Anton Nilsson

Anton Nilsson

Auto-generated profile, based on publicly available information

Anton Nilsson

Epidemiology, Population Studies and Infrastructures EPI@LUND, Department of Laboratory Medicine, Lund University , Lund

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2014 in American Economic Journal: Applied Economics
Verified
Journal Article
Closed access
PDF file
PLPetter Lundborg
Petter Lundborg
ANAnton Nilsson
Anton Nilsson
DRDan-Olof Rooth
Dan-Olof Rooth
We use the Swedish compulsory school reform to estimate the causal effect of parental education on sons’ outcomes. To this end, we use data from the Swedish military enlistment register on the entire population of males and consider outcomes, such as cognitive skills, noncognitive skills, and various dimensions of health at the age of 18. We find positive effects of maternal education on sons’ skills and health status but no effects of paternal education. One reason behind this result may be that the fathers affected by the reform did not face any labor market returns to their increased schooling. (JEL I21, J12, J24)

1 September 2014 in Journal of Health Economics
Verified
Journal Article
Closed access
PLPetter Lundborg
Petter Lundborg
ANAnton Nilsson
Anton Nilsson
DRDan-Olof Rooth
Dan-Olof Rooth

9 February 2021 in The Journal of Clinical Endocrinology & Metabolism
Verified
Journal Article
Open access
PDF file
ANAnton Nilsson
Anton Nilsson
KTKleoniki Tsoumani
Kleoniki Tsoumani
TPTereza Planck
Tereza Planck
Context/objective The aim of this study was to examine the effect of statins and other lipid-lowering agents on the development of Graves orbitopathy (GO) in patients with newly diagnosed Graves disease (GD). Methods Our sample included the full adult population of individuals living in Sweden with newly diagnosed GD between 2005 and 2018 (n = 34 894). We compared the GO incidence in statin users (n = 5574) and nonusers (n = 34 409) by applying Cox regression with a time-varying exposure variable. We adjusted for age, sex, and treatment for hyperthyroidism in the multivariate analyses. Results Periods of nonusage lasted for a median of 4.3 years (interquartile range [IQR] 1.2-8.4), whereas periods of usage lasted for a median of 4.7 years (IQR 2.0-8.1). Among statin users, 77.1% had used simvastatin, 28.9% atorvastatin, and 8.2% had used other statins. Statin users were found to be significantly less likely to develop GO. In the main analysis based on the full cohort, the unadjusted hazard ratio (HR) was 0.74 (CI 0.65-0.84, P < .001), whereas full adjustment altered the effect to 0.87 (CI 0.76-1.00, P = .04). The main results were largely driven by men; the fully adjusted HR was 0.78 (CI 0.58-1.04, P = .09) for men and 0.91 (CI 0.79-1.06, P = .24) for women. Lipid-lowering agents other than statins did not exhibit a similar protective effect. Conclusion In newly diagnosed patients with GD, treatment with statins may protect against the development of GO. Statins should be investigated in a clinical trial as a preventive treatment for GO in newly diagnosed patients with GD.

22 November 2020 in International Journal of Epidemiology
Verified
Journal Article
Open access
PDF file
ANAnton Nilsson
Anton Nilsson
CBCarl Bonander
Carl Bonander
USUlf Strömberg
Ulf Strömberg
JBJonas Björk
Jonas Björk
Background: Directed acyclic graphs (DAGs) are of great help when researchers try to understand the nature of causal relationships and the consequences of conditioning on different variables. One fundamental feature of causal relations that has not been incorporated into the standard DAG framework is interaction, i.e. when the effect of one variable (on a chosen scale) depends on the value that another variable is set to. In this paper, we propose a new type of DAG—the interaction DAG (IDAG), which can be used to understand this phenomenon. Methods: The IDAG works like any DAG but instead of including a node for the outcome, it includes a node for a causal effect. We introduce concepts such as confounded interaction and total, direct and indirect interaction, showing that these can be depicted in ways analogous to how similar concepts are depicted in standard DAGs. This also allows for conclusions on which treatment interactions to account for empirically. Moreover, since generalizability can be compromised in the presence of underlying interactions, the framework can be used to illustrate threats to generalizability and to identify variables to account for in order to make results valid for the target population. Conclusions: The IDAG allows for a both intuitive and stringent way of illustrating interactions. It helps to distinguish between causal and non-causal mechanisms behind effect variation. Conclusions about how to empirically estimate interactions can be drawn—as well as conclusions about how to achieve generalizability in contexts where interest lies in estimating an overall effect.

3 May 2021 in Thyroid®
Verified
Journal Article
Closed access
TPTereza Planck
Tereza Planck
FHFredric Hedberg
Fredric Hedberg
JCJan Calissendorff
Jan Calissendorff
ANAnton Nilsson
Anton Nilsson
Background: The prescription of liothyronine (LT3) to treat hypothyroidism is increasing worldwide; however, the long-term safety of LT3 use has yet to be determined. Previous studies have suggested a possible association between LT3 use and breast cancer. The aim of this study was to examine the effects of LT3 use on cancer incidence and mortality. Methods: Our sample included the full adult population of individuals living in Sweden with at least three purchases of thyroid hormone therapy between July 2005 and December 2017. Individual-level data on drug purchases were linked to registry data on cancer incidence and mortality. There were 575,461 individuals with at least three purchases, of which 11,147 had made at least three purchases of LT3, including combinations of levothyroxine (LT4) and LT3. Individuals were followed for a median follow-up time of 8.1 years. We applied Cox regression with a time-varying exposure variable, comparing LT3 users (individuals with at least three cumulative purchases of LT3) with LT4-only users (the rest). Outcomes included breast cancer incidence, any cancer incidence, all-cause mortality, any cancer mortality, and breast cancer mortality. We adjusted for age, sex, previous thyroid cancer, previous other cancer, use of antithyroid preparations, use of sex hormones, and dose in multivariate analyses. Results: Multivariate analyses produced a hazard ratio of 0.93 (95% confidence interval [0.75–1.15]) for breast cancer incidence (only females), 0.97 (0.87–1.08) for any cancer incidence, 0.69 (0.61–0.77) for all-cause mortality, 0.78 (0.62–0.98) for any cancer mortality, and 0.91 (0.50–1.66) for breast cancer mortality (only females). Conclusions: In this large, Swedish, long-term registry-based study, the use of LT3 did not lead to increased breast cancer incidence, any cancer incidence, all-cause mortality, any cancer mortality, or breast cancer mortality compared with LT4 use. Somewhat surprisingly, there was evidence of lower mortality in LT3 users in models adjusting for dose, potentially an artifact of underlying associations between dose and health status/diagnosis.

8 March 2022 in PLOS ONE
Verified
Journal Article
Open access
PDF file
CBCarl Bonander
Carl Bonander
ANAnton Nilsson
Anton Nilsson
JBJonas Björk
Jonas Björk
ABAnders Blomberg
Anders Blomberg
GEGunnar Engström
Gunnar Engström
USUlf Strömberg
Ulf Strömberg
Objectives: To study the value of combining individual- and neighborhood-level sociodemographic data to predict study participation and assess the effects of baseline selection on the distribution of metabolic risk factors and lifestyle factors in the Swedish CardioPulmonary bioImage Study (SCAPIS). Methods: We linked sociodemographic register data to SCAPIS participants (n = 30,154, ages: 50–64 years) and a random sample of the study’s target population (n = 59,909). We assessed the classification ability of participation models based on individual-level data, neighborhood-level data, and combinations of both. Standardized mean differences (SMD) were used to examine how reweighting the sample to match the population affected the averages of 32 cardiopulmonary risk factors at baseline. Absolute SMDs >0.10 were considered meaningful. Results: Combining both individual-level and neighborhood-level data gave rise to a model with better classification ability (AUC: 71.3%) than models with only individual-level (AUC: 66.9%) or neighborhood-level data (AUC: 65.5%). We observed a greater change in the distribution of risk factors when we reweighted the participants using both individual and area data. The only meaningful change was related to the (self-reported) frequency of alcohol consumption, which appears to be higher in the SCAPIS sample than in the population. The remaining risk factors did not change meaningfully. Conclusions: Both individual- and neighborhood-level characteristics are informative in assessing study selection effects. Future analyses of cardiopulmonary outcomes in the SCAPIS cohort can benefit from our study, though the average impact of selection on risk factor distributions at baseline appears small.