PEDIATRICS Vol. 118 No. 3 September 2006, pp. e669-e675 (doi:10.1542/10.1542/peds.2006-0116)
ARTICLE |
Adverse Perinatal Outcomes and Risk for Postpartum Suicide Attempt in Washington State, 1987–2001
a Harborview Injury Prevention and Research Center
b Department of Epidemiology, University of Washington, Seattle, Washington
c Center for Health Studies, Group Health Cooperative, Seattle, Washington
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. Postpartum suicide attempts are serious events with a significant impact on women and their families. Our objectives were to determine the temporal risk period for these events and to assess whether maternal complications and adverse infant outcomes are associated with risk for postpartum suicide attempt.
METHODS. We performed a case-control study that compared 520 women who were hospitalized for a postpartum suicide attempt with 2204 control women who were not hospitalized for a postpartum suicide attempt in Washington State from 1987 to 2001. We performed logistic regression to evaluate whether maternal complications and adverse infant outcome, after controlling for other risk factors, were associated with a hospitalization for a suicide attempt within 1 year after delivery.
RESULTS. Most attempts were the result of poisoning (63.6%). Suicide attempts were most frequent in the first and 12th months after delivery. Maternal complications including labor and delivery complications and cesarean delivery were not associated with risk for postpartum suicide attempt. After adjustment for age and marital status, fetal or infant death was associated with postpartum suicide attempt. Other adverse infant outcomes, including preterm delivery, low birth weight, and congenital malformations, were not associated with attempts.
CONCLUSIONS. Maternal complications were not associated with hospitalizations for suicide attempts in the year after delivery. However, fetal death or the death of an infant in the first year after delivery was strongly associated with hospitalization for a suicide attempt. These risk factors may be useful predictors for health care providers who care for women or children during the postpartum period.
Key Words: injury • postpartum care • suicide • pregnancy
Abbreviations: CHARS—Comprehensive Hospital Abstract Recording System • ICD-9-CM—International Classification of Diseases, Ninth Revision, Clinical Modification
Suicide and suicide attempts during the postpartum year are rare events. Studies of maternal mortality have reported that suicide accounts for between 2.7% and 15% of all deaths to women within 1 year postpartum.1–3 Postpartum suicide rates range from 0.5 to 5.9 per 100000 live births.4,5 In comparison, the suicide rate in the US female population for 2001 was 4.1 per 100000.6 Weiss7 found that pregnant women had a 38% decreased rate of hospitalization for self-inflicted injuries compared with reproductive-age, nonpregnant women. Although many more women attempt suicide than actually complete it,8 no population-based rates of postpartum suicide attempts are available in the literature.
In addition to determining suicide and suicide attempt rates, the evaluation of risk factors may provide some potential prevention strategies. Few studies have evaluated the obstetric and perinatal risk factors that are associated with suicide and suicide attempts among women who recently delivered a child. Gissler et al5 found that among postpartum women, those who were young and unmarried and had low socioeconomic status were at increased risk for postpartum suicide compared with older, married women of higher socioeconomic status. Several studies have suggested that certain pregnancy outcomes also may be risk factors, with higher rates of postpartum suicide among women who had a stillborn9 or a spontaneous or therapeutic abortion compared with women who had a live birth.5 Case-control studies have demonstrated an association between suicide and suicide attempts10–12 and sudden, intense, adverse personal experiences or losses such as the death of a close friend or relative, especially deaths from suicide. Little is known about the risk for a suicide attempt for a mother whose infant had an adverse outcome within the first postpartum year. In light of the limited research, we performed a population-based study to evaluate the association between maternal complications and adverse infant outcomes and the risk for attempting suicide postpartum.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Participants
We performed a case-control study to evaluate maternal complications and adverse infant outcomes as potential risk factors for postpartum women's attempting suicide in Washington State from 1987 to 2001. We linked state birth and fetal death certificate data with hospitalization discharge data to identify women who were hospitalized for an attempted suicide in the year after delivery of a live birth or fetal death. The hospital data were extracted from the Comprehensive Hospital Abstract Recording System (CHARS) from all Washington State nonfederal hospitals. Among all pregnant women who delivered in Washington State, 4.7% occurred in federal hospitals.
Postpartum women who were hospitalized for a suicide attempt (N = 520) were identified by linking the CHARS data of all women who had an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) external causation code (E code) of E959.0 through 959.9, indicating suicide attempt, to birth and fetal death certificates for the 364 days before hospitalization date. We selected a control group of postpartum women who had a live birth or fetal death during the study period and were not hospitalized for a suicide attempt during the postpartum year (N = 2204). Control subjects were randomly selected in a ratio of 4 control subjects to 1 case patient and frequency matched to case patients on year of live birth or fetal death.
We also linked state birth and fetal death certificate data with death certificates to identify women who died as a result of suicide within 1 year postpartum. Postpartum women who completed suicide (n = 17) were identified by linking all female death certificates that listed suicide as the cause of death (ICD-9-CM E codes 959.0–959.9) to birth and fetal death certificates for the 364 days before the date of death.
During the period 1987 to 2001, Washington State recorded 1182672 live births and 6647 fetal deaths. This study was approved by the Washington State Institutional Review Board.
Outcome Classification
We used ICD-9-CM E codes that were listed in the hospital discharge data to evaluate the method of suicide attempt. We classified method of suicide attempt as overdosing by tranquilizers and other psychotropic medications (E950.3), analgesics, antipyretics, and antirheumatic medications (E950.0), other specified drugs and medicinal substances (E950.4), barbiturates (E950.1), other sedatives and hypnotics (E950.2), and other and unspecified solid and liquid substances (E950.9); cutting and piercing instruments (E956); jumping from high place (E957); crashing of motor vehicle (E958.5); hanging (E953.0); extremes of cold (E958.3); and other methods, including jumping or lying before moving object (E958.0), burns or fire (E958.1), or other unspecified means (E958.9). We also evaluated the timing of the postpartum suicide attempt in relation to the delivery.
Exposure Classification
Before initiating our study, we hypothesized that specific maternal complications and adverse infant outcomes would be associated with postpartum suicide attempts. To evaluate these risk factors, we used the Washington State birth and fetal death certificates that were linked to the maternal and infant CHARS records that contained ICD-9-CM diagnosis and procedure codes for the delivery hospitalization. Since 1980, the Washington State birth and fetal death certificates have included a checklist for complications during pregnancy, method of delivery, complications of labor and delivery, and abnormal conditions of the newborn.
Maternal complications were evaluated using the check boxes on the birth and fetal death certificate. For some complications, we used ICD-9-CM codes that were available in the CHARS data set in addition to the check boxes. Maternal complications evaluated included pregnancy complications, labor and delivery complications, and cesarean delivery. Pregnancy complications included preterm labor at <37 weeks' gestation (ICD-9-CM codes 644.0 or 644.2 or use of tocolysis on the birth or fetal death certificate); premature rupture of membranes >24 hours or more before the onset of labor (ICD-9-CM code 658.2); placenta previa (ICD-9-CM codes 641.0–641.1 or as listed on the birth or fetal death certificate); chronic hypertension, preeclampsia, or eclampsia (ICD-9-CM codes 642.0–642.9 or as listed on the birth or fetal death certificate); and preexisting or gestational diabetes (ICD-9-CM codes 648.0 or 250.0–250.9 or as listed on the birth or fetal death certificate). Labor and delivery complications included placental abruption (ICD-9-CM code 641.2 or as listed on the birth or fetal death certificate), prolonged rupture of membranes (ICD-9-CM code 658.2), meconium at delivery, fetal distress (ICD-9-CM code 656.3 or as listed on the birth or fetal death certificate), and postpartum hemorrhage (ICD-9-CM codes 666.0–666.2). Cesarean delivery was evaluated (ICD-9-CM codes 74–74.2 in hospital discharge data or as listed on the birth or fetal death certificate).
Adverse infant outcomes investigated included preterm delivery (<37 weeks' gestational age), low birth weight (<2500 g), presence of any congenital malformations, and length of neonatal hospital stay (as listed in the CHARS record). Fetal death and infant death in the first year of life also were evaluated as adverse infant outcomes by linking the birth certificates for all case patients and control subjects to the fetal and infant death certificates. All infant deaths that occurred among the case patients were before the suicide attempt. We evaluated these deaths as separate variables and as a combined variable including either a fetal or an infant death. We evaluated other pregnancy factors, including gravidity, parity, prenatal smoking, trimester of pregnancy when prenatal care initiated, and use of the Special Supplemental Program for Women, Infants and Children.
We also evaluated reproductive risk factors among women with a pregnancy before the index pregnancy (multigravid women). Reproductive risk factors were evaluated using the check boxes on the birth and fetal death certificate and included a history of spontaneous abortion and history of therapeutic abortion. We also evaluated reproductive factors among women with a delivery before the index pregnancy (multiparous women). These risk factors included previous fetal death at >20 weeks' gestational age, previous live birth now deceased, previous preterm live birth, and number of years between the index birth and the live birth immediately before the index birth.
Statistical Analysis
We compared the demographic characteristics between the case and control women. We assessed proportions for the method of nonfatal and fatal suicide as classified by ICD-9-CM E codes. We evaluated the timing of postpartum suicide attempt using the attempt rate for each 1-month interval postpartum. We performed univariate analysis of each hypothesized factor and its association with risk for postpartum suicide attempt. We performed logistic regression analysis to estimate the odds ratios and 95% confidence intervals for the association between maternal complications, adverse infant outcomes, and pregnancy factors and postpartum suicide attempt. We built a multivariate regression model initially including only risk factors that were statistically significant in the univariate analysis and subsequently removing risk factors that no longer remained significant after entry into the multivariate model. We evaluated for potential interactions between age and parity, age and infant death, marital status and parity, and marital status and infant death using the likelihood ratio test with a significance level of P < .05 and found no significant interactions. Potential confounders we evaluated, included maternal age, maternal race, marital status, and family income. We adjusted our final regression model for age and marital status because these variables were associated with the majority of our exposures and our outcome and changed the crude odds ratio by >10%. We also performed 2 subanalyses of reproductive risk factors that were associated with postpartum suicide attempt using logistic regression: 1 among the multigravid women and another among the multiparous women. Because we identified only 17 cases of fatal postpartum suicide among our population during the study period, only descriptive information about these cases is reported. All statistical analyses were performed with Stata 8.0 (Stata Corp, College Station, TX).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We found that 520 postpartum women were hospitalized for a suicide attempt, resulting in a rate of 43.9 per 100000 live births. We also determined age-specific postpartum suicide attempt rates per 100000 live births: 85.1 for women 15 to 19 years, 58.2 for women 20 to 24 years, 38.6 for women 25 to 29 years, 27.6 for women 30 to 35 years, 21.1 for women 35 to 39 years, and 13.0 for women 40 to 49 years. Postpartum women who were hospitalized for suicide attempts were more likely to be <20 years of age, single, and black or American Indian and have <12 years of completed education and Medicaid for prenatal insurance compared with control women (Table 1).
|
Among the postpartum women who were hospitalized for a suicide attempt (Table 2), we found that the majority used medications such as tranquilizers or psychotropics (34.4%) or analgesics, antipyretics, or antirheumatics (29.2%). Fewer women used more violent mechanisms, such as a cutting or piercing instrument (6.2%), jumping from a high place (0.5%), or crashing a motor vehicle (0.4%).
|
We also evaluated the timing of the suicide attempt with respect to delivery. We found that suicide attempts were most frequent in the first month (10.5%) and the 12th month (10.0%) and least frequent in the second (4.0%) and third (6.3%) months postpartum (Fig 1). The remaining postpartum months had little variation in frequency of suicide attempts, ranging from 8.0% to 9.9%.
|
We evaluated maternal complications, adverse infant outcomes, and pregnancy risk factors for their association with suicide attempts (Table 3) and found that smoking; trimester of prenatal care initiation; use of Special Supplemental Program for Women, Infants and Children; gravidity; parity; preterm delivery; low birth weight; neonatal length of stay; infant death; and combined fetal or infant death were associated with risk for a suicide attempt. Our final multivariate model, adjusted for maternal age and marital status, showed that either a fetal or an infant death during the postpartum period was associated with a 3.1-fold increased risk for postpartum suicide attempt (Table 4). In addition, smoking during pregnancy was associated with a 2.7-fold increased risk, 2 previous deliveries was associated with a 1.8-fold increased risk, and 3 or more previous deliveries was associated with a 3.1-fold increased risk.
|
|
Our evaluation of reproductive risk factors among multigravid women showed no association between history of spontaneous or therapeutic abortion and risk for a suicide attempt. We also found no association between previous fetal death, previous liveborn now dead, previous preterm delivery, and number of years between the index birth and the live birth immediately before the index birth and risk for postpartum suicide attempt among multiparous women.
We found 17 cases of completed postpartum suicide (rate: 1.4 per 100000 live births). Nearly all women died before hospitalization; 3 women died in the hospital. Postpartum women who completed suicide most often used a firearm (53%) or hanging (12%). Other methods of suicide included use of medications such as tranquilizers or psychotropics (6%) or analgesics, antipyretics, or antirheumatics (6%); jumping from a high place (6%); or other methods (17%). The majority of postpartum women who completed suicide was 20 to 29 years (64%), with fewer being 30 to 34 years (18%) and 35 years and older (18%). Most were non-Hispanic white (82%), and the remaining were Asian (6%), Hispanic white (6%), and American Indian (6%). Nearly all were married (76%). The number of previous pregnancies among postpartum suicide cases varied with 41% having 1, 18% having 2, and 41% having 3 or more previous pregnancies.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We found that a postpartum suicide attempt occurs in
1 in 2276 live births in Washington State, with most women using poisoning with medications as the method for the attempt. Maternal complications, including labor and delivery complications and cesarean delivery, and previous spontaneous or therapeutic abortion were not associated with postpartum suicide attempt. A fetal or an infant death was strongly associated with an attempt. No studies of risk factors for postpartum suicide attempts have been performed to our knowledge, although a few studies have evaluated suicide attempts during pregnancy and suicide completion postpartum. Our study focused on pregnancy as well as labor and delivery complications and found no association with risk for postpartum suicide attempt. We also evaluated adverse infant outcomes and found that women who experienced a fetal or an infant death during the postpartum year had a threefold increased risk for suicide attempt. Similar to our findings, Lester et al13 reported that in a case series of 15 women who attempted suicide during pregnancy, some of the attempts seemed to be related to previous loss of a child. Appleby9 noted a sixfold increased risk for postpartum suicide associated with fetal death. Our results regarding fetal or infant death preceding a suicide attempt are in agreement with previous case-control studies of suicide and suicide attempts that show an association between sudden, intense, personal losses such as the death of a relative.10–12 These findings highlight the need for close postpartum follow-up by obstetricians, pediatricians, other primary care providers, and mental health professionals for women who experience an infant death.
Few previous studies have evaluated postpartum suicide attempt rates. Appleby and Turnbull14 determined a postpartum suicide attempt rate of 125 per 100000 postnatal women in a hospital-based study in London in 1991, although this rate was based on 5 women's attempting suicide during the year postpartum. Our rate is approximately one third of this rate but may be a more accurate estimate because it was population based for a 15-year period. Comparing the rate of hospitalization for suicide attempt of 63.7 per 100000 women among the US female population of similar age to our case patients15 with our suicide attempt hospitalization rate of 43.9 per 100000, the postpartum rate is one third lower. Appleby and Turnbull14 also reported a postpartum suicide attempt rate that was half the rate among nonpostpartum women.
Our finding that the majority of suicide attempts were poisonings is similar to a previous study by Appleby and Turnbull14 among postpartum women, as well as other studies in the nonpregnant female population.8,16,17 The use of tranquilizers or psychotropic medications indicates that case women may have had previous psychiatric morbidity. These results indicate that prenatal care providers should screen women for a history of psychiatric illness and use of psychotropic medications as a potential means of identifying postpartum women who are at risk for a suicide attempt. Women who are taking psychotropic medications should be followed carefully in the postpartum period by their obstetric and psychiatric providers.
Timing of postpartum suicide attempts has not been evaluated previously in the literature. Appleby9 noted an increase in the number of completed suicides in the first and the fifth months postpartum. We found that suicide attempts were most frequent in the first and 12th months postpartum. These findings highlight the need for obstetric providers to discuss mental health issues with their patients and emphasize the importance of contacting their provider if problems such as depression arise. In addition, pediatricians who interact with new mothers during well-infant evaluation throughout the first postpartum year as well as adult primary care providers and mental health professionals may play a key role in identifying mothers who are at risk for suicide attempt.
Our study had several limitations. Our evaluation of suicide attempts was limited to those that resulted in a hospitalization. Although we assumed that the majority of serious suicide attempts would result in a hospital stay, our study likely missed women who attempted suicide but were treated in an emergency department or outpatient setting. In addition, women may have been hospitalized after a suicide attempt but the CHARS diagnosis codes may not have indicated a suicide attempt because of insurance or social desirability factors. Limitations of coding and data entry in the CHARS and birth and fetal death certificates may have limited our ability to ascertain accurately demographic, pregnancy, and infant information. Studies have documented that birth weight, race, age, and number of previous live births are recorded on the birth and fetal death certificates with nearly complete accuracy,18 but prenatal and intrapartum complications, delivery method, and obstetric procedures are recorded less accurately.19,20 We attempted to minimize inaccuracies by use of multiple sources for identifying specific perinatal risk factors when possible.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We found that postpartum suicide attempts involve poisoning with medication, most commonly in the first and 12th months postpartum. Although maternal complications were not associated with hospitalization for suicide attempts, fetal or infant death was strongly associated with an attempt. Women who experience an infant death should receive careful monitoring and mental health services in the postpartum year. Additional studies are needed to evaluate the role of psychiatric morbidity during and after pregnancy as possible indicators of risk for postpartum suicide attempt.
|
ACKNOWLEDGMENTS |
|---|
This research was supported by a grant from the American Foundation for Suicide Prevention (New York, NY).
We thank the Washington State Department of Health as well as William O'Brien for supplying data analysis files.
|
FOOTNOTES |
|---|
Accepted Mar 8, 2006.
Address correspondence to Melissa A. Schiff, MD, MPH, Harborview Injury Prevention and Research Center, 325 Ninth Ave, Box 359960, Seattle, WA 98104. E-mail: mschiff{at}u.washington.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
This research was presented as a poster at the 3rd Western Maternal and Child Health Epidemiology Conference; Portland, OR; May 12–13, 2005; and as an oral presentation at the annual meeting of the American Foundation for Suicide Prevention; April 28–May 1, 2006; Seattle, WA.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Horon IL, Cheng D. Enhanced surveillance for pregnancy-associated mortality: Maryland, 1993–1998. JAMA. 2001;28 :1455 –1459
- Jocums SB, Berg CJ, Entman SS, Mitchell EF. Postdelivery mortality in Tennessee, 1989–1991. Obstet Gynecol. 1998;91 :766 –770[CrossRef][Web of Science][Medline]
- Hogberg U, Innala E, Sandstrom A. Maternal mortality in Sweden, 1980–1988. Obstet Gynecol. 1994;84 :240 –242[Web of Science][Medline]
- Kleiner GJ, Geston WM. Suicide in Pregnancy. London, England: PSG; 1984:23 –40
- Gissler M, Hemminki E, Lonnqvist J. Suicides after pregnancy in Finland, 1987–94: register linkage study.
BMJ. 1996;313
:1431
–1434
[Abstract/Free Full Text] - Anderson RN, Minino AM, Fingerhut LA, Warner M, Heinen MA. Deaths: injuries, 2001. Natl Vital Stat Rep. 2004;52(21) :1 –86
- Weiss HB. Pregnancy-associated injury hospitalizations in Pennsylvania. Ann Emerg Med. 1999;34 :626 –636[CrossRef][Web of Science][Medline]
- Miller M, Azrael D, Hemenway D. The epidemiology of case fatality rates for suicide in the Northeast. Ann Emerg Med. 2004;43 :723 –730[CrossRef][Web of Science][Medline]
- Appleby L. Suicide during pregnancy and in the first postnatal year.
BMJ. 1991;302
:137
–140
[Abstract/Free Full Text] - Benjaminsen S, Krarup G, Lauritsen R. Personality, parental rearing behavior and parental loss in attempted suicide: a comparative study. Acta Psychiatr Scand. 1990;82 :389 –397[Web of Science][Medline]
- Agerbo E. Midlife suicide risk, partner's psychiatric illness, spouse and child bereavement by suicide or other modes of death: a gender specific study.
J Epidemiol Community Health. 2005;59
:407
–412
[Abstract/Free Full Text] - Qin P, Mortensen PB. The impact of parental status on the risk of completed suicide.
Arch Gen Psychiatry. 2003;60
:797
–802
[Abstract/Free Full Text] - Lester D, Beck AT. Attempted suicide and pregnancy. Am J Obstet Gynecol. 1988;158 :1084 –1085[Web of Science][Medline]
- Appleby L, Turnbull G. Parasuicide in the first postnatal year. Psychol Med. 1995;25 :1087 –1090[Web of Science][Medline]
- Centers for Disease Control and Prevention, National Centers for Injury Prevention and Control. Web-based Injury Statistics Query and Reporting System (WISQARS); 2005. Available at: www.cdc.gov/ncipc/wisqars. Accessed February 13, 2006
- Murphy GE. Why women are less likely than men to commit suicide. Compr Psychicatry. 1998;39 :165 –175[CrossRef]
- Centers for Disease Control and Prevention. Nonfatal self-inflicted injuries treated in hospital emergency departments: United States, 2000. MMWR Morb Mortal Wkly Rep. 2002;51 :436 –438[Medline]
- Schoendorf KC, Parker JD, Batkhan LZ, Kiely JL. Comparability of the birth certificate and the 1988 Maternal and Infant Health Survey. Vital Health Stat 2. 1993;(116) :1 –19
- Parrish KM, Holt VL, Connell FA, Williams B, LoGerfo JP. Variations in the accuracy of obstetric procedures and diagnoses on birth records in Washington State, 1989.
Am J Epidemiol. 1993;138
:119
–127
[Abstract/Free Full Text] - Dobie SA, Baldwin LM, Rosenblatt RA, Fordyce MA, Andrilla CHA, Hart LG. How well do birth certificates describe the pregnancies they report? The Washington State experience with low-risk pregnancies. Matern Child Health J. 1998;2 :145 –154[CrossRef][Medline]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
