Pregnant Women With IBD Are More Likely to be Adherent to Biologic Therapies Than Other Medications

Sangmin Lee; Cynthia H. Seow; Kamala Adhikari; Amy Metcalfe


Aliment Pharmacol Ther. 2020;51(5):544-552. 

In This Article

Materials and Methods

This retrospective population-based observational cohort study is reported in accordance with "the REporting of studies Conducted using Observational Routinely collected health Data (RECORD) Statement".[16] Five de-identified routinely-collected health administrative databases maintained by Alberta Health Services and Alberta Health were used: (a) the Alberta Perinatal Health Program (2012–2015) database contains comprehensive perinatal information on all in-hospital and out-of-hospital births, (b) the Discharge Abstract Database (2011–2016) contains information on hospitalisations, (c) the National Ambulatory Care Reporting System (2011–2016) database contains information on emergency department visits, day surgeries and other ambulatory care services, (d) the Physician Claims (2011–2016) database contains information on out-patient fee-for-service billings, and (e) the Pharmaceutical Information Network (2011–2016) database contains information on prescription medications dispensed from community pharmacies only; this does not include any medications that were dispensed in-hospital. Both the Discharge Abstract Database and National Ambulatory Care Reporting System database utilise the International Classification of Diseases 10th Revision for Canada (ICD-10-CA) and the Canadian Classification of Health Intervention to code for diagnoses and procedures respectively. The Physician Claims database uses the International Classification of Diseases 9th Revision Clinical Modification (ICD-9-CM) and the Canadian Classification of Diagnostic, Therapeutic, and Surgical Procedures to code for diagnoses and procedures respectively. Deterministic linkage using the maternal personal health number and unique birth event identification number were used to link the records for mothers and their infants.

Women who were pregnant and delivered a live or stillborn infant ≥20 weeks of gestation between 2012 and 2015 in Alberta, Canada were identified from the Alberta Perinatal Health Program database. Only the first pregnancy was included if a woman had multiple pregnancies between 2012 and 2015. Amongst this population, women were classified as having IBD prior to pregnancy if within a 2-year period during any time before pregnancy they had ≥2 hospitalisation records, ≥2 emergency department visits or ≥4 physician claims with the following ICD-9-CM (CD: 555; UC: 556) or ICD-10-CA (CD: K500, K501, K508, K509; UC: K510, K512, K514, K515, K518, K519) codes.[17] Women were eligible for the study if they had ≥2 consecutive prescriptions of a specific class of IBD maintenance medication (ie, prescription re-filled within 30 days from their estimated last day of medication supply), indicating the physicians' intent for ongoing therapy.

The Pharmaceutical Information Network database uses the Anatomical Therapeutic Chemical Classification System based on the active ingredient of drugs according to the targeted organ or system.[18] Relevant maintenance IBD medications were included using the Pharmaceutical Information Network database; 5-aminosalicylates, antibiotics, biologic therapy, methotrexate, corticosteroids and thiopurines (Table S1). Women who were exclusively prescribed corticosteroids or antibiotic monotherapy during the year prior to pregnancy were excluded from further analyses because these medications are not routinely prescribed for maintenance of disease remission for IBD. Furthermore, women prescribed methotrexate during the year prior to pregnancy were also excluded, as methotrexate is contraindicated in women trying to conceive.[6] Both monotherapy and polytherapy were considered for analyses; however, for women on polytherapy, the analyses were simplified by converting polytherapy to "representative monotherapy" use by only including the most "intensive" therapy (ie, biologics > thiopurines>5-aminosalicylates). Medication prescriptions for rectal 5-aminosalicylates were also excluded.

Medication adherence was assessed by the prescription-based medication possession ratio, which is the proportion of medication supply dispensed assuming that the last prescription was not filled within the first and last dispensed date.[19,20] Adherence to medication was defined by a medication possession ratio ≥80% and non-adherence by a medication possession ratio <80%.[19,20] The pregnancy cohort for analysis only included women who were adherent to their medication in the year prior to pregnancy (n = 159) (Figure 1).

Figure 1.

Flow diagram of the selection of the study population. DAD, discharge abstract database; NACRS, National Ambulatory Care Reporting System; PC, physician claim; PHN, personal health number

The adverse pregnancy outcome of preterm birth (delivery <37 weeks) was recorded. Outcomes for health care utilisation after delivery included admission into the neonatal intensive care unit and neonatal extended length of stay in hospitals (vaginal delivery: >3 days; Caesarean section: >5 days).

Chi-squared tests were conducted to assess whether medication adherence pattern during the year prior to pregnancy and during pregnancy for women with IBD differed by drug class. Medication adherence pattern and drug class was also compared using chi-square tests for each trimester. Bonferroni correction was used for multiple hypothesis testing. Furthermore, the risk of adverse neonatal outcomes associated with medication adherence pattern for women with IBD compared to the women in the general obstetric population without IBD were assessed. E-value analyses was conducted to assess the potential impact unmeasured confounding may have on the observed associations.[21,22] Statistical significance was determined by α < 0.05. List-wise deletion was used to address missing data. All statistical analyses were conducted in STATA SE 14.0 (StataCorp LP).

Ethical Considerations

Ethics approval was obtained from the Conjoint Health Research Ethics Board at the University of Calgary, Alberta, Canada (REB17–0430).