Continued versus discontinued oxytocin stimulation in the active phase of labour (CONDISOX): individual management based on artificial intelligence - a secondary analysis.

BACKGROUND: Current guidelines regarding oxytocin stimulation are not tailored to individuals as they are based on randomised controlled trials. The objective of the study was to develop an artificial intelligence (AI) model for individual prediction of the risk of caesarean delivery (CD) in women with a cervical dilatation of 6 cm after oxytocin stimulation for induced labour. The model included not only variables known when labour induction was initiated but also variables describing the course of the labour induction. METHODS: Secondary analysis of data from the CONDISOX randomised controlled trial of discontinued vs. continued oxytocin infusion in the active phase of induced labour. Extreme gradient boosting (XGBoost) software was used to build the prediction model. To explain the impact of the predictors, we calculated Shapley additive explanation (SHAP) values and present a summary SHAP plot. A force plot was used to explain specifics about an individual's predictors that result in a change of the individual's risk output value from the population-based risk. RESULTS: Among 1060 included women, 160 (15.1%) were delivered by CD. The XGBoost model found women who delivered vaginally were more likely to be parous, taller, to have a lower estimated birth weight, and to be stimulated with a lower amount of oxytocin. In 108 women (10% of 1060) the model favoured either continuation or discontinuation of oxytocin. For the remaining 90% of the women, the model found that continuation or discontinuation of oxytocin stimulation affected the risk difference of CD by less than 5% points. CONCLUSION: In women undergoing labour induction, this AI model based on a secondary analysis of data from the CONDISOX trial may help predict the risk of CD and assist the mother and clinician in individual tailored management of oxytocin stimulation after reaching 6 cm of cervical dilation.

Using Explainable Artificial Intelligence to Predict Potentially Preventable Hospitalizations: A Population-Based Cohort Study in Denmark.

BACKGROUND: The increasing aging population and limited health care resources have placed new demands on the healthcare sector. Reducing the number of hospitalizations has become a political priority in many countries, and special focus has been directed at potentially preventable hospitalizations. OBJECTIVES: We aimed to develop an artificial intelligence (AI) prediction model for potentially preventable hospitalizations in the coming year, and to apply explainable AI to identify predictors of hospitalization and their interaction. METHODS: We used the Danish CROSS-TRACKS cohort and included citizens in 2016-2017. We predicted potentially preventable hospitalizations within the following year using the citizens' sociodemographic characteristics, clinical characteristics, and health care utilization as predictors. Extreme gradient boosting was used to predict potentially preventable hospitalizations with Shapley additive explanations values serving to explain the impact of each predictor. We reported the area under the receiver operating characteristic curve, the area under the precision-recall curve, and 95% confidence intervals (CI) based on five-fold cross-validation. RESULTS: The best performing prediction model showed an area under the receiver operating characteristic curve of 0.789 (CI: 0.782-0.795) and an area under the precision-recall curve of 0.232 (CI: 0.219-0.246). The predictors with the highest impact on the prediction model were age, prescription drugs for obstructive airway diseases, antibiotics, and use of municipality services. We found an interaction between age and use of municipality services, suggesting that citizens aged 75+ years receiving municipality services had a lower risk of potentially preventable hospitalization. CONCLUSION: AI is suitable for predicting potentially preventable hospitalizations. The municipality-based health services seem to have a preventive effect on potentially preventable hospitalizations.

Risk of asbestosis, mesothelioma, other lung disease or death among motor vehicle mechanics: a 45-year Danish cohort study.

INTRODUCTION: The risk of asbestosis, malignant mesothelioma and lung cancer among motor vehicle mechanics is of concern because of potential exposure to chrysotile asbestos during brake, clutch and gasket repair and maintenance. Asbestos has also been used in insulation and exhaust systems. METHODS: We examined the long-term risk of incident mesothelioma, lung cancer, asbestosis and other lung diseases and mortality due to mesothelioma, lung cancer, asbestosis and other lung diseases in a nationwide cohort of all men registered as motor vehicle mechanics since 1970 in Denmark. This was compared with the corresponding risk in a cohort of male workers matched 10:1 by age and calendar year, with similar socioeconomic status (instrument makers, dairymen, upholsterers, glaziers, butchers, bakers, drivers, farmers and workers in the food industry, trade or public services). RESULTS: Our study included 138 559 motor vehicle mechanics (median age 24 years; median follow-up 20 years (maximum 45 years)) and 1 385 590 comparison workers (median age 25 years; median follow-up 19 years (maximum 45 years)). Compared with other workers, vehicle mechanics had a lower risk of morbidity due to mesothelioma/pleural cancer (n=47 cases) (age-adjusted and calendar-year-adjusted HR=0.74 (95% CI 0.55 to 0.99)), a slightly increased risk of lung cancer (HR=1.09 (95% CI 1.03 to 1.14)), increased risk of asbestosis (HR=1.50 (95% CI 1.10 to 2.03)) and a chronic obstructive pulmonary disease risk close to unity (HR=1.02 (95% CI 0.99 to 1.05)). Corresponding HRs for mortality were 0.86 (95% CI 0.64 to 1.15) for mesothelioma/pleural cancer, 1.06 (95% CI 1.01 to 1.12) for lung cancer, 1.79 (95% CI 1.10 to 2.92) for asbestosis, 1.06 (95% CI 0.86 to 1.30) for other lung diseases caused by external agents and 1.00 (95% CI 0.98 to 1.01) for death due to all causes. CONCLUSIONS: We found that the risk of asbestosis was increased among vehicle mechanics. The risk of malignant mesothelioma/pleural cancers was not increased among vehicle mechanics.

Dietary folate intake and fecundability in two preconception cohorts.

STUDY QUESTION: To what extent is dietary folate intake and total folate intake (dietary and supplemental intakes) associated with fecundability, the per cycle probability of conception? SUMMARY ANSWER: Preconception dietary folate intake was positively associated with fecundability in a monotonic pattern. WHAT IS KNOWN ALREADY: Supplemental folic acid has been associated with improved fertility, but little is known about the relation between dietary folate and fecundability. STUDY DESIGN, SIZE, DURATION: A prospective cohort study including 9559 women trying to conceive without fertility treatment and enrolled in the period 2013-2020. PARTICIPANTS/MATERIALS, SETTING, METHODS: We used data from two internet-based prospective cohort studies of pregnancy planners from Denmark, where folic acid fortification is not performed (SnartForældre.dk (SF); n = 3755) and North America, where the food supply is fortified with folic acid (Pregnancy Study Online (PRESTO); n = 5804). Women contributed menstrual cycles at risk until they reported conception or experienced a censoring event. We used proportional probabilities regression models to compute fecundability ratios (FRs) and 95% CI, adjusting for potential confounders. MAIN RESULTS AND THE ROLE OF CHANCE: Compared with a dietary folate intake ≥400 µg/day, the adjusted FRs for women in SF were 0.92 (95% CI: 0.85-0.99) for intake 250-399 µg/day, and 0.80 (95% CI: 0.68-0.94) for intake of <250 µg/day. The corresponding FRs in PRESTO were 0.95 (95% CI: 0.89-1.01) and 0.81 (95% CI: 0.65-1.00). Compared with the highest level of total folate intake (diet folate ≥400 µg/day plus folic acid supplementation), in both cohorts fecundability was lowest among women with the lowest dietary intake <250 µg/day dietary folate and no supplementation (FR: 0.76, 95% CI: 0.59-0.98 [SF] and 0.49, 95% CI: 0.31-0.77 [PRESTO]). Further, total intake dietary folate <250 µg/day plus supplementation was associated with reduced fecundability for SF participants (FR; 0.79, 95% CI: 0.65-0.98) and for PRESTO participants (FR; 0.92, 95% CI: 0.72-1.16). LIMITATIONS, REASONS FOR CAUTION: It is unknown whether dietary folate and folic acid intake affect fecundability on its own or if there is an interaction with other micronutrients provided in healthy diet. Thus, the observed associations may not reflect dietary folate intake alone, but overall healthy diet. WIDER IMPLICATIONS OF THE FINDINGS: Recommendations for preconception dietary folate intake and folic acid supplementation are of importance not only to prevent neural tube defects but also to enhance fecundability. STUDY FUNDING/COMPETING INTEREST(S): The study was supported by the National Institute of Child Health and Human Development (R01-HD086742). The authors report no competing interests. TRIAL REGISTRATION NUMBER: N/A.

Seasonal patterns in fecundability in North America and Denmark: a preconception cohort study.

STUDY QUESTION: To what extent does fecundability vary across seasons? SUMMARY ANSWER: After accounting for seasonal patterns in pregnancy planning, we observed higher fecundability in the fall and lower fecundability in the spring, particularly at lower latitudes. WHAT IS KNOWN ALREADY: In human populations, there are strong seasonal patterns of births that vary across geographic regions and time periods. However, previous studies of seasonality and fecundity are limited because they examine season of birth rather than season of conception and therefore neglect to account for seasonal variation in initiating attempts to conceive or pregnancy loss or differences in gestational length. STUDY DESIGN, SIZE, DURATION: We conducted a preconception cohort study of 14 331 women residing in North America (June 2013-May 2018: n = 5827) and Denmark (June 2007-May 2018: n = 8504). Participants were attempting to conceive without fertility treatment and had been attempting pregnancy for ≤6 menstrual cycles at enrolment. PARTICIPANTS/MATERIAL, SETTING, METHODS: We collected information on season of each pregnancy attempt using last menstrual period dates over the study period. Pregnancy was reported on female bi-monthly follow-up questionnaires. We fit log-binomial models with trigonometric regression to examine periodic variation in fecundability. We accounted for seasonal variation in initiation of pregnancy attempts by including indicator variables for menstrual cycle of attempt in the regression models. MAIN RESULTS AND THE ROLE OF CHANCE: Initiation of pregnancy attempts peaked in September, with stronger seasonality in North America than in Denmark (48 vs. 16% higher probability initiating attempts in September compared with March). After accounting for seasonal variation in initiation of pregnancy attempts, we observed modest seasonal variation in fecundability, with a peak in the late fall and early winter in both cohorts, but stronger peak/low ratios in North America (1.16; 95% confidence interval [CI]: 1.05, 1.28) than in Denmark (1.08; 95% CI: 1.00, 1.16). When we stratified the North American data by latitude, we observed the strongest seasonal variation in the southern USA (peak/low ratio of 1.45 [95% CI: 1.14, 1.84]), with peak fecundability in late November. LIMITATIONS, REASONS FOR CAUTION: We estimated menstrual cycle dates between follow-up questionnaires, which may have introduced exposure misclassification, particularly when women skipped follow-up questionnaires. We were unable to measure seasonally varying factors that may have influenced fecundability, including ambient temperature, vitamin D levels or infectious disease. WIDER IMPLICATIONS OF THE FINDINGS: An understanding of how fecundability varies across seasons could help identify factors that can impair reproductive function. Neglecting to account for seasonal variation in initiation of pregnancy attempts could bias estimates of seasonal patterns in fecundability. This is the first preconception cohort study to examine seasonal variation in fecundability after accounting for seasonality in initiation of pregnancy attempts. Fecundability was highest in the fall and lowest in the spring, with stronger effects in southern latitudes of North America, suggesting that seasonal exposures may affect fecundity. STUDY FUNDING/COMPETING INTEREST(S): This research was funded by the Eunice K. Shriver National Institute of Child Health and Human Development (R21-050264, R01-HD060680, R21-HD072326 and R01-HD086742) and the Danish Medical Research Council (271-07-0338). The authors declare no conflicts of interest.

Nitrosatable drug exposure during pregnancy and risk of stillbirth.

PURPOSE: Nitrosatable drugs can react with nitrite in the stomach and form N-nitroso compounds. Exposure to nitrosatable drugs has been associated with congenital malformations and preterm birth, but use during pregnancy as a cause of fetal death is not well-known. We examined if prenatally nitrosatable drug use is associated with risk of stillbirth. METHODS: A nationwide cohort was conducted using 554 844 women with singleton and first recorded pregnancies regardless of previous pregnancy history from the Danish Medical Birth Register from 1996 to 2015. Exposure was recorded by use of the Danish National Prescription Register and defined as women who had redeemed a prescribed nitrosatable drug in the first 22 weeks of pregnancy. The reference group was women with no redeemed prescribed nitrosatable drug in this time period. We categorized nitrosatable drugs as secondary amines, tertiary amines, and amides. Cox hazard regression was used to estimate crude and adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs) for stillbirth. RESULTS: Among the 84 720 exposed women, 348 had a stillbirth compared with 1690 stillbirths among the 470 124 unexposed women. Women who used any prescribed nitrosatable drug were more likely to have a stillbirth compared with unexposed women (aHRs 1.24; 95% CI, 1.03-1.49). CONCLUSION: Nitrosatable drug use during the first 22 weeks of pregnancy might increase risk of stillbirth. The findings should be interpreted cautiously because of important unmeasured factors that might influence the observed association, including maternal vitamin C intake, dietary, and other sources of nitrate/nitrite intake.

Validating registry data on statins prescriptions by blood measurements.

PURPOSE: To validate prescription registry data as a measurement of adherence to statins through a direct method using assays for selected statins in serial blood samples collected from two prospective cohorts of Danish colorectal cancer patients. METHODS: We linked information on statin prescriptions from the Aarhus University Prescription Database with the cancer cohorts from Aalborg University Hospital. For statin-prescribed patients, we calculated a prescription window covering the anticipated duration of the prescription. For each statin-prescribed patient with at least one blood sample in a prescription window, we selected without replacement a never-statin-prescribed patient matched on sex, age, and calendar year of surgery. Each of the selected blood samples were analyzed using assays to detect statins. We calculated the positive and negative predictive value of the prescription registry reporting using the assay result as the gold standard. RESULTS: We identified 73 ever-statin-prescribed patients with a total of 253 blood samples and 74 blood samples among never-statin-prescribed patients. The positive predictive value for prescribed patients, with presence of statins in at least one blood sample as the gold standard, was 93% (95% CI, 86%-97%) and the negative predictive value was 93% (95% CI, 86%-97%). Stratified results did not reveal substantial differences in predictive values. Fifty-two (71%) of the statin-prescribed patients had statins in every blood sample, suggesting continuous adherence. CONCLUSION: This study showed a high adherence with treatment with statins among colorectal cancer patients.

Dietary Fat Intake and Fecundability in 2 Preconception Cohort Studies.

The association between dietary fat and fertility is not well studied. We evaluated intakes of total fat, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, trans fatty acids (TFA), ω-3 fatty acids, and ω-6 fatty acids in relation to fecundability in Danish and North American preconception cohort studies. Women who were attempting to become pregnant completed a validated food frequency questionnaire at baseline. Pregnancy status was updated bimonthly for 12 months or until pregnancy. Fecundability ratios (FR) and 95% confidence intervals were estimated using multivariable proportional probabilities regression. Intakes of total fat and saturated, monounsaturated, polyunsaturated, and ω-6 fatty acids were not appreciably associated with fecundability. TFA intake was associated with reduced fecundability in North American women (for the fourth quartile vs. the first, FR = 0.86, 95% confidence interval (CI): 0.71, 1.04) but not Danish women (for the fourth quartile vs. the first, FR = 1.04, 95% CI: 0.86, 1.25), though intake among Danish women was low. In North America, ω-3 fatty acid intake was associated with higher fecundability, but there was no dose-response relationship (among persons who did not use fish oil supplements: for the fourth quartile vs. the first, FR = 1.40, 95% CI: 1.13, 1.73); no association was found in Danish women, among whom low intake was rare. In the present study, high TFA intake and low ω-3 fatty acid intake were associated with reduced fecundity.

Impact of Allogeneic Stem Cell Transplantation in First Complete Remission in Acute Myeloid Leukemia: A National Population-Based Cohort Study.

To examine the outcomes of allogeneic stem cell transplantation (HSCT) in first complete remission (CR1) compared with chemotherapy alone in a population-based setting, we identified a cohort of patients with acute myeloid leukemia (AML) aged 15 to 70 years diagnosed between 2000 and 2014 in Denmark. Using the Danish National Acute Leukemia Registry, we compared relapse risk, relapse-free survival (RFS), and overall survival (OS) between patients with unfavorable cytogenetic features receiving postremission therapy with conventional chemotherapy only versus those undergoing HSCT in CR1. To minimize immortal time bias, we performed Cox proportional hazards regression, included date of allogeneic HSCT as a time-dependent covariate, and stratified the results by age (<60 or ≥60 years) and cytogenetic risk group. Overall, 1031 patients achieved a CR1. Of these, 196 patients (19%) underwent HSCT. HSCT was associated with a lower relapse rate (24% versus 49%) despite a similar median time to relapse (287 days versus 265 days). In all subgroups, the risk of relapse was lower and both RFS and OS were superior in recipients of HSCT (OS, adjusted mortality ratios: all patients, .54 [95% confidence interval (CI), .42-.71]; patients age <60 years, .58 [95% CI, .42-.81]; patients age ≥60 years, .42 [95% CI, .26-.69]; patients with intermediate-risk cytogenetics, .63 [95% CI, .43-.87]; patients with adverse-risk cytogenetics, .40 [95% CI, .24-.67]). In conclusion, in this population-based nationwide cohort study, HSCT was associated with improved survival in both younger and older patients and in patients with both intermediate and adverse cytogenetic risk.

Associations of Statin Use With Colorectal Cancer Recurrence and Mortality in a Danish Cohort.

In earlier studies of the influence of hydroxymethylglutaryl-coenzyme A reductase inhibitors (also known as statins) on colorectal cancer prognosis, investigators reported a reduced rate of cancer-specific mortality. Studies of recurrence are few and small. Using data from Danish registries, we followed 21,152 patients diagnosed with stage I-III colorectal cancer from 2001 to 2011. We estimated the association between statin use in the preceding year and cancer recurrence, cancer-specific mortality, and all-cause mortality rates. We identified 5,036 recurrences, 7,084 deaths from any cause, and 4,066 deaths from colorectal cancer. After adjustment for potential confounders, statin use was not associated with recurrence (adjusted hazard ratio (aHR) = 1.01, 95% confidence interval (CI): 0.93, 1.09), but it was associated with death from colorectal cancer (aHR = 0.72, 95% CI: 0.65, 0.79) and death from any cause (aHR = 0.72, 95% CI: 0.67, 0.76). Statin use in the year preceding recurrence was associated with a reduced risk of cancer-specific mortality (aHR = 0.83, 95% CI: 0.74, 0.92) but also a reduced risk of death from any other cause (aHR = 0.78, 95% CI: 0.61, 1.00). Statin use was not associated with a reduced rate of colorectal cancer recurrence, but it was associated with a reduced rate of cancer-specific mortality, which suggests that there is no cancer-directed benefit; therefore, there is no basis to prescribe statins to colorectal cancer patients who do not have cardiovascular indications.

Female exposure to phthalates and time to pregnancy: a first pregnancy planner study.

STUDY QUESTION: Is female exposure to phthalate metabolites associated with reduced fecundity, as estimated by prolonged time to pregnancy (TTP)? SUMMARY ANSWER: Female exposure to monoethyl phthalate (MEP) but not monobutyl phthalate (MBP), monobenzyl phthalate (MBzP) and monoethylhexyl phthalate (MEHP) was associated with a longer TTP. WHAT IS KNOWN ALREADY: Male exposure to phthalates is potentially associated with adverse effects on human fecundity in epidemiological studies, but little is known about the potential effects on female reproduction. STUDY DESIGN SIZE AND DURATION: A cohort study with prospective data based on 229 women from a Danish cohort of 430 first pregnancy planning couples enrolled in 1992-1994. In 2009, urinary analyses of phthalate metabolites were performed on stored urine samples from this cohort. PARTICIPANTS/MATERIALS, SETTING AND METHODS: We analyzed MEP, MBP, MBzP and MEHP in female morning spot urine samples collected daily during the first 10 days of menstrual cycles after discontinuation of contraception. The exposure assessment was based on the mean of two measurements from each woman collected in a period of 6 menstrual cycles. We used Cox regression with discrete time to estimate fecundability ratios (FRs) and 95% CI in relation to the average urine metabolite concentration exposure level, controlled for age and BMI, and the time-varying variables smoking and alcohol. MAIN RESULT AND ROLE OF CHANCE: Urinary concentration of MEP was associated with a decreased fecundity (adjusted FR 0.79; 95% CI: 0.63; 0.99) corresponding to a 21% decreased probability of conception for each natural log (ln) unit increase in MEP. No significant association with TTP was found for MBP, MBzP and MEHP. LIMITATIONS REASONS FOR CAUTION: Subfertile women were overrepresented in the study population due to exclusion of 77 high fertile women who became pregnant in the first cycle when urine collection began. WIDER IMPLICATIONS OF THE FINDINGS: Our results suggest that female exposure to MEP may have an adverse effect on female fecundity, but these findings need to be replicated in a larger and newer cohort study with sufficient exposure contrast if the use of diethyl phthalate (DEP) and thereby MEP in the future potentially should be regulated in cosmetics and industrial consumer products. STUDY FUNDING/COMPETING INTERESTS: The original data collected were founded by Aarhus University Research Foundation, the Danish Medical Research Council and the Danish Medical Health Insurance Foundation. There are no conflicts of interest to be declared. TRIAL REGISTRATION NUMBER: N/A.

Descriptive characteristics of colon and rectal cancer recurrence in a Danish population-based study.

BACKGROUND: Recurrence is a common outcome among patients that have undergone an intended curative resection for colorectal cancer. However, data on factors that influence colorectal cancer recurrence are sparse. We report descriptive characteristics of both colon and rectal cancer recurrence in an unselected population. MATERIAL AND METHODS: We identified 21,152 patients with colorectal cancer diagnosed between May 2001 and December 2011 and registered with the Danish Colorectal Cancer Group. Recurrences were identified in 3198 colon and 1838 rectal cancer patients during follow-up. We calculated the frequency, proportion, and incidence rates of colon and rectal cancer recurrence within descriptive categories, and the cumulative five- and ten-year incidences of recurrence, treating death as a competing risk. We used a Cox proportional hazard model to calculate hazard ratios (HR) and 95% confidence intervals (CI). RESULTS: Recurrence risk was highest in the first three years of follow-up. Patients <55 years old at initial diagnosis (incidence rate for colon: 7.2 per 100 person-years; 95% CI: 6.5-7.9; rectum: 8.1 per 100 person-years; 95% CI: 7.2-9.0) and patients diagnosed with stage III cancer (colon HR: 5.70; 95% CI: 4.61-7.06; rectal HR: 7.02; 95% CI: 5.58-8.82) had increased risk of recurrence. Patients diagnosed with stage III cancer from 2009 to 2011 had a lower incidence of recurrence than those diagnosed with stage III cancer in the years before. Cumulative incidences of colon and rectal cancer recurrence were similar for both cancer types among each descriptive category. CONCLUSIONS: In this population, increases in colorectal cancer recurrence risk were associated with younger age and increasing stage at diagnosis. Cumulative incidence of recurrence did not differ by cancer type. Descriptive characteristics of colon and rectal cancer recurrence may help to inform patient-physician decision-making, and could be used to determine adjuvant therapies or tailor surveillance strategies so that recurrence may be identified early, particularly within the first 3 years of follow-up.

Costs and Efficiency of Online and Offline Recruitment Methods: A Web-Based Cohort Study.

BACKGROUND: The Internet is widely used to conduct research studies on health issues. Many different methods are used to recruit participants for such studies, but little is known about how various recruitment methods compare in terms of efficiency and costs. OBJECTIVE: The aim of our study was to compare online and offline recruitment methods for Internet-based studies in terms of efficiency (number of recruited participants) and costs per participant. METHODS: We employed several online and offline recruitment methods to enroll 18- to 45-year-old women in an Internet-based Danish prospective cohort study on fertility. Offline methods included press releases, posters, and flyers. Online methods comprised advertisements placed on five different websites, including Facebook and Netdoktor.dk. We defined seven categories of mutually exclusive recruitment methods and used electronic tracking via unique Uniform Resource Locator (URL) and self-reported data to identify the recruitment method for each participant. For each method, we calculated the average cost per participant and efficiency, that is, the total number of recruited participants. RESULTS: We recruited 8252 study participants. Of these, 534 were excluded as they could not be assigned to a specific recruitment method. The final study population included 7724 participants, of whom 803 (10.4%) were recruited by offline methods, 3985 (51.6%) by online methods, 2382 (30.8%) by online methods not initiated by us, and 554 (7.2%) by other methods. Overall, the average cost per participant was €6.22 for online methods initiated by us versus €9.06 for offline methods. Costs per participant ranged from €2.74 to €105.53 for online methods and from €0 to €67.50 for offline methods. Lowest average costs per participant were for those recruited from Netdoktor.dk (€2.99) and from Facebook (€3.44). CONCLUSIONS: In our Internet-based cohort study, online recruitment methods were superior to offline methods in terms of efficiency (total number of participants enrolled). The average cost per recruited participant was also lower for online than for offline methods, although costs varied greatly among both online and offline recruitment methods. We observed a decrease in the efficiency of some online recruitment methods over time, suggesting that it may be optimal to adopt multiple online methods.

Brief Report: Cesarean Delivery and Subsequent Fecundability.

BACKGROUND: Studies have shown that cesarean delivery is associated with fewer subsequent births relative to vaginal delivery, but it is unclear whether confounding by pregnancy intention or indication for surgery explained these results. We evaluated the association between cesarean delivery and subsequent fecundability among 910 primiparous women after singleton live birth. METHODS: In a cohort of Danish women planning pregnancy (2007-2012), obstetrical history was obtained via registry linkage; time-to-pregnancy and covariate data were collected via questionnaire. Fecundability ratios (FRs) and 95% confidence intervals (CIs) were adjusted for potential confounders. RESULTS: Relative to spontaneous vaginal delivery, emergency cesarean delivery with cephalic presentation showed little association with fecundability (FR = 1.0, 95% CI = 0.83, 1.3), but cesarean delivery with breech presentation (FR = 0.72, 95% CI = 0.53, 0.97) and planned cesarean delivery with cephalic presentation (FR = 0.51, 95% CI = 0.25, 1.0) were associated with reduced fecundability. CONCLUSIONS: The cesarean-fecundability association varied by previous fetal presentation and emergency status.

Evaluation of Selection Bias in an Internet-based Study of Pregnancy Planners.

Selection bias is a potential concern in all epidemiologic studies, but it is usually difficult to assess. Recently, concerns have been raised that internet-based prospective cohort studies may be particularly prone to selection bias. Although use of the internet is efficient and facilitates recruitment of subjects that are otherwise difficult to enroll, any compromise in internal validity would be of great concern. Few studies have evaluated selection bias in internet-based prospective cohort studies. Using data from the Danish Medical Birth Registry from 2008 to 2012, we compared six well-known perinatal associations (e.g., smoking and birth weight) in an internet-based preconception cohort (Snart Gravid n = 4,801) with the total population of singleton live births in the registry (n = 239,791). We used log-binomial models to estimate risk ratios (RRs) and 95% confidence intervals (CIs) for each association. We found that most results in both populations were very similar. For example, maternal obesity was associated with an increased risk of delivering a macrosomic infant in Snart Gravid (RR = 1.5; 95% CI: 1.2, 1.7) and the total population (RR = 1.5; 95% CI: 1.45, 1.53), and maternal smoking of >10 cigarettes per day was associated with a higher risk of low birth weight (RR = 2.7; 95% CI: 1.2, 5.9 vs. RR = 2.9; 95% CI: 2.6, 3.1) in Snart Gravid and the total population, respectively. We cannot be certain that our results would apply to other associations or different populations. Nevertheless, our results suggest that recruitment of reproductive aged women via the internet may be no more prone to selection bias than traditional methods of recruitment.

Low-Dose Aspirin or Nonsteroidal Anti-inflammatory Drug Use and Colorectal Cancer Risk: A Population-Based, Case-Control Study.

BACKGROUND: A recent comprehensive review concluded that additional research is needed to determine the optimal use of aspirin for cancer prevention. OBJECTIVE: To assess associations between the use of low-dose aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs) and colorectal cancer risk. DESIGN: Population-based, case-control study. SETTING: Northern Denmark. PATIENTS: Patients with first-time colorectal cancer in northern Denmark between 1994 and 2011. Population control participants were selected by risk set sampling. MEASUREMENTS: Data on drug use, comorbid conditions, and history of colonoscopy were obtained from prescription and patient registries. Use of low-dose aspirin (75 to 150 mg) and nonaspirin NSAIDs was defined according to type, estimated dose, duration, and consistency of use. RESULTS: Among 10 280 case patients and 102 800 control participants, the adjusted odds ratios (ORs) for colorectal cancer associated with ever use (≥2 prescriptions) of low-dose aspirin and nonaspirin NSAIDs were 1.03 (95% CI, 0.98 to 1.09) and 0.94 (CI, 0.90 to 0.98), respectively. Continuous long-term use (≥5 years) of low-dose aspirin was associated with a 27% reduction in colorectal cancer risk (OR, 0.73 [CI, 0.54 to 0.99]), whereas the overall OR for cumulative long-term use (continuous or noncontinuous) was close to unity. Nonaspirin NSAID use was associated with a substantial reduction in colorectal cancer risk, particularly for long-term, high-intensity use (average defined daily dose ≥0.3) of agents with high cyclooxygenase-2 selectivity (OR, 0.57 [CI, 0.44 to 0.74]). LIMITATIONS: Data were unavailable on over-the-counter purchases of high-dose aspirin and low-dose ibuprofen or NSAID dosing schedules, there were several comparisons, and the authors were unable to adjust for confounding by some risk factors. CONCLUSION: Long-term, continuous use of low-dose aspirin and long-term use of nonaspirin NSAIDs were associated with reduced colorectal cancer risk. Persons who continuously used low-dose aspirin comprised only a small proportion of the low-dose aspirin users. PRIMARY FUNDING SOURCE: Danish Cancer Society, Aarhus University Research Foundation.

A validated algorithm to ascertain colorectal cancer recurrence using registry resources in Denmark.

Colorectal cancer recurrences are difficult to ascertain accurately and efficiently. We developed and validated an algorithm to identify recurrences that uses Danish medical registries. The algorithm uses metastasis and chemotherapy codes in the Danish National Patient Registry and codes indicating cancer recurrence in the Danish Pathology Registry. We applied the algorithm to a cohort (n = 21,246) of colorectal cancer patients diagnosed 2001-2011 and followed through 2012. In a cohort (n = 355) of two groups of actively followed patients, we compared the imputed recurrence data with recurrences diagnosed by regular follow-up. We compared cumulative incidence curves of imputed recurrence in local and regional stage patients from the large cohort, and of imputed and diagnosed recurrences in the actively followed cohort. In the 355 members of the actively followed cohort, our algorithm correctly identified 60 of 63 recurrences [sensitivity = 95%; 95% confidence interval (CI) 87-99%] and misclassified only 10 of 292 without recurrence (specificity = 97%; 95% CI 94-98%). Cumulative incidence curves showed that members of the large cohort with regional disease had much higher incidence of imputed recurrence than those with local disease. In the actively followed cohort, the cumulative incidence of recurrence overlapped substantially when recurrence was imputed by our algorithm or using the follow-up data. Despite some limitations regarding ambiguous pathology codes, our algorithm showed excellent performance against actively followed recurrence data, and the expected relation between recurrence risk and cancer stage. It can be used in the Danish registries and adapted to similar registries elsewhere.

Maternal Recall Error in Retrospectively Reported Time-to-Pregnancy: an Assessment and Bias Analysis.

BACKGROUND: Epidemiologic studies of fecundability often use retrospectively measured time-to-pregnancy (TTP), thereby introducing potential for recall error. Little is known about how recall error affects the bias and precision of the fecundability odds ratio (FOR) in such studies. METHODS: Using data from the Danish Snart-Gravid Study (2007-12), we quantified error for TTP recalled in the first trimester of pregnancy relative to prospectively measured TTP among 421 women who enrolled at the start of their pregnancy attempt and became pregnant within 12 months. We defined recall error as retrospectively measured TTP minus prospectively measured TTP. Using linear regression, we assessed mean differences in recall error by maternal characteristics. We evaluated the resulting bias in the FOR and 95% confidence interval (CI) using simulation analyses that compared corrected and uncorrected retrospectively measured TTP values. RESULTS: Recall error (mean = -0.11 months, 95% CI -0.25, 0.04) was not appreciably associated with maternal age, gravidity, or recent oral contraceptive use. Women with TTP > 2 months were more likely to underestimate their TTP than women with TTP ≤ 2 months (unadjusted mean difference in error: -0.40 months, 95% CI -0.71, -0.09). FORs of recent oral contraceptive use calculated from prospectively measured, retrospectively measured, and corrected TTPs were 0.82 (95% CI 0.67, 0.99), 0.74 (95% CI 0.61, 0.90), and 0.77 (95% CI 0.62, 0.96), respectively. CONCLUSIONS: Recall error was small on average among pregnancy planners who became pregnant within 12 months. Recall error biased the FOR of recent oral contraceptive use away from the null by 10%. Quantitative bias analysis of the FOR can help researchers quantify the bias from recall error.

Pre-gravid oral contraceptive use in relation to birth weight: a prospective cohort study.

Few studies have evaluated the association between pregravid oral contraceptive (OC) use and birth weight, and findings have been conflicting. We conducted a prospective cohort study of 5921 pregnancy planners in Denmark to evaluate recency, duration, and type of OC used before conception in relation to infant birth weight. Participants completed online questionnaires and reported detailed information on contraceptive history and covariates at baseline. Participants completed bimonthly follow-up questionnaires to update their pregnancy status for up to 12 months or until conception occurred. Birth weight data were ascertained from the Danish Medical Birth Registry for 4046 live births delivered by study participants between 2008 and 2010. We used multivariable linear and log-binomial regression analyses to control for confounding. Mean birth weight was higher among women who had used OCs within 0-1 months (mean difference = 97 g, CI 26, 168) or 2-6 months (mean difference = 40 g, CI -5, 85) before conception, compared with more than 12 months before conception. Mean birth weight was lower among women who had used OCs for long durations (mean difference comparing ≥12 with <4 years of OC use = -85 g, CI -158, -11). Our findings indicate that pregravid OC use within 6 months of conception may be associated with a small increase in birth weight, but that long duration of use may have the opposite effect. Results were stronger among male infants, among 2nd and 4th generation OC users, and among users of OCs with a higher estrogen dose.

Short look-back periods in pharmacoepidemiologic studies of new users of antibiotics and asthma medications introduce severe misclassification.

PURPOSE: The aim of this study was to quantify the effect of the look back period on the misclassification of new users of antibiotics and asthma medications. METHODS: We included all children born in Denmark from 1995 through 2006 and all prescriptions of antibiotics and asthma medication from 1995 through 2011. The study period was 2007 through 2011. True new users redeemed their first prescription in the study period whereas prior users redeemed their first prescription before the study period. Look-back periods ranged from 30 days up to 12 years prior to the study period, and we defined new users as those without a prescription in the look-back period. The relative misclassification (RM) was estimated as the number of defined new users divided by the number of true new users. RESULTS: For antibiotics, the RM decreased from 4.75 for look-back periods of 30 days to 2.36 for 2 years and 1.33 for 5 years. For asthma medication, the RM decreased from 2.53 for look-back periods of 30 days to 1.48 for 2 years and 1.20 for 5 years. Older age, male gender, and absence of treatment-related diagnoses were associated with higher RM. CONCLUSIONS: Studies applying the new user design are strongly dependent on the available information on prescriptions. For drug classes with intermittent use such as asthma medications, even a 2-year look-back period produced severe misclassification. Excluding children with a prior treatment-related diagnosis can reduce the level of misclassification.