Umeå University's logo

umu.sePublications
Change search
Link to record
Permanent link

Direct link
Waernbaum, Ingeborg
Alternative names
Publications (10 of 30) Show all publications
Gorbach, T., de Luna, X., Waernbaum, I. & Karvanen, J. (2023). Contrasting identifying assumptions of average causal effects: robustness and semiparametric efficiency. Journal of machine learning research, 24(197), 1-65
Open this publication in new window or tab >>Contrasting identifying assumptions of average causal effects: robustness and semiparametric efficiency
2023 (English)In: Journal of machine learning research, ISSN 1532-4435, E-ISSN 1533-7928, Vol. 24, no 197, p. 1-65Article in journal (Refereed) Published
Abstract [en]

Semiparametric inference on average causal effects from observational data is based on assumptions yielding identification of the effects. In practice, several distinct identifying assumptions may be plausible; an analyst has to make a delicate choice between these models. In this paper, we study three identifying assumptions based on the potential outcome framework:  the back-door assumption, which uses pre-treatment covariates, the front-door assumption, which uses mediators, and the two-door assumption using pre-treatment covariates and mediators simultaneously. We provide the efficient influence functions and the corresponding semiparametric efficiency bounds that hold under these assumptions, and their combinations. We demonstrate that neither of the identification models provides uniformly the most efficient estimation and give conditions under which some bounds are lower than others. We show when semiparametric estimating equation estimators based on influence functions  attain the bounds, and study the robustness of the estimators to misspecification of the nuisance models. The theory is complemented with simulation experiments on the finite sample behavior of the estimators. The results obtained are relevant for an analyst facing a choice between several plausible identifying assumptions and corresponding estimators. Our results show that this choice implies a trade-off between efficiency and robustness to misspecification of the nuisance models. 

Keywords
causal inference, efficiency bound, robustness, back-door, front-door
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:umu:diva-190082 (URN)
Funder
Forte, Swedish Research Council for Health, Working Life and Welfare, 2018-00852Swedish Research Council, 2018-02670Swedish Research Council, 2016-00703Marianne and Marcus Wallenberg Foundation, 2015.0060Academy of Finland, 311877
Available from: 2021-12-03 Created: 2021-12-03 Last updated: 2023-09-04Bibliographically approved
Eikenboom, A. M., Le Cessie, S., Waernbaum, I., Groenwold, R. H. H. & De Boer, M. G. J. (2022). Quality of Conduct and Reporting of Propensity Score Methods in Studies Investigating the Effectiveness of Antimicrobial Therapy. Open Forum Infectious Diseases, 9(4), Article ID ofac110.
Open this publication in new window or tab >>Quality of Conduct and Reporting of Propensity Score Methods in Studies Investigating the Effectiveness of Antimicrobial Therapy
Show others...
2022 (English)In: Open Forum Infectious Diseases, E-ISSN 2328-8957, Vol. 9, no 4, article id ofac110Article in journal (Refereed) Published
Abstract [en]

Background: Propensity score methods are becoming increasingly popular in infectious disease medicine to correct for confounding in observational studies. However, applying and reporting propensity score techniques correctly requires substantial knowledge of these methods. The quality of conduct and reporting of propensity score methods in studies investigating the effectiveness of antimicrobial therapy is yet undetermined.

Methods: A systematic review was performed to provide an overview of studies (2005-2020) on the effectiveness of antimicrobial therapy that used propensity score methods. A quality assessment tool and a standardized quality score were developed to evaluate a subset of studies in which antibacterial therapy was investigated in detail. The scale of this standardized score ranges between 0 (lowest quality) and 100 (excellent).

Results: A total of 437 studies were included. The absolute number of studies that investigated the effectiveness of antimicrobial therapy and that used propensity score methods increased 15-fold between the periods 2005-2009 and 2015-2019. Propensity score matching was the most frequently applied technique (65%), followed by propensity score-adjusted multivariable regression (25%). A subset of 108 studies was evaluated in detail. The median standardized quality score per year ranged between 53 and 61 (overall range: 33-88) and remained constant over the years.

Conclusions: The quality of conduct and reporting of propensity score methods in research on the effectiveness of antimicrobial therapy needs substantial improvement. The quality assessment instrument that was developed in this study may serve to help investigators improve the conduct and reporting of propensity score methods.

Place, publisher, year, edition, pages
Oxford University Press, 2022
Keywords
antimicrobial therapy, infectious diseases, propensity score methods
National Category
Infectious Medicine
Identifiers
urn:nbn:se:umu:diva-194356 (URN)10.1093/ofid/ofac110 (DOI)000775226500008 ()35355895 (PubMedID)2-s2.0-85128167951 (Scopus ID)
Available from: 2022-05-03 Created: 2022-05-03 Last updated: 2022-05-03Bibliographically approved
Toppe, C., Möllsten, A., Waernbaum, I., Schön, S., Gudbjörnsdottir, S., Landin-Olsson, M. & Dahlquist, G. (2019). Decreasing Cumulative Incidence of End-Stage Renal Disease in Young Patients With Type 1 Diabetes in Sweden: a 38-Year Prospective Nationwide Study. Diabetes Care, 42(1), 27-31
Open this publication in new window or tab >>Decreasing Cumulative Incidence of End-Stage Renal Disease in Young Patients With Type 1 Diabetes in Sweden: a 38-Year Prospective Nationwide Study
Show others...
2019 (English)In: Diabetes Care, ISSN 0149-5992, E-ISSN 1935-5548, Vol. 42, no 1, p. 27-31Article in journal (Refereed) Published
Abstract [en]

Objective: Diabetic nephropathy is a serious complication of type 1 diabetes. Recent studies indicate that end-stage renal disease (ESRD) incidence has decreased or that the onset of ESRD has been postponed; therefore, we wanted to analyze the incidence and time trends of ESRD in Sweden.

Research design and methods: In this study, patients with duration of type 1 diabetes >14 years and age at onset of diabetes 0–34 years were included. Three national diabetes registers were used: the Swedish Childhood Diabetes Register, the Diabetes Incidence Study in Sweden, and the National Diabetes Register. The Swedish Renal Registry, a national register on renal replacement therapy, was used to identify patients who developed ESRD.

Results: We found that the cumulative incidence of ESRD in Sweden was low after up to 38 years of diabetes duration (5.6%). The incidence of ESRD was lower in patients with type 1 diabetes onset in 1991–2001 compared to onset in 1977–1984 and 1985–1990, independently of diabetes duration.

Conclusion: The risk of developing ESRD in Sweden in this population is still low and also seems to decrease with time.

Place, publisher, year, edition, pages
American Diabetes Association, 2019
National Category
Pediatrics Urology and Nephrology
Identifiers
urn:nbn:se:umu:diva-153168 (URN)10.2337/dc18-1276 (DOI)000453904900014 ()30352897 (PubMedID)2-s2.0-85059071263 (Scopus ID)
Funder
Swedish Research Council, 0753Västerbotten County Council
Available from: 2018-11-08 Created: 2018-11-08 Last updated: 2023-03-24Bibliographically approved
Waernbaum, I., Dahlquist, G. & Lind, T. (2019). Perinatal risk factors for type 1 diabetes revisited: a population-based register study. Diabetologia, 62(7), 1173-1184
Open this publication in new window or tab >>Perinatal risk factors for type 1 diabetes revisited: a population-based register study
2019 (English)In: Diabetologia, ISSN 0012-186X, E-ISSN 1432-0428, Vol. 62, no 7, p. 1173-1184Article in journal (Refereed) Published
Abstract [en]

Aims/hypothesis: Single-centre studies and meta-analyses have found diverging results as to which early life factors affect the risk of type 1 diabetes during childhood. We wanted to use a large, nationwide, prospective database to further clarify and analyse the associations between perinatal factors and the subsequent risk for childhood-onset type 1 diabetes using a case–control design.

Methods: The Swedish Childhood Diabetes Register was linked to the Swedish Medical Birth Register and National Patient Register, and 14,949 cases with type 1 diabetes onset at ages 0–14 years were compared with 55,712 matched controls born from the start of the Medical Birth Register in 1973 to 2013. After excluding confounders (i.e. children multiple births, those whose mother had maternal diabetes and those with a non-Nordic mother), we used conditional logistic regression analyses to determine risk factors for childhood-onset type 1 diabetes. We used WHO ICD codes for child and maternal diagnoses.

Results: In multivariate analysis, there were small but statistically significant associations between higher birthweight z score (OR 1.08, 95% CI 1.06, 1.10), delivery by Caesarean section (OR 1.08, 95% CI 1.02, 1.15), premature rupture of membranes (OR 1.08, 95% CI 1.01, 1.16) and maternal urinary tract infection during pregnancy (OR 1.39, 95% CI 1.04, 1.86) and the subsequent risk of childhood-onset type 1 diabetes. Birth before 32 weeks of gestation was associated with a lower risk of childhood-onset type 1 diabetes compared with full-term infants (OR 0.54, 95% CI 0.38, 0.76), whereas birth between 32 and 36 weeks’ gestation was associated with a higher risk (OR 1.24, 95% CI 1.14, 1.35). In subgroup analyses (birth years 1992–2013), maternal obesity was independently associated with subsequent type 1 diabetes in the children (OR 1.27, 95% CI 1.15, 1.41) and rendered the association with Caesarean section non-significant. In contrast to previous studies, we found no association of childhood-onset type 1 diabetes with maternal–child blood-group incompatibility, maternal pre-eclampsia, perinatal infections or treatment of the newborn with phototherapy for neonatal jaundice. The proportion of children with neonatal jaundice was significantly higher in the 1973–1982 birth cohort compared with later cohorts.

Conclusions/interpretation: Perinatal factors make small but statistically significant contributions to the overall risk of childhood-onset type 1 diabetes. Some of these risk factors, such as maternal obesity, may be amendable with improved antenatal care. Better perinatal practices may have affected some previously noted risk factors over time.

Place, publisher, year, edition, pages
Springer, 2019
Keywords
Birthweight, Case-control study, Diabetes mellitus type 1, Perinatal risk factors, Urinary tract infection
National Category
Endocrinology and Diabetes Pediatrics
Identifiers
urn:nbn:se:umu:diva-161440 (URN)10.1007/s00125-019-4874-5 (DOI)000471176200008 ()31041471 (PubMedID)2-s2.0-85065230302 (Scopus ID)
Funder
Swedish Research Council, 2016-00703Swedish Research Council, 2014-07531
Available from: 2019-07-10 Created: 2019-07-10 Last updated: 2023-03-24Bibliographically approved
Pingel, R. & Waernbaum, I. (2017). Correlation and Efficiency of Propensity Score-based Estimators for Average Causal Effects. Communications in statistics. Simulation and computation, 46(5), 3458-3478
Open this publication in new window or tab >>Correlation and Efficiency of Propensity Score-based Estimators for Average Causal Effects
2017 (English)In: Communications in statistics. Simulation and computation, ISSN 0361-0918, E-ISSN 1532-4141, Vol. 46, no 5, p. 3458-3478Article in journal (Refereed) Published
Abstract [en]

Propensity score based-estimators are commonly used to estimate causal effects in evaluationresearch. To reduce bias in observational studies researchers might be tempted to include many, perhaps correlated, covariates when estimating the propensity score model. Taking into account that the propensity score is estimated, this study investigates how the efficiency of matching, inverse probability weighting and doubly robust estimators change under the case of correlated covariates. Propositions regarding the large sample variances under certain assumptions on the data generating process are given. The propositions are supplemented by several numerical large sample and finite sample results from a wide range of models. The results show that the covariate correlations may increase or decrease the variances of the estimators. There are several factors that influence how correlation affects the variance of the estimators, including the choice of estimator, the strength of the confounding towards outcome and treatment, and whether a constant or non-constant causal effect is present.

Keywords
Doubly robust, Inverse probability, central nervous system, Matching, Observational study
National Category
Probability Theory and Statistics
Research subject
Statistics
Identifiers
urn:nbn:se:umu:diva-113905 (URN)10.1080/03610918.2015.1094091 (DOI)000402090000010 ()2-s2.0-85008156821 (Scopus ID)
Funder
Riksbankens Jubileumsfond, P11-0814:1
Available from: 2016-01-05 Created: 2016-01-05 Last updated: 2023-03-23Bibliographically approved
Persson, E., Häggström, J., Waernbaum, I. & de Luna, X. (2017). Data-driven algorithms for dimension reduction in causal inference. Computational Statistics & Data Analysis, 105, 280-292
Open this publication in new window or tab >>Data-driven algorithms for dimension reduction in causal inference
2017 (English)In: Computational Statistics & Data Analysis, ISSN 0167-9473, E-ISSN 1872-7352, Vol. 105, p. 280-292Article in journal (Refereed) Published
Abstract [en]

In observational studies, the causal effect of a treatment may be confounded with variables that are related to both the treatment and the outcome of interest. In order to identify a causal effect, such studies often rely on the unconfoundedness assumption, i.e., that all confounding variables are observed. The choice of covariates to control for, which is primarily based on subject matter knowledge, may result in a large covariate vector in the attempt to ensure that unconfoundedness holds. However, including redundant covariates can affect bias and efficiency of nonparametric causal effect estimators, e.g., due to the curse of dimensionality. In this paper, data-driven algo- rithms for the selection of sufficient covariate subsets are investigated. Under the assumption of unconfoundedness we search for minimal subsets of the covariate vector. Based on the framework of sufficient dimension reduction or kernel smoothing, the algorithms perform a backward elim- ination procedure testing the significance of each covariate. Their performance is evaluated in simulations and an application using data from the Swedish Childhood Diabetes Register is also presented.

Keywords
covariate selection, marginal co-ordinate hypothesis test, matching, kernel smoothing, type 1 diabetes mellitus
National Category
Probability Theory and Statistics
Research subject
Statistics
Identifiers
urn:nbn:se:umu:diva-80696 (URN)10.1016/j.csda.2016.08.012 (DOI)000385604500019 ()2-s2.0-84987932350 (Scopus ID)
Funder
Swedish National Infrastructure for Computing (SNIC), SNIC 2016/1-2Swedish Research Council, 2013-672Swedish Research Council, 07531Riksbankens Jubileumsfond, P11-0814:1
Available from: 2013-09-24 Created: 2013-09-24 Last updated: 2023-03-23Bibliographically approved
Persson, E., Waernbaum, I. & Lind, T. (2017). Estimating marginal causal effects in a secondary analysis of case-control data. Statistics in Medicine, 36(15), 2404-2419
Open this publication in new window or tab >>Estimating marginal causal effects in a secondary analysis of case-control data
2017 (English)In: Statistics in Medicine, ISSN 0277-6715, E-ISSN 1097-0258, Vol. 36, no 15, p. 2404-2419Article in journal (Refereed) Published
Abstract [en]

When an initial case-control study is performed, data can be used in a secondary analysis to evaluate the effect of the case-defining event on later outcomes. In this paper, we study the example in which the role of the event is changed from a response variable to a treatment of interest. If the aim is to estimate marginal effects, such as average effects in the population, the sampling scheme needs to be adjusted for. We study estimators of the average effect of the treatment in a secondary analysis of matched and unmatched case-control data where the probability of being a case is known. For a general class of estimators, we show the components of the bias resulting from ignoring the sampling scheme and demonstrate a design-weighted matching estimator of the average causal effect. In simulations, the finite sample properties of the design-weighted matching estimator are studied. Using a Swedish diabetes incidence register with a matched case-control design, we study the effect of childhood onset diabetes on the use of antidepressant medication as an adult.

Place, publisher, year, edition, pages
Hoboken: Wiley-Blackwell, 2017
Keywords
design-weighted estimation, matched case-control study, propensity score
National Category
Probability Theory and Statistics Public Health, Global Health, Social Medicine and Epidemiology
Research subject
Statistics
Identifiers
urn:nbn:se:umu:diva-94965 (URN)10.1002/sim.7277 (DOI)000402799900007 ()28276084 (PubMedID)2-s2.0-85014944257 (Scopus ID)
Funder
Swedish Research Council, 07531Riksbankens Jubileumsfond, P11-0814:1
Available from: 2014-10-20 Created: 2014-10-20 Last updated: 2024-07-02Bibliographically approved
Pazzagli, L., Möllsten, A. & Waernbaum, I. (2017). Marginal structural model to evaluate the joint effect of socioeconomic exposures on the risk of developing end-stage renal disease in patients with type 1 diabetes: a longitudinal study based on data from the Swedish Childhood Diabetes Study Group. Annals of Epidemiology, 27(8), 479-484
Open this publication in new window or tab >>Marginal structural model to evaluate the joint effect of socioeconomic exposures on the risk of developing end-stage renal disease in patients with type 1 diabetes: a longitudinal study based on data from the Swedish Childhood Diabetes Study Group
2017 (English)In: Annals of Epidemiology, ISSN 1047-2797, E-ISSN 1873-2585, Vol. 27, no 8, p. 479-484Article in journal (Refereed) Published
Abstract [en]

Purpose: Diabetic nephropathy is a severe complication of type 1 diabetes (T1D) that may lead to renal failure and end-stage renal disease (ESRD) demanding dialysis and transplantation. The aetiology of diabetic nephropathy is multifactorial and both genes and environmental and life style related factors are involved. In this study we investigate the effect of the socioeconomic exposures unemployment and receiving income support on the development of ESRD in T1D patients, using a marginal structural model in comparison with standard logistic regression models.

Methods: The study is based on the Swedish Childhood Diabetes Register which in 1977 started to register patients developing T1D before 15 years of age. In the analyses we include patients born between 1965 and 1979, developing diabetes between 1977 and 1994, followed until 2013 (n=4034). A marginal structural model (MSM) was fitted to adjust for both baseline and time-varying confounders.

Results: The main results of the analysis indicate that being unemployed for more than one year and receiving income support are risk factors for the development of ESRD. Multiple exposure over time to these risk factors increases the risk associated with the disease.

Conclusions: Using a MSM is an advanced method well suited to investigate the effect of exposures on the risk of complications of a chronic disease with longitudinal data. The results show that socioeconomic disadvantage increases the risk of developing ESRD in patients with type 1 diabetes.

Place, publisher, year, edition, pages
New York: Elsevier, 2017
Keywords
Socioeconomic disparities, Type 1 diabetes, end stage renal disease, marginal structural model
National Category
Public Health, Global Health, Social Medicine and Epidemiology
Research subject
Epidemiology
Identifiers
urn:nbn:se:umu:diva-138138 (URN)10.1016/j.annepidem.2017.07.003 (DOI)000411770700004 ()28935026 (PubMedID)2-s2.0-85043353288 (Scopus ID)
Funder
The Royal Swedish Academy of SciencesSwedish Research Council, project number 0753
Available from: 2017-08-14 Created: 2017-08-14 Last updated: 2023-03-24Bibliographically approved
Waernbaum, I. & Dahlquist, G. (2016). Low mean temperature rather than few sunshine hours are associated with an increased incidence of type 1 diabetes in children. European Journal of Epidemiology, 31(1), 61-65
Open this publication in new window or tab >>Low mean temperature rather than few sunshine hours are associated with an increased incidence of type 1 diabetes in children
2016 (English)In: European Journal of Epidemiology, ISSN 0393-2990, E-ISSN 1573-7284, Vol. 31, no 1, p. 61-65Article in journal (Refereed) Published
Abstract [en]

The well-known north-south gradient and the seasonal variability in incidence of childhood type1 diabetes indicate climatological factors to have an effect on the onset. Both sunshine hours and a low temperature may be responsible. In the present study we tried to disentangle these effects that tend to be strongly connected.

Exposure data were sunshine hours and mean temperature respectively obtained from eleven meteorological stations in Sweden which were linked to incidence data from geographically matched areas. Incident cases during 1983-2008 were retrieved from the population based Swedish childhood diabetes register. We used generalized additive models to analyze the incidence as a function of mean temperature and hours of sun adjusted for the time trend, age and sex.

In our data set the correlation between sun hours and temperature was weak (r=0.36) implying that it was possible to estimate the effect of these variables in a regression model. We fit a general additive model with a smoothing term for the time trend. In the model with sun hours we found no significant effect on T1 incidence (p=0.17) whereas the model with temperature as predictor was significant (p=0.05) when adjusting for the time trend, sex and age. Adding sun hours in the model where mean temperature was already present did not change the effect of temperature.

There is an association with incidence of type1 diabetes in children and low mean temperature independent of a possible effect of sunshine hours after adjustment for age, sex and time trend. The findings may mirror the cold effect on insulin resistance and accords with the hypothesis that overload of an already ongoing beta cell destruction may accelerate disease onset.

Place, publisher, year, edition, pages
Springer, 2016
Keywords
Climate, Incidence, Risk factors, Time trend adjustment, Type 1 diabetes mellitus
National Category
Public Health, Global Health, Social Medicine and Epidemiology Endocrinology and Diabetes Pediatrics
Identifiers
urn:nbn:se:umu:diva-101214 (URN)10.1007/s10654-015-0023-8 (DOI)000370376600007 ()2-s2.0-84958780689 (Scopus ID)
Funder
Swedish Research Council, 07531Riksbankens Jubileumsfond, P11-0814:1
Available from: 2015-03-25 Created: 2015-03-25 Last updated: 2023-03-23Bibliographically approved
Häggström, J., Persson, E., Waernbaum, I. & de Luna, X. (2015). CovSel: An R Package for Covariate Selection When Estimating Average Causal Effects. Journal of Statistical Software, 68(1), 1-20
Open this publication in new window or tab >>CovSel: An R Package for Covariate Selection When Estimating Average Causal Effects
2015 (English)In: Journal of Statistical Software, E-ISSN 1548-7660, Vol. 68, no 1, p. 1-20Article in journal (Refereed) Published
Abstract [en]

We describe the R package CovSel, which reduces the dimension of the covariate vector for the purpose of estimating an average causal effect under the unconfoundedness assumption. Covariate selection algorithms developed in De Luna, Waernbaum, and Richardson (2011) are implemented using model-free backward elimination. We show how to use the package to select minimal sets of covariates. The package can be used with continuous and discrete covariates and the user can choose between marginal co-ordinate hypothesis tests and kernel-based smoothing as model-free dimension reduction techniques.

Keywords
causal inference, dimension reduction, dr, np, R
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:umu:diva-111952 (URN)10.18637/jss.v068.i01 (DOI)000366013800001 ()2-s2.0-84949921188 (Scopus ID)
Available from: 2015-11-26 Created: 2015-11-26 Last updated: 2023-10-03Bibliographically approved
Organisations

Search in DiVA

Show all publications