Our study cohort includes all live births in Denmark between 1997 and 2007. All subjects and information about them were gathered from three Danish national registries and Statistics Denmark [9
]: The Danish Fertility Database [10
], The Danish National Hospital Register [11
], and The Danish National Prescription Register [12
]. From these registers and Statistics Denmark, information at individual level was gathered via the unique personal identification number given to all Danish residents at birth or upon immigration [13
From The Danish Fertility Database we identified all births in Denmark from January 1st, 1997 to December 31st, 2007 and the personal id numbers of the respective mothers. We also gathered information about maternal parity, date of conception, and date of delivery from this database. The date of conception is based on ultrasound estimates and information on date of last menstrual period. Gestational age was defined as starting on the day of conception, which was defined as the first day of the last menstrual period plus 14 days. This definition of gestational age should not be confused with the “classical gestational age” ranging from the first day of the last menstrual period until birth.
The Danish National Hospital Register has since 1977 included individual level data on all discharges from Danish hospitals, and since 1994 also individual level data on all outpatient visits [11
]. From this register we acquired the pregnant mothers smoking habits according to the international classification of diseases (ICD-10, Danish revision), codes DUT00-99.
From two subsets held by Statistics Denmark (The Population's Education Register and The Income Statistics Register) we identified the mothers' highest level of completed education and their annual household income at birth year.
The Danish National Prescription Register contains information on all redeemed prescriptions from Danish pharmacies since 1995 [12
]. We identified all prescriptions redeemed by the mothers in the study period via their personal id number. To identify exposure we used the Anatomical Therapeutic Chemical (ATC) classification; P02CX01 for pyrvinium and P02CA01 for mebendazole [14
Pyrvinium is also available over-the-counter in Danish pharmacies; why an estimation of the redeemed prescription for pyrvinium could underestimate the true exposure.
The date of redemption of a prescription for mebendazole or pyrvinium was used to identify exposure, which was classified according to first, second, and third trimester. First trimester was defined as the period between day of conception and day 84 of pregnancy, second trimester between day 85 and 196 of pregnancy, and third trimester between day 197 of pregnancy, and birth.
Cases were divided into three groups: exposed to pyrvinium, exposed to mebendazole, and exposed to both pyrvinium and mebendazole.
Treatment against pinworm infections consists of two doses given two weeks apart to prevent reinfection. The second treatment may therefore extend into the following trimester.
To analyze differences in patterns of redemption before and after pregnancy, we furthermore identified women exposed to pyrvinium and mebendazole 0–3 and 6–9 months before and after pregnancy.
To assess the development of exposure to pyrvinium or mebendazole over the years we studied the incidence for each year in the study period.
Maternal characteristics (age, parity, smoking, income, and education) are presented as frequencies with percentages. Differences between categorical variables for the exposed versus unexposed were assessed by Chi-square (χ2) tests. We considered two-sided P values <0.05 to be statistically significant. To estimate if maternal characteristics were associated with being exposed to either pyrvinium or mebendazole we performed multivariable logistic regressions with odds ratios (OR) and 95% confidence intervals (CI) as results. We adjusted for age, parity, smoking, income, and education.
We classified five age groups for maternal age at conception: <20, 20–24, 25–29, 30–35, and >35 years (0% missing data). Parity was defined as the number of previous live births, and classified into four groups: 0, 1, 2, and >2 birth(s) (<1% missing data).
We classified highest level of completed education into three groups: low, medium, and high (4.0% missing data). Annual household income during the year of birth was classified into quartiles (<1% missing data). Finally, smoking was classified into smokers and nonsmokers (6.8% missing data). Statistical analyses and data management were all performed using SAS 9.2 (SAS Institute Inc., Cary, NC, USA).
Ethical approval is not required for register-based studies in Denmark. All personal information held in the registers was encrypted and analyzed on computers held by Statistics Denmark. The study has been approved by The Danish Data Protection Agency (no. 2008-41-2517).