Over 80 million people in the United States are at risk for suicide because of depression and other psychiatric diseases.1 Furthermore, about 31,000 Americans die each year by suicide, with a crude rate of 12 suicides per 100,000 people.1,2 In 2002, 132,353 individuals were hospitalized after suicide attempts, and 116,639 were treated in emergency departments and released.2 Women attempt suicide 3 times more frequently than men, and suicide is the fourth leading cause of female mortality in women of reproductive age.3 Although previous studies have shown a lower risk of suicide during pregnancy compared with the nonpregnant state, the rate for women in general is high, with approximately 16 females attempting suicide per day.4–7 Perinatal suicides have been studied with regard to their association with peri- and postpartum depression and other psychiatric disorders,8,9 but a MEDLINE search (English language from January 1965 to October 2005) using the key words “suicide,” “pregnancy,” and “pregnancy outcomes” revealed few studies that provide information on the maternal and fetal outcomes associated with an attempted suicide. Consequently, our goal was to identify characteristics for a population at risk for attempting suicide and to examine the maternal and neonatal outcomes influenced by the suicide attempt.
MATERIALS AND METHODS
A computerized database with maternal and neonatal hospital discharge information linked to birth and death certificates was used and contained records for 4,833,286 hospital obstetric births within the State of California between 1991 and 1999. The Vital Statistics-Patient Discharge Database is compiled from hospitals reporting to the California Office of Statewide Health Planning and Development. This linkage of maternal and neonatal hospital discharge records has been shown to be successful in 98% of cases.10 This study was approved by the California Office of Statewide Health Planning and Development and by the University of California Davis Institutional Review Board.
All pregnancies complicated by suicide attempts requiring hospitalization were identified using external causation codes (E 950–959). The method used to attempt suicide was also documented through the database. The control population was composed of patients who did not attempt suicide. The Vital Statistics-Patient Discharge Database does not contain information on women who were not hospitalized or for those who delivered at less than 20 weeks of gestation. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes were used to identify maternal and neonatal outcomes. Specifically, the ICD-9 codes used included antepartum hemorrhage, blood transfusion, cesarean delivery, infection, placental abruption, premature labor, premature rupture of membranes (PROM), and maternal death. Fetal, neonatal, and infant codes included fetal distress, fetal, neonatal, and infant death, low birth weight (LBW, < 2,500 g), premature delivery (< 37 weeks), and respiratory distress syndrome (RDS). Maternal substance abuse, defined as use of legal (tobacco, alcohol) or illegal drugs, not prescription medications, was also identified.
Women who attempted suicide during pregnancy were compared with the control population. Further analysis based on the timing of delivery was performed with subsets of the study group. Specifically, women who delivered at the time of initial hospitalization and those who were discharged and delivered at a later hospitalization were further assessed. The data were analyzed by performing univariable analysis (χ2) and multivariable logistic regression (SAS 8.2; SAS Institute, Cary, NC). Odd ratios and 95% confidence intervals were calculated. Because confounding factors are known to affect outcomes, the odds ratios were controlled for maternal age, marital status, race, parity, education, substance abuse, prenatal care, and insurance type. Statistical significance in all calculations was defined as P < .05.
There were 4,833,286 obstetric deliveries from 1991 to 1999 in California. Using external causation codes, 2,132 of these deliveries (0.4 per 1,000 pregnancies) were identified as being to women hospitalized for attempted suicide beyond the 19th week of pregnancy. These women made up the study group, and the remaining 4,831,154 deliveries made up the control group. Further analysis of the study group showed that 31 women required delivery at the time of initial hospitalization and the other 2,101 women did not require delivery at the initial hospitalization but delivered at a later date. Ninety percent of the women who required immediate delivery delivered within 48 hours of attempted suicide, whereas the group that delivered at a later date delivered, on average, 144 days from the date of attempted suicide.
The demographic information for the control and attempted suicide groups is shown in Table 1. In comparison to the control group, women who attempted suicide were young (< 30 years old), single, multiparous, less educated, and MediCal (Medicaid) recipients, and they were more often African American. As well, substance abuse was more common in women who attempted suicide. The mean gestational age was 272 days for all suicide attempters and 274 days for the controls. In the subanalysis, suicide attempters requiring delivery at that hospitalization had a mean gestational age of 260 days, whereas the delayed-delivery group had a mean gestational age of 274 days. In an attempt to evaluate the impact of psychiatric illnesses, we found that 0.01% (606 of 4,830,554) of controls and 0.56% (12 of 2,131) of suicide attempters had codings for psychiatric illness. Although the suicide attempters had a significantly (P < .001) higher chance of having a psychiatric illness, these data are felt to reflect undercoding, as previously described,11 and are thought to be unreliable. Other studies12,13 have established that the prevalence of psychiatric illness during pregnancy is about 9–10%.
Our data showed that 86% of women attempted suicide by ingestion of a solid or liquid, primarily by drug overdose or poisoning with a corrosive substance (Table 2). This was significantly higher in the women who delivered at a later date. The influence of suicide attempt on maternal and neonatal outcomes is outlined in Table 3. Compared with the control population, maternal outcomes in the attempted suicide group showed increases in premature labor, cesarean delivery, and a requirement for blood transfusion. Neonatal outcomes showed increases in LBW and RDS. A separate analysis of LBW, controlling for gestational age, revealed that LBW remained significant (P < .001, odds ratio [OR] 1.67, 95% confidence interval [CI] 1.41–1.98) in the attempted suicide group when compared with controls. Respiratory distress syndrome was directly related to gestational age (OR 0.164, 95% CI 0.160–0.167). There was no increased risk of maternal mortality in any of the study groups when compared with controls.
In a separate analysis of the subgroups, the group that delivered immediately (Table 4) demonstrated an increase in premature labor. Neonatal outcomes in this group were significant for higher rates of premature delivery and RDS. In addition, there was almost a 5-fold increase in neonatal and infant death. Unlike the immediate-delivery group, the group that delivered at a later date had higher rates of cesarean delivery and blood transfusion (P < .001, OR 2.04, 95% CI 1.25–3.35), along with premature labor. Also unlike the immediate-delivery group, neither neonatal nor infant death was increased, but LBW and RDS were increased. This later delivery group had outcomes exactly the same as the main study group. After controlling for gestational age in this later delivery group, LBW remained significant (P < .001, OR 1.68, 95% CI 1.41–1.99). On the other hand, after controlling for gestational age in the immediate delivery, LBW was no longer significant (P = .29, OR 1.81, 95% CI 0.60–5.48).
The prevalence of suicide among reproductive-aged women is such that practicing obstetricians can expect to care for someone who has attempted or will attempt suicide. We set out to describe identifiers for women who attempt suicide during pregnancy. From our demographic findings, young, single, multiparous, African-American women who are less educated and have poor prenatal care are more likely to attempt suicide. Kessler et al14 reported that the risk of attempted suicide was similar among different ethnic groups, but according to our findings, it appears that African-American women attempted suicide more often. This difference may be due to a variance between the 2 populations. In search of more specific identifiers, we analyzed for an association between suicide attempt and substance abuse or psychiatric illness. Our data revealed that 26% of women attempting suicide were substance abusers. On the other hand, an association with psychiatric illnesses was not established from our study because the data were believed to be unreliable. An association between suicide attempt during pregnancy and psychiatric illnesses has been established in other studies.12,13 Therefore, the best potential identifiers of women likely to attempt suicide during pregnancy are a history of substance abuse or psychiatric illness.
Our evaluation of maternal and neonatal outcomes revealed that women attempting suicide during pregnancy were at increased risk for preterm labor, cesarean delivery, need for blood transfusion, LBW, and RDS. Although cesarean delivery was increased, primary cesarean was not, indicating that these were mostly due to repeat cesarean. The increased need for blood transfusion is not clear, but possibly reflects a group with poor nutrition with resulting anemia. Others have published studies of attempted suicide during pregnancy, but there has been little focus on maternal and neonatal outcomes.8,9 Flint et al8 found an increased risk of miscarriage but no increase in congenital anomalies, premature birth, or low birth weight. Contrary to this, Czeizel et al9 found significantly lower birth weights, earlier gestational age at birth, and a higher infant death rate in women attempting suicide when compared with matched controls. Although contradictory, both studies focused primarily on the teratologic effects of the drug overdose rather than maternal or neonatal outcomes. Our study was more consistent with Czeizel et al, demonstrating increases in LBW and RDS. Although these women, similar to those in other studies,15–20 were more likely to have premature labor, they did not go on to deliver. We do not know whether interventions (tocolysis, steroids, bed rest) were implemented in an attempt to prolong pregnancy. We do know that RDS was directly related to gestational age and LBW persisted, regardless of gestational age, suggesting a mechanism of placental insufficiency. We speculate that there may be an effect of trauma leading to a subchorionic bleed or partial placental abruption, which in turn progresses into chronic abruption and placental insufficiency,21 thus resulting in LBW. Schizophrenia and affective disorders themselves have also been shown to be associated with antepartum hemorrhage, abruption, and LBW.15
From the subanalysis, patients in the later delivery group, as expected, had outcomes similar to those in the main attempted-suicide group, and these are discussed together above. On the other hand, the immediate-delivery group, demonstrated some differences in outcomes. There were increases in premature labor and delivery with a 2-week earlier mean gestational age, RDS, and a higher rate of neonatal and infant deaths. Therefore, it appears that the suicide event initiated a cascade that resulted in immediate delivery, perhaps without the benefit of steroids for fetal lung maturity (90% of deliveries were < 48 hours from time of hospital admission). The consequences of this cascade included neonatal and infant death. The number of cases in this group is small, but this outcome is catastrophic and suggests significant trauma might have been caused by the suicide attempt. Previous reports of traumatic injuries in pregnancy consistently show an increased risk of mortality to the fetus compared with the mother22–25 and a fetal-to-maternal death ratio as high as 9 to 1.22 This group did not show an increase in LBW after controlling for gestational age. This again suggests that, in the absence of delivery, pregnancies complicated by a suicide attempt acquire a placental insufficiency leading to LBW.
There are several limitations to this study because of its retrospective design and use of discharge diagnosis codes. In general, these data may be subject to miscoded diagnoses or erroneously classified diagnoses. This analysis, when evaluating for relevant history, is limited. We do not have access to information about prior pregnancy complications, such as preterm delivery. In addition, our study does have a potential selection bias. That is, women who delivered at less than 20 weeks of gestation and women who did not come to the hospital (Coroner's case) or women who were not admitted are not included in the database. Therefore, the subset of women not admitted may provide interesting information regarding outcomes, but this information cannot be ascertained. Despite these limitations, the strength of our study is the focus on maternal and neonatal outcomes and the large number of cases in a heterogeneous population.
Based on our findings, the immediate-delivery group may not give care providers time for intervention because these women present, deliver, and face the consequences of the delivery. On the other hand, if the intervention focuses on prevention, then the associated outcomes may be avoided. Prevention is best undertaken by identifying the “at risk” group (women with current psychiatric disorders or substance abuse or with a history of these) and by counseling them about the potential adverse outcomes if suicide is attempted. In addition, these data can help provide an argument for continuing psychiatric medications during pregnancy, especially now when many of the antidepressants (selective serotonin reuptake inhibitors) are being scrutinized in regard to their use during pregnancy.26,27
1. Goldsmith, SK, Pellmar, TC, Kleinman, AM, Bunney, WE, editors. Reducing suicide: a national imperative. Washington, DC: National Academies Press; 2002.
2. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web-based Injury Statistics Query and Reporting System (WISQARS). Available at: http://www.cdc.gov/ncipc/wisqars/
. Retrieved February 28, 2006.
3. Krug EG, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, editors. World report on violence and health. Available at: http://www.who.int/violence_injury_prevention/violence/world_report/en/
. Retrieved February 28, 2006.
4. Arias E, Anderson RN, Kung HC, Murphy SL, Kochanek KD. Deaths: final data for 2001. Natl Vital Stat Rep 2003;52:1–115
5. Marzuk PM, Tardiff K, Leon AC, Hirsch CS, Portera L, Hartwell N, et al. Lower risk of suicide during pregnancy. Am J Psychiatry 1997;154:122–3.
6. Appleby L. Suicide during pregnancy and in the first postnatal year. BMJ 1991;302:137–40.
7. Gissler M, Hemminki E, Lonnqvist J. Suicides after pregnancy in Finland, 1987–94: register linkage study. BMJ 1996;313:1431–4.
8. Flint C, Larsen H, Nielsen GL, Olsen J, Sorensen HT. Pregnancy outcome after suicide attempt by drug use: a Danish population-based study. Acta Obstet Gynecol Scand 2002;81:516–22.
9. Czeizel AE, Tomcsik M, Timar L. Teratologic evaluation of 178 infants born to mothers who attempted suicide by drugs during pregnancy. Obstet Gynecol 1997;90:195–201.
10. Herrchen B, Gould JB, Nesbitt TS. Vital statistics linked birth/infant death and hospital discharge record linkage for epidemiological studies. Comput Biomed Res 1997;30:290–305.
11. Kelly RH, Danielson BH, Zatzick DF, Haan MN, Anders TF, Gilbert WM, et al. Chart-recorded psychiatric diagnoses in women giving birth in California in 1992. Am J Psychiatry 1999;156:955–7.
12. Gotlib IH, Whiffen VE, Mount JH, Milne K, Cordy NI. Prevalence rates and demographic characteristics associated with depression in pregnancy and the postpartum. J Consult Clin Psychol 1989;57:269–74.
13. O'Hara MW. Social support, life events, and depression during pregnancy and the puerperium. Arch Gen Psychiatry 1986 43:569–73.
14. Kessler RC, Berglund P, Borges G, Nock M, Wang PS. Trends in suicide ideation, plans, gestures, and attempts in the United States, 1990–1992 to 2001–2003. JAMA 2005;293:2487–95.
15. Jablensky AV, Morgan V, Zubrick SR, Bower C, Yellachich LA. Pregnancy, delivery, and neonatal complication in a population cohort of women with schizophrenia and major affective disorders. Am J Psychiatry 2005;162:79–91.
16. Steer RA, Scholl TO, Hediger ML, Fischer RL. Self-reported depression and negative pregnancy outcomes. J Clin Epidemiol 1992;45:1093–9.
17. Dayan J, Creveuil C, Herlicoviez M, Herbel C, Baranger E, Savoye C, Thouin A. Role of anxiety and depression in the onset of spontaneous preterm labor. Am J Epidemiol 2002;155:293–301.
18. Hedegaard M, Henriksen TB, Sabroe S, Secher NJ. Psychological distress in pregnancy and preterm delivery. BMJ 1993;307:234–9.
19. Orr ST, James SA, Blackmore Prince C. Maternal prenatal depressive symptoms and spontaneous preterm births among African-American women in Baltimore, Maryland. Am J Epidemiol 2002;156:797–802.
20. Hoffman S, Hatch MC. Depressive symptomatology during pregnancy: evidence for an association with decreased fetal growth in pregnancies of lower social class women. Health Psychol 2000;19:535–43.
21. Williams JK, McClain L, Rosemurgy AS, Colorado NM. Evaluation of blunt abdominal trauma in the third trimester of pregnancy: maternal and fetal considerations. Obstet Gynecol 1990;75:33–7.
22. Hoff WS, D'Amelio LF, Tinkoff GH, Lucke JF, Rhodes M, Diamond DL, et al. Maternal predictors of fetal demise in trauma during pregnancy. Surg Gynecol Obstet 1991;172:175–80.
23. Shah KH, Simons RK, Holbrook T, Fortlage D, Winchell RJ, Hoyt DB. Trauma in pregnancy: maternal and fetal outcomes. J Trauma 1998;45:83–6.
24. Weiss HB, Songer TJ, Fabio A. Fetal deaths related to maternal injury. JAMA 2001;286:1863–8.
25. Ikossi DG, Lazar AA, Morabito D, Fildes J, Knudson MM. Profile of mothers at risk: an analysis of injury and pregnancy loss in 1,195 trauma patients [published erratum appears in J Am Coll Surg 2005;200:482]. J Am Coll Surg 2005;200:49–56.
26. Costei AM, Kozer E, Ho T, Ito S, Koren G. Perinatal outcome following third trimester exposure to paroxetine. Arch Pediatr Adolesc Med 2002;156:1129–32.
27. Kallen B. Neonate characteristics after maternal use of antidepressants in late pregnancy. Arch Pediatr Adolesc Med 2004;158:312–6.