Simulation-based sample size calculations of marginal proportional means models for recurrent events with competing risks.

In randomised controlled trials, the outcome of interest could be recurrent events, such as hospitalisations for heart failure. If mortality rates are non-negligible, both recurrent events and competing terminal events need to be addressed when formulating the estimand and statistical analysis is no longer trivial. In order to design future trials with primary recurrent event endpoints with competing risks, it is necessary to be able to perform power calculations to determine sample sizes. This paper introduces a simulation-based approach for power estimation based on a proportional means model for recurrent events and a proportional hazards model for terminal events. The simulation procedure is presented along with a discussion of what the user needs to specify to use the approach. The method is flexible and based on marginal quantities which are easy to specify. However, the method introduces a lack of a certain type of dependence. This is explored in a sensitivity analysis which suggests that the power is robust in spite of that. Data from a randomised controlled trial, LEADER, is used as the basis for generating data for a future trial. Finally, potential power gains of recurrent event methods as opposed to first event methods are discussed.

Oxidative Stress-Induced Damage to RNA and DNA and Mortality in Individuals with Psychiatric Illness.

Importance: All-cause mortality and the risk for age-related medical disease is increased in individuals with psychiatric illness, but the underlying biological mechanisms are not known. Oxidative stress on nucleic acids (DNA and RNA; NA-OXS) is a molecular driver of aging and a potential pathophysiological mechanism in a range of age-related disorders. Objective: To study the levels of markers of NA-OXS in a large cohort of community-dwelling individuals with and without psychiatric illness and to evaluate their association with prospective all-cause mortality. Design, Setting, and Participants: This cohort study used a combined cohort of participants from 2 population-based health studies: the Danish General Suburban Population Study (January 2010 to October 2013) and nondiabetic control participants from the Vejle Diabetes Biobank study (March 2007 to May 2010). Individual history of psychiatric illness was characterized using register data on psychiatric diagnoses and use of psychotropic drugs before baseline examination. Urinary markers of systemic RNA (8-oxo-7,8-dihydroguanosine [8-oxoGuo]) and DNA (8-oxo-7,8-dihydro-2'-deoxyguanosine [8-oxodG]) damage from oxidation were measured by ultraperformance liquid chromatography-tandem mass spectrometry. Cox proportional hazard regression models were applied for survival analyses, using register-based all-cause mortality updated to May 2023. The follow-up time was up to 16.0 years. Exposures: History of psychiatric illness. Main Outcomes and Measures: Mortality risk according to psychiatric illness status and 8-oxoGuo or 8-oxodG excretion level. Results: A total of 7728 individuals were included (3983 [51.5%] female; mean [SD] age, 58.6 [11.9] years), 3095 of whom (40.0%) had a history of psychiatric illness. Mean (SD) baseline 8-oxoGuo was statistically significantly higher in individuals with psychiatric illness than in those without (2.4 [1.2] nmol/mmol vs 2.2 [0.9] nmol/mmol; P < .001), whereas 8-oxodG was not. All-cause mortality was higher in the psychiatric illness group vs the no psychiatric illness group (hazard ratio [HR], 1.44; 95% CI, 1.27-1.64; P < .001) and increased sequentially with each increasing tertile of 8-oxoGuo excretion in both groups to an almost doubled risk in the psychiatric illness/high 8-oxoGuo group compared to the no psychiatric illness/low 8-oxoGuo reference group (HR, 1.99; 95% CI, 1.58-2.52; P < .001). These results persisted after adjustment for a range of potential confounders and in a sensitivity analysis stratified for sex. Conclusions and Relevance: This study establishes systemic oxidative stress-induced damage to RNA as a potential mechanism in the accelerated aging observed in psychiatric disorders and urinary 8-oxoGuo as a potentially useful marker of mortality risk in individuals with psychiatric illness.

Body height and spinal pain in adolescence: a cohort study from the Danish National Birth Cohort.

OBJECTIVES: To investigate how body height and trajectories of height from infancy through childhood and adolescence were associated with spinal pain in pre- and late adolescence. METHODS: This prospective study included 43,765 individuals born into The Danish National Birth Cohort (DNBC) from 1996 to 2003. DNBC-data were linked with health and social data identified from Statistics Denmark registers. Spinal pain was self-reported in both the 11-year- and 18-year follow-up of DNBC and classified according to severity. Body height was measured from birth and onwards and further modelled as distinct developmental height trajectories by using latent growth curve modelling. Associations were estimated by using multinomial logistic regression models. RESULTS: Taller body height in childhood and adolescence was associated with approximately 20% increased likelihood of spinal pain in pre- and late adolescence among girls compared to their peers in the normal height group. For boys, taller body height was associated with spinal pain by late adolescence only. Spinal pain in pre-adolescence almost doubled the likelihood of spinal pain in late adolescence regardless of body height at age 18. Height trajectories confirmed the relationship for girls with the tall individuals being most likely to have spinal pain in both pre- and late adolescence. CONCLUSION: Tall body height during childhood and adolescence predisposes to spinal pain among girls in both pre-and late adolescence, and among boys in late adolescence. Body height is a contributing factor to the pathogenesis of spinal pain in adolescence; however, the mechanisms may be related to growth velocity, but for now uncertain.

The effect of the SAFE intervention on post-discharge suicidal behavior: a quasi-experimental study using propensity score matching.

OBJECTIVES: The risk of suicidal behavior after discharge from psychiatric admission is high. The aim of this study was to examine whether the SAFE intervention, an implementation of a systematic safer discharge procedure, was associated with a reduction in suicidal behavior after discharge. METHODS: The SAFE intervention was implemented at Mental Health Center Copenhagen in March 2018 and consisted of three systematic discharge procedures: (1) A face-to-face meeting between patient and outpatient staff prior to discharge, (2) A face-to-face meeting within the first week after discharge, and (3) Involvement of relatives. Risk of suicide attempt at six-month post-discharge among patients discharged from the SAFE intervention was compared with patients discharged from comparison mental health centers using propensity score matching. RESULTS: 7604 discharges took place at the intervention site, which were 1:1 matched with discharges from comparison sites. During the six months of follow-up, a total of 570 suicide attempts and 25 suicides occurred. The rate of suicide attempt was 11,652 per 100,000 person-years at the SAFE site, while it was 10,530 at comparisons sites. No observable difference in suicide attempt 1.10 (95% CI: 0.89-1.35) or death by suicide (OR = 1.27; 95% CI:0.58-2.81) was found between sites at 6-month follow-up. CONCLUSION: No difference in suicidal behavior between the sites was found in this pragmatic study. High rates of suicidal behavior were found during the 6-months discharge period, which could suggest that a preventive intervention should include support over a longer post-discharge period than the one-week follow-up offered in the SAFE intervention.

Exposure to residential mold and dampness and the associations with respiratory tract infections and symptoms thereof in children in high income countries: A systematic review and meta-analyses of epidemiological studies.

BACKGROUND: Multiple reviews have been conducted on the associations between residential mold and dampness and respiratory outcomes in children, with few specifically investigating respiratory tract infections (RTIs). OBJECTIVE: We aimed to review and synthesize the available epidemiological literature on mold and dampness and risk of RTIs and respiratory symptoms compatible with RTIs in children living in high-income countries. METHOD: We performed a systematic search of literature available from MEDLINE, Embase, and Web of Science for observational studies. We conducted meta-analyses using two-level random effects (RE) and multi-level random effects (ML) models for contrasts of three exposure and three outcome categories, including multiple estimates reported by single studies. We report central estimates for pooled odds ratios (OR) and 95 % confidence intervals (CI).We conducted a risk of bias assessment using the Joanna Briggs Initiative (JBI) checklists for cross-sectional, case-control, and cohort studies. We additionally report on cumulative meta-analyses, leave-one-out analyses of single estimates, subgroup analyses by study quality and study design and inclusion of all effect estimates. RESULTS: Of the 932 studies initially screened by title and abstract, we included 30 studies with 267 effect estimates that met the inclusion criteria. Most were cross-sectional (n = 22), with fewer cohort (n = 5) and case-control (n = 3) studies. Most of the studies were according to the bias assessment of poor or fair quality (n = 24). The main meta-analyses generally provided similar results regardless of statistical model and central estimates ranged from OR 1.28 (95 % CI; 1.08, 1.53) for dampness and RTIs to OR 1.76 (95 % CI; 1.64, 1.88) for mold and respiratory symptoms. Most analyses were of moderate heterogeneity. Funnel plots did not indicate strong publication bias. CONCLUSION: Our results are compatible with a weak to moderate effect of residential mold and or dampness on risk of RTIs in children in high-income countries. However, these results are based primarily on cross-sectional studies.

Lifetime Incidence of Treated Mental Health Disorders and Psychotropic Drug Prescriptions and Associated Socioeconomic Functioning.

Importance: Few studies have estimated the lifetime incidence of mental health disorders and the association with socioeconomic functioning. Objective: To investigate whether the lifetime incidence of treated mental health disorders is substantially higher than previously reported and estimate associations with long-term socioeconomic difficulties. Design, Setting, and Participants: This nationwide population-based register linkage study includes a randomly selected sample of 1.5 million individuals from the population of Denmark from 1995 to 2018. Data were analyzed from May 2022 to March 2023. Main Outcomes and Measures: Lifetime incidence of any treated mental health disorder in the general population was estimated from birth to age 100 years taking into account the competing risk of all-cause death and associations with socioeconomic functioning. Register measures were (1) from hospitals, a diagnosis of any mental health disorder at an inpatient/outpatient hospital contact; (2) from hospitals and prescription statistics, any mental health disorder/psychotropic prescription, including a hospital-contact diagnosis, or any psychotropic medication prescribed by physicians, including general practitioners or private psychiatrists; and (3) socioeconomic functioning as indicated by highest educational achievement, employment, income, residential status, and marital status. Results: Among a sample of 462 864 individuals with any mental health disorder, the median (IQR) age was 36.6 years (21.0-53.6 years), 233 747 (50.5%) were male, and 229 117 (49.5%) were female. Of these, 112 641 were registered with a hospital-contact mental health disorder diagnosis and 422 080 with a prescription of psychotropic medication. The cumulative incidence of a hospital-contact mental health disorder diagnosis was 29.0% (95% CI, 28.8-29.1), 31.8% (95% CI, 31.6-32.0) for females, and 26.1% (95% CI, 25.9-26.3) for males. When also considering psychotropic prescriptions, the cumulative incidence of any mental health disorder/psychotropic prescription was 82.6% (95% CI, 82.4-82.6), 87.5% (95% CI, 87.4-87.7) for females, and 76.7% (95% CI, 76.5-76.8) for males. Socioeconomic difficulties were associated with mental health disorder/psychotropic prescriptions, including lower income (hazard ratio [HR], 1.55; 95% CI, 1.53-1.56), increased unemployment or disability benefit (HR, 2.50; 95% CI, 2.47-2.53), and a greater likelihood of living alone (HR, 1.78; 95% CI, 1.76-1.80) and being unmarried (HR, 2.02; 95% CI, 2.01-2.04) during long-term follow-up. These rates were confirmed in 4 sensitivity analyses with the lowest being 74.8% (95% CI, 74.7-75.0) (1) by using varying exclusion periods, (2) by excluding prescriptions of anxiolytics and quetiapine that may be used for off-label indications, (3) by defining any mental health disorder/psychotropic prescription as any hospital-contact mental health disorder diagnosis or any psychotropic medication prescribed at least 2 times, and (4) by excluding individuals with somatic diagnoses for which psychotropics may be prescribed off-label. Conclusions and Relevance: This registry study of data from a large representative sample of the Danish population showed that the majority of individuals either received a diagnosis of a mental health disorder or were prescribed psychotropic medication during their lifetime, which was associated with subsequent socioeconomic difficulties. These findings may help change our understanding of normalcy and mental illness, reduce stigmatization, and further prompt rethinking the primary prevention of mental illness and future mental health clinical resources.

Specific prediction of mortality by oxidative stress-induced damage to RNA vs. DNA in humans.

Modifications of nucleic acids (DNA and RNA) from oxidative stress is a potential driver of aging per se and of mortality in age-associated medical disorders such as type 2 diabetes (T2D). In a human cohort, we found a strong prediction of all-cause mortality by a marker of systemic oxidation of RNA in patients with T2D (n = 2672) and in nondiabetic control subjects (n = 4079). The finding persisted after the adjustment of established modifiers of oxidative stress (including BMI, smoking, and glycated hemoglobin). In contrast, systemic levels of DNA damage from oxidation, which traditionally has been causally linked to both T2D and aging, failed to predict mortality. Strikingly, these findings were subsequently replicated in an independent general population study (n = 3649). The data demonstrate a specific importance of RNA damage from oxidation in T2D and general aging.

Expected life years compared to the general population.

For cohorts with long-term follow-up, the number of years lost due to a certain disease yields a measure with a simple and appealing interpretation. Recently, an overview of the methodology used for this goal has been published, and two measures have been proposed. In this work, we consider a third option that may be useful in settings in which the other two are inappropriate. In all three measures, the survival of the given dataset is compared to the expected survival in the general population which is calculated using external mortality tables. We thoroughly analyze the differences between the three measures, their assumptions, interpretation, and the corresponding estimators. The first measure is defined in a competing risk setting and assumes an excess hazard compared to the population, while the other two measures also allow estimation for groups that live better than the general population. In this case, the observed survival of the patients is compared to that in the population. The starting point of this comparison depends on whether the entry into the study is a hazard changing event (e.g., disease diagnosis or the age at which the inclusion criteria were met). Focusing on the newly defined life years difference measure, we study the estimation of the variance and consider the possible challenges (e.g., extrapolation) that occur in practice. We illustrate its use with a dataset of French Olympic athletes. Finally, an efficient R implementation has been developed for all three measures which make this work easily available to subsequent users.

Socio-economic functioning in patients with bipolar disorder and their unaffected siblings - results from a nation-wide population-based longitudinal study.

BACKGROUND: Few studies have reported real-life data on socio-economic functioning in patients with bipolar disorder and their unaffected first-degree relatives. METHODS: We used Danish nation-wide population-based longitudinal register linkage to investigate socio-economic functioning in 19 955 patients with bipolar disorder, their 13 923 siblings and 20 sex, age and calendar-matched control individuals from the general population. Follow-up was from 1995 to 2017. RESULTS: Patients with a diagnosis of bipolar disorder had lower odds of having achieved the highest educational level [OR 0.75 (95% confidence interval (CI) 0.73-0.77)], being employed [OR 0.16 (95% CI 0.159-0.168)], having achieved the 80% highest quartile of income [OR 0.33 (95% CI 0.32-0.35)], cohabitating [OR 0.44 (95% CI 0.43-0.46)] and being married [OR 0.54 (95% CI 0.52-0.55)] at first contact to hospital psychiatry as inpatient or outpatient compared with control individuals from the general population. Similarly, siblings to patients with bipolar disorder had a lower functioning within all five socio-economic areas than control individuals. Furthermore, patients and partly siblings showed substantially decreased ability to enhance their socio-economic functioning during the 23 years follow-up compared to controls. CONCLUSIONS: Socio-economic functioning is substantially decreased in patients with bipolar disorder and their siblings and does not improve during long-term follow-up after the initial hospital contact, highlighting a severe and overlooked treatment gap.

Discussion of "A formal causal interpretation of the case-crossover design".

Discussion on "A formal causal interpretation of the case-crossover design" by Zach Shahn, Miguel A. Hernan, and James M. Robins.

Maternal stress during pregnancy and gestational duration: A cohort study from the Danish National Birth Cohort.

BACKGROUND: Preterm birth is one of the most important contributors to neonatal mortality and morbidity. Experiencing stress during pregnancy may increase the risk of adverse birth outcomes, including preterm birth. This association has been observed in previous studies, but differences in measures used limit comparability. OBJECTIVE: The objective of the study was to investigate the association between two measures of maternal stress during pregnancy, life stress and emotional distress, and gestation duration. METHODS: Women recruited in the Danish National Birth Cohort from 1996 to 2002, who provided information on their stress level during pregnancy and expecting a singleton baby, were included in the study. We assessed the associations between the level of life stress and emotional distress in quartiles, both collected at 31 weeks of pregnancy on average, and the rate of giving birth using Cox regression within intervals of the gestational period. RESULTS: A total of 80,991 pregnancies were included. Women reporting moderate or high levels of life stress vs no stress had a higher rate of giving birth earlier within all intervals of gestational age (e.g. high level: 27-33 weeks: hazard ratio (HR) 1.38, 95% confidence interval (CI) 1.04, 1.84; 34-36 weeks: 1.10, 95% CI 0.97, 1.25; 37-38 weeks: 1.21, 95% CI 1.15, 1.28). These associations between life stress and preterm birth were mainly driven by pregnancy worries. For emotional distress, a high level of distress was associated with shorter length of gestation in the preterm (27-33 weeks: 1.38, 95% CI 1.02, 1.86; 34-36 weeks: 1.05, 95% CI 0.91, 1.19) and early term (1.11, 95% CI 1.04, 1.17) intervals. CONCLUSIONS: Emotional distress and life stress were shown to be associated with gestational age at birth, with pregnancy-related stress being the single stressor driving the association. This suggests that reverse causality may, at least in parts, explain the earlier findings of stress as a risk factor for preterm birth.

Estimating the effect of donor sex on red blood cell transfused patient mortality: A retrospective cohort study using a targeted learning and emulated trials-based approach.

Background: Observational studies determining the effect of red blood cell (RBC) donor sex on recipient mortality have been inconsistent. Emulating hypothetical randomized target trials using large real-world data and targeted learning may clarify potential adverse effects. Methods: In this retrospective cohort study, a RBC transfusion database from the Capital Region of Denmark comprising more than 900,000 transfusion events defined the observational data. Eligible patients were minimum 18 years, had received a leukocyte-reduced RBC transfusion, and had no history of RBC transfusions within the past year at baseline. The doubly robust targeted maximum likelihood estimation method coupled with ensembled machine learning was used to emulate sex-stratified target trials determining the comparative effectiveness of exclusively transfusing RBC units from either male or female donors. The outcome was all-cause mortality within 28 days of the baseline-transfusion. Estimates were adjusted for the total number of transfusions received on each day k, hospital of transfusion, calendar period, patient age and sex, ABO/RhD blood group of the patient, Charlson comorbidity score, the total number of transfusions received prior to day k, and the number of RBC units received on each day k from donors younger than 40 years of age. Findings: Among 98,167 adult patients who were transfused between Jan. 1, 2008, and Apr. 10, 2018, a total of 90,917 patients (54.6% female) were eligible. For male patients, the 28-day survival was 2.06 percentage points (pp) (95 % confidence interval [CI]: 1.81-2.32, P<0.0001) higher under treatment with RBC units exclusively from male donors compared with exclusively from female donors. In female patients, exclusively transfusing RBC units from either male or female donors increased the 28-day survival with 0.64pp (0.52-0.76, P<0.0001), and 0.62pp (0.49-0.75, P<0.0001) compared with the current practice, respectively. No evidence of a sex-specific donor effect was found for female patients (0.02pp [-0.18-0.22]). The sensitivity analyses showed that a large unknown causal bias would have to be present to affect the conclusions. Interpretation: The results suggest that a sex-matched transfusion policy may benefit patients. However, a causal interpretation of the findings relies on the assumption of no unmeasured confounding, treatment consistency, positivity, and minimal model misspecifications. Funding: Novo Nordisk Foundation and the Innovation Fund Denmark.

Association of Oxidative Stress-Induced Nucleic Acid Damage With Psychiatric Disorders in Adults: A Systematic Review and Meta-analysis.

Importance: Nucleic acid damage from oxidative stress (NA-OXS) may be a molecular mechanism driving the severely increased morbidity and mortality from somatic causes in adults with psychiatric disorders. Objective: To systematically retrieve and analyze data on NA-OXS across the psychiatric disorder diagnostic spectrum. Data Sources: The PubMed, Embase, and PsycINFO databases were searched from inception to November 16, 2021. A hand search of reference lists of relevant articles was also performed. Study Selection: Key study inclusion criteria in this meta-analysis were as follows: adult human study population, measurement of any marker of DNA or RNA damage from oxidative stress, and either a (1) cross-sectional design comparing patients with psychiatric disorders (any diagnosis) with a control group or (2) prospective intervention. Two authors screened the studies, and 2 senior authors read the relevant articles in full and assessed them for eligibility. Data Extraction and Synthesis: The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines were followed. Two authors performed data extraction independently, and a senior coauthor was consulted in cases of disagreement. Data were synthesized with random-effects and multilevel meta-analyses. Main Outcomes and Measures: The predefined hypothesis was that individuals with psychiatric disorders have increased NA-OXS levels. The main outcome was the standardized mean differences (SMDs) among patients and controls in nucleic acid oxidation markers compared across diagnostic groups. Analyses were divided into combinations of biological matrices and nucleic acids. Results: Eighty-two studies fulfilled the inclusion criteria, comprising 205 patient vs control group comparisons and a total of 10 151 patient and 10 532 control observations. Overall, the data showed that patients with psychiatric disorders had higher NA-OXS levels vs controls across matrices and molecules. Pooled effect sizes ranged from moderate for urinary DNA markers (SMD = 0.44 [95% CI, 0.20-0.68]; P < .001) to very large for blood cell DNA markers (SMD = 1.12 [95% CI, 0.69-1.55; P < .001). Higher NA-OXS levels were observed among patients with dementias followed by psychotic and bipolar disorders. Sensitivity analyses excluding low-quality studies did not materially alter the results. Intervention studies were few and too heterogenous for meaningful meta-analysis. Conclusions and Relevance: The results of this meta-analysis suggest that there is an association with increased NA-OXS levels in individuals across the psychiatric disorder diagnostic spectrum. NA-OXS may play a role in the somatic morbidity and mortality observed among individuals with psychiatric disorders.

Inference for transition probabilities in non-Markov multi-state models.

Multi-state models are frequently used when data come from subjects observed over time and where focus is on the occurrence of events that the subjects may experience. A convenient modeling assumption is that the multi-state stochastic process is Markovian, in which case a number of methods are available when doing inference for both transition intensities and transition probabilities. The Markov assumption, however, is quite strict and may not fit actual data in a satisfactory way. Therefore, inference methods for non-Markov models are needed. In this paper, we review methods for estimating transition probabilities in such models and suggest ways of doing regression analysis based on pseudo observations. In particular, we will compare methods using land-marking with methods using plug-in. The methods are illustrated using simulations and practical examples from medical research.

Effectiveness of medical treatment for bipolar disorder regarding suicide, self-harm and psychiatric hospital admission: between- and within-individual study on Danish national data.

BACKGROUND: Mood stabilisers are the main treatment for bipolar disorder. However, it is uncertain which drugs have the best outcomes. AIMS: To investigate whether rates of suicide, self-harm and psychiatric hospital admission in individuals with bipolar disorder differ between mood stabilisers. METHOD: A cohort design was applied to people aged ≥15 years who were diagnosed with bipolar disorder and living in Denmark during 1995-2016. Treatment with lithium, valproate, other mood stabilisers and antipsychotics were compared in between- and within-individual analyses, and adjusted for sociodemographic characteristics and previous self-harm. RESULTS: A total of 33 337 individuals with bipolar disorder were included (266 900 person-years). When compared with individuals not receiving treatment, those receiving lithium had a lower rate of suicide (hazard ratio 0.40, 95% CI 0.31-0.51). When comparing treatment and non-treatment periods in the same individuals, lower rates of self-harm were found for lithium (hazard ratio 0.74, 95% CI 0.61-0.91). Lower rates of psychiatric hospital admission were found for all drug categories compared with non-treatment periods in within-individual analyses (P<0.001). The low rates of self-harm and hospital admission for lithium in within-individual analyses were supported by results of between-individual analyses. CONCLUSIONS: Lithium was associated with lower rates of suicide, self-harm and psychiatric hospital readmission in all analyses. With respect to suicide, lithium was superior to no treatment. Although confounding by indication cannot be excluded, lithium seems to have better outcomes in the treatment of bipolar disorder than other mood stabilisers.

The impact of the initial and second national COVID-19 lockdowns on mental health in young people with and without pre-existing depressive symptoms.

BACKGROUND: The evidence on mental health during COVID-19 evolved fast, but still little is known about the long-lasting impact of the sequential lockdowns. We examine changes in young people's mental health from before to during the initial and second more prolonged lockdown, and whether women and those with pre-existing depressive symptoms were disproportionally impacted. METHODS: Participants reported on mental health indicators in an ongoing 18-year data collection in the Danish National Birth Cohort and in a COVID-19 survey, including 8 data points: 7 in the initial lockdown, and 1 year post. Changes in quality of life (QoL), mental well-being, and loneliness were estimated with random effect linear regressions on longitudinal data (N = 32,985), and linear regressions on repeated cross-sections (N = 28,579). FINDINGS: Interim deterioration in mental well-being and loneliness was observed during the initial lockdown, and only in those without pre-existing depressive symptoms. During the second lockdown, a modest deterioration was again observed for mental well-being and loneliness. QoL likewise only declined among those without pre-existing symptoms, where women showed a greater decline than men. QoL did not normalise during the initial lockdown and remained at lower levels during the second lockdown. These findings were not replicated in the repeated cross-sections. INTERPRETATION: Except for an interim decrease in mental health, and only in those without pre-existing depressive symptoms, this study's findings do not suggest a substantial detrimental impact of the lockdowns.

Analyzing differences between restricted mean survival time curves using pseudo-values.

Hazard ratios are ubiquitously used in time to event analysis to quantify treatment effects. Although hazard ratios are invaluable for hypothesis testing, other measures of association, both relative and absolute, may be used to fully elucidate study results. Restricted mean survival time (RMST) differences between groups have been advocated as useful measures of association. Recent work focused on model-free estimates of the difference in restricted mean survival through follow-up times, instead of focusing on a single time horizon. The resulting curve can be used to quantify the association in time units with a simultaneous confidence band. In this work a model-based estimate of the curve is proposed using pseudo-values allowing for possible covariate adjustment. The method is easily implementable with available software and makes possible to compute a simultaneous confidence region for the curve. The pseudo-values regression using multiple restriction times is in good agreement with the estimates obtained by standard direct regression models fixing a single restriction time. Moreover, the proposed method is flexible enough to reproduce the results of the non-parametric approach when no covariates are considered. Examples where it is important to adjust for baseline covariates will be used to illustrate the different methods together with some simulations.

A causal inference approach to compare leukaemia treatment outcome in the absence of randomization and with dependent censoring.

BACKGROUND: One cause of poor outcomes in children of low-income countries affected by acute lymphoblastic leukaemia (ALL) is loss to follow-up due to abandonment of treatment. Assuming this type of loss to follow-up as independent censoring, as in standard Kaplan-Meier estimates, ignores the likely association of abandonment with biologic and socio-economic factors related to outcome. Moreover, when comparing treatment protocols adopted in different time periods, possible imbalances in patients' characteristics must be considered. We aim to compare the outcome of children enrolled in two subsequent protocols for ALL treatment (2000-2007 and 2008-2015) in Honduras, taking both dependent censoring due to abandonment of treatment and imbalances between patient characteristics into account. METHODS: Marginal structural models based on inverse probability of treatment and censoring (IPTC) weighting allow the estimation of potential event-free survival (EFS) as if no abandonment of treatment occurred and the whole cohort was exposed, or not, to both protocols. An Aalen additive model and a logistic-regression model were used to build abandonment and treatment weights, respectively. RESULTS: The two protocols recruited 514 and 717 patients. Measured baseline covariates in both protocols were gender, age, white blood cell count, central nervous system involvement, tumour histology and socio-economic status. The potential EFS is slightly higher under the more recent protocol in the first 3 years but no difference is estimated in the long period [survival difference at 5 years (95% confidence interval) = 0.1% (-0.97%; 1.13%)]. Both protocols would allow reducing the event rate by 12-13% if there was no abandonment of treatment. CONCLUSIONS: Using IPTC weighting, we found a similar potential effect of the two treatment protocols if the imbalance due to the different distribution of potential confounders and to abandonment of therapy was removed.

Methodological challenges in the analysis of recurrent events for randomised controlled trials with application to cardiovascular events in LEADER.

Analysis of recurrent events is becoming increasingly popular for understanding treatment effects in randomised controlled trials. The analysis of recurrent events can improve efficiency and capture disease burden compared to standard time-to-first event analyses. However, the added knowledge about the multi-state process comes at the cost of modelling complexity. High mortality rates can complicate matters even more. A case study using data from a randomised controlled trial, LEADER, is presented to highlight interpretation of common methods as well as potential pitfalls when analysing recurrent events in the presence of a competing risk. The presented methods either target features of the underlying intensity functions or marginal traits of a multi-state process which includes terminal events or not. In particular, approaches to handle death as a part of an event and as a competing risk are discussed. A new method targeting the marginal mean function for a composite endpoint, which includes both death as a component and as a competing risk, will be introduced. Finally, recommendations for how to capture meaningful treatment effects in randomised controlled trials when analysing recurrent and terminal events will be made.

Pregnancy in women with MS: Impact on long-term disability accrual in a nationwide Danish Cohort.

BACKGROUND: Pregnancy is considered to influence the disease course in women with multiple sclerosis (MS). OBJECTIVE: The aim of this study was to investigate the effect of pregnancy on long-term disability accrual in women with MS. METHODS: The Danish Multiple Sclerosis Registry (DMSR) was used to identify women diagnosed with clinically isolated syndrome or relapsing-remitting MS. Cox models with pregnancy as a time-dependent exposure and propensity score (PS) models were used to evaluate time to reach confirmed Expanded Disability Status Scale (EDSS) score of 4 and 6. RESULTS: A total of 425 women became parous and 840 remained nulliparous. When including pregnancy as a time-dependent exposure, a non-significant association with time to reach EDSS 4 (hazard ratio (HR) 0.86, 95% confidence interval (CI) 0.61-1.20) and EDSS 6 (HR 0.70, 95% CI 0.40-1.20) was found. Correspondingly, the PS model showed no association with pregnancy on time to reach EDSS 4 (HR 0.85, 95% CI 0.56-1.28). CONCLUSION: This study concludes that pregnancy does not affect long-term disability accumulation.