Ascertainment of medication use
In the first follow-up questionnaire at gestational week 17, participants were asked to complete a 6-weeks paper-based medication diary during gestational weeks 19-24 or 26-31 [22]. Through these diaries, participants reported on the exact name of the medication used, daily dose, and specific dates of use. To decrease participant burden, participants could write down the name of any medication used on a daily basis on the first page of the diary. A reminder was sent if the diary was not returned within two weeks after the last day covered by the diary.
In the Netherlands, most people visit only one or two pharmacies and all pharmacies use computerized dispensing records. Through the ‘Landelijk Schakelpunt’, to which 96% of Dutch pharmacies are connected [27], information on medications dispensed is exchanged between pharmacies, making their dispensing records virtually complete. For PRIDE Study participants who provided consent, data on medications dispensed in the three months before pregnancy until six months after the estimated date of delivery were requested from the pharmacies the women reported visits to. From these records, we abstracted data on medications with use in the 6-weeks covered by the medication diary, including name of the medication, start and stop dates, frequency of use, and dosage prescribed.
Medications were coded using the Anatomical Therapeutic Chemical (ATC) classification system [28]. We excluded vaccines (ATC code J06 and J07) and vitamins (A11, A12, B03B), as these may not be considered as medication by the participants. Furthermore, medications that are predominantly used over-the counter (acetaminophen [N02BE01], antacids [A02AD01], and xylometazoline [R01AA07]) and medications not indicated for use during pregnancy (contraceptives [G03A], fertility medication [G03CA and G03GA], and tranexamic acid [B02AA02]) were excluded. We classified the medications in three mutually exclusive categories after the classification of Bakker et al. [29]: (1) medication for chronic conditions, (2) medications for occasional and short-time use, and (3) pregnancy-related medication (Supplemental Table 1). Medications for chronic conditions did not necessarily have to be taken on a regular or daily basis, but may have been used on an as-needed basis as well.
Medication adherence
Adherence to prescribed medications was assessed with three measures. Actual use was defined as a dispensed medication of which use was reported in the diary. The definitions of initiation and implementation as described by the ABC European adherence initiative taxonomy were used [30]. For medications first dispensed in the diary period (i.e. no use in the three months preceding the diary period according to the pharmacy records), we assessed non-initiation and initiation time (difference between dispensing date and date of first reported use in the diary). Implementation adherence was calculated as the proportion of days with at least the correct number of doses taken, with the pharmacy records as reference standard. Medications that were prescribed on an as-needed basis were excluded from the analyses on initiation and implementation.