FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |   Search Term  Advanced Search

Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe

Previous Volume 329:850-854 September 16, 1993 Number 12 Next

Abstract Letters Add to Personal Archive Add to Citation Manager Notify a Friend E-mail When Cited PubMed Citation

ABSTRACT

Background Perinatal substance exposure has been linked to many neonatal and obstetrical complications. There have been few population-based epidemiologic studies to identify the prevalence and demographic profiles associated with drugs, alcohol, and smoking during pregnancy.

Methods We studied a population sample selected according to a multistage probability sampling design to estimate the prevalence of perinatal substance exposures in California in 1992. Urine samples from 29,494 women presenting for delivery in 202 hospitals were coded and screened for toxins; the results of toxicology screening were later linked by code number to the subjects' demographic variables and their reported use of tobacco and prescribed drugs. Urinary toxicologic tests provide conservative estimates because they can detect only very recent substance use.

Results The weighted prevalence for perinatal substance exposure was 5.16 percent for the use of one or more drugs, 6.72 percent for alcohol (analyzed independently), and 8.82 percent for self-reported smoking. The percentage of women testing positive for any drug, including alcohol, was 11.35 percent. Estimates for racial and ethnic groups varied widely. Black women had the highest prevalence of total drug use (14.22 percent), alcohol use (11.58 percent), cocaine use (7.79 percent), and tobacco use (20.12 percent). Most drug exposures occurred among white non-Hispanic and Hispanic women. White non-Hispanic women had the second highest prevalence rate for the use of one or more drugs (6.79 percent) and self-reported tobacco use (14.82 percent). Hispanic women had the second highest prevalence rate for alcohol (6.87 percent).

Conclusions In California in 1992, there were 67,361 estimated perinatal exposures to one or more drugs, including alcohol, and 52,346 self-reported exposures to tobacco. These findings have clinical and public health implications.

Few population-based epidemiologic studies have accurately identified the prevalence and demographic patterns associated with licit and illicit drug use during pregnancy^6,7. In South Carolina, a recent study indicated that 12.1 percent of women who gave birth were using drugs or alcohol, as determined by urine testing⁸. In Rhode Island, where peripartum urine samples were tested only for illicit drugs, the prevalence rate was 7.5 percent⁹. Other prevalence studies, including those from which national estimates have been derived, have involved a single county or one or a few hospitals. These studies have usually been limited to minority-group women of low socioeconomic status¹⁰. The many small studies that have been conducted in large, urban hospitals that serve poor and ethnically diverse populations have found high rates of licit and illicit drug use^11,12,13. Variability in study design and drug-testing methods makes it difficult to compare these findings. Large population studies of representative samples are required to produce reliable estimates of perinatal substance exposure.

Methods

Study Sample

The 29,494 subjects in this study were selected according to a multistage probability sampling design. Data were collected from March through October 1992. The population characteristics used in our analyses were based on those of women delivering 607,000 infants in California between July 1, 1991, and June 30, 1992. A statewide list of 297 maternity hospitals with 10 or more deliveries during the previous fiscal year constituted the sampling frame of 593,487 births (98 percent of births statewide). Because of design and feasibility considerations, birthing centers, federal hospitals, and hospitals performing deliveries only on an emergency basis were not included.

Hospitals were arrayed according to size in each stratum (i.e., large-county or small-county grouping), and 202 hospitals were selected by random sampling. Ten hospitals refused to participate; each was replaced by a hospital with a comparable size, patient racial or ethnic composition, and type of ownership. The number of subjects sampled from each hospital selected was directly proportional to the number of births within the hospital.

To adjust for any disproportionalities due to slight oversampling or undersampling according to the size or patient racial or ethnic composition, values were weighted to conform the outcomes to the racial and ethnic distributions of births in each hospital in the period 1991-1992. For statewide estimates, appropriate weights were applied to conform the total sample to the racial and ethnic distribution of births in the state in 1991-1992. These procedures ensured that the study sample was representative of the defined population. Since sampling took place primarily in the spring and summer, possible seasonal fluctuations in drug use would not be detected, as reported elsewhere¹³.

Subjects were selected in a manner designed to minimize selection bias and to ensure the anonymity of those from whom urine specimens were obtained. Starting on a given day, nurses were instructed to test all admitted patients until the sampling fraction for the hospital was met. Nurses collected urine specimens and basic descriptive information from each subject and recorded this information on a code sheet that contained no personally identifying information. The same code number was used on the code sheet and the urine-specimen label.

In accordance with national standards of nursing care, all patients in California hospitals are asked at the time of admission whether they currently smoke. There is no standard phrasing for this question. The answer does not indicate the frequency or extent of smoking. Information on patients' smoking was recorded on the code sheets; it was missing for 6.4 percent of the subjects in the sample. The direction of bias, if any, due to the missing data could not be judged.

Procedures and safeguards were established to ensure that no personally identifying information could be linked to the results of urine testing. The research protocol was reviewed and approved by the Committee for the Protection of Human Subjects, Health and Welfare Agency, State of California, and by the Human Subjects Review Committee of the University of California, Berkeley, as well as by institutional review boards at the individual hospitals.

Urine Testing

The urine specimens were sent to a laboratory certified by the National Institute of Drug Abuse (PharmChem, Menlo Park, Calif.). Table 1 lists the tests performed and the detection periods for each substance. Each specimen was assayed by personnel who did not know its origin or the subject's demographic characteristics. They used enzyme-multiplied immunoassay techniques for all drugs and an enzymatic assay for alcohol (test materials for these assays were manufactured by Syva, a subsidiary of Syntex, Palo Alto, Calif.). If a screening test was negative -- i.e., if it failed to detect the presence of drugs or alcohol -- the test results were reported to be negative and no further testing was conducted. If an enzyme-multiplied immunoassay was positive for any drug except a cannabinoid (marijuana), the result was confirmed by gas chromatography; this technique is commonly used by analytical toxicologists to confirm the presence of drugs or their metabolites in urine or other biologic fluids. If an assay was positive for a cannabinoid, the result was confirmed by high-performance thin-layer chromatography; the result of this technique correlates highly with that of gas chromatography, and the sensitivity of the test is comparable. When an immunoassay is combined with an appropriate confirmation assay that is chemically independent, the likelihood that a drug will be correctly identified is greater than 99 percent.

View this table: [in this window] [in a new window]   Table 1. Drug and Alcohol Testing Procedures.

 
The pharmacologic characteristics of alcohol differentiate it from other drugs. Its concentration in blood, breath, and urine can be estimated according to body weight if the amount ingested and the time elapsed are known; alcohol is rapidly absorbed into the bloodstream and distributed throughout body water. If an average-sized person in good health drinks 6 oz of beer (170 ml), 2 oz of wine (60 ml), or 0.5 oz of distilled spirits (15 ml), urine collected 1 to 1 1/2 hours later should contain at least 10 mg of alcohol per deciliter, a level used as the cutoff value for this study. A person who has one or two drinks at night and urinates after awakening in the morning would have a lower urinary alcohol level and not test positive. A person who consumes at least 1 oz of distilled spirits 2 to 2 1/2 hours before urine collection would test positive. Alcohol cannot be measured accurately in urine specimens containing glucose; to avoid confounding, we considered such specimens to be negative for alcohol.

Our testing methods had limitations. Urine toxicology screening is commonly used in studies of this nature because it is less expensive than other biologic-marker tests, such as analysis of meconium, and more accurate than self-reports of drug use. The sensitivity of urine testing is constrained by the limited detection period for each substance. Therefore, the actual prevalence of drug or alcohol use over the course of a pregnancy is likely to be underestimated. When other testing methods are used, such as meconium analysis, which has a longer detection period, prevalence estimates are likely to differ¹⁴.

Statistical Analysis

Calculations of standard errors took the sampling design into account by weighting values to conform the data to the distributions of births for the period 1991-1992 within hospitals, within regions, and statewide. The 95 percent confidence interval for each subgroup reflected the observed percentage of positive tests and the standard error of the estimate. To facilitate subgroup comparisons, differences between proportions were tested, with the Student-Newman-Keuls ranges adjustment (a two-tailed test) for multiple comparisons, in analyses of prevalence according to racial or ethnic group and the duration of prenatal care. The comparisons of the prevalence of tobacco use included the exact t-test value and probability for each substance.

Results

We present three types of data that profile drug use among pregnant women in California: prevalence data, estimates of the total number of perinatal drug exposures, and the percentage of subjects using a particular drug who belonged to each racial or ethnic group. Table 2 shows weighted statewide prevalence estimates for the sample as a whole and for each racial or ethnic group, including standard errors and the results of statistical tests of group differences. The overall estimate (Table 2, "any drug") is the percentage of subjects testing positive for one or more illicit or licit drugs listed in Table 1 that were not prescribed by a physician. Estimates for alcohol use and self-reported tobacco use are shown separately, as are those for marijuana, cocaine, opiates, and amphetamines that were used with relative frequency; estimates for other substances, such as barbiturates, benzodiazepines, methadone, and phencyclidine, are not shown separately because of their negligible prevalence.

View this table: [in this window] [in a new window]   Table 2. Prevalence of Perinatal Drug Exposures, According to Racial or Ethnic Group (California, March through October 1992).

 
The overall estimate of the statewide prevalence of perinatal drug use was 5.16 percent, and the estimated prevalence of alcohol use was 6.72 percent. The prevalence of positive tests for one or more drugs, including alcohol, was 11.35 percent. Only about 0.5 percent of pregnant women tested positive for more than one drug. The prevalence of tobacco use as measured by the subjects' reports was 8.82 percent.

There were wide variations in prevalence estimates for racial or ethnic groups. For example, 14.22 percent of black women were positive for one or more drugs, 11.58 percent were positive for alcohol, and 20.12 percent were positive for self-reported tobacco use -- rates significantly higher (P<0.05) than those for any other racial or ethnic group. For cocaine use, 7.79 percent of black women tested positive, as compared with 0.6 percent or less of women from the other racial or ethnic groups. White non-Hispanic women had the second highest prevalence estimate for the use of one or more drugs, 6.79 percent, and tobacco use, 14.82 percent. Hispanic women had the second highest prevalence estimate for alcohol, 6.87 percent. In general, Asian women had lower prevalence estimates for drug and tobacco use than did women from the other racial or ethnic groups.

We also calculated prevalence estimates for perinatal substance exposure in relation to maternal smoking and the duration of prenatal care. Women who did not receive prenatal care had a higher prevalence estimate for the use of all drugs than women who received care (Table 3). Prenatal care that began during the first trimester of pregnancy was associated with lower prevalence estimates for the use of most drugs, especially cocaine (0.46 percent). One in four women with no prenatal care tested positive for any drug; 11.50 percent tested positive for cocaine, 6.98 percent for amphetamines, and 5.72 percent for marijuana, and 28.61 percent reported tobacco use. All differences between the prenatal-care groups reported in Table 3 were significant (P<0.05).

View this table: [in this window] [in a new window]   Table 3. Prevalence of Perinatal Drug Exposures, According to Duration of Prenatal Care (California, March through October 1992).

 
Women who reported that they smoked during pregnancy had higher prevalence rates (P<0.001) for all drugs -- approximately 10 times higher for marijuana, 22 times higher for cocaine, and 21 times higher for amphetamine use -- than women who did not smoke (Table 4). Smoking has been found to be a risk factor for drug use among pregnant women¹¹.

View this table: [in this window] [in a new window]   Table 4. Prevalence of Perinatal Drug Exposures, According to Self-Reported Tobacco Use during Pregnancy (California, March through October 1992).

 
Rates for racial or ethnic groups within exposure categories are shown in Figure 1. A group may have a low prevalence rate yet represent a significant proportion of all women testing positive if the group constitutes a large fraction of pregnant women: white non-Hispanic women had a prevalence of 14.82 percent for smoking; they constituted 36.0 percent of the study population and 62.2 percent of all smokers. Black women, who had the highest prevalence rate for any drug, accounted for only 21.3 percent of all women testing positive for any drug. Hispanic and Asian women were proportionately underrepresented with respect to any drug and tobacco use, whereas 45.7 percent of all women testing positive for alcohol were Hispanic. In a separate analysis of the study sample, we found that one of every three women who tested positive for alcohol did not speak English.

View larger version (64K): [in this window] [in a new window]   Figure 1. Study Population and Perinatal Exposure to Alcohol, Drugs, and Tobacco, According to Racial or Ethnic Group. Percentages do not add to 100 because potential subjects whose racial or ethnic group was unknown or too small to allow meaningful estimates were excluded from analysis.

 
We used the survey data to estimate the number of childbearing women in California who were likely to have had perinatal substance exposures in 1992, when about 600,000 women gave birth. The totals were 67,361 estimated exposures to one or more drugs, including alcohol, 39,882 exposures to alcohol, 30,565 exposures to licit or illicit drugs other than alcohol, and 52,346 exposures to tobacco, on the basis of self-reports.

Discussion

From this large survey of perinatal substance exposure, we estimate that 11.35 percent of maternity patients at California hospitals in 1992 (approximately 67,000 women) had used a licit or illicit drug or alcohol within hours or days of delivery. About 1 in 11 patients reported that they currently smoked. Among black women, the frequency of substance use was disproportionately high as compared with other groups, especially the use of alcohol, tobacco, and cocaine. These estimates are likely to be quite conservative because urine tests can detect only recent substance use.

Our findings have clinical and public health implications. The high frequency of substance use suggests that communities with large black populations should be targeted for extensive clinical and educational interventions, which should include prenatal counseling and residential treatment programs specifically designed for pregnant women. Educational campaigns aimed at black adolescents to prevent substance use could help reduce the number of infants who are exposed to drugs, alcohol, or tobacco in utero. The opportunity for intervening before adulthood seems especially promising because in early adolescence black teenagers have lower rates of drug use than teenagers in other racial or ethnic groups¹⁵.

Public health interventions should aim to forestall alcohol and tobacco use during pregnancy in all racial and ethnic groups. Much popular and political concern has focused on illicit drug use during pregnancy, especially the use of "crack" cocaine. However, the use of alcohol and tobacco during pregnancy was far more common. Like alcohol, tobacco has often not been addressed in other studies of perinatal substance exposure.

Physicians should describe to women of childbearing age the potential risks of alcohol to fetal growth and development. A large number of Hispanic women in California are immigrants. Many speak primarily Spanish and have difficulty understanding English. The relatively high use of alcohol during pregnancy by persons in this ethnic group suggests that medical care professionals need training in effectively communicating to them information about the risks of alcohol. Hispanic women often visit public prenatal clinics or hospitals, where consultation time is limited. As compared with white non-Hispanic women, they may receive prenatal care later in pregnancy. In addition, they may be unaware of warnings about prenatal exposure to alcohol that are found on the labels of alcoholic beverages or are given by the media or public health sources.

The level of hazard associated with drinking small amounts of alcohol during pregnancy is controversial. According to the American Academy of Pediatrics, no level of exposure to alcohol can be considered safe until more definitive information about the health effects of low levels of exposure on the fetus is available¹⁶. The academy takes this position because of the often irremediable effects of heavy drinking during pregnancy. Fetal alcohol syndrome is one of the most common identifiable causes of mental retardation. Women who drink alcohol in moderation during pregnancy are at increased risk of giving birth to infants with growth retardation. In our study, we reported any woman as positive for alcohol use if she had drunk at least 6 oz of beer, 2 oz of wine, or 0.5 oz of distilled spirits in the period immediately before she was admitted as a maternity patient, or had drunk larger quantities of alcohol more than a few hours before admission.

Smoking was associated with higher prevalence rates for the use of alcohol and all the drugs tested for. Similarly, starting prenatal care late in pregnancy or a lack of prenatal care was associated with illicit drug use, suggesting that personal disorganization, fear of detection of alcohol and drug use, and a lack of health insurance or access to care may be serious barriers to care.

Although targeted interventions are important, balanced public health policies are also needed. Even though black women had the highest rates of perinatal substance use, nonblack women constituted four fifths of all women in California who tested positive for any licit or illicit drug other than alcohol. About three fifths of all smokers were white non-Hispanic women. A careful assessment of epidemiologic findings such as these, especially when this information is linked to specific regions and types of hospitals, can provide a rational basis for program planning, development of clinical interventions, professional training, and public health policy.

Conducted under a contract with the Department of Alcohol and Drug Programs, Office of Perinatal Substance Abuse, State of California, by the School of Public Health, University of California at Berkeley.

We are indebted to Pete Wilson, governor of California, and Andrew Mecca, director, California Department of Alcohol and Drug Programs, for their interest and support; to Pat Porter, Jim Williams, and Tony Bole of the Western Consortium for Public Health for their organization of field data and specimen-gathering activities; and to Mary Brown, California Office of Perinatal Substance Abuse, for her guidance.

Source Information

From the University of California, Berkeley (W.A.V.); San Diego State University, San Diego (B.K.); California State University, San Bernardino (J.H.); and the Western Consortium for Public Health, Berkeley (A.N.) -- all in California.

Address reprint requests to Professor Vega at the School of Public Health, Center for Community Health, Warren Hall, University of California, Berkeley, CA 94720.

References

1. Finnegan LP. Drug addiction and pregnancy: the newborn. In: Chasnoff IJ, ed. Drugs alcohol, pregnancy and parenting. Boston: Kluwer Academic, 1988:59-71. 
2. Oro AS, Dixon SD. Perinatal cocaine and methamphetamine exposure: maternal and neonatal correlates. J Pediatr 1987;111:571-578. [Medline]
3. Dixon SD. Effects of transplacental exposure to cocaine and methamphetamine on the neonate. West J Med 1989;150:436-442. [Medline]
4. Chasnoff IJ, Bussey ME, Savich R, Stack CM. Perinatal cerebral infarction and maternal cocaine use. J Pediatr 1986;108:456-459. [CrossRef][Medline]
5. California vital statistics -- 1991. Sacramento, Calif.: State Department of Health Services, 1992. (Report no. 92-12001.)
6. Chasnoff IJ. Drug use and women: establishing a standard of care. Ann N Y Acad Sci 1989;562:208-210. [Medline]
7. Chasnoff IJ, Landress HJ, Barrett ME. The prevalence of illicit-drug and alcohol use during pregnancy and discrepancies in mandatory reporting in Pinellas County, Florida. N Engl J Med 1990;322:1202-1206. [Abstract]
8. Nalty D. 1991 South Carolina Prevalence Study of Drug Use among Women Giving Birth: report of the South Carolina Commission on Alcohol and Drug Abuse, October 23, 1991. Columbia: South Carolina Commission on Alcohol and Drug Abuse, 1991.
9. Hollinshead WH, Griffin JF, Scott HD, Burke ME. Statewide prevalence of illicit drug use by pregnant women -- Rhode Island. MMWR Morb Mortal Wkly Rep 1990;39:225-227. [Medline]
10. Khalsa JH, Gfroerer J. Epidemiology and health consequences of drug abuse among pregnant women. Semin Perinatol 1991;15:265-270. [Medline]
11. Gillogley KM, Evans AT, Hansen RL, Samuels SJ, Batra KK. The perinatal impact of cocaine, amphetamine, and opiate use detected by universal intrapartum screening. Am J Obstet Gynecol 1990;163:1535-1542. [Medline]
12. Frank DA, Zuckerman BS, Amaro H, et al. Cocaine use during pregnancy: prevalence and correlates. Pediatrics 1988;82:888-895. [Free Full Text]
13. 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. [Abstract]
14. Ostrea EM Jr, Brady M, Gause S, Raymundo AL, Stevens M. Drug screening of newborns by meconium analysis: a large-scale, prospective, epidemiologic study. Pediatrics 1992;89:107-113. [Free Full Text]
15. Vega WA, Zimmerman RS, Warheit GJ, Apospori E, Gil AG. Risk factors for early adolescent drug use in four ethnic and racial groups. Am J Public Health 1993;83:185-189. [Free Full Text]
16. American Academy of Pediatrics Committee on Substance Abuse and Committee on Children with Disabilities. Fetal alcohol syndrome and fetal alcohol effects. Pediatrics 1993;91:1004-1006. [Medline]

Abstract Letters Add to Personal Archive Add to Citation Manager Notify a Friend E-mail When Cited PubMed Citation

Related Letters:

Prevention of Preeclampsia with Low-Dose Aspirin
Moya F. R., Germain A., Sibai B. M.
Extract | Full Text  
N Engl J Med 1994; 330:794-795, Mar 17, 1994. Correspondence

This article has been cited by other articles:

• Kraft, W. K., Gibson, E., Dysart, K., Damle, V. S., LaRusso, J. L., Greenspan, J. S., Moody, D. E., Kaltenbach, K., Ehrlich, M. E. (2008). Sublingual Buprenorphine for Treatment of Neonatal Abstinence Syndrome: A Randomized Trial. Pediatrics 122: e601-e607 [Abstract] [Full Text]  
• Westover, A. N., McBride, S., Haley, R. W. (2007). Stroke in Young Adults Who Abuse Amphetamines or Cocaine: A Population-Based Study of Hospitalized Patients. Arch Gen Psychiatry 64: 495-502 [Abstract] [Full Text]  
• Perreira, K. M., Cortes, K. E. (2006). Race/Ethnicity and Nativity Differences in Alcohol and Tobacco Use During Pregnancy. AJPH 96: 1629-1636 [Abstract] [Full Text]  
• Vega, W. A., Sribney, W. M., Achara-Abrahams, I. (2003). Co-Occurring Alcohol, Drug, and Other Psychiatric Disorders Among Mexican-Origin People in the United States. AJPH 93: 1057-1064 [Abstract] [Full Text]  
• Ventura, S. J., Hamilton, B. E., Mathews, T. J., Chandra, A. (2003). Trends and Variations in Smoking During Pregnancy and Low Birth Weight: Evidence From the Birth Certificate, 1990-2000. Pediatrics 111: 1176-1180 [Abstract] [Full Text]  
• Glik, D., Halpert-Schilt, E., Zhang, W. (2001). Narrowcasting Risks of Drinking during Pregnancy among African American and Latina Adolescent Girls. Health Promot Pract 2: 222-232 [Abstract]  
• Finch, B. K. (2001). Nation of Origin, Gender, and Neighborhood Differences in Past-Year Substance Use among Hispanics and Non-Hispanic Whites. Hispanic Journal of Behavioral Sciences 23: 88-101 [Abstract]  
• Jayakody, R., Danziger, S., Pollack, H. (2000). Welfare Reform, Substance Use, and Mental Health. Journal of Health Politics, Policy and Law 25: 623-652 [Abstract]  
• Robinson, R., Iida, H., O'Brien, T. P., Pane, M. A., Traystman, R. J., Gleason, C. A. (2000). Comparison of cerebrovascular effects of intravenous cocaine injection in fetal, newborn, and adult sheep. Am. J. Physiol. Heart Circ. Physiol. 279: H1-H6 [Abstract] [Full Text]  
• Yu, R. C.-T., Lee, T.-C., Wang, T.C., Li, J.-H. (1999). Genetic toxicity of cocaine. Carcinogenesis 20: 1193-1199 [Abstract] [Full Text]  
• Watson, C. S., White, S. E., Homan, J. H., Kimura, K. A., Brien, J. F., Fraher, L., Challis, J. R. G., Bocking, A. D. (1999). Increased cerebral extracellular adenosine and decreased PGE2 during ethanol-induced inhibition of FBM. J. Appl. Physiol. 86: 1410-1420 [Abstract] [Full Text]  
• Escobar, J. I. (1998). Immigration and Mental Health: Why Are Immigrants Better Off?. Arch Gen Psychiatry 55: 781-782 [Full Text]  
• MacMahon, J. R. (1997). Perinatal Substance Abuse: The Impact of Reporting Infants to Child Protective Services. Pediatrics 100: e1-e1 [Abstract] [Full Text]  
• Kwong, T. C., Ryan, R. M. (1997). Detection of intrauterine illicit drug exposure by newborn drug testing. Clin. Chem. 43: 235-242 [Abstract] [Full Text]  
• Tsay, C. H., Partridge, J. C., Villarreal, S. F., Good, W. V., Ferriero, D. M. (1996). Neurologic and Ophthalmologic Findings in Children Exposed to Cocaine in Utero. J Child Neurol 11: 25-30 [Abstract]  
• Yawn, B. P., Thompson, L. R., Lupo, V. R., Googins, M. K., Yawn, R. A. (1994). Prenatal Drug Use in Minneapolis^St Paul, Minn: A 4-Year Trend. Arch Fam Med 3: 520-527 [Abstract]  
• Mason, E., Vega, W. A. (1994). Perinatal Substance Exposures in California. NEJM 330: 794-794 [Full Text]