Growing evidence suggests that the human microbiome plays an important role in maternal and child health, yet many microbiome studies can be methodologically challenging and vulnerable to confounding and bias, particularly when assessing medication-related effects. Prescription drugs are among the strongest modifiable factors associated with microbiome composition at the population level, but little is known about the long-term health consequences of microbiome-modulating drugs when used before and during pregnancy, beyond the traditional focus on toxicity and teratogenicity. This project therefore aims to quantify associations between microbiome-modulating prescription drug exposure and adverse outcomes in the mother, the newborn, and the child. The project is designed as a population-based, register-linked cohort study including all live births and stillbirths in Sweden from July 1, 2006, onwards. Pregnancies are identified via the Swedish Medical Birth Registry, while drug exposures are captured through the Swedish Prescribed Drug Register (nationwide coverage since July 2005), enabling complete ascertainment of dispensed prescriptions during pregnancy and the three months before conception. Linkage across registries is possible using the Swedish personal identity number3. Additional outcomes and covariate information are obtained from the Patient Register (in- and outpatient specialist care) and the Causes of Death Register. Drug exposure is defined using dispensed prescriptions classified by ATC codes and evaluated across key windows: pre-pregnancy, each trimester, and a 90-day postpartum (lactation) period. The project prioritizes drug classes with documented or suspected microbiome effects and/or common use during pregnancy, including systemic antibacterials (J01), gynecological anti-infectives (G01), gastric-acid inhibitors (A02B: proton pump inhibitors (PPIs) and H2-receptor antagonists), Helicobacter pylori eradication regimens (defined by combined PPI and multi-antibiotic classes), diabetes drugs (including metformin), and selected nervous system drugs (antiepileptics, psycholeptics including benzodiazepines, and antidepressants). Outcomes are grouped into maternal, neonatal, and early-childhood endpoints. Maternal outcomes include pre-eclampsia, gestational diabetes, excessive gestational weight gain, and antenatal or postnatal depression. Neonatal outcomes include preterm birth, stillbirth, and small/large for gestational age. Childhood outcomes focus on early-onset disease (before age 10), including type 1 diabetes, allergic disease (asthma/eczema), neurodevelopmental conditions (e.g., autism/ADHD), gastrointestinal disorders, obesity and childhood infections. Analyses will compare exposed and unexposed pregnancies using multivariable regression (primarily logistic regression), with time-window-specific and combined exposure definitions. Where appropriate, particularly for long-term childhood outcomes with variable follow-up, survival models (e.g., Cox regression) may be used. Confounding control includes maternal age, parity, comorbidities, socioeconomic factors, smoking/snus, body mass index, pregnancy characteristics, co-medication use, and proxies for confounding by indication via diagnoses from the Medical Birth and Patient registers. Notably, leveraging nationwide Swedish registry linkages provides exceptional population coverage (>1 million deliveries) and statistical power, but the reliance on dispensed prescriptions and routinely collected clinical data may limit granularity for some exposures, indications, and rare outcomes.