Original Article

Journal of Perinatology (2005) 25, 631–637. doi:10.1038/sj.jp.7211378; published online 18 August 2005

Low Birth Weight and Preterm Births: Etiologic Fraction Attributable to Prenatal Drug Exposure

Henrietta S Bada MD, MPH1, Abhik Das PhD2, Charles R Bauer MD3, Seetha Shankaran MD4, Barry M Lester PhD5, Charlotte C Gard MS2, Linda L Wright MD6, Linda LaGasse PhD5 and Rosemary Higgins MD6

  1. 1Department of Pediatrics (H.S.B.), College of Medicine and College of Public Health, University of Kentucky, Lexington, KY, USA
  2. 2Research Triangle Institute International (A.D., C.C.G.), Research Triangle Park, NC, USA
  3. 3Department of Pediatrics (C.R.B.), University of Miami, Miami, FL, USA
  4. 4Department of Pediatrics (S.S.), Wayne State University, Detroit, MI, USA
  5. 5Department of Pediatrics (B.M.L., L.L.), Brown University Medical School, Providence, RI, USA
  6. 6National Institute of Child Health and Human Development (L.L.W., R.H.), National Institutes of Health, Bethesda, MD, USA

Correspondence: HS Bada, MD, MPH, Department of Pediatrics, University of Kentucky Chandler Medical Center, Room MS477, 800 Rose Street, Lexington, KY 40536, USA





To determine the factors that would increase the likelihood of outcomes: low birth weight (LBW), preterm births and intrauterine growth restriction (IUGR).



Secondary data analysis from a multi-center study. Risk factors for each outcome were derived from logistic regression models. Odds ratios (OR), 95% confidence intervals, and population-attributable risk proportions (PAR%) were estimated.



Prenatal cocaine exposure increased the likelihood of LBW (OR: 3.59), prematurity (OR: 1.25), and IUGR (OR: 2.24). Tobacco, but not marijuana, significantly influenced these outcomes. Alcohol had an effect on LBW and IUGR. Etiologic fractions (PAR%) attributable to tobacco for LBW, prematurity, and IUGR were 5.57, 3.66, and 13.79%, respectively. With additional drug exposure including cocaine, estimated summary PAR% increased to 7.20% (LBW), 5.68% (prematurity), and 17.96% (IUGR).



Disease burden for each outcome increases with each added drug exposure; however, etiologic fraction attributable to tobacco is greater than for cocaine.



Low birth weight (LBW) and prematurity rates serve as indicators of our nation's health. In contrast to the marked decline in neonatal mortality,1 the rates of LBW and preterm births continue to increase.2 Furthermore, the disparities persist in the rates between black and white infants. Disorders related to these health indicators also account for the top causes of neonatal mortality.1

LBW infants born near term or at term may show characteristics of intrauterine growth restriction (IUGR); that is, low weight or small for gestational age, and IUGR accounts for a large proportion of LBW infants. Reported causes of IUGR include maternal medical conditions, pregnancy disorders, and fetal complications, factors that may also result in preterm births. A major risk factor for preterm birth is infection of the genital tract, but its treatment may delay but not necessarily prevent preterm delivery.3 Social morbidities have also been associated with LBW, prematurity, and IUGR; the high prevalence of drug use during pregnancy renders it important to address the influence of drug use on these outcomes.4, 5 In the US in 2002, an estimated 405,000 women or 17.3% of those pregnant between the ages of 15 to 44 years reported smoking during pregnancy.6 An estimated 213,000 (9.1%) drank alcohol while pregnant and 77,000 (3.3%) used illicit drugs.6

We previously reported that prenatal cocaine, tobacco, or alcohol use during pregnancy was associated with a decrease in birth weight evident after 32 weeks' gestation.7 Therefore, it would be important to assess the extent to which prenatal drug use would increase the risk of LBW, prematurity, and IUGR. Using the large database of the Maternal Lifestyle Study (MLS),8 it was our objective to determine the risk factors for each of these outcomes. We hypothesized that legal and illegal drug use during pregnancy would increase the likelihood of LBW, prematurity, and IUGR, and that the combination of legal and illegal drug use would explain a significant etiologic fraction of these outcomes. From the public health perspective, estimation of the fraction of disease risk that could be eliminated for outcomes LBW, prematurity, and IUGR, especially if a combination of risk factors can be prevented, would help direct and support preventive efforts to achieve the goals of Healthy People 2010.9



The MLS is conducted in four centers of the National Institute of Child Health and Human Development Neonatal Research Network (Brown University, Providence, RI; University of Miami, Miami, FL; The University of Tennessee, Memphis, TN; and Wayne State University, Detroit, MI). The Research Triangle Institute, Research Triangle Park, NC, serves as the data-coordinating center, a role assumed in the early phase of the study by the Biostatistics Coordinating Center of George Washington University, Washington, DC. The study had approval from the Institutional Review Board of each participating institution, and informed consent was obtained if enrollment criteria were met. Mother–infant dyads were recruited if an infant was from a singleton birth, and had birth weight greater than or equal to500 g and gestational age <43 weeks. We have reported previously a detailed description of study methods and subjects.8

Mothers had a brief interview regarding tobacco, alcohol, and drug use during pregnancy. Tobacco use was categorized into no smoking, <1/2 pack/day, or greater than or equal to1/2 pack/day, and alcohol consumption into no alcohol, <1 drink/month, 1–3 drinks/month, or heavy use (greater than or equal to1 drink/week), and whether >5 drinks were consumed on any given day (binge). Mothers were also asked about their pre-pregnancy weight, date of first prenatal care (PNC), and number of visits. Data were abstracted from maternal and infant medical records. We dichotomized PNC into intermediate/no PNC and adequate PNC, based on Kessner's Index.10 We defined abnormal reproductive history as parity >5 or previous preterm births or abortions >2. Infants were considered cocaine-exposed if mothers admitted to cocaine use during pregnancy or meconium analysis yielded cocaine metabolites. Cocaine-negative infants were those born to mothers who denied cocaine use, with confirmation by a negative meconium analysis. Opiate exposure was similarly determined.

We evaluated outcome measures: LBW and prematurity, since these are the routinely monitored maternal and child health indicators nationwide. Also, attention to these outcomes has increased because of the reported association between birth weight and the origins of adult onset diseases. LBW was defined as birth weight <2500 g. Prematurity referred to births before 37 completed weeks' gestation based on the best obstetric estimate. We also evaluated IUGR as an outcome since it can account for a large number of LBW among those born near-term or at-term gestation. IUGR was defined as birth weight less than 10th percentile for a specific gestational age based on gender-specific US national reference for fetal growth.11 For this analysis, we considered gestational age determined from the best obstetric estimate.

Statistical Methods

Bivariate analyses by t-test or chi2 test compared maternal and infant characteristics between LBW and non-LBW, preterm and non-preterm, and IUGR and non-IUGR groups. Logistic regression (LR) models were used to obtain adjusted odds ratios (OR) and 95% confidence intervals (CI) for each risk factor associated with each outcome. Potential risk factors evaluated included: maternal medical and obstetric complications, any hospitalization during pregnancy, maternal weight gain during pregnancy, PNC, maternal age, Medicaid insurance, and infant's gender, and race. In the LR models, we always adjusted for clinical site, and legal and illegal drug use (tobacco, marijuana, alcohol, cocaine, and opiates). Based on preliminary analysis, an interaction term between cocaine and gestational age (weeks) was included in the models for outcomes, LBW, and IUGR. A backward selection algorithm with an exit P-value greater than or equal to0.10 was used to eliminate those variables that were not statistically significant.

On risk factors that may be amenable to interventions, we estimated the population-attributable fraction or population-attributable risk proportion (PAR%).12 Summary PAR% for a combination of risk factors was also estimated. Adjusted PAR% estimates were derived for each outcome using an approach adapted from Kooperberg and Petitti.13 The technique relies on the LR model and for each risk factor, provides a PAR% estimate that is adjusted for all the other factors in the model. In addition, we developed a bootstrap approach to derive approximate 95% CI for each PAR% estimate.14 Statistical analyses were carried out using SAS statistical software (SAS Institute Inc., Carey, NC).



Of 19,079 maternal–infant dyads screened for enrollment, 16,988 (89%) met eligibility criteria. Of those eligible, consent was obtained in 11,811 (70%). A negative maternal history of cocaine or opiate use was unconfirmed in 3174 because of insufficient or no meconium collected. These dyads were excluded from analysis and their characteristics are compared in Table 1 to the remaining 8637 dyads included in the analysis. LBW, prematurity, and IUGR rates were comparable between those excluded and included in the analysis. Additionally, 10 were excluded in the analysis of prematurity and IUGR because of missing gestational age. Gender was unknown in two infants; they were excluded in the analysis of IUGR. In all, 2032 infants had birth weight <2500 g, 2128 were born preterm, and 1324 had IUGR.

Table 2 shows maternal and infant characteristics of the LBW, non-LBW, preterm, non-preterm, IUGR, and non-IUGR groups. The LBW, preterm, and IUGR groups had significantly higher proportions of mothers on Medicaid, with intermediate or no prenatal care, who smoked cigarettes, had greater than or equal to1 drink/week, and used cocaine, opiates, or marijuana during pregnancy compared to their counterparts. We noted higher proportions of black infants in the LBW, preterm, and IUGR groups as well. The proportion of female infants in the LBW group was higher than in the non-LBW group. There were no gender differences between preterm and non-preterm or between IUGR and non-IUGR groups.

Risk Factors Associated with LBW

The results of the LR models are shown in Table 3. Black and female infants were more likely to be born LBW compared, respectively, to white and male infants. Chronic hypertension, pre-eclampsia, abnormal amniotic fluid volume, and history of tuberculosis were significantly associated with increased risk of LBW. Median maternal weight gain in our population was 29 lb, consistent with the published targeted mean prenatal weight gain of 30 lb (range: 25 to 35 lb) at term gestation.15 Maternal weight gain during pregnancy moderated the risk for LBW; a 10-lb weight gain would make it around 12 times less likely for a child to be born with LBW. The risk of LBW decreased with increasing gestational age, with the odds for LBW decreasing by more than half for each week increase in gestational age among those unexposed to cocaine. Increasing gestational age, however, magnified the cocaine effect, with likelihood of LBW increasing among cocaine-exposed infants at higher gestational age. Marijuana was not associated with LBW; opiate had some effect. Smoking greater than or equal to1/2 pack/day and heavy alcohol use during pregnancy increased the LBW risk. Table 4 shows that the etiologic fraction of LBW attributable to tobacco was 5.57% (95% CI 3.06, 7.91%). Estimated summary PAR% was 7.18% (95% CI 4.16, 9.92%) when alcohol use was added, and changed minimally with addition of cocaine and opiates.

Risk Factors Associated with Prematurity

Table 3 also shows the adjusted OR (95% CI) for the significant risk factors associated with prematurity. Black infants were more likely than white infants to be born at preterm gestation. Except for chronic hypertension, maternal disorders associated with LBW also increased the risk of preterm delivery. Maternal lifestyle choices, including cocaine use, smoking, and intermediate or no PNC, significantly increased the risk of prematurity. Alcohol, marijuana, and opiate did not increase the risk of prematurity. Alcohol consumption of <1 drink a month was associated with decreased risk of prematurity. At its median value (29 lb), maternal weight gain had a similar magnitude of effect in decreasing the risk of prematurity, as was noted with LBW. The estimated etiologic fraction of prematurity attributable to maternal tobacco use was 3.66%. This estimate increased to 5.62% with addition of cocaine (Table 4).

Risk Factors Associated with IUGR

Factors significantly associated with IUGR are also shown in Table 3. Being small for gestational age was more likely to occur at later gestation. Black infants were almost 1.6 times more likely to be born IUGR than white infants. Maternal disorders associated with LBW were also significant risk factors for IUGR. Cocaine, heavy alcohol, and any tobacco use increased the risk of IUGR, with the effect of cocaine increasing at later gestational ages. There was a suggestion of an opiate effect on IUGR as well, with an OR (95% CI) of 1.43 (0.97, 2.09). The magnitude of effect of maternal weight gain during pregnancy is similar to its effect on prematurity.

If tobacco use during pregnancy were prevented, the estimated etiologic fraction of IUGR eliminated would be 13.79% (95% CI 10.08, 17.33%). Table 4 shows that this summary PAR% estimate increased slightly with addition of alcohol, and went up to 17.96% (95% CI 13.43, 22.06%) upon adding cocaine and opiates.



Most studies have shown an association between gestational cocaine or opiate exposure and outcomes: LBW, prematurity, and IUGR. Shiono et al.16 found no association between cocaine use and LBW or prematurity; however, the prevalence of cocaine use in their population was only 2%. Our findings of smaller etiologic fractions attributable to cocaine and opiates for the outcomes we evaluated may be explained by a lower rate of illegal drug use compared to rates of tobacco and alcohol use in our population. However, the effects of antenatal cocaine or opiate exposure should not be minimized. Infants exposed to cocaine and or opiates in utero are more likely to have medical complications, neurological manifestations, central and autonomic nervous systems signs, and abnormal neurobehavioral findings.17, 18 Long-term outcomes have included language problems and lower IQ scores than comparison children.19 Additionally, significant economic, legal, and social issues surround illegal drug use during pregnancy.20

The reported association between alcohol and prematurity is less strong than the association between alcohol and LBW or IUGR.21 Moreover, we found that low alcohol consumption during pregnancy was associated with a lower risk of prematurity. McDonald et al.22 also found a significant but slight decreased risk of LBW, prematurity, and IUGR, with small alcohol use relative to abstainers. However, these findings should not be translated to imply support for alcohol use during pregnancy, since prenatal alcohol use is significantly associated with fetal alcohol syndrome and a spectrum of alcohol-related effects on the developing brain, besides LBW and IUGR.

More women smoke while pregnant than those who drink alcohol or use illegal drugs.6 A high prevalence of smoking remains even with existing reports on the teratogenic effects of tobacco on the developing fetal brain, manifested as abnormal neonatal neurological signs,17 abnormal behavior in infancy and childhood, and impaired cognitive development.23 Our estimates on the etiologic fractions of LBW, prematurity, and IUGR, that are attributable to tobacco support health policies, directed to assuring that states cover tobacco cessation and treatment programs for pregnant women and prevention programs for children and their families. Furthermore, policies need to consider the link between strategies for tobacco prevention/cessation and PNC. Women are more likely to quit smoking, with intervention initiated early in pregnancy. PNC could also impact not only prematurity but also LBW. When PNC was accessed by cocaine users, improvement in birth weight was noted.24

Reports show higher rates of LBW and prematurity in black women compared to white women.2 We found that black mothers were more likely to have infants with LBW, preterm gestation, or IUGR. Also, higher rates of medical risk factors, for example, eclampsia, hematologic disorders, abnormal amniotic fluid volume, and chronic hypertension, are reported in births from black women compared to white women.2 We found these same medical factors to be associated with LBW, prematurity, or IUGR. The unclear etiology of most of these medical conditions renders it difficult to institute specific preventive measures, and limits the significance of estimating the etiologic fraction of each outcome attributable to medical conditions. However, early and periodic PNC could screen for these disorders and timely management can minimize their impact on LBW and prematurity. Early PNC provides the opportunity for initiating drug prevention and treatment, and for pregnant women to receive nutritional counseling. Maternal nutrition can impact birth weight; maternal weight gain during pregnancy has been positively correlated with birth weight.25 In our population, we did find weight gain during pregnancy to be associated with the outcomes we studied.

Our study differs from other studies in that the large number enrolled enabled us to determine not only the impact of licit and illicit drugs on LBW, prematurity, and IUGR but also control for multiple confounders, not possible with smaller studies. With the large number, we were able to determine the effect of each prenatal drug exposure, independent of other substance exposures. Also different from previous studies is our determination of summary PAR%, which combines the risks attributable to poly drug use. In the MLS, we defined the variables at the onset and data were obtained with training, certification, and masking of research staff to the infant's drug exposure status. Illegal drug use was confirmed by a highly sensitive meconium testing.

As to the limitations, we did not obtain biomarkers for exposure to tobacco or alcohol; this may result in misclassification. However, patients are less likely to deny use of legal substances such as tobacco and alcohol. Moreover, investigators26 have found antenatal and post-partum maternal reports on substance use during pregnancy to be reasonably accurate. Retrospective reports on smoking are consistent with antenatal report, and there is no systematic tendency for women to understate the amount of alcohol used during pregnancy.26 Also, we found in our population that, even with illicit drug use, the probability of confirmation for cocaine in meconium was only 0.02, given that the mother denied use on interview.8

Our definition of IUGR was based on birth weight for a specific gestational age, rather than from antenatal evaluation of fetal growth parameters. The large number in MLS precluded a systematic prenatal monitoring of fetal growth and a high proportion of mothers, especially in the exposed cohort, had limited or no prenatal care.

Our population was drawn from four urban sites, with a large number of very LBW infants, and not necessarily a representative population, limiting the generalizability of our findings. We did enroll a large percentage of black mothers and their children, with a large proportion on Medicaid insurance. Evaluating outcomes in our study population provided an added perspective on racial disparities in LBW and prematurity rates. Because of enrollment from only singleton births, it was not possible to determine the impact of multiple births on the outcomes evaluated. The exclusion from analysis of maternal–infant dyads having lower rates of smoking, marijuana, and alcohol use, and a higher proportion with adequate prenatal care, may have resulted in an overestimation of the impact of legal and illegal drug use on the outcomes we studied. However, differences between groups in percentages of marijuana, tobacco, and alcohol use were small, and there were no significant differences in rates of LBW, prematurity, or IUGR between those excluded and included in the analysis.

Our use of PAR% does not connote any projection to national estimates. It provided us an idea of the magnitude of reduction in these adverse birth outcomes that may be achieved in predominantly urban, minority, and low SES populations such as the one we have for this study. In such disadvantaged populations with higher rates of LBW, prematurity, and IUGR, and where disparities in these outcomes are most persistent, we believe that the PAR% estimates computed here for risk factors amenable to interventions are of significant relevance to public health-policy makers and practitioners.

In summary, a significant burden of LBW, prematurity, and IUGR is attributable to legal and illegal drug use during pregnancy. The increase in magnitude of the impact on each outcome with the combination of these risk factors supports the need for a multi-faceted approach in health promotion and prevention programs. Public health programs directed to smoking prevention and cessation must continue, or perhaps be expanded, with ongoing monitoring for their effectiveness, if we are to reverse the trends in LBW and preterm birth rates.



  1. Minino A, Arias E, Kochanek K, Murphy S & Smith B. Deaths: Final Data for 2000 Hyattsville, MD: National Center for Health Statistics 2002;.
  2. Martin J, Hamilton B, Ventura S, Menacker F, Park M & Sutton P. Births: Final Data for 2001 Hyattsville, MD: National Center For Health Statistics 2002;.
  3. Goldenberg RL, Hauth JC & Andrews WW. Intrauterine infection and preterm delivery. N Engl J Med 2000; 342: 1500−1507. | Article | PubMed | ISI | ChemPort |
  4. Zuckerman B, Frank DA & Hingson R et al. Effects of maternal marijuana and cocaine use on fetal growth. N Engl J Med 1989; 320: 762−768. | PubMed | ChemPort |
  5. Eyler FD, Behnke M, Conlon M, Woods NS & Wobie K. Birth outcome from a prospective, matched study of prenatal crack/cocaine use: I. Interactive and dose effects on health and growth. Pediatrics 1998; 101: 229−237. | Article | PubMed | ChemPort |
  6. SAMSHA. Overview of Findings from the 2002 National Survey on Drug Use and Health Rockville, MD: Office of Applied Statistics 2003;.
  7. Bada HS, Das A & Bauer CR et al. Gestational cocaine exposure and intrauterine growth: maternal lifestyle study. Obstet Gynecol 2002; 100: 916−924. | Article | PubMed |
  8. Lester BM, ElSohly M & Wright LL et al. The maternal lifestyle study: drug use by meconium toxicology and maternal self-report. Pediatrics 2001; 107: 309−317. | Article | PubMed | ChemPort |
  9. Healthy People 2010. With Understanding and Improving Health and Objectives for Improving Health United States Department of Health and Human Services. Washington, DC: US Government Printing Office 2000;.
  10. Kessner D. Infant Death: An Analysis by Maternal Risk and Health Care Washington, DC: Institute of Medicine National Academy of Sciences 1973;.
  11. Alexander GR, Himes JH, Kaufman RB, Mor J & Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996; 87: 163−168. | Article | PubMed | ChemPort |
  12. Northridge ME. Public health methods—attributable risk as a link between causality and public health action. Am J Public Health 1995; 85: 1202−1204. | PubMed | ChemPort |
  13. Kooperberg C & Petitti DB. Using logistic regression to estimate the adjusted attributable risk of low birthweight in an unmatched case−control study. Epidemiology 1991; 2: 363−366. | PubMed | ChemPort |
  14. Efron B & Tibshirani R. An Introduction to the Bootstrap New York, NY: Chapman & Hill/CRC 1998;.
  15. Nutrition during pregnancy. Institute of Medicine Committee on Nutritional Status During Pregnancy and Lactation Washington, DC: National Academy Press 1990;.
  16. Shiono PH, Klebanoff MA & Nugent RP et al. The impact of cocaine and marijuana use on low birth weight and preterm birth: a multicenter study. Am J Obstet Gynecol 1995; 172: 19−27. | Article | PubMed | ChemPort |
  17. Bada HS, Bauer CR & Shankaran S et al. Central and autonomic system signs with in utero drug exposure. Arch Dis Child Fetal Neonatal Ed 2002; 87: F106−F112. | Article | PubMed | ChemPort |
  18. Eyler FD, Behnke M, Conlon M, Woods NS & Wobie K. Birth outcome from a prospective, matched study of prenatal crack/cocaine use: II. Interactive and dose effects on neurobehavioral assessment. Pediatrics 1998; 101: 237−241. | Article | PubMed | ChemPort |
  19. Singer LT, Arendt R, Minnes S, Salvator A, Siegel AC & Lewis BA. Developing language skills of cocaine-exposed infants. Pediatrics 2001; 107: 1057−1064. | Article | PubMed | ChemPort |
  20. DeVille KA & Kopelman LM. Moral and social issues regarding pregnant women who use and abuse drugs. Obstet Gynecol Clin North Am 1998; 25: 237−254. | Article | PubMed | ChemPort |
  21. Windham GC, Fenster L, Hopkins B & Swan SH. The association of moderate maternal and paternal alcohol consumption with birthweight and gestational age. Epidemiology 1995; 6: 591−597. | PubMed | ChemPort |
  22. McDonald AD, Armstrong BG & Sloan M. Cigarette, alcohol, and coffee consumption and prematurity. Am J Public Health 1992; 82: 87−90. | PubMed | ChemPort |
  23. Weitzman M, Byrd RS, Aligne CA & Moss M. The effects of tobacco exposure on children's behavioral and cognitive functioning: implications for clinical and public health policy and future research. Neurotoxicol Teratol 2002; 24: 397−406. | Article | PubMed | ChemPort |
  24. Racine A, Joyce T & Anderson R. The association between prenatal care and birth weight among women exposed to cocaine in New York City. JAMA 1993; 270: 1581−1586. | Article | PubMed | ChemPort |
  25. Scholl TO, Hediger ML, Ances IG, Belsky DH & Salmon RW. Weight gain during pregnancy in adolescence: predictive ability of early weight gain. Obstet Gynecol 1990; 75: 948−953. | PubMed | ChemPort |
  26. Jacobson SW, Chiodo LM, Sokol RJ & Jacobson JL. Validity of maternal report of prenatal alcohol, cocaine, and smoking in relation to neurobehavioral outcome. Pediatrics 2002; 109: 815−825. | Article | PubMed |


The study was carried out with support from NIH National Institute of Child Health and Human Development through cooperative agreements and interagency agreement with the National Institute on Drug Abuse (NIDA), Administration on Children, Youth and Families (ACYF), and Center for Substance Abuse Treatment (CSAT). Participating Institutions, grant awards, investigators, and key research personnel include: Brown University, U10 HD 27904, N01-HD-2-3159 (Barry M. Lester, PhD; Susan Schibler, RN; Melissa Ambrosia, RN; Linda LaGasse, PhD); University of Miami, U10 HD 21397 (Charles R. Bauer, MD; Wendy Griffin, RN; Elizabeth Jacque, RN); University of Tennessee, U10 HD 21415 (Henrietta S. Bada, MD; Marilyn Williams, MSW; Tina Hudson, BSN), Wayne State University, U10 HD 21385 (Seetha Shankaran, MD; Eunice Woldt, MSN, Jay Ann Nelson, BSN), RTI, International, U01 HD 36790 (W. Kenneth Poole, PhD; Abhik Das, PhD; Jane Hammond, PhD); NICHD (Linda L. Wright, MD; Rosemary Higgins, MD) and NIDA (Vincent L. Smeriglio, PhD).