Database

The data used in this study came from 2 nationwide population-based data sets. The first was the Taiwan National Health Insurance Research Dataset (NHIRD), published by Taiwan National Health Research Institute. Taiwan began its National Health Insurance (NHI) program in 1995. Currently more than 22 million citizens, representing more than 98% of the Taiwanese population, are enrolled in this program. The NHIRD comprises all medical claims for enrollees covered under the program as well as registries of contracted medical facilities, board-certified physicians, and beneficiaries. Therefore, the NHIRD, 1 of the largest nationwide population-based data sets available, provides a unique opportunity to examine the risk of adverse pregnancy outcomes among women with herpangina.

The second data set was Taiwan’s birth certificate registry. The government in Taiwan mandates all births be registered, so the birth certificate data are accurate and comprehensive. The completeness and validity of Taiwan’s birth registry has been verified. Taiwan’s birth certificate registry includes data consisting of birth dates for both infants and their parents, gestational week at birth, infant birthweight, sex, parity, place of birth, parental educational levels, and maternal marital status.

These 2 data sets were linked by the mother’s and infant’s unique personal identification numbers with assistance from the Bureau of the NHI. All personal identifiers were encrypted by the Bureau of NHI before release to the researchers. Because the NHIRD consists of deidentified secondary data released to the public for research purposes, this study was exempt from full review by the internal review board.

Study sample

The study features a study cohort and a comparison cohort. A total of 242 pregnant women who had singleton live births between 2001 and 2003 in Taiwan (there were 473,529 pregnant women who had singleton live births during the study period) were identified as having ever visited ambulatory care centers for the treatment of herpangina (any International Classification of Diseases, Ninth Revision, Clinical Modification code 0740) during their pregnancies. If a mother had more than 1 singleton birth between 2001 and 2003, we included only the first 1 for the study sample, regardless of herpangina infection status.

The comparison group was extracted from the remaining 471,603 mothers. We randomly selected 1936 mothers (8 for every mother with herpangina, in accordance with principles proposed by Rothman and Hennessy et al ) and used the SAS statistical package (SAS Institute, Cary, NC) to match the study group in terms of age (<20, 20-24, 25-29, 30-34 and ≥35 years old) and the year of delivery. Ultimately, 2718 women were included in our study sample.

Variables of interest

The outcome variables of this study were all dichotomous and included LBW (<2500 vs ≥2500 g), preterm delivery (<37 weeks vs ≥37 weeks), and SGA (birthweight below the 10th percentile for gestational age). The independent variable of interest was whether a mother had been diagnosed with herpangina during the index pregnancy.

In this study, we also adjusted for the characteristics related to adverse pregnancy outcomes reported by previous studies in the regression modeling. These included factors relating to the mother (age, the highest educational level [elementary school or lower, junior high school, senior high school, college or above] and marital status ), the infant (sex and parity ), and monthly family income.

Monthly income was grouped into the following 4 categories: less than New Taiwan (NT) $15,000, NT$15,000-NT30,000, NT$30,001-NT50,000, NT$50,001 or greater (US$1.00 = NT$33.00 in 2003). In addition, comorbid chronic medical conditions that may increase the risk of adverse pregnancy outcomes, such as hypertension, diabetes, anemia, coronary heart disease (CHD), and hyperlipidemia, were also adjusted for in the regression modeling.

Statistical analysis

In this study, we used the SAS statistical package (version 8.2) to perform all analyses. The χ ² tests were conducted to investigate the differences in the characteristics of mother and infant between mothers with herpangina and unaffected mothers. Conditional logistic regression analyses, which were conditioned on maternal age and the year of delivery, were performed to examine the risk of LBW, preterm birth, and SGA for mothers with herpangina and unaffected mothers, after adjusting for characteristics of the mother (age, the highest educational level, and marital status), infant (sex and parity) and monthly family income and diabetes, hypertension, anemia, CHD, and hyperlipidemia. A 2-sided P < .05 was considered statistically significant.