Inverse probability weighting for self-selection bias correction in the investigation of social inequality in mortality.

BACKGROUND: Empirical evaluation of inverse probability weighting (IPW) for self-selection bias correction is inaccessible without the full source population. We aimed to: (i) investigate how self-selection biases frequency and association measures and (ii) assess self-selection bias correction using IPW in a cohort with register linkage. METHODS: The source population included 17 936 individuals invited to the Copenhagen Aging and Midlife Biobank during 2009-11 (ages 49-63 years). Participants counted 7185 (40.1%). Register data were obtained for every invited person from 7 years before invitation to the end of 2020. The association between education and mortality was estimated using Cox regression models among participants, IPW participants and the source population. RESULTS: Participants had higher socioeconomic position and fewer hospital contacts before baseline than the source population. Frequency measures of participants approached those of the source population after IPW. Compared with primary/lower secondary education, upper secondary, short tertiary, bachelor and master/doctoral were associated with reduced risk of death among participants (adjusted hazard ratio [95% CI]: 0.60 [0.46; 0.77], 0.68 [0.42; 1.11], 0.37 [0.25; 0.54], 0.28 [0.18; 0.46], respectively). IPW changed the estimates marginally (0.59 [0.45; 0.77], 0.57 [0.34; 0.93], 0.34 [0.23; 0.50], 0.24 [0.15; 0.39]) but not only towards those of the source population (0.57 [0.51; 0.64], 0.43 [0.32; 0.60], 0.38 [0.32; 0.47], 0.22 [0.16; 0.29]). CONCLUSIONS: Frequency measures of study participants may not reflect the source population in the presence of self-selection, but the impact on association measures can be limited. IPW may be useful for (self-)selection bias correction, but the returned results can still reflect residual or other biases and random errors.

Seasonality and sun exposure in incidence of major depression, bipolar disorder, and first-time use of antidepressant medication.

INTRODUCTION: Seasonality in depressive and bipolar disorders, are recognized in the ICD-10/11 and DSM-5 diagnostic systems. The existence of a seasonal pattern of hospital diagnosis of major depression, bipolar disorder and prescription of antidepressant medications has not been evaluated in the Danish population. METHODS: We retrieved date and year for all first-time hospital contacts with depression or bipolar disorder between 1999 and 2019, registered in the Danish National Patient Registry. Depression was defined using the ICD-10 F32-F33 codes, and for bipolar disorder the F30 or F31 codes. Date and year of all first-time purchases of antidepressant medications with ATC codes (N06A) between 1999 and 2021 were retrieved from the Danish National Prescription Registry, containing information on all prescribed drugs dispensed at pharmacies since 1995. Data on sunlight hours from 2012 to 2021 were retrieved from the Danish Metrological Institute. RESULTS: Incidences of hospital diagnoses as well as purchases of medication varied with month and season. The monthly variations were larger for antidepressant medication and smallest for bipolar disorder. The multiple linear regression analysis showed that number of first-time diagnoses of depression or bipolar disorder did not correlate with season. For antidepressant medication the number of first-time prescriptions was significantly lower in summer compared to the winter season. CONCLUSION: This study found a seasonal variation of first-time prescriptions of antidepressant medication. We did not find a seasonal variation in first-time hospital diagnoses. Further research looking into depression severity, polarity of bipolar illness episodes, lag-time for sunlight exposure, and specific parts of the yearly photoperiods should be conducted.

Cognitive ability, education, height and body mass index in relation to risk of schizophrenia and mortality following its diagnosis.

This study examines the hypotheses that the traits of higher IQ, longer education and taller height are associated with lower risk of death as compared to traits of low IQ, short education, and short height in men with schizophrenia compared to men without schizophrenia. In total, 937,919 men born 1939-59 and 1983-1997 with information from conscription were followed for incident schizophrenia in Danish registries. Higher levels of cognitive ability, longer education, and taller height were associated with fewer cases of schizophrenia. In a sub-sample of 652,368 men with information on body mass index, underweight was associated with more and overweight and obesity were associated with fewer cases of schizophrenia compared with normal weight. Higher cognitive ability, longer education, and taller height were associated with fewer deaths from both natural and unnatural causes in both men with and without schizophrenia. Underweight was associated with more deaths from natural and unnatural causes, whereas overweight and obesity were associated with more deaths from natural causes and fewer deaths from unnatural causes in both groups of men. Due to interaction, tall height and long educational duration were associated with fewer deaths from natural causes, and obesity was associated with fewer deaths from unnatural causes among men with schizophrenia compared to men without. In conclusion, traits in young adulthood are associated with higher mortality in men with and without schizophrenia, but traits of long educational duration and obesity seem to be especially important for lower mortality in men with schizophrenia.

Risk of Bowel Obstruction after Hysterectomy for Benign Indication According to Surgical Method in Denmark, 1984-2013.

STUDY OBJECTIVE: To estimate the risk of bowel obstruction (BO) after hysterectomy for benign indications depending on the surgical method (abdominal, vaginal, or laparoscopic) and identify risk factors for adhesive BO. DESIGN: A national registry-based cohort. SETTING: Danish hospitals during the period 1984-2013. PATIENTS: Danish women who underwent hysterectomy for benign indications (N = 125 568). INTERVENTIONS: Abdominal hysterectomies were compared with vaginal hysterectomies, laparoscopic hysterectomies, and minimally invasive (vaginal and laparoscopic) hysterectomies. MEASUREMENTS AND MAIN RESULTS: The incidence of BO according to the surgical method was compared using Cox proportional hazard regression. The covariates included were the time period, age, concomitant operations, previous abdominal surgery or disease, and socioeconomic factors. In a subanalysis (n = 35 712 women) of the period 2004-2013, detailed information from the Danish Hysterectomy Database enabled the inclusion of patient-, surgery-, and complication-related covariates. The overall crude incidence of BO was 17.4 of 1000 hysterectomies (2196 incident cases). The 10-year cumulative incidence of BO differed among the surgical routes (abdominal, 1.7%; laparoscopic, 1.4%; and vaginal, 0.9%). In multiple-adjusted analyses, the risk of BO was higher after abdominal hysterectomy than after vaginal (hazard ratio 1.64 [95% confidence interval, 1.39-1.93]) and minimally invasive (vaginal or laparoscopic) hysterectomy (hazard ratio 1.54 [1.33-1.79]). Additional pre-existing risk factors for BO at the time of hysterectomy were increased age, low education, low income, smoking, high American Society of Anesthesiologists comorbidity score, history of infertility, abdominal infection, and previous abdominal surgery (apart from cesarean section), penetrating lesions in abdominal organs, or operative adhesiolysis. Perioperative risk factors at the time of hysterectomy included concomitant removal of the ovaries, adhesiolysis, blood transfusion, readmission, and overall presence of perioperative complications. CONCLUSION: Abdominal hysterectomy is associated with a 54% higher risk of BO than minimally invasive (laparoscopic or vaginal) hysterectomy.

Postmenopausal hormone therapy and cognition in twins.

OBJECTIVE: Mild cognitive impairment may be caused by pathophysiological changes occurring decades prior to symptom development. It has been hypothesised that oestrogen can prevent such changes. We aimed to investigate the association between postmenopausal hormone therapy and cognition in Danish female twins and to examine differences in this association before and after publication of the findings from the Women's Health Initiative study in 2002. STUDY DESIGN: This study includes cognitive assessment of 4510 twins aged 50+ years. Information on hormone therapy was obtained through Danish health registries. The association between current hormone therapy use and cognition was analysed in twins aged 50+ using both cross-sectional, intrapair and longitudinal analysis, adjusting for age, education, social class, and unobserved familial confounding. RESULTS: Cross-sectionally, systemic HT users aged 70+ had a significantly lower cognitive function than non-users, whereas systemic HT users aged 50-69 did not differ from non-users before 2002. Longitudinal data in younger twins aged 50-69 showed a significantly lower cognitive function in systemic HT users after 2002 compared to non-users. Systemic HT users aged 70+ showed that the lower cognitive function was most explicit before 2002, whereas after 2002 the cognitive function was closer to non-users. Twins aged 50-69 who changed from systemic HT to local HT after 2002, or dropped it altogether, performed cognitively better. CONCLUSIONS: Our findings cautiously indicate a change in the association between cognition and hormone therapy use after 2002, which suggests an alteration in the hormone therapy user profile in the wake of the 2002 WHI publication.

Adverse Pregnancy Outcomes and Subsequent First-Time Use of Psychiatric Treatment Among Fathers in Denmark.

Importance: Becoming a first-time parent is a major life-changing event and can be challenging regardless of the pregnancy outcome. However, little is known how different adverse pregnancy outcomes affect the father's risk of psychiatric treatment post partum. Objective: To examine the associations of adverse pregnancy outcomes with first-time psychiatric treatment in first-time fathers. Design, Setting, and Participants: This nationwide cohort study covered January 1, 2008, to December 31, 2017, with a 1-year follow-up completed December 31, 2018. Data were gathered from Danish, nationwide registers. Participants included first-time fathers with no history of psychiatric treatment. Data were analyzed from August 1, 2022, to February 20, 2024. Exposures: Adverse pregnancy outcomes including induced abortion, spontaneous abortion, stillbirth, small for gestational age (SGA) and not preterm, preterm with or without SGA, minor congenital malformation, major congenital malformation, and congenital malformation combined with SGA or preterm compared with a full-term healthy offspring. Main Outcomes and Measures: Prescription of psychotropic drugs, nonpharmacological psychiatric treatment, or having a psychiatric hospital contact up to 1 year after the end of the pregnancy. Results: Of the 192 455 fathers included (median age, 30.0 [IQR, 27.0-34.0] years), 31.1% experienced an adverse pregnancy outcome. Most of the fathers in the study had a vocational educational level (37.1%). Fathers experiencing a stillbirth had a significantly increased risk of initiating nonpharmacological psychiatric treatment (adjusted hazard ratio [AHR], 23.10 [95% CI, 18.30-29.20]) and treatment with hypnotics (AHR, 9.08 [95% CI, 5.52-14.90]). Moreover, fathers experiencing an early induced abortion (≤12 wk) had an increased risk of initiating treatment with hypnotics (AHR, 1.74 [95% CI, 1.33-2.29]) and anxiolytics (AHR, 1.79 [95% CI, 1.18-2.73]). Additionally, late induced abortion (>12 wk) (AHR, 4.46 [95% CI, 3.13-6.38]) and major congenital malformation (AHR, 1.36 [95% CI, 1.05-1.74]) were associated with increased risk of nonpharmacological treatment. In contrast, fathers having an offspring being born preterm, SGA, or with a minor congenital malformation did not have a significantly increased risk of any of the outcomes. Conclusions and Relevance: The findings of this Danish cohort study suggest that first-time fathers who experience stillbirths or induced abortions or having an offspring with major congenital malformation had an increased risk of initiating pharmacological or nonpharmacological psychiatric treatment. These findings further suggest a need for increased awareness around the psychological state of fathers following the experience of adverse pregnancy outcomes.

Use of antipsychotic medication, benzodiazepines, and psychiatric hospitalization in cannabis-related versus cannabis-unrelated schizophrenia - a nationwide, register-based cohort study.

BACKGROUND: Evidence suggests that cannabis may be a causal factor for development of schizophrenia. We aimed to investigate whether use of antipsychotic medication, benzodiazepines, and psychiatric service use differs among patients with schizophrenia depending on whether psychosis was precipitated by a diagnosis of cannabis use disorder (CUD). METHODS: We utilized the nationwide Danish registries to identify all individuals with an incident diagnosis of schizophrenia from 1995 to 2016. We also collected information on whether first CUD diagnosis preceded schizophrenia and thus defined a group of potentially cannabis-related schizophrenia. We compared the cannabis-related schizophrenia group both with all non-cannabis-related patients with schizophrenia and with non-cannabis-related patients with schizophrenia that were propensity-score matched to cases using a range of potentially confounding variables. RESULTS: We included 35 714 people with incident schizophrenia, including 4116 (11.5%) that were cannabis-related. In the unmatched-comparison analyses, there were no clear differences over time in use of antipsychotics and benzodiazepines related to whether the diagnosis of schizophrenia was cannabis-related. After propensity-score matching, use of antipsychotics and benzodiazepines was significantly lower among cannabis-related cases of schizophrenia. In the unmatched comparison, the cannabis-related group had significantly more days admitted than the non-cannabis-related group. This was markedly attenuated after propensity-score matching. CONCLUSIONS: Our findings indicate the importance of considering cannabis-related cases of schizophrenia as a potentially distinct disorder in terms of prognosis. It is unclear, however, if these differences are due to different biological types of schizophrenia being compared or if they rather indicate behavioral differences such as reduced adherence and treatment-seeking.

Adult offspring's education and parental mortality: A nationwide cohort study of the mediating role of lifestyle-related diseases.

Aim: The mechanisms behind the association between adult offspring's socioeconomic position and their parents' mortality are not well understood. This study investigates lifestyle-related diseases as a potential mediating pathway between adult offspring's education and parental mortality. Methods: This nationwide register-based cohort study consists of 963,742 older adults aged 65 years between 2000 and 2018. Lifestyle-related diseases were measured between 60 and 65 years and those with prior lifestyle-related diseases were excluded. Natural Effect Models were performed to assess potential mediation through lifestyle-related diseases of the association between offspring's education and parental mortality measured by additive hazard estimates with 95% confidence intervals (CIs). Results: Between 60 and 65 years, 150,501 (15.6%) older adults were diagnosed with lifestyle-related diseases and 149,647 (15.5%) died during follow-up. Compared with having offspring with long education, short education was associated with 631 (95% CI: 555; 707) and 581 (95% CI: 525; 638) additional deaths per 100,000 person-years for women and men, respectively, of which 15.4% (95% CI: 9.0; 21.6) and 16.8% (95% CI: 14.6; 18.9) were mediated by lifestyle-related diseases. The corresponding numbers for medium education were 276 (95% CI: 205; 347) and 299 (95% CI: 255; 343) with 26.2% (95% CI: 12.0; 40.6) and 27.6% (95% CI: 25.1; 31.8) mediated by lifestyle-related diseases. Conclusions: Lifestyle-related diseases accounted for 15-28% of the association between offspring's education and parental mortality for both men and women.

The Danish Aging and Cognition (DanACo) cohort.

BACKGROUND: With aging populations worldwide, identification of predictors of age-related cognitive decline is becoming increasingly important. The Danish Aging and Cognition Cohort (DanACo) including more than 5000 Danish men was established to investigate predictors of age-related cognitive decline from young adulthood to late mid-life. CONSTRUCTION AND CONTENT: The DanACo cohort was established through two separate data collections with identical designs involving a follow-up examination in late mid-life of men for whom intelligence test scores were available from their mandatory conscription board examination. The cohort consists of 5,183 men born from 1949 through 1961, with a mean age of 20.4 years at baseline and a mean age of 64.4 years at follow-up. The baseline measures consisted of height, weight, intelligence test score and educational level collected at the conscription board examination. The follow-up assessment consisted of a re-administration of the same intelligence test and a comprehensive questionnaire covering socio-demographic factors, lifestyle, and health-related factors. The data were collected in test sessions with up to 24 participants per session. Using the unique personal identification number assigned to all Danes, the cohort has been linked to data from national administrative and health registers for prospectively collected data on socioeconomic and health-related factors. UTILITY AND DISCUSSION: The DanACo cohort has some major strengths compared to existing cognitive aging cohorts such as a large sample size (n = 5,183 men), a validated global measure of cognitive ability, a long retest interval (mean 44.0 years) and the availability of prospectively collected data from registries as well as comprehensive questionnaire data. The main weakness is the low participation rate (14.3%) and that the cohort consists of men only. CONCLUSION: Cognitive decline is a result of a summary of factors across the life-course. The DanACo cohort is characterized by a long retest interval and contains data on a wealth of factors across adult life which is essential to establish evidence on predictors of cognitive decline. Moreover, the size of the cohort ensures sufficient statistical power to identify even relatively weak predictors of cognitive decline.

Sedative drug prescription patterns in Danish adults from 2002 through 2021. A register-based cohort study.

OBJECTIVE: Pharmacological treatment strategies for insomnia seem to vary, and there is lack of knowledge about how sedative drugs are used in a real-world setting. We investigated changes in sedative drug prescription patterns in Danish adults who initiated treatment between 2002 and 2016. METHODS: All adults with a first-time purchase of a sedative drug registered in the Danish National Prescription Register from 2002 through 2016 were followed for five years between 2002 and 2021 for subsequent prescriptions of sedative drugs, death, or emigration. Sedative drugs were classified into anxiolytic benzodiazepines (N05BA), hypnotic benzodiazepines (N05CD), Z-drugs (N05CF), melatonin (N05CH01), promethazine (R06AD), and low-dose quetiapine (N05AH04). Analyses were stratified on time: 2002-2006, 2007-2011, and 2012-2016. RESULTS: A total of 842,880 individuals purchased their first sedative drug between 2002 and 2016. Most of them (40.0%) initiated treatment between 2002 and 2006, whereas 29.2% initiated treatment in 2012-2016. In 2002-2006, anxiolytic benzodiazepines (46.4%), Z-drugs (42.8%), and hypnotic benzodiazepines (5.4%) were the most common first treatment. This pattern changed over time with a gradual increase in the use of melatonin, promethazine, and low-dose quetiapine, which in 2011-2016 accounted for 27% of all first treatments. During the five years from first prescription, around 27% shifted to a different sedative drug. This percentage increased slightly over time, but over time the first shift to another drug class was most often to a Z-drug or anxiolytic benzodiazepine. Few individuals (5.8%) had more than one shift and the third choice seemed randomly distributed across all other drug classes. CONCLUSION: Sedative drug prescriptions are distributed on different drug classes, with Z-drugs and anxiolytic benzodiazepines as the most frequent first treatment, and second choice in case of shift.

Mental healthcare utilisation among Danish formerly deployed military personnel and their civilian counterparts: a cohort study.

Background: Prior studies comparing the mental healthcare utilisation (MHU) of Danish formerly deployed military personnel (FDP) with the general population have not included data on psychotherapy through the Defence or talking therapy with the general practitioner. This study included these and several other data sources in a comprehensive comparison of MHU between Danish FDP and civilians.Methods: First-time deployed military personnel (N = 10,971) who had returned from a mission to Kosovo, Afghanistan, Iraq or Lebanon between January 2005 and July 2017 were included. A sex and birth-year-matched civilian reference group was randomly drawn from the entire Danish non-deployed population (N = 253,714). Furthermore, a sub-cohort, including male FDP and civilians deemed eligible for military service, was defined. These cohorts were followed up in military medical records and registers covering the primary and secondary civilian health sectors from 2005 to 2018, and the rates of MHU were compared.Results: Approximately half of the initial help-seeking for FDP took place through the Defence (49.4%), and the remainder through the civilian healthcare system. When help-seeking through the Defence was not included, MHU was significantly lower among FDP in the main cohort during the first two years (IRR = 0.84, 95% CI: [0.77, 0.92]) compared to civilians. When help-seeking through the Defence was included, MHU was significantly higher among FDP compared to civilians both in the first two years of follow-up (IRR = 2.01, 95% CI: [1.89, 2.13]) and thereafter (IRR = 1.18, 95% CI: [1.13, 1.23]). In the sub-cohort, these differences were even more pronounced both in the first two years of follow-up and thereafter.Conclusions: MHU was higher among Danish FDP compared to civilians only when data from the Defence was included. The inclusion of data on both civilian and military healthcare services is necessary to evaluate the full impact of deployment on MHU among Danish FDP.

This study compared mental healthcare utilisation among Danish deployed military personnel and civilians.Most personnel sought help first through the Defence.When all data sources were included, mental healthcare utilisation was significantly higher among military personnel.

Early cerebral amyloid-ß accumulation and hypermetabolism are associated with subtle cognitive deficits before accelerated cerebral atrophy.

AIMS: Alzheimer's disease (AD) is characterized by the accumulation of amyloid beta (Aß) in the brain. The deposition of Aß is believed to initiate a detrimental cascade, including cerebral hypometabolism, accelerated brain atrophy, and cognitive problems-ultimately resulting in AD. However, the timing and causality of the cascade resulting in AD are not yet fully established. Therefore, we examined whether early Aß accumulation affects cerebral glucose metabolism, atrophy rate, and age-related cognitive decline before the onset of neurodegenerative disease. METHODS: Participants from the Metropolit 1953 Danish Male Birth Cohort underwent brain positron emission tomography (PET) imaging using the radiotracers [11C]Pittsburgh Compound-B (PiB) (N = 70) and [18F]Fluorodeoxyglucose (FDG) (N = 76) to assess cerebral Aß accumulation and glucose metabolism, respectively. The atrophy rate was calculated from anatomical magnetic resonance imaging (MRI) scans conducted presently and 10 years ago. Cognitive decline was examined from neurophysiological tests conducted presently and ten or 5 years ago. RESULTS: Higher Aß accumulation in AD-critical brain regions correlated with greater visual memory decline (p = 0.023). Aß accumulation did not correlate with brain atrophy rates. Increased cerebral glucose metabolism in AD-susceptible regions correlated with worse verbal memory performance (p = 0.040). CONCLUSIONS: Aß accumulation in known AD-related areas was associated with subtle cognitive deficits. The association was observed before hypometabolism or accelerated brain atrophy, suggesting that Aß accumulation is involved early in age-related cognitive dysfunction. The association between hypermetabolism and worse memory performance may be due to early compensatory mechanisms adapting for malfunctioning neurons by increasing metabolism.