Gastroschisis Trends and Ecologic Link to Opioid Prescription Rates — United States, 2006–2015

Tyiesha D. Short, MPH; Erin B. Stallings, MPH; Jennifer Isenburg, MSPH; Leslie A. O'Leary, PhD; Mahsa M. Yazdy, PhD; Michele K. Bohm, MPH; Mary Ethen, MPH; Xiaoli Chen, PhD; Tri Tran, MPH; Deborah J. Fox, MPH; Jane Fornoff, PhD; Nina Forestieri, MPH; Emily Ferrell, MPH; Glenda M. Ramirez, MPH; Jamie Kim, MPH; Jing Shi, MS; Sook Ja Cho, PhD; Kirstan Duckett, MPH; Norm Nelson, MS; Katherine Zielke, MPH; Kristen St. John, MPH; Brennan Martin, MPH; Carolina Clark, MD; My-Phuong Huynh, MPH; Colin Benusa, MPH; Jennita Reefhuis, PhD

Disclosures

Morbidity and Mortality Weekly Report. 2019;68(2):31-36.

Abstract and Introduction

Introduction

Prevalence of gastroschisis, a serious birth defect of the abdominal wall resulting in some of the abdominal contents extending outside the body at birth, has been increasing worldwide.^[1,2] Gastroschisis requires surgical repair after birth and is associated with digestive and feeding complications during infancy, which can affect development. Recent data from 14 U.S. states indicated an increasing prevalence of gastroschisis from 1995 to 2012.^[1] Young maternal age has been strongly associated with gastroschisis, but research suggests that risk factors such as smoking, genitourinary infections, and prescription opioid use also might be associated.^[3–5] Data from 20 population-based state surveillance programs were pooled and analyzed to assess age-specific gastroschisis prevalence during two 5-year periods, 2006–2010 and 2011–2015, and an ecologic approach was used to compare annual gastroschisis prevalence by annual opioid prescription rate categories. Gastroschisis prevalence increased only slightly (10%) from 2006–2010 to 2011–2015 (prevalence ratio = 1.1, 95% confidence interval [CI] = 1.0–1.1), with the highest prevalence among mothers aged <20 years. During 2006–2015, the prevalence of gastroschisis was 1.6 times higher in counties with high opioid prescription rates (5.1 per 10,000 live births; CI = 4.9–5.3) and 1.4 times higher where opioid prescription rates were medium (4.6 per 10,000 live births; CI = 4.4–4.8) compared with areas with low prescription rates (3.2 per 10,000 live births; CI = 3.1–3.4). Public health research is needed to understand factors contributing to the association between young maternal age and gastroschisis and assess the effect of prescription opioid use during pregnancy on this pregnancy outcome.

CDC requested annual data from U.S. population-based birth defects surveillance programs to assess the prevalence of gastroschisis during 2006–2015. The case definition for gastroschisis was based on the British Pediatric Association Classification of Diseases code (756.71), the International Classification of Diseases, Ninth Revision, Clinical Modification code (756.79 before October 1, 2009, and 756.73 thereafter because 756.79 was a shared code with omphalocele), or the International Classification of Diseases, Tenth Revision, Clinical Modification code (Q79.3 after October 1, 2015). Gastroschisis cases included all pregnancy outcomes (i.e., live births, fetal deaths, terminations, and unspecified nonlive births). The total number of live births in the same catchment area were used as denominators.

Twenty states* provided data on gastroschisis by year, maternal age group, and maternal race/ethnicity. Births from these 20 state surveillance programs accounted for approximately 47% of all U.S. births. To provide a sufficient number of subjects for each comparison category, birth years were pooled into two 5-year periods (2006–2010 and 2011–2015). For each year during 2006–2015, IQVIA Xponent^† provided CDC with county-specific opioid prescription rate categories (low = <57.2 opioid prescriptions per 100 persons per year; medium = 57.2–82.3; high = 82.4–112.5; and very high = >112.5).^[6] The IQVIA county-specific opioid prescription rates were calculated by dividing the number of opioid prescriptions in each county by the U.S. Census county-level population estimates for each year. CDC provided these county opioid prescription levels to each participating birth defects surveillance program, which used them to ascertain the total number of gastroschisis cases and total number of live births each year in the state's counties with low, medium, high, and very high opioid prescribing rates. Because gastroschisis prevalence was not found to be significantly different in areas where opioid prescribing rates were high and very high, these two categories were combined and are referred to as high for the remainder of this report. Surveillance programs aggregated gastroschisis data by year and opioid prescribing level; county-specific gastroschisis information on individual cases was not reported to CDC.

Prevalence of gastroschisis was calculated as number of gastroschisis cases (among all birth outcomes) divided by the total number of live births, and is presented as prevalence per 10,000 live births for each year and each 5-year period, by maternal age group and race/ethnicity. Exact Poisson methodology was used to calculate CIs.^[7] Statistical software was used for all analyses, including to generate prevalence ratios (PRs) for each maternal age and race/ethnicity category and overall. Linear trends in gastroschisis prevalence by maternal age from 2006 to 2015 were examined using the Cochran-Armitage test. In the ecologic analysis, PRs were calculated by dividing the prevalence of gastroschisis in areas with high and medium prescription rates by those with low rates for each calendar year and over the entire study period.

During 2006–2010, among 8,342,741 live births, 3,489 gastroschisis cases (4.2 per 10,000 live births; CI = 4.0–4.3) were reported; during 2011–2015, among 9,359,005 live births, 4,166 (4.5 per 10,000 live births; CI = 4.3–4.6) were reported (PR = 1.1, CI = 1.0–1.1) (Table). Gastroschisis prevalence was higher among infants born to non-Hispanic white mothers and Hispanic mothers than among those born to non-Hispanic black mothers in most maternal age categories (<20, 20–24, and 25–29 years). From 2006 to 2015, a linear increase in the prevalence of gastroschisis was observed in three of the four maternal age categories (Figure 1). Although gastroschisis prevalence was highest among infants born to mothers aged <20 years in each year, there was no significant linear increase.

(Enlarge Image)

Figure 1.

Trends in gastroschisis prevalence, by maternal age group — 20 states, 2006–2015*
* States contributing to the figure: Arizona, CDC/Georgia (Metropolitan Atlanta Congenital Defects Program), Illinois, Kansas, Kentucky, Louisiana (2010–2015), Massachusetts, Minnesota, Nebraska, New Jersey, New York, North Carolina, Ohio (2010–2015), Rhode Island, South Carolina (2010–2015), Tennessee (2010–2015), Texas, Utah, Vermont (2009–2015), and Virginia. Data were provided from 2006 to 2015 unless otherwise noted.

During 2006–2015, prevalences of gastroschisis in areas where opioid prescription rates were high (5.1 per 10,000 live births; CI = 4.9–5.3) and medium (4.6 per 10,000 live births; CI = 4.4–4.8) were 1.6 and 1.4 times higher, respectively, than were those in areas where opioid prescription rates were low (3.2 per 10,000 live births; CI = 3.1–3.4). PRs fluctuated over time, but stayed above 1.0 for each included study year (Figure 2). Within maternal age strata, higher gastroschisis PRs for high versus low opioid prescription rates were observed among mothers aged >25 years (<20 years: PR = 1.1, CI = 1.0–1.2; 20–24 years: 1.2, CI = 1.1–1.4; 25–29 years: 1.6, CI = 1.4–1.8; ≥30 years: 1.6, CI = 1.3–1.9).

(Enlarge Image)

Figure 2.

Trends in gastroschisis prevalence ratio, by year* and annual opioid prescription rate category^† — 20 states, 2006–2015^§
* Overall prevalence ratio for medium opioid prescription rate category and high opioid prescription rate category versus low opioid prescription rate category for each year of the study period 2006–2015.
^† Opioid prescription rate categories include medium (57.2–82.3 prescriptions per 100 persons) and high (≥82.4 prescriptions per 100 persons). The low opioid prescription rate category (<57.2 prescriptions per 100 persons) was used as the reference group.
^§ States contributing to the figure: Arizona, CDC/Georgia (Metropolitan Atlanta Congenital Defects Program), Illinois, Kansas, Kentucky, Louisiana (2010–2015), Massachusetts, Minnesota, Nebraska, New Jersey, New York, North Carolina, Ohio (2010–2015), Rhode Island, South Carolina (2010–2015), Tennessee (2010–2015), Texas, Utah, Vermont (2009–2015), and Virginia. Data were provided from 2006 to 2015 unless otherwise noted.

1 2

Next Section

Abstract and Introduction
Discussion
References
Sidebar

* The 20 states that provided data on gastroschisis and the years for which data were provided were Arizona, CDC/Georgia (Metropolitan Atlanta Congenital Defects Program), Illinois, Kansas, Kentucky, Louisiana (2010–2015), Massachusetts, Minnesota, Nebraska, New Jersey, New York, North Carolina, Ohio (2010–2015), Rhode Island, South Carolina (2010–2015), Tennessee (2010–2015), Texas, Utah, Vermont (2009–2015), and Virginia. Data were provided from 2006 to 2015 unless otherwise noted.
^† The IQVIA Xponent provides estimates of the number of opioid prescriptions dispensed in the United States based on a sample of approximately 59,000 pharmacies, which represent 88% of all prescriptions in the United States.

Table. Gastroschisis cases, gastroschisis prevalence, and prevalence ratio (PR), by maternal age group and race/ethnicity for two periods — 20 U.S. states, 2006–2015*
Maternal age group (yrs),^† race/ethnicity	2006–2010		2011–2015		2006–2015		PR^¶ (95% CI)
Maternal age group (yrs),^† race/ethnicity	No. of cases	Prevalence^§ (95% CI)	No. of cases	Prevalence^§ (95% CI)	No. of cases	Prevalence^§ (95% CI)	PR^¶ (95% CI)
<20 yrs
White, non-Hispanic	461	17.1 (15.6–18.7)	420	17.2 (15.6–18.9)	881	17.1 (16.0–18.3)	1.0 (0.9–1.1)
Black, non-Hispanic	172	9.0 (7.7–10.5)	148	9.4 (8.0–11.1)	320	9.2 (8.2–10.3)	1.0 (0.8–1.3)
Hispanic	489	14.7 (13.4–16.1)	425	17.5 (15.9–19.2)	914	15.9 (14.9–16.9)	1.2 (1.0–1.4)**
A/PI or AI/AN, non-Hispanic	48	26.0 (19.2–34.5)	36	25.6 (18.0–35.5)	84	25.8 (20.6–32.0)	1.0 (0.6–1.5)
Total ^††	1,194	14.5 (13.7–15.3)	1,055	15.7 (14.8–16.7)	2,249	15.0 (14.4–15.7)	1.1 (1.0–1.2)
20–24 yrs
White, non-Hispanic	676	7.9 (7.3–8.5)	998	10.4 (9.8–11.1)	1,674	9.2 (8.8–9.7)	1.3 (1.2–1.5) **
Black, non-Hispanic	169	4.4 (3.8–5.2)	246	5.4 (4.8–6.1)	415	5.0 (4.5–5.5)	1.2 (1.0–1.5)
Hispanic	457	7.0 (6.4–7.7)	507	8.7 (8.0–9.5)	964	7.8 (7.4–8.4)	1.2 (1.1–1.4) **
A/PI or AI/AN, non-Hispanic	52	7.6 (5.7–10.0)	56	8.2 (6.2–10.7)	108	7.9 (6.5–9.6)	1.1 (0.7–1.6)
Total ^††	1,389	7.0 (6.6–7.4)	1,859	8.9 (8.5–9.3)	3,248	8.0 (7.7–8.2)	1.3 (1.2–1.4)**
25–29 yrs
White, non-Hispanic	298	2.5 (2.2–2.8)	495	3.4 (3.1–3.7)	793	3.0 (2.8–3.2)	1.4 (1.2–1.6)**
Black, non-Hispanic	51	1.6 (1.2–2.1)	89	2.3 (1.8–2.8)	140	2.0 (1.7–2.3)	1.4 (1.0–2.0)
Hispanic	153	2.5 (2.1–2.9)	188	3.2 (2.8–3.7)	341	2.8 (2.5–3.2)	1.3 (1.0–1.6)**
A/PI or AI/AN, non-Hispanic	18	1.2 (0.7–1.9)	38	2.3 (1.6–3.1)	56	1.8 (1.3–2.3)	1.9 (1.1–3.3)**
Total ^††	536	2.3 (2.1–2.5)	828	3.1 (2.9–3.4)	1,364	2.8 (2.6–2.9)	1.3 (1.2–1.5)**
≥30 yrs
White, non-Hispanic	147	0.8 (0.7–0.9)	242	1.1 (0.9–1.2)	389	0.9 (0.9–1.0)	1.3 (1.1–1.6)**
Black, non-Hispanic	28	0.8 (0.5–1.1)	46	1.0 (0.7–1.3)	74	0.9 (0.7–1.1)	1.2 (0.8–2.0)
Hispanic	51	0.7 (0.6–1.0)	83	1.1 (0.9–1.3)	134	0.9 (0.8–1.1)	1.5 (1.0–2.1)**
A/PI or AI/AN, non-Hispanic	14	0.5 (0.3–0.9)	21	0.6 (0.4–0.9)	35	0.6 (0.4–0.8)	1.2 (0.6–2.3)
Total ^††	249	0.8 (0.7–0.9)	404	1.0 (0.9–1.1)	653	0.9 (0.8–1.0)	1.3 (1.1–1.6)**
All ages
White, non-Hispanic	1,616	3.9 (3.7–4.1)	2,161	4.4 (4.2–4.6)	3,777	4.2 (4.0–4.3)	1.1 (1.1–1.2)**
Black, non-Hispanic	432	3.5 (3.1–3.8)	532	3.6 (3.3–3.9)	964	3.5 (3.3–3.8)	1.0 (0.9–1.2)
Hispanic	1,183	5.2 (4.9–5.5)	1,207	5.5 (5.2–5.9)	2,390	5.4 (5.1–5.6)	1.1 (1.0–1.2)**
A/PI or AI/AN, non-Hispanic	132	2.6 (2.2–3.1)	152	2.6 (2.2–3.0)	284	2.6 (2.3–2.9)	1.0 (0.8–1.2)
Total ^††	3,489	4.2 (4.0–4.3)	4,166	4.5 (4.3–4.6)	7,655	4.3 (4.2–4.4)	1.1 (1.0–1.1)**

Abbreviations: A/PI = Asian/Pacific Islander; AI/AN = American Indian/Alaska Native; CI = confidence interval.
* States contributing to the table: Arizona, CDC/Georgia (Metropolitan Atlanta Congenital Defects Program), Illinois, Kansas, Kentucky, Louisiana (2010–2015), Massachusetts, Minnesota, Nebraska, New Jersey, New York, North Carolina, Ohio (2010–2015), Rhode Island, South Carolina (2010–2015), Tennessee (2010–2015), Texas, Utah, Vermont (2009–2015), and Virginia. Data were provided from 2006 to 2015 unless otherwise noted.
^† Cases missing information on maternal age are not included in this table.
^§ Prevalence per 10,000 live births.
^¶ Unrounded prevalence for 2011–2015 divided by the unrounded prevalence for 2006–2010.
** Denotes a statistically significant confidence interval.
^†† Total includes non-Hispanic white, non-Hispanic black, Hispanic, non-Hispanic A/PI, non-Hispanic AI/AN, and other/unknown maternal race/ethnicity.