Extremely Low Birth Weight and Infant Mortality Ratesin the United States

WHAT’S KNOWN ON THIS SUBJECT: Infant and neonatal mortalityrates in the United States decreased markedly during thetwentieth century but have not decreased notably during recentyears. There has been an increase in preterm and low birthweight births in recent years.

WHAT THIS STUDY ADDS: The lack of decrease in infant andneonatal mortality rates in recent years is due in large part to theincreasing proportion of preterm and low birth weight infants,particularly infants ,500 g.

abstractOBJECTIVE: Infant mortality rates (IMR) and neonatal mortality rates(NMR) in the United States have not decreased recently. The purposeof this study was to determine the contributions of birth weight andgestational age subgroups to the IMR and NMR in the United States.

METHODS: We used the most recent (1983–2005) US linked birth andinfant death data and simple regression analysis to determine thecontributions of specific birth weight and gestational age subgroupsto trends in IMR and NMR.

RESULTS: IMR and NMR decreased between 1983 and 2005 for all birthweight and gestational age subgroups. There was an increase in birthsof very low birth weight infants from 1.2% to 1.5% (P, .001) over thisperiod. The proportion of very low birth weight–infant deaths in-creased from 42.9% to 54.8%, resulting in recent nonsignificantdeclines in IMR and NMR. The proportion of live-birth infants ,500 gincreased from 0.12% to 0.18% (P , .001). The adjusted IMR and NMRover time (excluding infants,500 g) have steeper declining trends thanthe ones including infants ,500 g. The changes in overall IMR and NMRin recent years (2000–2005) are not statistically significant. However, theadjusted IMR and NMR trends during this time are highly significant.

CONCLUSIONS: The increased proportions of infants,500 g and otherlow birth weight infants contribute greatly to the lack of a decrease inIMR and NMR from 2000 to 2005, although birth weight– and gesta-tional age–specific IMR and NMR continue to decrease. Pediatrics2013;131:855–860

AUTHORS: Carissa Lau,a Namasivayam Ambalavanan, MD,a

Hrishikesh Chakraborty, DrPH,b Martha S. Wingate, DrPH,c

and Waldemar A. Carlo, MDa

aDepartment of Pediatrics, University of Alabama at Birmingham,and Children’s of Alabama, Birmingham, Alabama;bDepartment of Epidemiology and Biostatistics, University ofSouth Carolina, Columbia, South Carolina; and cDepartment ofHealth Care Organization and Policy, School of Public Health,University of Alabama at Birmingham, Birmingham, Alabama

KEY WORDSbirth rate trends, infants, infant mortality trends, vital statistics,low birth weight, prenatal care delivery, United Statesepidemiology

ABBREVIATIONSIMR—infant mortality ratesNMR—neonatal mortality ratesVLBW—very low birth weight

Ms Lau assisted in the conceptualization and design of the study,carried out the initial analysis, designed data collectioninstruments, collected data, critically reviewed the manuscript,and approved the final manuscript as submitted; DrAmbalavanan designed data collection instruments, collecteddata, critically reviewed the manuscript, and approved the finalmanuscript as submitted; Dr Chakraborty assisted in theconceptualization and design of the study, carried out the initialanalysis, designed data collection instruments, criticallyreviewed the manuscript, and approved the final manuscript assubmitted; Dr Wingate assisted in the conceptualization anddesign of the study, critically reviewed the manuscript, andapproved the final manuscript as submitted; and Dr Carloconceptualized and designed the study, drafted the initialmanuscript, designed data collection instruments, collecteddata, critically reviewed the manuscript, and approved the finalmanuscript as submitted.

www.pediatrics.org/cgi/doi/10.1542/peds.2012-2471

doi:10.1542/peds.2012-2471

Accepted for publication Feb 1, 2013

Address correspondence to Waldemar A. Carlo, MD, Departmentof Pediatrics, University of Alabama at Birmingham, 9380 Womenand Infants Center, 1700 6th Avenue South, Birmingham, AL 35249-7335. E-mail: wcarlo@peds.uab.edu

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2013 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they haveno financial relationships relevant to this article to disclose

FUNDING: Supported by a grant from the Perinatal Health andHuman Development Program of the University of Alabama atBirmingham and the Children’s of Alabama Centennial Fund.

PEDIATRICS Volume 131, Number 5, May 2013 855

ARTICLE

by guest on April 29, 2016Downloaded from

From 1915 to 2008, the reported infantmortality rates (IMR) in the UnitedStates decreased from 99.9 deaths to6.6 deaths per 1000 live births.1 Neo-natal mortality rate (NMR) decreasedfrom 20.5 deaths in 19502 to 4.3 deathsin 2008 per 1000 live births.1 Althoughthe IMR and NMR in some other high-income countries have continued todecrease,3 IMR and NMR in the UnitedStates have not decreased notablyfrom 2000 to the most recent data in2008, ranging from 6.6–7.0 and 4.3–4.7deaths per 1000 live births, respec-tively,1,4 thereby leading to a worseninginternational rank.5

The recent plateau in infant mortalitymay be related to the increasing pro-portion of preterm (,37 weeks ges-tation) and low birth weight (,2500 g)deliveries. From 1983 to 2005, pretermbirth rates increased from 9.0% to12.7%.6 Additionally, the percentage oflow birth weight infant births in-creased from 6.8% in 1983 to 8.2% in2005.6 Preterm birth and low birthweight are among the most frequentcauses of infant and neonatal death inthe United States.7 Between 2000 and2005, total preterm births increased by9%, which accounted for a large per-centage of infant deaths.8 Increases inthe Alabama and Delaware IMR havebeen attributed to a rise in very lowbirth weight infant mortality.8 In 1993,a similar study in Canada noted thatincreased reporting of live infants bornweighing ,500 g had negatively af-fected its IMR.10 An increase in pretermbirths may in part account for the lackof improvement in IMR and NMR, andthus analysis of nationwide data wouldbe useful.

Thepurposeof thisstudywastoexaminethe contribution of narrow birth weightcategories (250-g and 500-g intervals) tothe IMR and NMR of the United States.Specifically, we evaluated the effect ofbirthweight on the overall United StatesIMR and NMR. By examining narrow

ranges of birth weights, it would bepossible todetermine their proportionalcontributions to IMR and NMR. We hy-pothesized that the increasednumberofinfants born weighing ,500 g has dis-proportionately influenced IMR andNMR.

METHODS

We obtained data from the NationalCenter for Health Statistics linked birthand infant death cohort data files6 forall the years available during the pe-riod 1983–2005. Years 1992–1994 wereunavailable and were therefore ex-cluded. To determine the birth weight–specific neonatal and infant mortalityrates, data were analyzed by the fol-lowing weight subgroups used in thedatabase: $3500 g, 3000 to 3499 g,2500 to 2999 g, 2000 to 2499 g, 1500 to1999 g, 1250 to 1499 g, 1000 to 1249 g,750 to 999 g, 500 to 749 g, and,500 g.Infants with unknown birth weightwere excluded. For each year, the fol-lowing calculations were performedfor each weight subgroup: percentageof live births, percentage of infantdeaths, percentage of neonatal deaths,IMR, and NMR. The same analysis wasdone for gestational age subgroups inthe database: ,28 weeks, 28 to 31weeks, 32 to 35 weeks, 36 weeks, 37 to39 weeks, 40 weeks, 41 weeks, and 42weeks. By determining the percentagesof births and deaths, we were able toobserve how each subgroup’s contri-bution has changed over time.

Simple regression analysiswas used toanalyze the trends of IMR and NMR overtime. We computed adjusted rates,which did not include infants born,500 g, as well as the actual rate,which included all birth weight sub-groups. We evaluated and comparedthe trends over 2 time periods: 1983–1999 and 2000–2005. These periodswere selected on the basis of recentanalyses that support a lack of de-crease in IMR during this the lastdecade.11 Analysis of this cut off allowed

us to observe whether the mortalityrates had changed significantly overtime. We also analyzed the proportion ofinfants born,500 g to total births overthe same 2 periods and over the entireperiod and their effect on IMR and NMR.This analysis allowed us to determinethe contribution of these births to IMRand NMR. A P value of ,.05 was con-sidered significant.

RESULTS

Over the study years, there was an in-creasing trend (P , .001) in the num-ber of infants in the lower birth weightsubgroups (,1500 g) with a corre-sponding decreasing trend (P , .001)in infants$3500 g (Fig 1). From 1983 to2005, the contribution of very low birthweight infants (VLBW, ,1500 g) to thetotal number of infant births increasedfrom 1.2% to 1.5% (43 284–63 030), andthe proportion of live birth infants,500 g increased from 0.12% to 0.18%(4444–7274), both P , .001, while thecontribution of those .3500 g de-creased from 40.1% to 35.1%. Thecontribution of VLBW infants to deathsincreased from 42.9% to 54.8% duringthis period. Similarly, there was an in-creased proportion of preterm infantsborn, with a corresponding decreasein postterm and term infants (Fig 2).The various subgroups of VLBW infantsand infants born at ,28 weeks in-creasingly contributed to infant deaths(Figs 3 and 4).

Both the IMR and NMR showed a sig-nificant decline from 1983 to 1999and showed a nonsignificant decliningtrend from 2000 to 2005 (Fig 5). IMR andNMR decreased between 1983 and 2005for all birth weight and gestational agesubgroups. The analysis showed thatduring the period 1983–1999, the totalIMR was declining significantly (P ,.001), but during 2000–2005 it had nosignificant declining trend (P = .28).The gap between the IMR and NMRbecame smaller, as neonatal deaths

856 LAU et alby guest on April 29, 2016Downloaded from

contributed more to the IMR. When theIMR was adjusted so that it did notinclude births or deaths of infants,500 g, there was an improvement inthe United States IMR and NMR (Fig 5).The adjusted IMR decrease declinedduring 2000–2005 (P = .006). The dif-ference between the adjusted and un-adjusted IMR widened in the recentyears as more infants ,500 g werebeing registered. NMR decreased sig-nificantly during 1983–1999 (P, .001)but not during 2000–2005 (P = .26). Theadjusted NMR, however, decreasedthroughout both the first (P , .001)and the second (P = .002) analyzed timeperiods (Fig 4). Infants ,500 g hada bigger impact on the NMR thaninfants in other weight subgroups.

DISCUSSION

This study shows that IMR and NMR inthe United States have not decreasedrecently, but this is due to increases inlive birth registrations of smaller andmore immature infants, particularly in-fants with birth weights ,500 g. IMRand NMR have continued to decreasewhen birth weight– and gestational age–specific analyses are done. The narrow-ing gap between IMR and NMR is due tothe increased proportion of extremelylow birth weight and preterm infants.

This study has some intrinsic limita-tions. The data are retrospective andonly published through 2005. Data col-lection of selected variables at the statelevel varied over this time period.12

There is possible misclassification be-tween early infant deaths and fetaldeaths12 but this is difficult to ascertainfrom the database. The unavailable co-hort data from 1992 to 1994 could affectthe statistical analysis of that timeperiod.

TheeffectofVLBWinfantson IMRhasbeenreported individually for Delaware9 andAlabama7 as well as in Canada.10 In theDelaware study, investigators reportedthat the increased IMR was due to anincrease in VLBW infant mortality, while

FIGURE 1Percentage contributions of infants in each VLBW subgroup to total births for 1983–2005. The con-tribution of infants in each VLBW subgroup increased during the study period.

FIGURE 2Percentage contributions of infants in gestational age subgroups to total births for 1983–2005.

ARTICLE

PEDIATRICS Volume 131, Number 5, May 2013 857by guest on April 29, 2016Downloaded from

in the Alabama study it was concludedthat the increase in infant mortality wasdue to infants ,500 g. In Canada, anincrease in IMR noted from 1992 to 1993was reported to be due to increasedregistration of infants,500 g.10

A previous national study in the UnitedStates identified preterm birth as themost frequent cause of infant death.13

The current analysis focused on birthweight, but birth weight and gesta-tional age are highly correlated.14 Theincreasing proportion of low birthweight infant births and deaths sug-gests an increasing contribution ofpreterm births to IMR and NMR. In anin-depth analysis of causes of infantmortality, preterm birth was found tobe the most frequent cause in theUnited States, accounting for at least34% of the deaths in 2002 and twiceas many as that determined usingstandard coding procedures.13 Thecurrent study suggests that the con-tribution of prematurity to IMR andNMR has been increasing in recentyears, particularly due to the mostpreterm infants, and accounts for thesignificant declining trends in IMR andNMR in recent years.

In the United States, there has beena rise in preterm birth rates due toinduced preterm birth while sponta-neous preterm births have declined,15

but these data are not likely to be dueto the infants ,500 g, as they consti-tute a minute proportion (currentlyabout 0.15%) of the births (Fig 1). Therise in induced preterm births may in-dicate obstetrical successes,15 but thebenefits for perinatal mortality andmorbidity need to be demonstrated. Im-portantly, 23% of late-preterm infantsdo not have a recorded indication fordelivery noted on the birth certificate.16

In addition to prematurity, the long-standing racial and ethnic disparitiesin infant outcomes also account for thelackofdecrease in infantmortality in theUnited States.11 The continuing increase

FIGURE 3Percentage contribution of infants in each birth weight subgroup to infant mortality for 1983–2005. Thesubgroups that contributed themost to infantmortality were the VLBW subgroups despite their relativelysmall contribution to births. Infants ,500 g contributed the most (.20%) to deaths in recent years.

FIGURE 4Percentage contribution of infants in gestational age subgroups to infantmortality for 1983–2005. Infants,28 weeks contributed the most to infant mortality, and their contribution increased over the years.

858 LAU et alby guest on April 29, 2016Downloaded from

in twin, triplet, and higher-order birthsin the United States16 accounts for someof the increase in lower birth weightand preterm infants. To our knowledge,this is the first birth weight–specificanalysis that identifies the determinantcontribution of infants ,500 g to theIMR and NMR in the national data trendsin the United States.

Although we focused the study on na-tional linked birth and infant deathcohort data, we also examined linkedbirth and infant death period data. Theevaluations of the period linked set ofdata returned very similar results andverified that the changes in IMR andNMR from 1999 to 2005 were not sig-nificant unless infants ,500 g wereexcluded.

It is possible that increased reportingaccounts for the increased number ofinfants born weighing ,500 g in the

database. The World Health Organi-zation defines live birth as the ex-pelled product of conception showingevidence of life regardless of the du-ration of pregnancy. In contrast, inmany countries, including many high-income countries, the definition oflive birth requires a birth weight of500 g or more.

In a study of live birth reporting in in-dustrialized countries, the proportionof live births under 500 g varied from,1 per 10 000 live births in Belgium,Ireland, Latvia, Poland, and the SlovakRepublic to .10 per 10 000 in Canadaand the United States.17 These in-ternational differences compromisethe validity of the rankings of neonataland infant mortality. This also explainsin part why the US IMR and NMRare higher than those of many othercountries. The Born-Alive Infant Pro-

tection Act, which defines a live birthwithout regard to gestation, may havehad a small impact after it became lawin 2002, but the effect of the increasedproportion of smaller and more im-mature infants occurred before.

Agestational ageanalysis indicates thattheprimaryreason for thehigher IMR inthe United States appears to be thehigher proportion of preterm births.5

A greater than 10-fold differential inthe classification of fetal versus first24-hour deaths has been reported be-tween states in the United States.Twenty states had a specific tendencyto classify these infants as eitherfetal deaths or live births/deaths.12 Atrend for more births being classi-fied as early neonatal deaths ratherthan fetal deaths could explain theincrease in very low birth weightinfants. The current study documentsthat there is an increase in the pro-portion of infants in the lower birthweight subgroups during the periodof over 2 decades. Whether this isdue to increased reporting or notcannot be determined by our studyas it is difficult to ascertain from thenational database. The 2009 US na-tional data indicate that prematu-rity rates have decreased duringthe previous 3 years, and that overalllow birth weight rates were not in-creasing.18

CONCLUSIONS

The current study shows that IMRand NMR have continued to decreaseduring recent years in the UnitedStates when birth weight–specificanalysis is done. The increasing pro-portion of infants in the lower birthweight groups is preventing largerdecreases in IMR and NMR. The extentto which the increased proportion oflower birth weight infants is due toreporting, versus actual increasedbirths, cannot be determined fromour analysis.

FIGURE 5Total IMR and NMR and adjusted (excluding infants,500 g) IMR and NMR for 1983–2005. Both IMR andNMR decreased significantly from 1983 to 1999, but there was a nonsignificant declining trend for2000–2005 in total. However, adjusted IMR and adjusted NMR declined significantly during the mostrecent period (2000–2005).

ARTICLE

PEDIATRICS Volume 131, Number 5, May 2013 859by guest on April 29, 2016Downloaded from

REFERENCES

1. Mathews TJ, Miniño AM, Osterman MJK,Strobino DM, Guyer B. Annual summary ofvital statistics: 2008. Pediatrics. 2011;127(1):146–157

2. Wegman ME. Annual summary of vital statis-tics—1965. Pediatrics. 1966;38(6):1063–1067

3. Sepkowitz S. International rankings of in-fant mortality and the United States’ vitalstatistics natality data collecting system—failure and success. Int J Epidemiol. 1995;24(3):583–588

4. Hoyert DL, Freedman MA, Strobino DM,Guyer B. Annual summary of vital statistics:2000. Pediatrics. 2001;108(6):1241–1255

5. MacDorman MF, Mathews TJ. Behind in-ternational rankings of infant mortality:how the United States compares withEurope. NCHS Data Brief. 2009;23:1–8

6. Centers for Disease Control and Pre-vention. Cohort linked birth–infant deathdata files. 1983–2005. Available at: www.cdc.gov/nchs/data_access/Vitalstatsonline.htm. Accessed August 9, 2012

7. MacDorman MF, Mathews TJ. Recent trendsin infant mortality in the United States.NCHS Data Brief. 2008;9:1–8

8. Phelan ST, Goldenberg R, Alexander G,Cliver SP. Perinatal mortality and its re-lationship to the reporting of low-birthweight infants. Am J Public Health.1998;88(8):1236–1239

9. Centers for Disease Control and Prevention(CDC). Increasing infant mortality amongvery low birthweight infants—Delaware,1994-2000. MMWR Morb Mortal Wkly Rep.2003;52(36):862–866

10. Joseph KS, Kramer MS. Recent trends inCanadian infant mortality rates: effect ofchanges in registration of live newbornsweighing less than 500 g. CMAJ. 1996;155(8):1047–1052

11. MacDorman MF, Mathews TJ; Centers forDisease Control and Prevention (CDC). Infantdeaths - United States, 2000-2007. MMWRSurveill Summ. 2011;60(Suppl):49–51

12. Ehrenthal DB, Wingate MS, Kirby RS. Variationby state in outcomes classification for de-liveries less than 500 g in the United States.Matern Child Health J. 2011;15(1):42–48

13. Callaghan WM, MacDorman MF, RasmussenSA, Qin C, Lackritz EM. The contribution ofpreterm birth to infant mortality rates in

the United States. Pediatrics. 2006;118(4):1566–1573

14. Mathews TJ, MacDorman MF. Infant mor-tality statistics from the 2007 period linkedbirth/infant death data set. Natl Vital StatRep. 2011;59(6):1–30

15. VanderWeele TJ, Lantos JD, Lauderdale DS.Rising preterm birth rates, 1989-2004:changing demographics or changing obstet-ric practice? Soc Sci Med. 2012;74(2):196–201

16. Reddy UM, Ko CW, Raju TN, Willinger M.Delivery indications at late-preterm gestationsand infant mortality rates in the UnitedStates. Pediatrics. 2009;124(1):234–240

17. Joseph KS, Liu S, Rouleau J, et al; Fetal andInfant Health Study Group of the CanadianPerinatal Surveillance System. Influence ofdefinition based versus pragmatic birthregistration on international comparisonsof perinatal and infant mortality: pop-ulation based retrospective study. BMJ.2012;344:e746

18. Kochanek KD, Kirmeyer SE, Martin JA,Strobino DM, Guyer B. Annual summary ofvital statistics: 2009. Pediatrics. 2012;129(2):338–348

WHAT YOU DRINK MIGHT CHANGE WHAT YOU CAN EAT: As a counselor at thesummer camp I had attended five years prior, I remember being surprised bya drastic change in the camp’s cuisine. Peanut butter and jelly sandwiches, oncea staple for many of us, had been replaced by soy-nut butter sandwiches. Thecamp had gone nut free to protect those with allergies. Most of us know that anincreasing number of individuals are reporting food allergies. The Centers forDisease Control has reported an 18% increase in all food sensitivities. Nuts, eggs,wheat, and shellfish are the more prevalent allergies in both children and adults.While there is no known cause for the rising prevalence of food allergies, newresearch suggests that pesticides found in tap water may contribute to theirdevelopment. As reported by CNN (Health: December 4, 2012), a recent studyfound a connection between dichlorophenols (chemicals used to sterilizedrinking water) and increased odds of food allergy development. The study in-cluded 2,211 individuals over the age of six and found that individuals withhighest urine levels of dichlorophenols had significantly greater chances of beingsensitized to one or more food allergens. The dichlorophenols may disrupt thenatural gut flora of our digestive tract and contribute to the development of anabnormal immune response to food. While this study adds another possibleconnection between our environment and the development of food sensitivities,more researchwill need to be completed before we can definitively determine thecause of allergic disease and its rising incidence in the western world. Until then,soy-nut butter and gluten-free cookies will, no doubt, become regulars on campmenus across the nation.

Noted by Leah H. Carr, BS, MS-III

860 LAU et alby guest on April 29, 2016Downloaded from

DOI: 10.1542/peds.2012-2471; originally published online April 1, 2013; 2013;131;855Pediatrics

Wingate and Waldemar A. CarloCarissa Lau, Namasivayam Ambalavanan, Hrishikesh Chakraborty, Martha S.

Extremely Low Birth Weight and Infant Mortality Rates in the United States  

ServicesUpdated Information &

/content/131/5/855.full.htmlincluding high resolution figures, can be found at:

References

/content/131/5/855.full.html#ref-list-1at:This article cites 16 articles, 9 of which can be accessed free

Citations /content/131/5/855.full.html#related-urls

This article has been cited by 2 HighWire-hosted articles:

Subspecialty Collections

/cgi/collection/maternal_fetal_medicine_subMaternal and Fetal Medicine

/cgi/collection/gynecology_subGynecology

/cgi/collection/neonatology_subNeonatology

/cgi/collection/fetus:newborn_infant_subFetus/Newborn Infantthe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

/site/misc/Permissions.xhtmltables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints /site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by guest on April 29, 2016Downloaded from

DOI: 10.1542/peds.2012-2471; originally published online April 1, 2013; 2013;131;855Pediatrics

Wingate and Waldemar A. CarloCarissa Lau, Namasivayam Ambalavanan, Hrishikesh Chakraborty, Martha S.

Extremely Low Birth Weight and Infant Mortality Rates in the United States  

  /content/131/5/855.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by guest on April 29, 2016Downloaded from