umu.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Jonsson, Håkan
Publications (10 of 81) Show all publications
Häggström, C., Jonsson, H., Bjørge, T., Nagel, G., Manjer, J., Ulmer, H., . . . Stocks, T. (2019). Linear age-course effects on the associations between body mass index, triglycerides, and female breast and male liver cancer risk: An internal replication study of 800,000 individuals. International Journal of Cancer
Open this publication in new window or tab >>Linear age-course effects on the associations between body mass index, triglycerides, and female breast and male liver cancer risk: An internal replication study of 800,000 individuals
Show others...
2019 (English)In: International Journal of Cancer, ISSN 0020-7136, E-ISSN 1097-0215Article in journal (Refereed) Epub ahead of print
Abstract [en]

Apart from the consistently observed differential association between obesity and breast cancer risk by menopausal status, the associations between obesity and other metabolic imbalances with risks of cancers have not been systematically investigated across the age‐course. We created two random 50–50% cohorts from six European cohorts comprising 813,927 individuals. In the “discovery cohort”, we used Cox regression with attained age as time‐scale and tested interactions between body mass index (BMI), blood pressure, plasma glucose, triglycerides and cholesterol, and attained age in relation to cancer risk. Results with a p‐value below 0.05 were additionally tested in the “replication cohort” where a replicated result was considered evidence of a linear interaction with attained age. These findings were investigated by flexible parametric survival models for any age‐plateaus in their shape of associations with cancer risk across age. Consistent with other studies, BMI was negatively related to breast cancer risk (n cases = 11,723) among younger (premenopausal) women. However, the association remained negative for several years after menopause and, although gradually weakening over age, the association became positive only at 62 years of age. This linear and positive age‐interaction was also found for triglycerides and breast cancer, and for BMI and triglycerides in relation to liver cancer among men (n cases = 444). These findings are unlikely to be due to chance owing to the replication. The linear age‐interactions in breast cancer may suggest an influence by other age‐related factors than menopause; however, further investigation of age‐related effect modifiers in both breast and liver cancer is needed.

Place, publisher, year, edition, pages
John Wiley & Sons, 2019
Keywords
age interaction, cancer risk, cohort study, metabolic factors, survival analysis
National Category
Cancer and Oncology
Identifiers
urn:nbn:se:umu:diva-157853 (URN)10.1002/ijc.32240 (DOI)30815851 (PubMedID)
Available from: 2019-04-04 Created: 2019-04-04 Last updated: 2019-04-05
diva2:1275153
Open this publication in new window or tab >>Estimation of overdiagnosis in breast cancer screening using a non-homogeneous multi-state model: a simulation study
2018 (English)In: Journal of Medical Screening, ISSN 0969-1413, E-ISSN 1475-5793, Vol. 25, no 4, p. 183-190Article in journal (Refereed) Published
Abstract [en]

Objectives: Overdiagnosis is regarded as a harm of screening. We aimed to develop a non-homogeneous multi-state model to consider the age-specific transition rates for estimation of overdiagnosis, to validate the model by a simulation study where the true frequency of overdiagnosis can be calculated, and to compare our estimate with the cumulative incidence method. Methods: We constructed a four-state model to describe the natural history of breast cancer. The latent disease progression and the observed states for each individual were simulated in a trial with biennial screening of women aged 51-69 and a control group of the same size without screening. We performed 100 repetitions of the simulation with one million women to evaluate the performance of estimates. A sensitivity analysis with reduced number of controls was performed to imitate the data from the service screening programme. Results Based on the 100 repetitions, the mean value of the true frequency of overdiagnosis was 12.5% and the average estimates by the cumulative incidence method and the multi-state model were 12.9% (interquartile range: 2.46%) and 13.4% (interquartile range: 2.16%), respectively. The multi-state model had a greater bias of overdiagnosis than the cumulative incidence method, but the variation in the estimates was smaller. When the number of unscreened group was reduced, the variation of multi-state model estimates increased. Conclusions: The multi-state model produces a proper estimate of overdiagnosis and the results are comparable with the cumulative incidence method. The multi-state model can be used in the estimation of overdiagnosis, and might be useful for the ongoing service screening programmes.

Place, publisher, year, edition, pages
Sage Publications, 2018
Keywords
Overdiagnosis, breast cancer screening, mammography, multi-state model, cumulative incidence method
National Category
Public Health, Global Health, Social Medicine and Epidemiology
Identifiers
urn:nbn:se:umu:diva-154901 (URN)10.1177/0969141317733294 (DOI)000452324300004 ()29153013 (PubMedID)
Available from: 2019-01-04 Created: 2019-01-04 Last updated: 2019-01-04Bibliographically approved
Wu, W.-Y. Y., Törnberg, S., Elfström, K. M., Liu, X., Nyström, L. & Jonsson, H. (2018). Overdiagnosis in the population-based organized breast cancer screening program estimated by a non-homogeneous multi-state model: a cohort study using individual data with long-term follow-up. Breast Cancer Research, 20, Article ID 153.
Open this publication in new window or tab >>Overdiagnosis in the population-based organized breast cancer screening program estimated by a non-homogeneous multi-state model: a cohort study using individual data with long-term follow-up
Show others...
2018 (English)In: Breast Cancer Research, ISSN 1465-5411, E-ISSN 1465-542X, Vol. 20, article id 153Article in journal (Refereed) Published
Abstract [en]

Background: Overdiagnosis, defined as the detection of a cancer that would not become clinically apparent in a woman’s lifetime without screening, has become a growing concern. Similar underlying risk of breast cancer in the screened and control groups is a prerequisite for unbiased estimates of overdiagnosis, but a contemporary control group is usually not available in organized screening programs.

Methods: We estimated the frequency of overdiagnosis of breast cancer due to screening in women 50–69 years old by using individual screening data from the population-based organized screening program in Stockholm County 1989–2014. A hidden Markov model with four latent states and three observed states was constructed to estimate the natural progression of breast cancer and the test sensitivity. Piecewise transition rates were used to consider the time-varying transition rates. The expected number of detected non-progressive breast cancer cases was calculated.

Results: During the study period, 2,333,153 invitations were sent out; on average, the participation rate in the screening program was 72.7% and the average recall rate was 2.48%. In total, 14,648 invasive breast cancer cases were diagnosed; among the 8305 screen-detected cases, the expected number of non-progressive breast cancer cases was 35.9, which is equivalent to 0.43% (95% confidence interval (CI) 0.10%–2.2%) overdiagnosis. The corresponding estimates for the prevalent and subsequent rounds were 15.6 (0.87%, 95% CI 0.20%–4.3%) and 20.3 (0.31%, 95% CI 0.07%–1.6%), respectively. The likelihood ratio test showed that the non-homogeneous model fitted the data better than an age-homogeneous model (P<0.001).

Conclusions: Our findings suggest that overdiagnosis in the organized biennial mammographic screening for women 50–69 in Stockholm County is a minor phenomenon. The frequency of overdiagnosis in the prevalent screening round was higher than that in subsequent rounds. The non-homogeneous model performed better than the simpler, traditional homogeneous model.

Place, publisher, year, edition, pages
BioMed Central, 2018
Keywords
Overdiagnosis, Breast cancer, Organized screening program, Mammography, Multi-state model
National Category
Cancer and Oncology
Identifiers
urn:nbn:se:umu:diva-155103 (URN)10.1186/s13058-018-1082-z (DOI)000453551900001 ()30558679 (PubMedID)
Funder
Swedish Research CouncilVästerbotten County Council
Available from: 2019-01-10 Created: 2019-01-10 Last updated: 2019-01-10Bibliographically approved
Teleka, S., Häggström, C., Nagel, G., Bjorge, T., Manjer, J., Ulmer, H., . . . Stocks, T. (2018). Risk of bladder cancer by disease severity in relation to metabolic factors and smoking: a prospective pooled cohort study of 800,000 men and women. International Journal of Cancer, 143(12), 3071-3082
Open this publication in new window or tab >>Risk of bladder cancer by disease severity in relation to metabolic factors and smoking: a prospective pooled cohort study of 800,000 men and women
Show others...
2018 (English)In: International Journal of Cancer, ISSN 0020-7136, E-ISSN 1097-0215, Vol. 143, no 12, p. 3071-3082Article in journal (Refereed) Published
Abstract [en]

Previous studies on metabolic factors and bladder cancer (BC) risk have shown inconsistent results and have commonly not investigated associations separately by sex, smoking, and tumor invasiveness. Among 811,633 participants in six European cohorts, we investigated sex‐specific associations between body mass index (BMI), mid‐blood pressure (BP, [systolic + diastolic]/2), plasma glucose, triglycerides, total cholesterol and risk of BC overall, non‐muscle invasive BC (NMIBC) and muscle invasive BC (MIBC). Among men, we additionally assessed additive interactions between metabolic factors and smoking on BC risk. During follow‐up, 2,983 men and 754 women were diagnosed with BC. Among men, triglycerides and BP were positively associated with BC risk overall (hazard ratio [HR] per standard deviation [SD]: 1.17 [95% confidence interval (CI) 1.06–1.27] and 1.09 [1.02–1.17], respectively), and among women, BMI was inversely associated with risk (HR: 0.90 [0.82–0.99]). The associations for BMI and BP differed between men and women (pinteraction ≤ 0.005). Among men, BMI, cholesterol and triglycerides were positively associated with risk for NMIBC (HRs: 1.09 [95% CI 1.01–1.18], 1.14 [1.02–1.25], and 1.30 [1.12–1.48] respectively), and BP was positively associated with MIBC (HR: 1.23 [1.02–1.49]). Among women, glucose was positively associated with MIBC (HR: 1.99 [1.04–3.81]). Apart from cholesterol, HRs for metabolic factors did not significantly differ between MIBC and NMIBC, and there were no interactions between smoking and metabolic factors on BC. Our study supports an involvement of metabolic aberrations in BC risk. Whilst some associations were significant only in certain sub‐groups, there were generally no significant differences in associations by smoking or tumor invasiveness.

Place, publisher, year, edition, pages
Wiley-Blackwell, 2018
Keywords
bladder cancer, metabolic factors, smoking, non-muscle invasive bladder cancer, muscle-invasive bladder cancer
National Category
Urology and Nephrology
Identifiers
urn:nbn:se:umu:diva-154026 (URN)10.1002/ijc.31597 (DOI)000451115900003 ()29756343 (PubMedID)
Funder
Swedish Research Council, 2015-02322
Available from: 2018-12-20 Created: 2018-12-20 Last updated: 2018-12-20Bibliographically approved
Glimelius, B., Melin, B., Enblad, G., Alafuzoff, I., Beskow, A., Ahlström, H., . . . Sjöblom, T. (2018). U-CAN: a prospective longitudinal collection of biomaterials and clinical information from adult cancer patients in Sweden. Acta Oncologica, 57(2), 187-194
Open this publication in new window or tab >>U-CAN: a prospective longitudinal collection of biomaterials and clinical information from adult cancer patients in Sweden
Show others...
2018 (English)In: Acta Oncologica, ISSN 0284-186X, E-ISSN 1651-226X, Vol. 57, no 2, p. 187-194Article in journal (Refereed) Published
Abstract [en]

Background: Progress in cancer biomarker discovery is dependent on access to high-quality biological materials and high-resolution clinical data from the same cases. To overcome current limitations, a systematic prospective longitudinal sampling of multidisciplinary clinical data, blood and tissue from cancer patients was therefore initiated in 2010 by Uppsala and Umea Universities and involving their corresponding University Hospitals, which are referral centers for one third of the Swedish population.

Material and Methods: Patients with cancer of selected types who are treated at one of the participating hospitals are eligible for inclusion. The healthcare-integrated sampling scheme encompasses clinical data, questionnaires, blood, fresh frozen and formalin-fixed paraffin-embedded tissue specimens, diagnostic slides and radiology bioimaging data.

Results: In this ongoing effort, 12,265 patients with brain tumors, breast cancers, colorectal cancers, gynecological cancers, hematological malignancies, lung cancers, neuroendocrine tumors or prostate cancers have been included until the end of 2016. From the 6914 patients included during the first five years, 98% were sampled for blood at diagnosis, 83% had paraffin-embedded and 58% had fresh frozen tissues collected. For Uppsala County, 55% of all cancer patients were included in the cohort.

Conclusions: Close collaboration between participating hospitals and universities enabled prospective, longitudinal biobanking of blood and tissues and collection of multidisciplinary clinical data from cancer patients in the U-CAN cohort. Here, we summarize the first five years of operations, present U-CAN as a highly valuable cohort that will contribute to enhanced cancer research and describe the procedures to access samples and data.

Place, publisher, year, edition, pages
Taylor & Francis, 2018
National Category
Cancer and Oncology
Identifiers
urn:nbn:se:umu:diva-144850 (URN)10.1080/0284186X.2017.1337926 (DOI)000423473200003 ()28631533 (PubMedID)
Available from: 2018-02-23 Created: 2018-02-23 Last updated: 2018-06-09Bibliographically approved
Stattin, P., Sandin, F., Birkebæk Thomsen, F., Garmo, H., Robinson, D., Franck Lissbrant, I., . . . Bratt, O. (2017). Association of Radical Local Treatment with Mortality in Men with Very High-risk Prostate Cancer: A Semiecologic, Nationwide, Population-based Study. European Urology, 72(1), 125-134
Open this publication in new window or tab >>Association of Radical Local Treatment with Mortality in Men with Very High-risk Prostate Cancer: A Semiecologic, Nationwide, Population-based Study
Show others...
2017 (English)In: European Urology, ISSN 0302-2838, E-ISSN 1873-7560, Vol. 72, no 1, p. 125-134Article in journal (Refereed) Published
Abstract [en]

Background: Current guidelines recommend androgen deprivation therapy only for men with very high-risk prostate cancer (PCa), but there is little evidence to support this stance.

Objective: To investigate the association between radical local treatment and mortality in men with very high-risk PCa.

Design, setting, and participants: Semiecologic study of men aged <80 yr within the Prostate Cancer data Base Sweden, diagnosed in 1998–2012 with very high-risk PCa (local clinical stage T4 and/or prostate-specific antigen [PSA] level 50–200 ng/ml, any N, and M0). Men with locally advanced PCa (local clinical stage T3 and PSA level <50 ng/ml, any N, and M0) were used as positive controls.

Intervention: Proportion of men who received prostatectomy or full-dose radiotherapy in 640 experimental units defined by county, diagnostic period, and age at diagnosis.

Outcome measurements and statistical analysis: PCa and all-cause mortality rate ratios (MRRs).

Results and limitations: Both PCa and all-cause mortality were half as high in units in the highest tertile of exposure to radical local treatment compared with units in the lowest tertile (PCa MRR: 0.51; 95% confidence interval [CI], 0.28–0.95; and all-cause MRR: 0.56; 95% CI, 0.33–0.92). The results observed for locally advanced PCa for highest versus lowest tertile of exposure were in agreement with results from randomized trials (PCa MRR: 0.75; 95% CI, 0.60–0.94; and all-cause MRR: 0.85; 95% CI, 0.72–1.00). Although the semiecologic design minimized selection bias on an individual level, the effect of high therapeutic activity could not be separated from that of high diagnostic activity.

Conclusions: The substantially lower mortality in units with the highest exposure to radical local treatment suggests that radical treatment decreases mortality even in men with very high-risk PCa for whom such treatment has been considered ineffective.

Patient summary: Men with very high-risk prostate cancer diagnosed and treated in units with the highest exposure to surgery or radiotherapy had a substantially lower mortality.

Place, publisher, year, edition, pages
Elsevier, 2017
Keywords
Prostate cancer, Very high-risk, Radical, Treatment, Semiecologic, Nationwide, Population-based, PCBaSe
National Category
Urology and Nephrology
Identifiers
urn:nbn:se:umu:diva-137374 (URN)10.1016/j.eururo.2016.07.023 (DOI)000403205900026 ()27481175 (PubMedID)
Available from: 2017-07-06 Created: 2017-07-06 Last updated: 2018-06-09Bibliographically approved
Burström, L., Jonsson, H., Björ, B., Hjalmarsson, U., Nilsson, T., Reuterwall, C. & Wahlström, J. (2016). Daily text messages used as a method for assessing low back pain among workers. Journal of Clinical Epidemiology (70), 45-51
Open this publication in new window or tab >>Daily text messages used as a method for assessing low back pain among workers
Show others...
2016 (English)In: Journal of Clinical Epidemiology, ISSN 0895-4356, E-ISSN 1878-5921, no 70, p. 45-51Article in journal (Refereed) Published
Abstract [en]

OBJECTIVES: To evaluate a method for collecting data concerning low back pain (LBP) using daily text messages and to characterize the reported LBP in terms of intensity, variability, and episodes.

STUDY DESIGN AND SETTING: We conducted a cohort study of LBP among workers used by a mining company. The participants were asked to answer the question "How much pain have you had in your lower back in the last 24 hours on a scale from 0 to 10, where 0 = no pain and 10 = the worst pain imaginable" once a day for 5 weeks, with this process being repeated 6 months later.

RESULTS: A total of 121 workers participated in the first period of data collection, and 108 participated in the second period. The daily response rate was 93% for both periods, and cluster analysis was shown to be a feasible statistical method for clustering LBP into subgroups of low, medium, and high pain. The daily text messages method also worked well for assessing the episodic nature of LBP.

CONCLUSION: We have demonstrated a method for repeatedly measuring of LBP using daily text messages. The data permitted clustering into subgroups and could be used to define episodes of LBP.

National Category
Occupational Health and Environmental Health
Identifiers
urn:nbn:se:umu:diva-111623 (URN)10.1016/j.jclinepi.2015.08.011 (DOI)000370676900006 ()26342444 (PubMedID)
Funder
Forte, Swedish Research Council for Health, Working Life and Welfare, 2011-0494
Available from: 2015-11-18 Created: 2015-11-18 Last updated: 2018-06-07Bibliographically approved
Björ, O., Jonsson, H., Damber, L., Burström, L. & Nilsson, T. (2016). Is outdoor work associated with elevated rates of cerebrovascular disease mortality?: a cohort study based on iron-ore mining. Journal of Occupational Medicine and Toxicology, 11, Article ID 40.
Open this publication in new window or tab >>Is outdoor work associated with elevated rates of cerebrovascular disease mortality?: a cohort study based on iron-ore mining
Show others...
2016 (English)In: Journal of Occupational Medicine and Toxicology, ISSN 1745-6673, E-ISSN 1745-6673, Vol. 11, article id 40Article in journal (Refereed) Published
Abstract [en]

BACKGROUND: A cohort study that examined iron ore mining found negative associations between cumulative working time employed underground and several outcomes, including mortality of cerebrovascular diseases. In this cohort study, and using the same group of miners, we examined whether work in an outdoor environment could explain elevated cerebrovascular disease rates.

METHODS: This study was based on a Swedish iron ore mining cohort consisting of 13,000 workers. Poisson regression models were used to generate smoothed estimates of standardized mortality ratios and adjusted rate ratios, both models by cumulative exposure time in outdoor work.

RESULTS: The adjusted rate ratio between employment classified as outdoor work ≥25 years and outdoor work 0-4 years was 1.62 (95 % CI 1.07-2.42). The subgroup underground work ≥15 years deviated most in occurrence of cerebrovascular disease mortality compared with the external reference population: SMR (0.70 (95 % CI 0.56-0.85)).

CONCLUSIONS: Employment in outdoor environments was associated with elevated rates of cerebrovascular disease mortality. In contrast, work in tempered underground employment was associated with a protecting effect.

Keywords
Poisson regression, Occupation, Standardized mortality ratio, Mortality, Cohort study, Mining
National Category
Public Health, Global Health, Social Medicine and Epidemiology
Identifiers
urn:nbn:se:umu:diva-125120 (URN)10.1186/s12995-016-0131-8 (DOI)000384698600001 ()27570536 (PubMedID)
Available from: 2016-09-06 Created: 2016-09-06 Last updated: 2018-06-07Bibliographically approved
Björ, O., Damber, L., Jonsson, H. & Nilsson, T. (2015). A comparison between standard methods and structural nested modelling when bias from a healthy worker survivor effect is suspected: an iron-ore mining cohort study. Occupational and Environmental Medicine, 72(7), 536-542
Open this publication in new window or tab >>A comparison between standard methods and structural nested modelling when bias from a healthy worker survivor effect is suspected: an iron-ore mining cohort study
2015 (English)In: Occupational and Environmental Medicine, ISSN 1351-0711, E-ISSN 1470-7926, Vol. 72, no 7, p. 536-542Article in journal (Refereed) Published
Abstract [en]

Objectives Iron-ore miners are exposed to extremely dusty and physically arduous work environments. The demanding activities of mining select healthier workers with longer work histories (ie, the Healthy Worker Survivor Effect (HWSE)), and could have a reversing effect on the exposure-response association. The objective of this study was to evaluate an iron-ore mining cohort to determine whether the effect of respirable dust was confounded by the presence of an HWSE. Methods When an HWSE exists, standard modelling methods, such as Cox regression analysis, produce biased results. We compared results from g-estimation of accelerated failure-time modelling adjusted for HWSE with corresponding unadjusted Cox regression modelling results. Results For all-cause mortality when adjusting for the HWSE, cumulative exposure from respirable dust was associated with a 6% decrease of life expectancy if exposed >= 15 years, compared with never being exposed. Respirable dust continued to be associated with mortality after censoring outcomes known to be associated with dust when adjusting for the HWSE. In contrast, results based on Cox regression analysis did not support that an association was present. Conclusions The adjustment for the HWSE made a difference when estimating the risk of mortality from respirable dust. The results of this study, therefore, support the recommendation that standard methods of analysis should be complemented with structural modelling analysis techniques, such as g-estimation of accelerated failure-time modelling, to adjust for the HWSE.

Keywords
Cox regression, occupation, standardized incidence ratio, standardized mortality ratio
National Category
Occupational Health and Environmental Health
Research subject
Occupational and Environmental Medicine
Identifiers
urn:nbn:se:umu:diva-81789 (URN)10.1136/oemed-2014-102251 (DOI)000356298200012 ()
Note

Originally published in manuscript form with the title A comparison between standard methods and structural nested modeling when adjusting for the healthy worker survivor effect. 

Available from: 2013-10-22 Created: 2013-10-22 Last updated: 2018-06-08Bibliographically approved
Hellquist, B. N., Czene, K., Hjälm, A., Nyström, L. & Jonsson, H. (2015). Effectiveness of population-based service screening with mammography for women ages 40 to 49 years with a high or low risk of breast cancer: socioeconomic status, parity, and age at birth of first child. Cancer, 121(2), 251-258
Open this publication in new window or tab >>Effectiveness of population-based service screening with mammography for women ages 40 to 49 years with a high or low risk of breast cancer: socioeconomic status, parity, and age at birth of first child
Show others...
2015 (English)In: Cancer, ISSN 0008-543X, E-ISSN 1097-0142, Vol. 121, no 2, p. 251-258Article in journal (Refereed) Published
Abstract [en]

BACKGROUND: Invitation to mammography screening of women aged 40 to 49 years is a matter of debate in many countries and a cost-effective alternative in countries without screening among women aged 40 to 49 years could be inviting those at higher risk. The relative effectiveness of mammography screening was estimated for subgroups based on the breast cancer risk factors parity, age at time of birth of first child, and socioeconomic status (SES).

METHODS: The SCReening of Young Women (SCRY) database consists of all women aged 40 to 49 years in Sweden between 1986 and 2005 and was split into a study and control group. The study group consisted of women residing in areas in which women aged 40 to 49 years were invited to screening and the control group of women in areas in which women aged 40 to 49 years were not invited to screening. Rate ratio (RR) estimates were calculated for 2 exposures: invitation and attendance.

RESULTS: There were striking similarities noted in the RR pattern for women invited to and attending screening and no statistically significant difference or trend in the RR was noted by risk group. The RR estimates increased by increasing parity for parity of 0 to 2 and ranged from 0.55 (95% confidence interval [95% CI], 0.38-0.79) to 0.79 (95% CI, 0.65-0.95) for attending women. The RR for women with high SES was lower than that for women with low SES (RR, 0.72 [95% CI, 0.60-0.86] and RR, 0.79 [95% CI, 0.63-0.99], respectively). For women aged 20 to 24 years at the time of the birth of their first child, the RR was 0.73 (95% CI, 0.58-0.91) and estimates for other ages were similar.

CONCLUSIONS: There was no statistically significant difference noted in the relative effectiveness of mammography screening by parity, age at the time of birth of the first child, or SES. Cancer 2014.

Keywords
mammography, cancer screening, breast cancer, mortality, risk factors, socioeconomic status (SES), parity
National Category
Cancer and Oncology
Identifiers
urn:nbn:se:umu:diva-95817 (URN)10.1002/cncr.29011 (DOI)000347540900012 ()25242087 (PubMedID)881251 (Local ID)881251 (Archive number)881251 (OAI)
Available from: 2014-11-05 Created: 2014-11-05 Last updated: 2019-02-15Bibliographically approved
Organisations

Search in DiVA

Show all publications