exploring the seasonality of birth defects in the new york

Exploring the Seasonality of Birth Defects in theNew York State Congenital Malformations Registry

Alissa R. Caton*

University at Albany, Department of Epidemiology and Biostatistics, Rensselaer, New York

Received 19 December 2011; Revised 8 February 2012; Accepted 10 February 2012

BACKGROUND: Examining seasonal patterns of birth defects may help to identify environmental risk fac-tors. Because the teratogenic window for most birth defects is during gestational weeks 3 to 8, investigatingexposures closer to the timing of conception is important. However, studies are usually based on month ofbirth, which is not the biologically relevant exposure period and does not account for differences in gesta-tional length. We aimed to determine whether the occurrence of birth defects varied by month of conceptionusing the population-based New York State Congenital Malformations Registry (CMR). METHODS: Wemerged live birth certificates (n 5 2,044,091) with CMR records for mothers residing in New York State,excluding New York City, for the years 1992 through 2006. We categorized birth defects according to theNational Birth Defects Prevention Network guidelines and performed Cochran-Armitage trend, Hewitt-Rog-erson, and Walter-Elwood tests on month of conception and chi-square tests on season of conception. Wegraphed seasonal distributions and seasonality test results. We performed stratified analyses by maternaland infant characteristics. RESULTS: Of 42 groups examined in the 15-year period, 24 (57%) had at least onestatistically significant test result, suggesting a trend or seasonal variation: Cochran-Armitage (18), Hewitt-Rogerson (17), Walter-Elwood (4), and chi-square (5). Ventricular septal defect showed the most consistentresults: Cochran-Armitage (p 5 0.0006), Hewitt-Rogerson (December to May; p 5 0.0130), Walter-Elwood(March 14; p 5 0.0027), and chi-square (winter; p 5 0.0046). Congenital cataract, pulmonary valve atresia/stenosis, coarctation of aorta, biliary atresia, and renal agenesis or hypoplasia had at least three significanttests. DISCUSSION: These results may help to generate hypotheses about environmental factors that vary byseason for further studies. Birth Defects Research (Part A) 00:000–000, 2012. � 2012 Wiley Periodicals, Inc.

Key words: birth defects; epidemiology; registry; seasonality; surveillance

INTRODUCTION

Although major birth defects occur in approximately3% of births (Centers for Disease Control and Prevention,2008), the causes of most birth defects are unknown(Nelson and Holmes, 1989). Examining seasonal patternsusing routinely collected birth defects surveillance datamay help us generate hypothesis about environmentalrisk factors that can be tested in analytic studies. Resultsof seasonality studies in the literature have been inconsis-tent by geography and time period and are often basedon month of birth, which is not the biologically relevantexposure period and does not account for differences inlength of gestation (Bound et al., 1989; Castilla et al.,1990; Siffel et al., 2005). For most birth defects, the criticalperiod of exposure is in the 3rd to 8th weeks after con-ception during organogenesis (Sadler, 2004); therefore, itis important to examine environmental factors closer tothe timing of conception and early pregnancy. Our objec-

tive was to determine whether the occurrence of birthdefects in upstate New York varied by month of concep-tion by using the population-based New York State Con-genital Malformations Registry (CMR).

MATERIALS AND METHODS

Appropriate institutional review board approvals weregranted to access New York State birth certificate datafrom vital records and birth defects data from the CMR.We linked live birth certificates (n 5 2,044,091) withCMR records for mothers residing in upstate New York(New York State, excluding New York City) at birth for

*Correspondence to: Alissa R. Caton, Department of Epidemiology and Bio-statistics, University at Albany, School of Public Health, One University Place,Room 131, Rensselaer, NY 12144-3456. E-mail: [email protected] online in Wiley Online Library (wileyonlinelibrary. com).DOI: 10.1002/bdra.23006

Birth Defects Research (Part A): Clinical and Molecular Teratology 00:000�000 (2012)

� 2012 Wiley Periodicals, Inc. Birth Defects Research (Part A) 00:000�000 (2012)

the years 1992 to 2006 to construct a 15-year live birthcohort. The CMR is a population-based registry thatreceives mandated reports on children who were born inNew York State and were diagnosed with birth defects,metabolic defects, or chromosomal anomalies up until 2years of age from hospitals and physicians. Hospitalaudits are conducted to capture the unreported cases.On-site hospital medical record audits documented thatthe CMR reports were greater than 90% correct (Wanget al., 2010), which is comparable to that of the Metro-politan Atlanta Congenital Defects Program, an activesurveillance system that is regarded as the gold standard(Honein and Paulozzi, 1999). The birth certificate containsdata on maternal and infant characteristics, such asmaternal age, race or ethnicity, education level, date oflast menses, infant date of birth, sex, birth weight, andgestational age. Roohan et al. (2003) assessed the accu-racy of the date of last menses recorded on the NewYork State birth certificate by checking medical records;they found the date to be correct 87% of the time and93% accurate within 1 week of the actual date.

Using the International Classification of Disease ClinicalModification, Ninth Revision, diagnosis codes from theCMR records, we grouped birth defects into 45 categoriesas defined in the guidelines of the National Birth DefectsPrevention Network (NBDPN; National Birth DefectsPrevention Network, 2010). We excluded three casegroups with fewer than 50 cases from our analyses: aniri-dia (n 5 9), common truncus (n 5 44), and bladderexstrophy (n 5 23). The remaining 42 case groupsselected for analyses are listed in Table 1.

We estimated date of conception by adding 14 days tothe maternal date of last menses. Based on this referencepoint, we categorized each birth into 12 months and fourseasons of conception: winter (December–February),spring (March–May), summer (June–August), and fall(September–November). Because there is a seasonal vari-ation in live births, it is important to account for that var-iation in analyses of the distribution of adverse birth out-comes (Darrow et al., 2009). To visualize the distributionsacross seasons, we plotted birth defects prevalence ratesper 10,000 live births (No. of birth defects 3 10,000 7No. of live births) by month and season of conception.We performed four statistical tests to examine trend andseasonal variation of the birth defects groups based ontiming of conception. According to the statistical methodsrecommended for analyzing the seasonal variation ofbirth defects proposed and described in detail by Siffelet al. (2005), we conducted the Cochran-Armitage trendtest, the Hewitt-Rogerson test, and the Walter-Elwoodtest using month of conception. Each test accounts forthe seasonal variation of live births. The Cochran-Armit-age test is used to detect an underlying trend in datawith binomial proportions (Cochran, 1954; Armitage,1955). The Hewitt-Rogerson test is a nonparametric testfor seasonality that ranks monthly prevalence rates fromlowest to highest, sums all possible sequences of six con-secutive months, and identifies the 6-month period withthe maximum rank sum (i.e., the Hewitt Score; Hewittet al., 1971; Rogerson, 1996). The test requires at least 6months with nonzero frequencies. The Walter-Elwoodtest is a parametric test for seasonality that uses a simpleharmonic curve and allows for a variable population atrisk (Walter and Elwood, 1975). The test estimates theamplitude of the seasonal variation, the date at which the

maximum occurs, and the goodness of fit. The testrequires at least 50 cases. Because New York has fourdistinct seasons, we also performed a chi-square testusing season of conception. The chi-square test is used totest for the relationship between two discrete variablesby comparing observed versus expected values (Cochran,1954). It is recommended that no more than one fifth ofcells have expected values less than 5. To perform theHewitt-Rogerson and Walter-Elwood tests, we used theStatistical Analysis Battery for Epidemiologic Research(SABER) version 1.96, developed by LM James anddownloaded from the Centers for Disease Control andPrevention (Atlanta, GA; http://www.cdc.gov/ncbddd/birthdefects/research-tools.html). Other analyses wereperformed in SAS version 9.1 (SAS Institute, Cary, NC).Because seasonal patterns of births varied by maternal

sociodemographics in a study in metropolitan Atlanta(Darrow et al., 2009), and because some studies havereported sex differences in seasonality of birth defects(Heikkila, 1984; Puri and Singh, 1995; Fraser and Gwyn,1998; Krost and Schubert, 2006; Liu et al., 2011), we alsoexamined seasonality within strata of maternal age (<20,20–34, �35 years), race or ethnicity (non-Hispanic white,non-Hispanic black, Hispanic), education level (<12, 12–15, �16 years), and infant sex (male, female). The Walter-Elwood test was limited to case groups with at least 50subjects, and the Hewitt-Rogerson and chi-square testswere limited to case groups with at least 20 subjects.

RESULTS

Of the 2,044,091 live births within the 1992 to 2006study period, 1,967,654 (96.3%) infants with dates of lastmenstrual period were included in the analyses. Livebirth prevalence rates per 10,000 live births and results ofthe four statistical tests for the selected birth defectsgroups are displayed in Table 1. Twenty-four (57%) ofthe 42 groups examined had at least one statistically sig-nificant test result, suggesting a trend or seasonal varia-tion. The Cochran-Armitage trend test was significantand suggested heterogeneity of monthly distributions fornine groups of congenital heart defects (i.e., tetralogy ofFallot, ventricular septal defect, atrial septal defect, endo-cardial cushion defect, pulmonary valve atresia or steno-sis, Ebstein anomaly, aortic valve stenosis, patent ductusarteriosus, and coarctation of the aorta) and nine otherbirth defects groups: hydrocephalus without spina bifida,anophthalmia or microphthalmia, congenital cataract, bil-iary atresia, renal agenesis/hypoplasia, lower limb reduc-tion, Down syndrome, trisomy 18, and fetal alcohol syn-drome. The Hewitt-Rogerson test was significant for cer-tain 6-month intervals in 17 birth defect groups:pulmonary valve atresia or stenosis, coarctation of theaorta, and amniotic bands (January–June); biliary atresia,lower limb reduction, and trisomy 18 (March–August);congenital megacolon (April–September); congenital cata-ract and renal agenesis or hypoplasia (June–November);endocardial cushion defect (July–December); hydrocepha-lus without spina bifida and transposition of the greatarteries (August–January); upper limb reduction (Septem-ber–February); anotia or microtia (October–March); andencephalocele, ventricular septal defect, and fetal alcoholsyndrome (December–May). The Walter-Elwood test wassignificant for ventricular septal defect (peak, March 14),pulmonary valve atresia/stenosis (peak, October 14),

2 CATON

Birth Defects Research (Part A) 00:000�000 (2012)

Tab

le1

SummaryofTrendan

dSeasonalityTests

forSelectedBirth

Defects

inUpstateNew

York,1992–2006

Group

Live

births

Rate

per

10,000

livebirths

Coch

rane-Arm

itag

eHew

itt-Rogerson

Walter-Elw

ood

Chi-square

pvalue,

2-sided

Hew

itt

score

Peak6-month

period

pvalue

Peak

pvalue,

centerofgravity

pvalue,

goodness

offit

Peak

season

pvalue

LiveBirths

1,967,65

4CentralNervousSystem

Anen

cephalus

920.47

0.12

948

Oct–M

ar0.09

0Feb

10.571

0.80

9W

inter

0.834

Spinabifidawithoutan

encephalus

359

1.82

0.43

844

Mar–A

ug

0.24

2Mar

20.976

0.46

1Spring

0.968

Hydrocephaluswithoutsp

inabifida

989

5.03

0.048a

51Aug–Jan

0.033a

Nov3

0.651

0.87

2W

inter

0.812

Encephalocele

970.49

0.06

851

Dec–M

ay0.033a

Feb

190.427

0.56

1W

inter

0.270

Microcephalus

634

3.22

0.15

448

Nov–A

pr

0.09

0Dec

190.773

0.93

8Fall

0.799

Eye

Anophthalmia,microphthalmia

990.50

0.013a

48Dec–M

ay0.09

0Mar

60.308

0.35

8Spring

0.101

Congen

ital

cataract

236

1.20

0.007a

53Jun–N

ov

0.013a

Mar

200.231

0.01

2Summer

0.027a

Ear

Anotia,

microtia

105

0.53

0.13

652

Oct–M

ar0.021a

Dec

100.668

0.67

8Fall

0.927

Cardiovascular

Transp

ositionofgreat

arteries

234

1.19

0.15

252

Aug–Jan

0.021a

Nov7

0.199

0.98

6Fall

0.645

TetralogyofFallot

386

1.96

0.020a

46Jul–Dec

0.15

5Nov1

0.681

0.16

6Fall

0.055

Ven

tricularseptaldefect

4923

25.02

0.001a

53Dec–M

ay0.013a

Mar

140.003a

0.16

8W

inter

0.005a

Atrialseptaldefect

2645

13.44

0.027a

48Jan–Jun

0.09

0Mar

270.196

0.04

4Spring

0.090

Endocard

ialcu

shiondefect

156

0.79

0.013a

56Jul–Dec

0.002a

Mar

280.060

0.00

1Fall

0.327

Pulm

onaryvalveatresiaorsten

osis

1408

7.16

0.010a

53Jan–Jun

0.013a

Oct

140.018a

0.00

3Spring

0.034a

Tricu

spid

atresiaorsten

osis

140

0.71

0.11

748

Sep

–Feb

0.09

0Jan7

0.741

0.34

6W

inter

0.413

Ebsteinan

omaly

960.49

0.049a

46Oct–M

ar0.15

5Nov24

0.179

0.33

7Fall

0.772

Aortic

valvesten

osis

269

1.37

0.042a

47Jan–Jun

0.12

0Oct

260.503

0.04

9W

inter

0.307

Hypoplastic

leftheart

syndrome

284

1.44

0.18

448

May

–Oct

0.09

0Feb

50.614

0.35

3Fall

0.620

Paten

tductusarteriosu

s(�

2500

gm)

1789

9.09

0.024a

49Feb

–Jul

0.06

6Nov23

0.125

0.05

9Summer

0.564

Coarctationofao

rta

706

3.59

0.008a

56Jan–Jun

0.002a

Oct

60.044a

0.00

5W

inter

0.048a

Orofacial

Cleftpalatewithoutcleftlip

1087

5.52

0.07

547

Nov–A

pr

0.12

0Feb

20.490

0.42

3W

inter

0.781

Cleftlipwithan

dwithoutcleftpalate

1509

7.67

0.13

946

Mar–A

ug

0.15

5Dec

170.780

0.07

4Summer

0.614

Choan

alatresia

312

1.59

0.08

548

Mar–A

ug

0.09

0Dec

180.777

0.16

2Summer

0.450

Gastrointestinal

Esophag

ealatresia,

tracheo

esophag

ealfistula

363

1.84

0.13

047

May

–Oct

0.12

0Dec

10.788

0.58

0W

inter

0.726

Rectalan

dlargeintestinal

atresiaorsten

osis

555

2.82

0.11

244

May

–Oct

0.24

2Feb

40.641

0.22

6Summer

0.763

Pyloricsten

osis

4026

10.46

0.10

545

Jun–N

ov

0.19

7NC

Summer

0.226

Hirschsp

rungdisease

(congen

ital

meg

acolon)

394

2.00

0.14

750

Apr–Sep

0.047a

Jan2

0.411

0.47

5Summer

0.711

Biliary

atresia

152

0.77

0.014a

56Mar–A

ug

0.002a

Dec

60.027a

0.00

1Summer

0.056

Genitourinary

Ren

alag

enesis

orhypoplasia

567

2.88

0.001a

51Jun–N

ov

0.033a

Mar

160.142

<0.00

1Summer

0.010a

Obstructivegen

itourinarydefect

5091

25.87

0.20

347

Jun–N

ov

0.12

0Mar

180.926

0.15

6Fall

0.610

Hyposp

adias,

epispad

ias

7623

75.68

0.27

342

Jan–Jun

0.35

0NC

Summer

0.561

Musculosk

eletal

Upper

limbreductiondeform

ity

381

1.94

0.05

750

Sep

–Feb

0.047a

Jan3

0.124

0.86

6W

inter

0.124

Lower

limbreductiondeform

ity

243

1.23

0.029a

51Mar–A

ug

0.033a

Nov11

0.246

0.10

5Spring

0.114

Gastroschisis

344

1.75

0.10

244

Mar–A

ug

0.24

3Feb

90.925

0.51

8Summer

0.908

Omphalocele

246

1.25

0.32

745

Nov–A

pr

0.19

7Nov22

0.861

0.70

9Fall

0.875

Congen

ital

hip

dislocation

2467

12.54

0.10

149

Mar–A

ug

0.06

6Dec

190.305

0.31

7Spring

0.191

3SEASONALITY OF BIRTH DEFECTS


coarctation of the aorta (peak, October 6), and biliaryatresia (peak, December 6); however, only the peak forventricular septal defect fell within the time periods indi-cated by the Hewitt-Rogerson and chi-square tests. Thechi-square test for season of conception was significantfor congenital cataract (summer peak), ventricular septaldefect (winter peak), pulmonary valve atresia/stenosis(spring peak), coarctation of the aorta (winter peak), andrenal agenesis/hypoplasia (summer peak).Figure 1 displays the results of the four statistical tests

graphically for the 17 birth defect groups with at leastone statistically significant seasonality test during 1992 to2006. Each panel in the figure provides a visual summaryof the results for one birth defect group. The bars repre-sent the rates by season of conception, with darker grayshading indicating a statistically significant chi-squaretest. The line shows the rates by month of conception,with solid black indicating statistical significance of theCochran-Armitage test for trend. The boxes on the x-axismark the peak 6-month period of conception with thesolid black indicating a statistically significant Hewitt-Rogerson test. Finally, the triangle near the top of theplot marks the peak month of conception with the solidblack indicating statistical significance of the Walter-Elwood test.In stratified analyses of maternal characteristics and

infant sex, each of the 25 birth defect groups that did notexhibit a seasonal pattern in the total live birth cohortdisplayed a positive seasonality test within at least onestratum. Among infants of mothers younger than 20years (7.5% of live births), seasonal patterns weredetected for tetralogy of Fallot, omphalocele, congenitalhip dislocation, diaphragmatic hernia, and Down syn-drome (Table 2). Infants of mothers 20 to 34 years old(75.2% of live births) displayed seasonal patterns formicrocephalus, atrial septal defect, Ebstein anomaly, andpatent ductus arteriosus. Seasonal variation in infants ofmothers 35 years and older (17.3% of live births) wasapparent for spina bifida, hypoplastic left heartsyndrome, esophageal atresia–tracheoesophageal fistula,pyloric stenosis, obstructive genitourinary defect, hypo-spadias or epispadias, congenital hip dislocation, andfetal alcohol syndrome.Infants of non-Hispanic white mothers (76.4% of live

births) demonstrated seasonality for microcephalus, atrialseptal defect, hypoplastic left heart syndrome, patentductus arteriosus, rectal and large intestinal atresia or ste-nosis, biliary atresia, renal agenesis or hypoplasia,omphalocele, and congenital hip dislocation (Table 3).Infants of non-Hispanic black mothers (10.4% of livebirths) displayed seasonality for omphalocele, congenitalhip dislocation, and diaphragmatic hernia. Infants of His-panic mothers (9.7% of live births) showed seasonal pat-terns for microcephalus, atrial septal defect, aortic valvestenosis, cleft palate, choanal atresia, rectal and large in-testinal atresia/stenosis, pyloric stenosis, and hypospa-dias or epispadias.Infants of those with less than a high school education

(14.2% of live births) displayed seasonality for tetralogyof Fallot, esophageal atresia–tracheoesophageal fistula,rectal and large intestinal atresia or stenosis, pyloric ste-nosis, biliary atresia, congenital hip dislocation, andDown syndrome (Table 4). Tetralogy of Fallot, patentductus arteriosus, gastroschisis, omphalocele, trisomy 13,and Down syndrome exhibited seasonal patterns in

Tab

le1

SummaryofTrendan

dSeasonalityTests

forSelectedBirth

Defects

inUpstateNew

York,1992–2006(C

ontinued

)

Group

Live

births

Rate

per

10,000

livebirths

Coch

rane-Arm

itag

eHew

itt-Rogerson

Walter-Elw

ood

Chi-square

pvalue,

2-sided

Hew

itt

score

Peak6-month

period

pvalue

Peak

pvalue,

centerofgravity

pvalue,

goodness

offit

Peak

season

pvalue

Diaphragmatic

hernia

346

1.76

0.173

45Jun–N

ov

0.19

7Feb

160.888

0.690

Summer

0.462

Chromosomal

Trisomy13

136

0.69

0.191

48Feb

–Jul

0.09

0Nov19

0.880

0.892

Winter

0.854

Downsyndrome(trisomy21

)22

5411

.46

0.04

6a45

Dec–M

ay0.19

5Mar

190.401

0.243

Winter

0.114

Trisomy18

202

1.03

0.00

9a51

Mar–A

ug

0.033a

Nov16

0.361

0.080

Summer

0.108

Other

Fetal

alcoholsyndrome

231

1.17

0.02

9a50

Dec–M

ay0.047a

Feb

140.071

0.202

Winter

0.089

Amnioticban

ds

790.40

0.115

50Jan–Jun

0.047a

Feb

70.562

0.809

Winter

0.709

aStatistically

significanttren

dorseasonalitytest

(p�

0.05).

NC,notcomputed.

4 CATON


infants of mothers with high school or some collegeeducation (55.3% of live births). Infants of women with acollege degree (30.5% of live births) demonstrated sea-

sonality for microcephalus, anophthalmia or microphthal-mia, tetralogy of Fallot, atrial septal defect, tricuspid atre-sia or stenosis, Ebstein anomaly, hypoplastic left heart

Figure 1. Graphical summaries of trend and seasonality tests for birth defects groups with at least one statistically significant seasonalitytest. The line shows the live birth prevalence rates by month of conception, with solid black indicating statistical significance (p � 0.05)of the Cochrane-Armitage test for trend. The boxes on the x-axis mark the peak 6-month period of conception, with solid black indicatinga statistically significant (p � 0.05) Hewitt-Rogerson test. The triangle near the top of the plot marks the peak month of conception, withsolid black indicating statistical significance (p � 0.05) of the Walter-Elwood test. The bars represent the rates by season of conceptionwith darker gray shading indicating a statistically significant (p � 0.05) chi-square test.



Tab

le2

SummaryofSeasonalityTests

byMaternal

AgeforSelectedBirth

Defects

inUpstateNew

York,1992–2006

<20

years

20–34years

�35years

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

Births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

LiveBirths

147,91

41,479,28

433

9,80


Anen

cephalus

10NC

NC

NC

65Jan–Jun

Jan10

Winter

17NC

NC

NC


encephalus

31Jun1–

Dec

1NC

Fall

269

Mar–A

ug

Dec

28Summer

59Sep

–Feb

aDec

18a

Wintera


inabifida

108

Oct–M

arNov21

Winter

728

Aug–Jan

aOct

16W

inter

152

Dec–M

ayApr1

Spring

Encephalocele

8NC

NC

NC

71Dec–M

ayFeb

6W

inter

18NC

NC

NC

Microcephalus

61Nov–A

pr

Jan27

Spring

466

Sep

–Feb

aDec

30Fall

107

Jul–Dec

Feb

26Summer

Eye


5NC

NC

NC

75Dec–M

ayFeb

21Spring

19NC

NC

NC

Congen

ital

cataract

21Nov1–

May

1NC

Summer

178

Jun–N

ova

Mar

10Summer

a37

Aug–Jan

NC

Winter

Ear

Anotia,

microtia

10NC

NC

NC

73May

–Oct

aJan14

Spring

22Nov–A

pra

NC

Winter

Cardiovascular

Transp

ositionofgreat

arteries

20Jul–Dec

aNC

Fall

169

Oct–M

arJan2

Fall

45Aug–Jan

aNC

Winter

TetralogyofFallot

29Jun15

–Dec

16NC

Falla

288

Dec–M

ayDec

1Spring

69Feb

–Jul

Mar

23Fall

Ven


347

Dec–M

aya

Mar

14W

inter

3629

Jan–Jun

Mar

14a

Wintera

946

Feb

–Jul

Oct

19Spring

Atrialseptaldefect

212

Aug–Jan

Jan4

Spring

1920

Jan–Juna

Mar

11Spring

511

Mar–A

ug

Spring

Endocard

ialcu

shiondefect

7NC

NC

NC

122

Aug–Jan

aOct

12Fall

27May

–Oct

NC

Fall

Pulm


osis

130

May

–Oct

aFeb

6Summer

1019

Jan–Juna

Oct

17a

Springa

258

Dec–M

ayFeb

24Spring

Tricu

spid

atresiaorsten

osis

13NC

NC

NC

102

Sep

–Feb

Mar

16W

inter

25Nov–A

pr

NC

Winter

Ebsteinan

omaly

6NC

NC

NC

76Dec–M

aya

Dec

11W

inter

14NC

NC

NC

Aortic

valvesten

osis

15NC

NC

NC

210

Jan–Jun

Oct

15W

inter

44Feb

–Jul

NC

Summer

Hypoplastic

leftheart

syndrome

18NC

NC

NC

230

Sep

–Feb

Feb

10Fall

36Apr–Sep

aNC

Spring

Paten

tductusarteriosu

s(�

2500

gm)

123

May

–Oct

Jan26

Fall

1313

Feb

–Jula

Nov16

Summer

353

Nov–A

pr

Oct

8Spring

Coarctationofao

rta

47Jan–Juna

NC

Winter

524

Dec–M

ayOct

7W

intera

134

Jan–Jun

Nov23

Summer

Orofacial


83May

–Oct

Dec

10W

inter

800

Jun–N

ov

Jan14

Winter

203

Sep

–Feb

Mar

8Fall

Cleftlipwithan

dwithoutcleftpalate

143

Jun–N

ov

Oct

17Fall

1134

Mar–A

ug

Nov26

Spring

231

Jun–N

ov

NC

Fall

Choan

alatresia

19NC

NC

NC

214

Aug–Jan

Mar

25Spring

79Mar–A

ug

Nov1

Spring

Gastrointestinal

Esophag

ealatresia,

tracheo

esophag

ealfistula

24Jul–Dec

NC

Winter

256

Dec–M

ayMar

1W

inter

83May

–Oct

aJan7

Summer

Rectalan

dlargeintestinal

atresiaorsten

osis

46Aug–Jan

NC

Spring

408

Jul–Dec

Feb

17Summer

101

May

–Oct

Jan10

Spring

Pyloricsten

osis

399

Nov–A

pr

Feb

17Spring

3052

Jun–N

ov

NC

Winter

573

Jun–N

ova

Mar

14Summer

Hirschsp

rungdisease

(congen

ital

meg

acolon)

39Nov–A

pr

NC

Spring

289

Apr–Sep

aJan6a

Summer

66Nov–A

pr

Dec

8W

inter

Biliary

atresia

22Feb

1–Aug1

NC

Spring

108

Mar–A

uga

Dec

11a

Summer

a22

Jan–Jun

NC

Winter

Genitourinary

Ren

alag

enesis

orhypoplasia

53Jul–Dec

aMar

17Summer

424

Jun–N

ov

Mar

26Summer

90Jun–N

ova

Mar

10Fall

Obstructivegen

itourinarydefect

341

Sep

–Feb

Mar

29Fall

3721

Apr–Sep

NC

Summer

1029

Sep

–Feb

aOct

24Fall

Hyposp

adias,

epispad

ias

555

Jan–Jun

Nov1

Summer

5720

Oct–M

arJan12

Summer

1352

Apr–Sep

aJan6

Summer

Musculosk

eletal

Upper

limbreductiondeform

ity

31Jan1–

Jul1

NC

Wintera

292

Oct–M

arJan7

Winter

58Aug–Jan

Oct

6Fall

Lower

limbreductiondeform

ity

20Mar

1–Sep

1aNC

Spring

193

Mar–A

uga

Nov16

Spring

30May

–Oct

NC

Spring

Gastroschisis

116

Mar–A

ug

Nov7

Summer

222

Jun–N

ov

Dec

1Fall

5NC

NC

NC

6 CATON


syndrome, patent ductus arteriosus, hypospadias or epis-padias, and omphalocele.Male infants (51.2% of live births) displayed seasonal

patterns for anencephalus, anophthalmia or microphthal-mia, Ebstein anomaly, hypoplastic left heart syndrome,cleft lip with and without cleft palate, omphalocele, con-genital hip dislocation, diaphragmatic hernia, trisomy 13,and Down syndrome (Table 5). Female infants (48.8% oflive births) demonstrated seasonality for spina bifida,patent ductus arteriosus, cleft lip with and without cleftpalate, pyloric stenosis, and trisomy 13.

DISCUSSION

To generate hypotheses about environmental causes ofbirth defects, we used a large, population-based birthdefects registry to examine the seasonality of specificbirth defects in geographically diverse upstate New York.A trend or seasonal pattern was detected in 24 (57%) ofthe 42 NBDPN surveillance birth defect groups examinedin the 15-year birth cohort. Eighteen birth defects groupshad a statistically significant Cochran-Armitage test fortrend, 17 groups had a significant Hewitt-Rogerson test,four groups had a significant Walter-Elwood test, andfive groups had a significant chi-square test. Ventricularseptal defect showed the most consistent results withfour statistically significant tests: Cochran-Armitage (p 50.0006), Hewitt-Rogerson (December–May; p 5 0.0130),Walter-Elwood (March 14; p 5 0.0027), and chi-square(winter; p 5 0.0046). Pulmonary valve atresia or stenosisand coarctation of the aorta tested positive in all fourtests; however, the peaks in the Walter-Elwood tests (Oc-tober 14 and October 6, respectively) fell outside the peri-ods identified by the Hewitt-Rogerson (January–Junepeaks for both) and chi-square tests (spring and winter,respectively). Both congenital cataract and renal agenesisor hypoplasia showed a 6-month peak in June to Novem-ber conceptions with the Hewitt-Rogerson test and asummer peak with the chi-square test. Biliary atresia hada significant Hewitt-Rogerson test (March–August peak)but a conflicting December 6 peak in the Walter-Elwoodtest.In the total live birth cohort for the study period, we

detected some evidence of seasonality in upstate NewYork for 17 groups of birth defects. These findings areconsistent with other seasonality analyses of ventricularseptal defect, endocardial cushion defect, and renal defect(Bound et al., 1989). There have been mixed results ofseasonality reported in the literature for encephalocele(Castilla et al., 1990; Roux et al., 2009), anotia and micro-tia (Castilla et al., 1990; Liu et al., 2011), pulmonary valveatresia or stenosis (Bound et al., 1989; Bosshardt et al.,2005; Siffel et al., 2005), biliary atresia (Caton et al., 2004;The et al., 2004; Wada et al., 2007; Yoon et al., 1997; Live-sey et al., 2009), and trisomy 18 (Gadow et al., 2006;Tonelli et al., 2006). In contrast, there have been only neg-ative reports of hydrocephalus (Sandahl, 1977b; Castillaet al., 1990), congenital cataract (Haargaard et al., 2005),transposition of the great arteries (Bound et al., 1989), co-arctation of the aorta (Bound et al., 1989; Tikkanen andHeinonen, 1993), and limb reduction (Bound et al., 1989).No literature on the seasonality of congenital megacolon,fetal alcohol syndrome, and amnionic bands was identi-fied.

Tab

le2


byMaternal

AgeforSelectedBirth

Defects

inUpstateNew

York,1992–2006(C

ontinued

)

<20

years

20–3

4years

�35years

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

Births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Omphalocele

26Nov–A

pra

NC

Spring

173

Sep

–Feb

Mar

15Summer

47Mar–A

ug

NC

Summer

Congen

ital

hip

dislocation

155

Feb

–Jula

Dec

1Summer

1899

Jun–N

ov

Feb

7Fall

413

Jan–Juna

Oct

15Spring

Diaphragmatic

hernia

24Nov1–

May

1aNC

Summer

264

Aug–Jan

NC

Summer

58Mar–O

ctJan17

Summer

Chromosomal

Trisomy13

9NC

NC

NC

89Feb

–Jul

Oct

23W

inter

38Jul–Dec

NC

Summer


)10

5Sep

–Feb

aJan9

Winter

1148

Nov–A

pr

Mar

5W

inter

1001

Apr–Sep

NC

Winter

Trisomy18

9NC

NC

NC

110

Sep

–Feb

Jan20

Summer

83Mar–A

ug

Jan9

Summer

Other

Fetal

alcoholsyndrome

13NC

NC

NC

163

Sep

–Feb

Jan20

Winter

54Dec–M

aya

Feb

23a

Springa

Amnioticban

ds

8NC

NC

NC

61Jan–Jun

Feb

13W

inter

10NC

NC

NC

aStatistically

significanttren

dorseasonalitytest

(p�

0.05).

NC,notcomputed.



Tab

le3


byMaternal

Race/

EthnicityforSelectedBirth

Defects

inUpstateNew

York,1992–2006

Non-H

ispan

icW

hite

Non-H

ispan

icBlack

Hispan

ic

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

LiveBirths

1,498,49

820

3,08

118

9,51


Anen

cephalus

69Dec–M

ayFeb

18W

inter

8NC

NC

NC

12NC

NC

NC


encephalus

275

May

–Oct

Feb

21Spring

30May

–Oct

NC

Winter

41Feb

–Jul

NC

Spring


inabifida

671

Jun–N

ova

Mar

15Summer

163

Nov–A

pra

Feb

25Spring

129

Sep

–Feb

Dec

8W

inter

Encephalocele

70Dec–M

ayMar

6W

inter

14NC

NC

NC

12NC

NC

NC

Microcephalus

411

Nov–A

pra

Jan27

Winter

143

Sep

–Feb

Dec

14W

inter

70Jun–N

ova

Feb

8aSummer

a

Eye


74Feb

–Jul

Oct

22Spring

10NC

NC

NC

10NC

NC

NC

Congen

ital

cataract

181

Jun–N

ov

Oct

7Summer

a32

Jun–N

ov

NC

Falla

15NC

NC

NC

Ear

Anotia,

microtia

68Oct–M

arOct

1W

inter

4NC

NC

NC

26Oct–M

arNC

Fall

Cardiovascular

Transp

ositionofgreat

arteries

188

Oct–M

ara

Dec

1W

inter

20Jul1–

–Jan

1NC

Winter

19NC

NC

NC

TetralogyofFallot

286

Jul–Dec

Dec

20Spring

46Apr–Sep

NC

Fall

36Oct–M

arNC

Spring

Ven


3763

Dec–M

aya

Mar

21a

Wintera

506

Feb

–Jul

Mar

13Spring

480

Jun–N

ov

NC

Fall

Atrialseptaldefect

1891

Jan–Juna

Mar

15Spring

370

Nov–A

pr

Dec

27Summer

281

Apr–Sep

aDec

29Spring

Endocard

ialcu

shiondefect

118

Jul–Dec

aMar

3Summer

17NC

NC

NC

16NC

NC

NC

Pulm


osis

967

Mar–A

uga

Oct

29a

Springa

260

Jan–Jun

Mar

13W

inter

130

Jan–Jun

Jan30

Winter

Tricu

spid

atresiaorsten

osis

98Sep

–Feb

Dec

11W

inter

20Jan1–

Jul1

NC

Winter

19NC

NC

NC

Ebsteinan

omaly

70Oct–M

arNov17

Fall

5NC

NC

NC

15NC

NC

NC

Aortic

valvesten

osis

227

Feb

–Jul

Nov16

Winter

13NC

NC

NC

20Jan–Juna

NC

Winter

Hypoplastic

leftheart

syndrome

210

Apr–Sep

aJan18

Fall

38Aug–Jan

NC

Winter

29Jan–Jun

NC

Fall

Paten

tductusarteriosu

s(�

2500

gm)

1172

Jan–Juna

Nov2

Summer

332

Jun–N

ova

Feb

9Falla

186

Mar–A

ug

Dec

6Summer

Coarctationofao

rta

572

Jan–Juna

Oct

4aW

inter

61Aug–Jan

Oct

29W

inter

52Mar–A

uga

Nov5

Spring

Orofacial


883

Nov–A

pr

Feb

8Fall

61Jan–Jun

Feb

9W

inter

95Feb

–Jula

Feb

20Summer

Cleftlipwithan

dwithoutcleftpalate

1220

Mar–A

ug

Jan3

Spring

105

Sep

–Feb

Nov17

Fall

131

Mar–A

ug

Dec

9Summer

Choan

alatresia

256

Mar–A

ug

Nov14

Summer

23Apr–Sep

NC

Spring

26Apr–Sep

aNC

Summer

Gastrointestinal

Esophag

ealatresia,

tracheo

esophag

ealfistula

303

May

–Oct

Dec

2W

inter

26May

–Oct

NC

Fall

26Mar–A

ug

NC

Summer

Rectalan

dlargeintestinal

atresiaorsten

osis

427

Jul–Dec

aFeb

20Fall

49May

–Oct

Oct

11Fall

51Jan–Juna

Mar

29a

Spring

Pyloricsten

osis

3364

Apr–Sep

Jan26

Summer

220

Jun–N

ov

Oct

24Summer

371

Oct–M

ara

Jan20

Winter

Hirschsp

rungdisease

(congen

ital

meg

acolon)

286

Apr–Sep

aJan19

Summer

62Oct–M

ara

Jan24

Winter

29Mar–A

uga

NC

Spring

Biliary

atresia

101

Mar–A

uga

Dec

1Summer

29Feb

–Jul

NC

Spring

14NC

NC

NC

Genitourinary

Ren

alag

enesis

orhypoplasia

433

Jun–N

ov

Apr1

Summer

a70

Jun–N

ova

Feb

11Summer

44Jun–N

ov

NC

Fall

Obstructivegen

itourinarydefect

3904

Feb

–Jul

NC

Fall

440

Aug–Jan

Mar

28Fall

552

Sep

–Feb

Oct

20W

inter

Hyposp

adias,

epispad

ias

6256

Jun–N

ov

Jan25

Summer

714

Feb

–Jul

Dec

18Fall

455

Jan–Juna

Nov3a

Summer

Musculosk

eletal

Upper

limbreductiondeform

ity

306

Aug–Jan

Dec

15W

inter

34Dec

1–Jun1a

NC

Wintera

28Sep

1–Mar

1NC

Winter

Lower

limbreductiondeform

ity

187

Mar–A

uga

Dec

1Spring

23Jan31

–Jul31

NC

Spring

23Feb

–Jul

NC

Fall

Gastroschisis

272

Mar–A

ug

Feb

11Summer

31Jun–N

ov

NC

Fall

34Apr–Sep

NC

Summer

Omphalocele

186

Nov–A

pra

Feb

15W

inter

38Jun–N

ova

NC

Fall

16NC

NC

NC

8 CATON


We found no evidence of seasonality for the othertwenty-five NBDPN birth defects groups in the entirelive birth cohort. These findings are in agreement withother negative seasonality findings for esophageal atresia(Kyyronene et al., 1988; Bound et al., 1989; Castilla et al.,1990), rectal atresia or stenosis (Kyyronene and Hem-minki, 1988; Bound et al., 1989; Castilla et al., 1990), ob-structive genitourinary defects (Bound et al., 1989), hypo-spadias (Castilla et al., 1990; Skriver et al., 2004), ompha-locele (Bound et al., 1989), and diaphragmatic hernia(Bound et al., 1989; Castilla et al., 1990; Torfs et al., 1992).Mixed results have been published for anencephaly (San-dahl, 1977a; Jorde et al., 1984; Fraser et al., 1986; Boundet al., 1989; Castilla et al., 1990; Siffel et al., 2005), spinabifida (Sandahl, 1977a; Jorde et al., 1984; Fraser et al.,1986; Bound et al., 1989; Castilla et al., 1990; Siffel et al.,2005; Beyer et al., 2011), aortic valve stenosis (Boundet al., 1989; Bosshardt et al., 2005; Siffel et al., 2005),hypoplastic left heart syndrome (Tikkanen and Heinonen,1994; Siffel et al., 2005; Eghtesady et al., 2011), oral clefts(Sandahl, 1977b; Coupland and Coupland, 1988; Boundet al., 1989; Castilla et al., 1990; Amidei et al., 1994; Fraserand Gwyn, 1998; Cooper et al., 2000; Siffel et al., 2005;Krost and Schubert, 2006; Elliott et al., 2008; Gregg et al.,2008; de la Vega and Lopez-Cepero, 2009; Chung et al.,2011), esophageal atresia (Kyyronene et al., 1988; Boundet al., 1989; Castilla et al., 1990), rectal atresia or stenosis(Kyyronene and Hemminki, 1988; Bound et al., 1989; Cas-tilla et al., 1990), obstructive genitourinary defects (Boundet al., 1989), hypospadias (Castilla et al., 1990; Skriveret al., 2004), gastroschisis (de la Vega and Lopez-Cepero,2009; Waller et al., 2010), omphalocele (Bound et al.,1989), congenital hip dislocation (Chen et al., 1970; Heik-kila, 1984; Bound et al., 1989; Siffel et al., 2005; Anandet al., 1992), diaphragmatic hernia (Bound et al., 1989;Castilla et al., 1990; Torfs et al., 1992), trisomy 13 (Gadowet al., 2006; Tonelli et al., 2006), and Down syndrome(Videbech and Nielsen, 1984; Bound et al., 1989; Castillaet al., 1990; Puri and Singh, 1995; Stolwijk et al., 1997;Morris et al., 1998; Gadow et al., 2006; Tonelli et al.,2006). Our findings differ from other reports of positiveseasonality tests for atrial septal defect (Bound et al.,1989), tricuspid atresia and stenosis (Bosshardt et al.,2005), and patent ductus arteriosus (Bound et al., 1989).No seasonality studies of microcephalus, anophthalmia,microphthalmia, tetralogy of Fallot, Ebstein anomaly,choanal atresia, and pyloric stenosis were detected.Because some studies have reported sex differences in

seasonality of birth defects, we performed stratified anal-yses by infant sex. Our findings of sex differences in sea-sonal patterns agree with earlier findings in cleft lip andpalate (Fraser and Gwyn, 1998; Krost and Schubert, 2006)and microtia (Liu et al., 2011). Whereas births in femaleswith congenital hip dislocation (Heikkila, 1984) andDown syndrome (Puri and Singh, 1995) displayed sea-sonal variation in prior studies, we found positiveHewitt-Rogerson tests in males for both groups. BecauseDarrow et al. (2009) showed that seasonal patterns ofbirths in metropolitan Atlanta varied by maternal socio-demographics, we also performed stratified analyses bymaternal age, race/ or ethnicity, and education level. Ourstratified analyses revealed additional seasonal patternsamong strata for the 25 birth defects groups that did notdemonstrate seasonality in the total live birth cohort.These findings suggest that inconsistencies in seasonality

Tab

le3


byMaternal

Race/

EthnicityforSelectedBirth

Defects

inUpstateNew

York,1992–2006(C

ontinued

)

Non-H

ispan

icW

hite

Non-H

ispan

icBlack

Hispan

ic

Group

Live

births

Hew

itt-

Rogerson

(peak6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peakseason)

Congen

ital

hip

dislocation

2058

Mar–A

uga

Dec

17Spring

101

Oct–M

ara

Dec

25W

inter

219

May

–Oct

Dec

25Summer

Diaphragmatic

hernia

259

Jun–N

ov

Apr1

Summer

38Apr–Sep

aNC

Summer

30Nov1–

May

1NC

Fall

Chromosomal

Trisomy13

94May

–Oct

Dec

8W

inter

19NC

NC

NC

16NC

NC

NC


)17

58Dec–M

ayMar

21W

inter

198

Jan–Jun

Mar

27W

inter

227

Apr–Sep

Jan23

Summer

Trisomy18

145

Apr–Sep

Dec

19Summer

32Nov–A

pra

NC

Winter

22Jan14

–Jul14

aNC

Spring

Other

Fetal

alcoholsyndrome

78Oct–M

ara

Jan8

Winter

138

Dec–M

ayFeb

23W

inter

9NC

NC

NC

Amnioticban

ds

59Jan–Jun

Jan23

Summer

15NC

NC

NC

3NC

NC

NC

aStatistically

significanttren

dorseasonalitytest

(p�

0.05).

NC,notcomputed.



Tab

le4


byMaternal

EducationforSelectedBirth

Defects

inUpstateNew

York,1992–2006

Lessthan

highschool(<

12years)

Highschoolorsomecolleg

e(12–

15years)

Colleg

edeg

ree(�

16years)

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak6-

month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

LiveBirths

274,47

91,069,926

589,11


Anen

cephalus

23Feb

–Jul

NC

Spring

40Nov–A

pr

NC

Winter

22Aug1–

Feb

1NC

Fall


encephalus

50Mar–A

ug

Nov1

Summer

217

Oct–M

arDec

21W

inter

84Oct–M

arMar

18Spring


inabifida

211

May

–Oct

Feb

17Summer

545

Aug–Jan

aOct

14Fall

208

Nov–A

pr

Feb

23Spring

Encephalocele

20Dec

1–Jun1a

NC

Winter

58Jan–Jun

Mar

12W

inter

17NC

NC

NC

Microcephalus

145

Dec–M

ayMar

11Spring

367

Nov–A

pr

Jan15

Winter

108

Jun–N

ova

Mar

21Fall

Eye


12NC

NC

NC

48May

–Oct

NC

Fall

35Dec–M

aya

NC

Springa

Congen

ital

cataract

36Jun–N

ova

NC

Summer

122

Sep

–Feb

Dec

30W

inter

66Jun–N

ova

Mar

7Summer

Ear

Anotia,

microtia

31Jul–Dec

aNC

Fall

46Mar–A

ug

NC

Summer

27Jan–Juna

NC

Winter

Cardiovascular

Transp

ositionofgreat

arteries

38Jan–Jun

NC

Spring

133

Aug–Jan

aOct

10Summer

58Aug–Jan

aNov26

Fall

TetralogyofFallot

65Jul–Dec

Nov25

Falla

204

Jul–Dec

aOct

16Fall

110

Mar–A

ug

Oct

26Springa

Ven


680

Dec–M

ayJan26

Winter

2,673

Jan–Juna

Oct

8Spring

1484

Dec–M

aya

Mar

3W

intera

Atrialseptaldefect

401

Apr–Sep

Dec

19Spring

1454

Jan–Jun

Oct

13Spring

735

Dec–M

aya

Feb

23W

inter

Endocard

ialcu

shiondefect

32Jun–N

ova

NC

Falla

79Jul–Dec

Nov9

Fall

37Apr–Sep

NC

Summer

Pulm


osis

242

Apr–Sep

Dec

3Summer

799

Jan–Juna

Oct

11W

inter

337

Mar–A

uga

Oct

3Springa

Tricu

spid

atresiaorsten

osis

17NC

NC

NC

84May

–Oct

Dec

10Fall

35Nov–A

pra

NC

Winter

Ebsteinan

omaly

16NC

NC

NC

47Mar–A

ug

NC

Spring

29Aug1–

Feb

1aNC

Fall

Aortic

valvesten

osis

29Jan1–

Jul1

NC

Winter

153

Jan–Jun

Oct

20W

inter

80May

–Oct

Jan9

Summer

Hypoplastic

leftheart

syndrome

39May

–Oct

NC

Fall

163

May

–Oct

Feb

15Summer

66Jan–Juna

Mar

10Spring

Paten

tductusarteriosu

s(�

2500

gm)

276

Apr–Sep

Oct

19W

inter

927

Apr–Sep

aJan3

Summer

555

Feb

–Jula

Oct

17Spring

Coarctationofao

rta

93Oct–M

arJan3

Fall

412

Jan–Juna

Oct

25a

Wintera

189

Nov–A

pra

Mar

3Spring

Orofacial


149

Jan–Jun

Feb

3W

inter

619

Nov–A

pr

Mar

13Fall

296

Dec–M

ayJan22

Winter

Cleftlipwithan

dwithoutcleftpalate

250

Dec–M

ayMar

13Spring

861

Mar–A

ug

Dec

29Summer

353

Jul–Dec

Nov8

Fall

Choan

alatresia

34Feb

–Jul

NC

Spring

168

Apr–Sep

Dec

26Summer

102

Mar–A

ug

Oct

10Spring

Gastrointestinal

Esophag

ealatresia,

tracheo

esophag

ealfistula

42Aug–Jan

aNC

Fall

196

May

–Oct

Dec

5Spring

122

Mar–A

ug

Oct

16W

inter

Rectalan

dlargeintestinal

atresiaorsten

osis

84Aug–Jan

aOct

5Fall

307

Jun–N

ov

Feb

9Summer

153

Feb

–Jul

Oct

23Spring

Pyloricsten

osis

679

Aug–Jan

aDec

17W

inter

2387

Feb

–Jul

NC

Summer

909

Jun–N

ov

Mar

19Summer

Hirschsp

rungdisease

(congen

ital

meg

acolon)

66Apr–Sep

Mar

28Spring

217

May

–Oct

aJan17

Summer

107

Nov–A

pr

NC

Summer

Biliary

atresia

35Feb

–Jula

NC

Spring

84Mar–A

uga

Dec

2a

Summer

28Sep

–Feb

NC

Winter

Genitourinary

Ren

alag

enesis,hypoplasia

94Jul–Dec

Mar

1Summer

303

Jun–N

ova

Mar

5Summer

a14

3Aug–Jan

Jan18

Winter

Obstructivegen

itourinarydefect

668

Sep

–Feb

Dec

4Fall

2610

Jun–N

ov

Feb

4Fall

1722

Apr–Sep

Jan19

Fall

Hyposp

adias,

epispad

ias

951

Mar–A

ug

Mar

19Summer

4148

Oct–M

arNC

Winter

2423

Mar–A

uga

Dec

18Spring

Musculosk

eletal

Upper

limbreductiondeform

ity

60Oct–M

arJan19

aW

intera

224

Sep

–Feb

aDec

9Fall

91Apr–Sep

Oct

23Summer

Lower

limbreductiondeform

ity

53Feb

–Jula

Oct

21a

Springa

124

Mar–A

ug

Oct

30Spring

58May

–Oct

Feb

3Fall

10 CATON


across studies can be partly attributed to sociodemo-graphic differences in the live birth populations.There are a number of strengths and limitations to this

type of analysis. The strengths are that we used a large,population-based birth defects registry with NBDPNbirth defects surveillance definitions and examined sea-sonality based on timing of conception. Using NBDPNdefinitions permits the analysis of birth defects groupsthat are more etiologically homogeneous than body sys-tem classifications and allows other birth defects surveil-lance systems to perform comparable analyses. Analyzingseasonal variation of month of conception controls fordifferences in length of gestation and is closer to the rele-vant biologic period of exposure for most birth defectsthan month of birth. Nevertheless, it is important to notethat the biologically relevant exposure period for somebirth defects can occur before conception (e.g., chromo-somal anomalies) or after the first trimester (e.g., amni-otic bands, fetal alcohol syndrome). The limitations arethat we examined seasonality among live births only, weestimated the date of conception using the last menstrualperiod as recorded on the birth certificate, seasonal pat-terns may vary by the selected time-frame, and chancefindings could result from multiple testing. Because ouranalyses were limited to live births, we were unable todetect patterns among birth defects that could result ininduced termination, miscarriage, or stillbirth. Strandet al. (2011) reviewed four studies of stillbirths anddescribed geographic variations in seasonal peaks. Ourpragmatic use of last menstrual period plus 14 days toestimate the date of conception may be inaccuratebecause of recording errors or poor recall of the date oflast menses and the assumption of a 28-day menstrualcycle with ovulation on day 14 (Lynch and Zhang, 2007).In our birth certificate data, 3.7% of live births and 4.8%of birth defects cases with missing dates for last men-strual period were identified. However, we expect boththe misclassification and missing data to be nondifferen-tially distributed across seasons. In an analysis of datafrom the Metropolitan Atlanta Congenital Defects Pro-gram (Siffel et al., 2005) and in Flyde of Lancashire(Bound et al., 1989), variations in the seasonality findingsacross selected time frames were noted.In this exploratory analysis without prior assumptions

of underlying seasonal patterns, we performed multiplestatistical tests on 42 groups of birth defects; some statis-tically significant findings may be the result of chance.We summarized the results of the four statistical testsgraphically for each birth defect to allow the reader tovisualize the consistency of the results. Siffel et al. (2005)describe the strengths and limitations of the Cochran-Armitage trend, the Hewitt-Rogerson test, and Walter-Elwood tests. The Hewitt-Rogerson test assumes nounderlying model for the seasonal pattern and is sensi-tive, specific, and robust in its detection of a 6-monthpeak period, whereas the Walter-Elwood test assumes asinusoidal pattern (i.e., a cyclic variation with one peakin the 12-month period) and measures the goodness-of-fitto that curve. The chi-square test of the four distinctiveseasons was a simple test of observed versus expectedvalues in a two-by-four contingency table.These seasonality tests and graphical summaries can

be incorporated into our ongoing birth defects surveil-lance activities in New York State. Based on our findings,the next steps are to generate hypotheses about potential

Tab

le4


byMaternal

EducationforSelectedBirth

Defects

inUpstateNew

York,1992–2006(C

ontinued

)

Lessthan

highschool(<

12years)

Highschoolorsomecolleg

e(12–

15years)

Colleg

edeg

ree(�

16years)

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak6-

month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Gastroschisis

106

Jun–N

ov

Feb

17Summer

202

Nov–A

pra

Feb

16Spring

29May

–Oct

NC

Summer

Omphalocele

36Jul1–

Jan1

NC

Summer

145

Jun–N

ova

Oct

12Fall

55Jan–Juna

Oct

27a

Springa

Congen

ital

hip

dislocation

274

Mar–A

uga

Nov28

Spring

1300

Feb

–Jul

Dec

4Spring

846

Jun–N

ov

Jan27

Fall

Diaphragmatic

hernia

60Nov–A

pr

Nov26

Winter

203

Jun–N

ov

Jan20

Summer

74Apr–Sep

Mar

12W

inter

Chromosomal

Trisomy13

20Feb

15–A

ug17

NC

Summer

69Mar–A

uga

Nov12

Summer

39Sep

–Feb

NC

Winter


)27

5Dec–M

ayFeb

25W

intera

1195

Jan–Juna

Oct

10W

inter

725

Nov–A

pr

Dec

24W

inter

Trisomy18

26Dec–M

ayNC

Spring

118

Mar–A

uga

Nov13

Summer

50Jun–N

ova

Mar

24Summer

a

Other

Fetal

alcoholsyndrome

91Sep

–Feb

Dec

28Fall

115

Dec–M

aya

Mar

8a

Winter

7NC

NC

NC

Amnioticban

ds

8NC

NC

NC

47Jun–N

ov

NC

Summer

21Nov1–

May

1aNC

Winter

aStatistically

significantseasonalitytest

(p�

0.05).

NC,notcomputed.



Tab

le5


byInfantSex

forSelectedBirth

Defects

inUpstateNew

York,1992–2006

Male

Fem

ale

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

Square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

LiveBirths

1,007,241

960,37


Anen

cephalus

37Nov–A

pra

NC

Winter

55Dec–M

ayMar

8Spring


encephalus

164

Dec–M

ayFeb

20Spring

195

May

–Oct

aFeb

23Summer


inabifida

562

Nov–A

pra

Jan27

Winter

427

Aug–Jan

aMar

18Fall

Encephalocele

49Dec–M

ayNC

Winter

48Dec–M

ayNC

Winter

Microcephalus

247

Jun–N

ov

NC

Fall

387

Nov–A

pr

Jan8

Winter

Eye


55Dec–M

aya

Feb

22Spring

a44

Jan–Jun

NC

Spring

Congen

ital

cataract

120

Jun–N

ova

Apr2a

Falla

116

Feb

–Jul

Dec

9Summer

a

Ear

Anotia,

microtia

65Oct–M

ara

Dec

12a

Winter

40Feb

–Jula

NC

Summer

Cardiovascular

Transp

ositionofgreat

arteries

152

Aug–Jan

aNov13

Winter

82Oct–M

arOct

27Summer

TetralogyofFallot

231

Dec–M

ayDec

29Spring

155

Jul–Dec

Mar

30Fall

Ven


2316

Dec–M

ayMar

9W

inter

2607

Dec–M

aya

Mar

18a

Wintera

Atrialseptaldefect

1317

Jan–Jun

Mar

16Spring

1328

Jan–Jun

Oct

12Spring

Endocard

ialcu

shiondefect

77Aug–Jan

aOct

21Fall

79Jul–Dec

aMar

10Fall

Pulm


osis

669

Dec–M

aya

Apr1

Springa

739

Jan–Juna

Nov3

Summer

Tricu

spid

atresiaorsten

osis

72Jan–Jun

Jan9

Winter

68May

–Oct

Jan7

Winter

Ebsteinan

omaly

54Oct–M

ara

Dec

17Spring

42Jul–Dec

NC

Fall

Aortic

valvesten

osis

179

Feb

–Jul

Nov8

Summer

90Dec–M

ayJan20

Winter

Hypoplastic

leftheart

syndrome

175

May

–Oct

aFeb

3Fall

109

Nov–A

pr

Feb

7W

inter

Paten

tductusarteriosu

s(�

2500

gm)

880

Nov–A

pr

Feb

9Fall

909

Mar–A

uga

Dec

8aSummer

a

Coarctationofao

rta

406

Jan–Juna

Oct

16Spring

300

Jan–Jun

Mar

22W

inter

Orofacial


469

Oct–M

arJan12

Winter

618

Jan–Jun

Feb

12W

inter

Cleftlipwithan

dwithoutcleftpalate

972

Jul–Dec

aOct

23Fall

537

Feb

–Jula

Nov9a

Springa

Choan

alatresia

152

Jul–Dec

Feb

23Summer

160

Mar–A

ug

Nov8

Spring

Gastrointestinal

Esophag

ealatresia,

tracheo

esophag

ealfistula

196

Apr–Sep

Dec

24Summer

167

Jan–Jun

Feb

6Spring

Rectalan

dlargeintestinal

atresiaorsten

osis

281

Apr–Sep

Jan29

Summer

274

Oct–M

arNC

Spring

Pyloricsten

osis

3286

Mar–A

ug

Mar

20Summer

740

Sep

–Feb

aOct

16Fall

Hirschsp

rungdisease

(congen

ital

meg

acolon)

273

Apr–Sep

aJan6

Spring

121

Feb

–Jul

Oct

1Summer

Biliary

atresia

73Mar–A

ug

Jan6

Summer

79Feb

–Jula

Nov22

aSpring

Genitourinary

Ren

alag

enesis

orhypoplasia

346

Jun–N

ova

Mar

21Summer

221

Jun–N

ova

Mar

11Summer

a

Obstructivegen

itourinarydefect

3589

Apr–Sep

Jan9

Fall

1502

Jun–N

ov

NC

Fall

Hyposp

adias,

epispad

ias

7623

Jan–Jun

NC

Summer

Musculosk

eletal

Upper

limbreductiondeform

ity

212

Oct–M

ara

Dec

28W

inter

169

Sep

–Feb

Jan4

Winter

Lower

limbreductiondeform

ity

145

Nov–A

pr

Mar

14Spring

96Mar–A

uga

Dec

9aSpringa

Gastroschisis

177

Mar–A

ug

Dec

19Summer

167

Sep

–Feb

Dec

25Fall

Omphalocele

118

Oct–M

ara

Dec

14W

inter

128

Apr–Sep

Dec

20Summer

12 CATON


etiologic factors that vary by season in New York, suchas meteorological factors, environmental chemicals, andmaternal infections. For example, influenza displays aseasonal and regional variation in New York (New YorkState Department of Health, 2011), and maternal feverhas been associated certain birth defects in animals andhumans (Edwards, 2006). A time series analysis could beperformed to assess the correlation between the seasonalpattern of influenza and the occurrence of birth defectsin infants. Alternatively, a case-control study could beperformed to examine birth defects in infants of motherswhose first trimesters occurred during influenza season.

REFERENCES

Amidei RL, Hamman RF, Kassebaum DK, et al. 1994. Birth prevalence ofcleft lip and palate in Colorado by sex distribution, seasonality, race/ethnicity, and geographic variation. Spec Care Dentist 14:233–240.

Anand JK, Moden I, Myles JW, et al. 1992. Incidence of neonatal hipinstability: are there seasonal variations? Acta Orthop Belg 58:205–208.

Armitage P. 1955. Tests for linear trends in proportions and frequencies.Biometrics 11:375–386.

Beyer DA, Diedrich K, Weichert J, et al. 2011. Seasonality of spina bifidain Northern Germany. Arch Gynecol Obstet 284:849–854.

Bosshardt D, Ajdacic-Gross V, Lang P, et al. 2005. Season of birth in val-vular heart disease. Paediatr Perinat Epidemiol 19:246–252.

Bound JP, Harvey PW, Francis BJ, et al. 1989. Seasonal prevalence ofmajor congenital malformations in the Fylde of Lancashire 1957–1981.J Epidemiol Community Health 43:330–342.

Castilla EE, Orioli IM, Lugarinho R, et al. 1990. Monthly and seasonalvariations in the frequency of congenital anomalies. Int J Epidemiol19:399–404.

Caton AR, Druschel CM, McNutt LA, et al. 2004. The epidemiology of ex-trahepatic biliary atresia in New York State, 1983–98. Paediatr PerinatEpidemiol 18:97–105.

Centers for Disease Control and Prevention. 2008. Update on overallprevalence of major birth defects—Atlanta, Georgia, 1978–2005. MorbMortal Wkly Rep 57:1–5.

Chen R, Weissman SL, Salama R, Klingberg MA. 1970. Congenital dislo-cation of the hip (CDH) and seasonality: the gestational age of vul-nerability to some seasonal factor. Am J Epidemiol 92:287–293.

Chung MK, Lao TT, Ting YH, et al. 2011. Is there seasonality in the inci-dence of oral-facial clefts? J Matern Fetal Neonatal Med. [Epub aheadof print]

Cochran ,WG. 1954. Some methods for strengthening the common chi-square tests. Biometrics 10:417–451.

Cooper ME, Stone RA, Liu Y, et al. 2000. Descriptive epidemiology ofnonsyndromic cleft lip with or without cleft palate in Shanghai,China, from 1980 to 1989. Cleft Palate Craniofac J 37:274–280.

Coupland MA, Coupland AI. 1988. Seasonality, incidence, and sex distri-bution of cleft lip and palate births in Trent Region, 1973–1982. CleftPalate J 25:33–37.

Darrow LA, Strickland MJ, Klein M, et al. 2009. Seasonality of birth andimplications for temporal studies of preterm birth. Epidemiology20:699–706.

de la Vega A, Lopez-Cepero R. 2009. Seasonal variations in the incidenceof some congenital anomalies in Puerto Rico based on the timing ofconception. P R Health Sci J 28:121–125.

Edwards MJ. 2006. Review: hyperthermia and fever during pregnancy.Birth Defects Res A Clin Mol Teratol 76:507–516.

Elliott RF, Jovic G, Beveridge M. 2008. Seasonal variation and regionaldistribution of cleft lip and palate in Zambia. Cleft Palate Craniofac J45:533–538.

Eghtesady P, Brar A, Hall M, et al. 2011. Seasonality of hypoplastic leftheart syndrome in the United States: a 10-year time-series analysis. JThorac Cardiovasc Surg 141:432–438.

Fraser FC, Frecker M, Allderdice P, et al. 1986. Seasonal variation of neuraltube defects in Newfoundland and elsewhere. Teratology 33:299–303.

Fraser FC, Gwyn A. 1998. Seasonal variation in birth date of childrenwith cleft lip. Teratology 57:93–95.

Gadow E, Petracchi F, Poletta FA, et al. 2006. De novo chromosomalabnormalities and month of conception. Data from the southernhemisphere. Prenat Diagn 26:1184–1186.

Gregg TA, Leonard AG, Hayden C, et al. 2008. Birth prevalence of cleftlip and palate in Northern Ireland (1981 to 2000). Cleft Palate Cranio-fac J 45:141–147.

Tab

le5


byInfantSex

forSelectedBirth

Defects

inUpstateNew

York,1992–2006(C

ontinued

)

Male

Fem

ale

Group

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

Square

(peak

season)

Live

births

Hew

itt-

Rogerson

(peak

6-month

period)

Walter-

Elw

ood

(peak)

Chi-

square

(peak

season)

Congen

ital

hip

dislocation

611

Jun–N

ova

Feb

13Summer

1856

Mar–A

ug

Nov25

Spring

Diaphragmatic

hernia

203

Jul–Dec

aNov9

Winter

141

Mar–A

ug

Dec

9Summer

Chromosomal

Trisomy13

73Apr–Sep

aDec

28Summer

63Dec–M

aya

Jan30

Winter


)11

87Nov–A

pra

Mar

4W

inter

1067

Apr–Sep

NC

Winter

Trisomy18

73Mar–A

uga

Oct

6Summer

129

Apr–Sep

Dec

22Summer

Other

Fetal

alcoholsyndrome

127

Dec–M

aya

Feb

16a

Springa

104

Feb

–Jul

Jan25

Fall

Amnioticban

ds

31Jan–Juna

NC

Winter

48Jun–N

ov

NC

Summer

aStatistically

significantseasonalitytest

(p�

0.05).

NC,notcomputed.



Haargaard B, Wohlfahrt J, Rosenberg T, et al. 2005. Risk factors for idio-pathic congenital/infantile cataract. Invest Ophthalmol Vis Sci46:3067–3073.

Heikkila E. 1984. Congenital dislocation of the hip in Finland. An epide-miologic analysis of 1035 cases. Acta Orthop Scand 55:125–129.

Hewitt D, Milner J, Csima A, et al. 1971. On Edwards’ criterion of seasonal-ity and a non-parametric alternative. Br J Prev Soc Med 25:174–176.

Honein MA, Paulozzi LJ. 1999. Birth defects surveillance: assessing the‘‘gold standard.’’ Am J Public Health 89:1238–1240.

Jorde LB, Fineman RM, Martin RA, et al. 1984. Epidemiology of neuraltube defects in Utah, 1940–1979. Am J Epidemiol 119:487–495.

Krost B, Schubert J. 2006. Influence of season on prevalence of cleft lipand palate. Int J Oral Maxillofac Surg 35:215–218.

Kyyronen P, Hemminki K. 1988. Gastro-intestinal atresias in Finland in1970–79, indicating time-place clustering. J Epidemiol CommunityHealth 42:257–265.

Liu L, Pan B, Lin L, et al. 2011. Seasonal variation in months of birth ofpatients with microtia in a Chinese population. Int J Pediatr Otorhi-nolaryngol 75:782–784.

Livesey E, Cortina Borja M, Sharif K, et al. 2009. Epidemiology of biliaryatresia in England and Wales (1999–2006). Arch Dis Child Fetal Neo-natal Ed 94:F451–F455.

Lynch CD, Zhang J. 2007. The research implications of the selection of agestational age estimation method. Paediatr Perinat Epidemiol 21Suppl 2:86–96.

Morris JK, Alberman E, Mutton D, et al. 1998. Is there evidence of cluster-ing in Down syndrome? Int J Epidemiol 27:495–498.

National Birth Defects Prevention Network. 2010. Selected birth defects datafrom population-based birth defects surveillance programs in the UnitedStates, 2003–2007. Birth Defects Res A Clin Mol Teratol 88:1062–1074.

Nelson K, Holmes LB. 1989. Malformations due to presumed spontaneousmutations in newborn infants. N Engl J Med 5; 320:19–23.

New York State Department of Health. 2011. New York State Departmentof Health 2010–2011 Flu Monitoring—Week Ending May 21, 2011.Available at: http://www.health.ny.gov/diseases/communicable/influenza/surveillance/2010–2011/archive/2011–05-21. Accessed December6, 2011.

Puri BK, Singh I. 1995. Season of birth in Down’s syndrome. Br J ClinPract 49:129–130.

Rogerson PA. 1996. A generalization of Hewitt’s test for seasonality. Int JEpidemiology 25:644–648.

Roohan PJ, Josberger RE, Acar J, et al. 2003. Validation of birth certificatedata in New York State. J Community Health 28:335–346.

Roux FE, Oucheng N, Lauwers-Cances V, et al. 2009. Seasonal variationsin frontoethmoidal meningoencephalocele births in Cambodia. J Neu-rosurg Pediatr 4:553–556.

Sadler TW. 2004. Langman’s medical embryology. 9th ed. Baltimore: Lip-pincott Williams and Wilkins.

Sandahl B. 1977a. Seasonal incidence of cleft lips and cleft palates in Swe-den, 1965–1974. Scand J Plast Reconstr Surg 11:39–43.

Sandahl B. 1977b. Seasonal incidence of some congenital malformations inthe central nervous system in Sweden, 1965–1972. Acta PaediatrScand 66:65–72.

Siffel C, Alverson CJ, Correa A, et al. 2005. Analysis of seasonal variationof birth defects in Atlanta. Birth Defects Res A Clin Mol Teratol73:655–662.

Skriver MV, Pedersen L, Stang P, et al. 2004. The month of birth does notaffect the risk of hypospadias. Eur J Epidemiol 19:1135–1136.

Stolwijk AM, Jongbloet PH, Zielhuis GA, et al. 1997. Seasonal variation inthe prevalence of Down syndrome at birth: a review. J EpidemiolCommunity Health 51:350–353.

Strand LB, Barnett AG, Tong S, et al. 2011. The influence of season andambient temperature on birth outcomes: a review of the epidemiolog-ical literature. Environ Res 111:451–462.

The NS, Honein MA, Caton AR, et al.; National Birth Defects PreventionStudy. 2007. Risk factors for isolated biliary atresia, National BirthDefects Prevention Study, 1997–2002. Am J Med Genet A 143:2274–2284.

Tikkanen J, Heinonen OP. 1993. Risk factors for coarctation of the aorta.Teratology 47:565–572.

Tikkanen J, Heinonen OP. 1994. Risk factors for hypoplastic left heart syn-drome. Teratology 50:112–117.

Tonelli M, Specchia C, Decarli A, et al. 2006. De novo chromosomalabnormalities and month of conception. Prenat Diagn 26:118–122.

Torfs CP, Curry CJ, Bateson TF, et al. 1992. A population-based study ofcongenital diaphragmatic hernia. Teratology 46:555–565.

Videbech P, Nielsen J. 1984. Chromosome abnormalities and season ofbirth. Hum Genet 65:221–231.

Wada H, Muraji T, Yokoi A, et al. 2007. Insignificant seasonal and geo-graphical variation in incidence of biliary atresia in Japan: a regionalsurvey of over 20 years. J Pediatr Surg 42:2090–2092.

Waller SA, Paul K, Peterson SE, Hitti JE. 2010. Agricultural-related chemi-cal exposures, season of conception, and risk of gastroschisis inWashington State. Am J Obstet Gynecol 202:241.e1–e6.

Walter SD, Elwood JM. 1975. A test for seasonality of events with a vari-able population at risk. Br J Prev Soc Med 29:18–21.

Wang Y, Cross PK, Druschel CM, et al. 2010. Hospital discharge data: canit serve as the sole source of case ascertainment for population-basedbirth defects surveillance programs? J Public Health Manag Pract16:245–251.

Yoon PW, Bresee JS, Olney RS, et al. 1997. Epidemiology of biliary atresia:a population-based study. Pediatrics 99:376–382.

14 CATON


exploring the seasonality of birth defects in the new york

Documents