Congenital Malformations: Agreement Between Diagnostic Codes in an Administrative Database and Mothers’ Reports

doi:10.1016/S1701-2163(16)34523-6

Journal of Obstetrics and Gynaecology Canada

Volume 32, Issue 6, June 2010, Pages 549-554

https://doi.org/10.1016/S1701-2163(16)34523-6 Get rights and content

Abstract

Objective

The validity of epidemiologic research on congenital malformations conducted using administrative databases relies heavily on the quality of diagnostic information. The goal of this study was to validate the diagnostic codes for major congenital malformations (MCMs) recorded in a medical service claims database against mother’s report obtained by questionnaire.

Methods

Using the Quebec Pregnancy Registry, we mailed a questionnaire to women who gave birth to a baby with an MCM and to a random sample of mothers whose infants did not have any malformation. We reviewed each infant’s database record for a diagnostic code for MCM, which we then compared with the corresponding mother’s report.

Results

Among the 3142 mother–infant pairs there was 60% agreement regarding the presence or absence of an MCM; for 456 babies, both sources reported the presence of an MCM. For 90% of these infants the two sources of information agreed with respect to the type of malformation.

Conclusion

Overall, mothers reported fewer infants with MCM than were recorded in the database; among those for whom both sources of information indicated the presence of an MCM, a very high proportion of diagnostic codes accorded with the mother’s description of the specific type of malformation.

Résumé

Objectif

La validité de la recherche épidémiologique sur les malformations congénitales menée au moyen de bases de données administratives repose grandement sur la qualité des renseignements diagnostiques. L’objectif de cette étude était de valider, en fonction du rapport de la mère obtenu par questionnaire, les codes diagnostiques des malformations congénitales majeures (MCM) consignés dans une base de données des réclamations de service médical.

Méthodes

Au moyen du Registre des grossesses du Québec, nous avons posté un questionnaire aux femmes qui ont donné naissance à un enfant présentant une MCM, ainsi qu’à un échantillon aléatoire de mères dont les enfants ne présentaient aucune malformation. Nous avons analysé le dossier de base de données de chacun de ces enfants afin d’y repérer un code diagnostique correspondant à la MCM, que nous avons alors comparé au rapport de la mère correspondant.

Résultats

Parmi les 3 142 paires mère-enfant, nous avons constaté une correspondance de 60 % à l’égard de la présence ou de l’absence d’une MCM; dans le cas de 456 enfants, les deux sources ont signalé la présence d’une MCM. Chez 90 % de ces enfants, les deux sources de renseignements indiquaient le même type de malformation.

Conclusion

De façon globale, les mères ont signalé moins d’enfants présentant une MCM que ce qui a été consigné dans la base de données; chez celles pour lesquelles les deux sources de renseignements indiquaient la présence d’une MCM, les codes diagnostiques correspondaient à la description du type particulier de malformation par la mère dans une proportion très élevée.

REFERENCES (13)

M. Wilchesky et al.
Validation of diagnostic codes within medical services claims
J Clin Epidemiol
(2004)
A. Bérard et al.
First trimester exposure to paroxetine and risk of cardiac malformations in infants: the importance of dosage
Birth Defects Res B Dev Reprod Toxicol
(2007)
B. Ofori et al.
Risk of congenital anomalies in pregnant users of statin drugs
Br J Clin Pharmacol
(2007)
E. Ramos et al.
Duration of antidepressant use during pregnancy and risk of major congenital malformations
Br J Psychiatry
(2008)
A. Vilain et al.
Agreement between administrative databases and medical charts for pregnancy-related variables among asthmatic women
Pharmacoepidemiol Drug Saf
(2008)
S. Kulaga et al.
Prescriptions filled during pregnancy for drugs with the potential of fetal harm
BJOG
(2009)

There are more references available in the full text version of this article.

Cited by (14)

New evidence for concern over the risk of birth defects from medications for nausea and vomitting of pregnancy
2019, Journal of Clinical Epidemiology
Citation Excerpt :
MCMs diagnosed in the first 6 months of life were identified in the RAMQ and MedEcho databases and defined according to ICD-9 and ICD-10 codes (Appendix Table S2). The use of ICD-9 and ICD-10 codes for the identification of MCMs in the QPC has been validated against medical patient charts (PPV of at least 80% and NPV of 93%) [23,24]. All organ systems were considered as well as overall MCMs.
The aim of the study was to quantify the risk of major congenital malformations (MCM) associated with first-trimester exposure to antiemetics.
Using the Quebec Pregnancy Cohort (1998–2015), first-trimester doxylamine–pyridoxine, metoclopramide, and ondansetron exposures were assessed for their association with MCM. Generalized estimating equations were used to estimate odds ratios (OR), adjusting for potential confounders (aOR).
Within 17 years of follow-up, the prevalence of antiemetic use during pregnancy increased by 76%. Within our cohort, 45,623 pregnancies were exposed to doxylamine–pyridoxine, 958 to metoclopramide, and 31 to ondansetron. Doxylamine–pyridoxine and metoclopramide use were associated with an increased risk of overall MCM (aOR 1.07, 95% confidence interval [CI]: 1.03–1.11; 3,945 exposed cases) and (aOR 1.27, 95% CI: 1.03–1.57; 105 exposed cases), respectively. Doxylamine–pyridoxine exposure was associated with increased risks of spina bifida (aOR 1.87, 95% CI: 1.11–3.14; 23 exposed cases), nervous system (aOR 1.25, 95% CI: 1.06–1.47; 225 exposed cases), and musculoskeletal system defects (aOR 1.08, 95% CI: 1.02–1.14; 1,735 exposed cases). Metoclopramide exposure was associated with an increased risk of genital organ defects (aOR 2.26, 95% CI: 1.14–4.48; 10 exposed cases). No statistically significant association was found between ondansetron exposure and the risk of overall MCM.
First-trimester doxylamine–pyridoxine and metoclopramide exposure was associated with a significantly increased risk of overall and specific MCM.
Low-to-moderate prenatal alcohol consumption and the risk of selected birth outcomes: A prospective cohort study
2015, Annals of Epidemiology
Citation Excerpt :
Birth length and head circumference were analyzed as continuous outcomes using the lowest 10th percentile according to Centers for Disease Control and Prevention standards [28] to define low birth length and reduced head circumference. Additional clinical outcomes included major selected congenital malformations [29]; Apgar score less than 7 at 1 and 5 minutes; ventilation (including need for continuous positive airway pressure); placement in the neonatal intensive care unit (NICU; observation or admission); and neonatal jaundice. Demographic covariates included maternal age, ethnicity, marital status, education, parity, and employment.
To estimate whether low-to-moderate prenatal alcohol exposure is associated with selected birth outcomes.
Low-to-moderate prenatal alcohol drinking and effects on low birthweight, preterm delivery, intrauterine growth restriction, and selected neonatal outcomes were evaluated among 4496 women and singleton infants. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multivariable logistic regression, controlling for confounding variables.
Early pregnancy drinking was associated with reduced odds of low birthweight, OR, 0.66 (95% CI, 0.46–0.96) and birth length less than 10th percentile, OR, 0.74 (95% CI, 0.56–0.97). Drinking during the first 3 months showed lower odds for birth length and head circumference less than 10th percentile, OR, 0.56 (95% CI, 0.36–0.87) and OR, 0.69 (95% CI, 0.50–0.96), respectively. Third trimester drinking was associated with lower odds for low birthweight, OR, 0.56 (95% CI, 0.34–0.94) and preterm delivery, OR, 0.60 (95% CI, 0.42–0.87).
Our results suggest low-to-moderate alcohol exposure during early and late gestation is not associated with increased risk of low birthweight, preterm delivery, intrauterine growth restriction, and most selected perinatal outcomes.
Quebec pregnancy cohort: Prevalence of medication use during gestation and pregnancy outcomes
2014, Therapie
L’exposition aux médicaments durant la grossesse est fréquente. La cohorte des grossesses du Québec (CGQ) a été créée pour étudier l’effet des médicaments pris pendant la grossesse.
Quatre bases de données administratives du Québec, Canada ont été jumelées – médicales, pharmaceutiques, hospitalières, démographiques, scolaires. Les grossesses incluses étaient couvertes par le régime d’assurance médicaments du Québec 12 mois avant et jusqu’à la fin de la grossesse (36 % des femmes de 15 – 45 ans).
L’analyse inclut 97 680 grossesses. La prévalence d’utilisation des médicaments était 74 % avant, 56 % pendant et 80 % suivant la grossesse. Les médicaments les plus utilisés durant la grossesse étaient : antibiotiques (47 %), antiémétiques (23 %) et anti-inflammatoires non stéroïdiens (AINS) [17 %]. Les utilisatrices étaient plus susceptibles d’avoir une fausse-couche, une naissance prématurée, un enfant avec une malformation congénitale ou une dépression post-partum que les non-utilisatrices (p < 0,01).
La prise de médicaments demeure élevée en grossesse. La CGQ est un excellent outil pour la recherche pharmaco-épidémiologique périnatale.
Many women are exposed to medications during pregnancy. The Quebec Pregnancy Cohort (QPC) is a prospective population-based cohort which includes all data on pregnancies and children between January 1997 and December 2008.
We linked four administrative databases in Quebec, Canada: RAMQ (medical and pharmaceutical), MEDECHO (hospitalizations), ISQ (births/deaths), and MELS (Ministry of Education). Pregnancies included were covered by the Quebec prescription drug insurance plan (36% of women aged 15–45 years) from 12 months prior until the end of pregnancy.
We analyzed 97,680 pregnancies. Prevalence of medication use was 74% pre-pregnancy, 56% during pregnancy, and 80% post-pregnancy. Most frequently used medications during pregnancy were antibiotics (47%), antiemetic drugs (23%), and non-steroïdal anti-inflammatory drugs (NSAIDs) [17%]. Medication users were more likely to have spontaneous abortions, preterm births, children with congenital malformations and postpartum depression than non-users (p < 0.01).
Medications are commonly used during pregnancy. The QPC is a powerful tool for perinatal pharmacoepidemiological research.
Antiepileptic drug use during pregnancy: Perinatal outcomes
2011, Seizure
Citation Excerpt :
Gestational age, previously validated,26 was obtained from hospital charts, which enabled us to calculate exact timing of AED exposure during pregnancy. Finally, diagnoses of major congenital malformations have previously been validated.36 Limitations of this study include the use of exposure data based on prescription fillings, which might not necessarily reflect actual intake.
This study was undertaken to (1) measure the frequency of AED monotherapy or polytherapy during pregnancy and AED discontinuation prior to pregnancy in a cohort of women with treated epilepsy; and (2) describe the frequency of major congenital malformations according to maternal use of AED during pregnancy.
A cohort of epileptic pregnant women was identified within the Quebec Pregnancy Registry and was divided into three groups based on maternal AED use during pregnancy: AED monotherapy, AED polytherapy and no AED use.
Of the 349 pregnancies meeting eligibility criteria, 79.6% were exposed to AED monotherapy and 5.8% to polytherapy during pregnancy; 14.6% discontinued AED prior to pregnancy. The most commonly used AEDs were carbamazepine (29.9%) and valproic acid (19.7%); the most common AED polytherapy combination was carbamazepine combined with clobazam (2.5%). Of 111 deliveries in the group of women on monotherapy during pregnancy, 9.9% (n = 11) were born with major congenital malformations; in the group of women treated with polytherapy, 19.0% (n = 8 over 42) of babies had major congenital malformations compared to 20.0% in women who discontinued AEDs prior to pregnancy.
This study demonstrates that the majority of women suffering from epilepsy were treated with monotherapy rather than polytherapy during pregnancy. While most used other agents, an important number of women continued to use valproate in pregnancy despite the long standing evidence of its teratogenicity and increasing evidence of its neuro-toxicity to the fetus.
Induced abortions in women with systemic lupus erythematosus
2021, Lupus
Can we rely on pharmacy claims databases to ascertain maternal use of medications during pregnancy?
2017, Birth Defects Research

View all citing articles on Scopus

: Competing Interests: None declared

View full text

OBSTETRICSCongenital Malformations: Agreement Between Diagnostic Codes in an Administrative Database and Mothers’ Reports

Abstract

Objective

Methods

Results

Conclusion

Résumé

Objectif

Méthodes

Résultats

Conclusion

J Clin Epidemiol

First trimester exposure to paroxetine and risk of cardiac malformations in infants: the importance of dosage

Birth Defects Res B Dev Reprod Toxicol

Risk of congenital anomalies in pregnant users of statin drugs

Br J Clin Pharmacol

Duration of antidepressant use during pregnancy and risk of major congenital malformations

Br J Psychiatry

Agreement between administrative databases and medical charts for pregnancy-related variables among asthmatic women

Pharmacoepidemiol Drug Saf

Prescriptions filled during pregnancy for drugs with the potential of fetal harm

BJOG

OBSTETRICS
Congenital Malformations: Agreement Between Diagnostic Codes in an Administrative Database and Mothers’ Reports