01, 95% CI [0.83, 1.23]). However, sensitivity analyses based on the Swedish data suggested that working hours (full-time, part-time) and work absence may modify the association (birth weight: estimate = -37.66 g, 95% CI [-61.11, -14.22], LBW: OR = 1.49, 95% CI [1.22, 1.81], PTB: OR = 1.44, 95% CI [1.23, 1.70]).Conclusions: Overall, the findings do not support strong associations between occupational noise exposure and adverse birth outcomes. Based on sensitivity analysis, additional data on working hours and work absence may modify the relationship in future studies.
- 5 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessOccupational Noise Exposure During Pregnancy and Birth Outcomes: A Large-Scale Cohort Study in Nordic Countries
Objectives: Adverse birth outcomes such as low birth weight (LBW), small for gestational age (SGA) and preterm birth (PTB), are major indicators of perinatal health, yet evidence on the effects of occupational noise during pregnancy remains limited.Methods: This multi-country register-based study (n=590,454) examined associations between occupational noise exposure estimated using a job-exposure matrix with daily 8-hour equivalent sound levels (<70, ≥70–<75, ≥75–<80, ≥80 dBA) and adverse birth outcomes in metropolitan areas of Denmark, Finland, Norway, and Sweden.Results: Results across the four countries were inconsistent. Indications of higher odds of LBW, SGA, and PTB, particularly in Norway and Sweden were seen, though these associations attenuated after full adjustment. Meta-analyses of fully adjusted models indicated no pooled associations between occupational noise exposure (≥80 dBA category) and birth weight (estimate = –0.35 g, 95% CI [–14.58, 13.88]), LBW (OR = 1.03, 95% CI [0.88, 1.20]), SGA (OR = 1.07, 95% CI [0.88, 1.29]), or PTB (OR = 1.01, 95% CI [0.83, 1.23]). However, sensitivity analyses based on the Swedish data suggested that working hours (full-time, part-time) and work absence may modify the association (birth weight: estimate = -37.66 g, 95% CI [-61.11, -14.22], LBW: OR = 1.49, 95% CI [1.22, 1.81], PTB: OR = 1.44, 95% CI [1.23, 1.70]).Conclusions: Overall, the findings do not support strong associations between occupational noise exposure and adverse birth outcomes. Based on sensitivity analysis, additional data on working hours and work absence may modify the relationship in future studies.
- Open Access
- 9 December 2023
- Published by Oxford University Press (OUP) in Annals of Work Exposures and Health
- Vol. 68 (2) , 146-154
An updated job-exposure matrix for occupational noise: development and validation
Objectives: The aim of this study was to create a quantitative job-exposure matrix (JEM) for noise including a large set of measurements for the Swedish workforce, a detailed exposure-level assessment, spanning over an extensive time period from 1970 to 2014. Methods: The JEM was developed by 2 teams, each with an experienced occupational hygienist and an occupational safety engineer. Each pair assessed the exposure using measurements performed and reported by occupational hygienists, occupational safety engineers, or similar, from 1970 to 2014. The measurements included either the original LAeq(8h) measurements or an LAeq(8h) levels calculated from partial measurements of the working day, provided that the measurement targeted a regular task usually performed during a full workday. The collection of measurement reports was done in 2008 and 2012 by contacting clinics working in the area of occupational health or occupational safety engineers and their submitted reports were added to our own material. Noise exposure assessments were inserted at the appropriate time period for the relevant job family. The final matrix was developed in a consensus procedure and the validity was investigated by comparison of the 2 team’s individual results. Results: The noise JEM contains 321 job families with information regarding occupational noise from 1970 to 2014. The time-period label has a 5-yr scale starting in 1970. The estimated average 8 h (TWA) noise level in decibels [dB(A)] for every job family and 5-yr period was coded as 1: <70 dB(A), 2: 70 to 74 dB(A), 3: 75 to 79 dB(A), 4: 80 to 84 dB(A) or 5: 85(+) dB(A). The validation showed no systematic difference in relative position and very high agreement in the ordering of paired ordinal classifications. The JEM has also successfully been applied in several epidemiological studies. Conclusions: We present a JEM for occupational noise using Swedish data from 1970 to 2014 with a higher degree of sensitivity in assessed noise exposure compared with the previously existing version. Repeated application of the JEM, in epidemiological studies, has shown consistent results and contributed to yielding important findings.
- 9 December 2023
- Published by Oxford University Press (OUP) in Annals of Work Exposures and Health
- Vol. 68 (2) , 146-154
Open AccessAn updated job-exposure matrix for occupational noise: development and validation
Objectives: The aim of this study was to create a quantitative job-exposure matrix (JEM) for noise including a large set of measurements for the Swedish workforce, a detailed exposure-level assessment, spanning over an extensive time period from 1970 to 2014. Methods: The JEM was developed by 2 teams, each with an experienced occupational hygienist and an occupational safety engineer. Each pair assessed the exposure using measurements performed and reported by occupational hygienists, occupational safety engineers, or similar, from 1970 to 2014. The measurements included either the original LAeq(8h) measurements or an LAeq(8h) levels calculated from partial measurements of the working day, provided that the measurement targeted a regular task usually performed during a full workday. The collection of measurement reports was done in 2008 and 2012 by contacting clinics working in the area of occupational health or occupational safety engineers and their submitted reports were added to our own material. Noise exposure assessments were inserted at the appropriate time period for the relevant job family. The final matrix was developed in a consensus procedure and the validity was investigated by comparison of the 2 team’s individual results. Results: The noise JEM contains 321 job families with information regarding occupational noise from 1970 to 2014. The time-period label has a 5-yr scale starting in 1970. The estimated average 8 h (TWA) noise level in decibels [dB(A)] for every job family and 5-yr period was coded as 1: <70 dB(A), 2: 70 to 74 dB(A), 3: 75 to 79 dB(A), 4: 80 to 84 dB(A) or 5: 85(+) dB(A). The validation showed no systematic difference in relative position and very high agreement in the ordering of paired ordinal classifications. The JEM has also successfully been applied in several epidemiological studies. Conclusions: We present a JEM for occupational noise using Swedish data from 1970 to 2014 with a higher degree of sensitivity in assessed noise exposure compared with the previously existing version. Repeated application of the JEM, in epidemiological studies, has shown consistent results and contributed to yielding important findings.
- Open Access
- 6 April 2023
- Published by MDPI AG in Proceedings
- Vol. 86 (1) , 2
Hand-Arm Vibration Exposure Trends among the Workforce in Sweden
The aim of this research was to study hand-arm vibration (HAV) exposure trends in the workforce in Sweden by using a Job-Exposure matrix (JEM). All individuals employed during 1980 and 2010 with an occupational code were included. The daily eight-hour equivalent HAV exposure values were divided into three exposure categories. During the study period, the proportion of workers exposed above the action value had decreased, as well as the proportion of workers exposed to any HAV. In 2010, 4% of the workforce in Sweden were exposed to HAV above the action value.
- 6 April 2023
- Published by MDPI AG in Proceedings
- Vol. 86 (1) , 2
Open AccessHand-Arm Vibration Exposure Trends among the Workforce in Sweden
The aim of this research was to study hand-arm vibration (HAV) exposure trends in the workforce in Sweden by using a Job-Exposure matrix (JEM). All individuals employed during 1980 and 2010 with an occupational code were included. The daily eight-hour equivalent HAV exposure values were divided into three exposure categories. During the study period, the proportion of workers exposed above the action value had decreased, as well as the proportion of workers exposed to any HAV. In 2010, 4% of the workforce in Sweden were exposed to HAV above the action value.
- Open Access
- 1 May 2018
- Published by Elsevier in Chemosphere
- Vol. 198, 274-280
An improved method for determining dermal exposure to polycyclic aromatic hydrocarbons
- 1 May 2018
- Published by Elsevier in Chemosphere
- Vol. 198, 274-280
Open AccessAn improved method for determining dermal exposure to polycyclic aromatic hydrocarbons
- 1 September 2016
- Published by BMJ
- Vol. 73, A73-A74
O38-5 A job-exposure matrix for occupational noise – development, validation and update
Noise is one of the most common occupational hazardous exposures. In addition to hearing loss and tinnitus increasing evidence show that noise exposure is also associated with stress reactions, and cardiovascular effects. A Recent study (Selander et al. EHP 2015 Dec 8.) using an earlier version of our noise JEM, showed that noise exposure can negatively impact fetuses during pregnancy. We have developed, updated and validated a job exposure matrix for occupational noise using some 270 measurement reports from 1970 through 2014. The matrix contains 321 job families, as defined by the Nordic Occupational Classification system (NYK 85/90), and classifies 8 hour average noise exposure using a five grade scale 1: <70 dB(A), 2: 70 < 75 dB(A), 3: 75 < 80 dB(A), 4: 80 < 85 dB(A), 5: ≥85 dB(A). Exposure estimates were developed for every 5-year period during 1970–2014, yielding exposure development over 44 years. The matrix also includes assessments of the probability for impact noise exposure using a 4 grade scale1. Yes, for sure; 2. Yes, probably; 3. Yes, possibly; 4. Unlikely. Exposures were estimated by two separate teams, each comprised by one occupational hygienists and one experienced occupational safety engineer. Agreements were statistically analysed using Svensson’s non-parametric methods in terms of relative position (RP) and relative concentration (RC) Disagreements were discussed and a consensus was reached for the final version. In the final phase preliminary results indicate good agreement between the two teams concerning noise exposure levels, without systematic differences. However, impact noise source data were scarce and the team assessments were more split giving weaker agreement and a systematic difference. The consensus matrix will be made available to collaborators. A previous JEM, based on approximately half the number of reports and a 3-grade classification was published 2013. (Sjöström et al Ann Occup Hyg. 2013 Jul;57(6):774-83)
- 1 September 2016
- Published by BMJ
- Vol. 73, A73-A74
O38-5 A job-exposure matrix for occupational noise – development, validation and update
Noise is one of the most common occupational hazardous exposures. In addition to hearing loss and tinnitus increasing evidence show that noise exposure is also associated with stress reactions, and cardiovascular effects. A Recent study (Selander et al. EHP 2015 Dec 8.) using an earlier version of our noise JEM, showed that noise exposure can negatively impact fetuses during pregnancy. We have developed, updated and validated a job exposure matrix for occupational noise using some 270 measurement reports from 1970 through 2014. The matrix contains 321 job families, as defined by the Nordic Occupational Classification system (NYK 85/90), and classifies 8 hour average noise exposure using a five grade scale 1: <70 dB(A), 2: 70 < 75 dB(A), 3: 75 < 80 dB(A), 4: 80 < 85 dB(A), 5: ≥85 dB(A). Exposure estimates were developed for every 5-year period during 1970–2014, yielding exposure development over 44 years. The matrix also includes assessments of the probability for impact noise exposure using a 4 grade scale1. Yes, for sure; 2. Yes, probably; 3. Yes, possibly; 4. Unlikely. Exposures were estimated by two separate teams, each comprised by one occupational hygienists and one experienced occupational safety engineer. Agreements were statistically analysed using Svensson’s non-parametric methods in terms of relative position (RP) and relative concentration (RC) Disagreements were discussed and a consensus was reached for the final version. In the final phase preliminary results indicate good agreement between the two teams concerning noise exposure levels, without systematic differences. However, impact noise source data were scarce and the team assessments were more split giving weaker agreement and a systematic difference. The consensus matrix will be made available to collaborators. A previous JEM, based on approximately half the number of reports and a 3-grade classification was published 2013. (Sjöström et al Ann Occup Hyg. 2013 Jul;57(6):774-83)
- 4 February 2013
- Published by Oxford University Press (OUP) in Annals of Work Exposures and Health
- Vol. 57 (6) , 774-783
A Job-Exposure Matrix for Occupational Noise: Development and Validation
To develop a job-exposure matrix (JEM) for occupational noise in Sweden and to estimate its validity. The JEM, developed by a group of experienced occupational hygienists, contains 321 job families with information regarding occupational noise from 1970 to 2004. The occupational noise information derives from measurements collected from different sources. The time period label has a 5-year scale starting in 1970. The estimated average 8h (TWA) noise level in decibel [dB(A)] for every 5-year period was coded either as <75 dB(A), 75–84 dB(A), or ≥85 dB(A) and the risk of peak level exposure assessed. The validity of the JEM is tested, using Svensson’s non-parametric methods based on classification consensus, reached by a second group of occupational hygienists. Validation results show ~ 80% agreement and no systematic differences, in classification, between the two different groups of occupational hygienists, classifying the occupational noise exposure. However, classification of peak level exposure did show a systematic difference in relative position. The results will give more power to the JEM that it gives a good general estimate for the occupational noise levels in Sweden for different job families during 1970–2004. We, thus, intend to use it in further studies and also make it available to collaborators.
- 4 February 2013
- Published by Oxford University Press (OUP) in Annals of Work Exposures and Health
- Vol. 57 (6) , 774-783
A Job-Exposure Matrix for Occupational Noise: Development and Validation
To develop a job-exposure matrix (JEM) for occupational noise in Sweden and to estimate its validity. The JEM, developed by a group of experienced occupational hygienists, contains 321 job families with information regarding occupational noise from 1970 to 2004. The occupational noise information derives from measurements collected from different sources. The time period label has a 5-year scale starting in 1970. The estimated average 8h (TWA) noise level in decibel [dB(A)] for every 5-year period was coded either as <75 dB(A), 75–84 dB(A), or ≥85 dB(A) and the risk of peak level exposure assessed. The validity of the JEM is tested, using Svensson’s non-parametric methods based on classification consensus, reached by a second group of occupational hygienists. Validation results show ~ 80% agreement and no systematic differences, in classification, between the two different groups of occupational hygienists, classifying the occupational noise exposure. However, classification of peak level exposure did show a systematic difference in relative position. The results will give more power to the JEM that it gives a good general estimate for the occupational noise levels in Sweden for different job families during 1970–2004. We, thus, intend to use it in further studies and also make it available to collaborators.