- Email: [email protected]
Gestational and Postnatal Cortisol Profiles of Women With Posttraumatic Stress Disorder and the Dissociative Subtype
RESEARCH 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
Gestational and Postnatal Cortisol Profiles of Women With Posttraumatic Stress Disorder and the Dissociative Subtype Julia S. Seng, Yang Li, James J. Yang, Anthony P. King, Lisa M. Kane Low, Mickey Sperlich, Heather Rowe, Q3 Hyunhwa Lee, Maria Muzik, Julian D. Ford, and Israel Liberzon
Correspondence Julia S. Seng, PhD, CNM, FAAN, University of Michigan, School of Nursing, 400 N. Ingalls St., Room 3220, Ann Arbor, MI 48109. [email protected]
ABSTRACT
Keywords adverse childhood events cortisol developmental origins of health and disease dissociation dissociative subtype gestation posttraumatic stress disorder toxic stress
Participants: Women expecting their first infants who fit with one of four cohorts: a nonexposed control group, a trauma-exposed control group, a group with PTSD, and a group with the dissociative subtype of PTSD.
Objective: To test the hypothesis that women with posttraumatic stress disorder (PTSD) have greater salivary cortisol levels across the diurnal curve and throughout gestation, birth, and the postpartum period than women who do not have PTSD. Design: Prospective, longitudinal, biobehavioral cohort study. Setting: Prenatal clinics at academic health centers in the Midwest region of the United States.
Methods: In the first half of pregnancy, 395 women provided three salivary cortisol specimens on a single day for diurnal data. A subsample of 111 women provided three salivary cortisol specimens per day, 12 times, from early pregnancy to 6 weeks postpartum for longitudinal data. Trauma history, PTSD, and dissociative symptoms were measured via standardized telephone diagnostic interviews with the use of validated epidemiologic measures. Generalized estimating equations were used to determine group differences. Results: Generalized estimating equations were used to show that women with the dissociative subtype of PTSD had the highest and flattest gestational cortisol level curves. The difference was greatest in early pregnancy, when participants in the dissociative subtype group had cortisol levels 8 times greater in the afternoon and 10 times greater at bedtime than those in the nonexposed control group. Conclusion: Women with the dissociative subtype of PTSD, a complex form associated with a history of childhood maltreatment, may have toxic levels of cortisol that contribute to intergenerational patterns of adverse health outcomes.
JOGNN,
-, -–-;
2017. https://doi.org/10.1016/j.jogn.2017.10.008 Accepted October 2017
Julia S. Seng, PhD, CNM, FAAN, is a professor in the School of Nursing, University of Michigan, Ann Arbor, MI.
(Continued)
The authors report no conflict of interest or relevant financial relationships.
http://jognn.org
P
erinatal outcomes are indicative of the health of a society’s population (Allen, 2008; Risnes et al., 2011). Maternal psychosocial stress has been implicated as one of five pathways to the adverse perinatal outcome of premature birth (Green et al., 2005). It is also associated with lower birth weight (Shannon, King, & Kennedy, 2007) and is a risk factor in the fetus for early lifespan morbidity and mortality (Reynolds, Labad, Buss, Ghaemmaghami, & Raikkonen, 2013). Biological mediators of social determinants of health, including psychosocial stressors, are a priority focus for research to redress preterm birth (Lackritz et al., 2013).
Posttraumatic stress disorder (PTSD) is a severe form of psychological stress that affects 8% of pregnant women (Seng et al., 2010). However, PTSD has only recently been studied in relation to perinatal outcomes. Recent studies in large prenatal clinic samples showed significant associations of PTSD with preterm birth or shorter gestation, especially in current as opposed to lifetime remitted PTSD, PTSD with depression (Seng, Low, Sperlich, Ronis, & Liberzon, 2011; Shaw et al., 2014; Yonkers et al., 2014), and PTSD that is subsequent to childhood maltreatment (Seng et al., 2011) or military sexual trauma (Shaw et al., 2014).
ª 2017 AWHONN, the Association of Women’s Health, Obstetric and Neonatal Nurses. Published by Elsevier Inc. All rights reserved.
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
1
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
RESEARCH
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168
Cortisol in Women With PTSD and Dissociative Subtype
Posttraumatic stress disorder is associated with adverse perinatal outcomes. Elevated cortisol is a plausible mechanism for adverse outcomes and may affect maternal prenatal physiology.
Yang Li, MS, is a doctoral candidate in the School of Nursing, University of Michigan, Ann Arbor, MI. James J. Yang, PhD, is an associate research scientist in the Department of Psychiatry, University of Michigan, Ann Arbor, MI. Anthony P. King, PhD, is a research assistant professor in the Department of Psychiatry, University of Michigan, Ann Arbor, MI. Lisa M. Kane Low, PhD, CNM, FACNM, is an associate professor in the School of Nursing, University of Michigan, Ann Arbor, MI. Mickey Sperlich, PhD, MSW, CPM, is an assistant professor of social work, University at Buffalo (SUNY), Buffalo, NY.
With the release of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013), a diagnosis of PTSD may include a dissociative subtype characterized by altered perception of the environment in the form of derealization or depersonalization. In an epidemiologic analysis of 25,000 adults across 26 nations, 14.4% of those with PTSD fit the dissociative subtype classification (Stein et al., 2013). Dissociation has been associated with more severe PTSD that is diagnosed earlier in the lifespan and is associated with a greater number of lifetime trauma exposures, including history of childhood maltreatment or cumulative adverse childhood experiences (Liotti, 2004; Shonkoff et al., 2012; Stein et al., 2013). Thus, dissociation is an important indicator of severity and chronicity. It also may be a more proximal predictor of shorter gestation than the previously identified risk factor of childhood maltreatment (Seng et al., 2011).
associated with adverse perinatal outcomes (Green et al., 2005), slower infant mental development (Davis & Sandman, 2010), and adverse consequences to lifespan health (Reynolds, 2013), cortisol is an important biomarker for perinatal and developmental research. Data about the effects of PTSD on the HPA axis in pregnant women are very limited. In a recent study on cortisol levels in women at 21 and 34 weeks gestation that was focused on depression but entered trauma history and PTSD as covariates, researchers found no evidence to indicate an effect of PTSD after depression was taken into account (O’Connor et al., 2014). One recent study of gestational cortisol levels in women’s hair showed effects of childhood abuse in one analysis (Schreier, Enlow, Ritz, Gennings, & Wright, 2015) and cumulative trauma in another (Schreier et al., 2016). Depression and PTSD were covariates used to adjust the model in the latter analysis, but the researchers did not report the relative effects. The dissociative subtype of PTSD, a new diagnosis in 2013 (American Psychiatric Association, 2013), requires investigation in perinatal research. Our objective was to contribute to the literature on perinatal psychosocial stress by describing cortisol levels across the diurnal curve for 1 day in the first half of pregnancy and across the longitudinal gestational curve, including time points during pregnancy, birth, and the postnatal period, in women with PTSD, including a subset with the dissociative subtype. We examined four cohorts: a nonexposed control group, a resilient trauma-exposed control group, a group with PTSD (PTSD-only), and a group with the dissociative subtype of PTSD (PTSD-D). Our hypotheses were that (a) women with PTSD have greater diurnal levels of cortisol and (b) these greater levels of cortisol persist across gestation and the postnatal period. We further hypothesized that women with the dissociative subtype would have the greatest levels of diurnal and gestational cortisol.
Julian D. Ford, PhD, ABPP, is a professor in the Department of Psychiatry, University of Connecticut School of Medicine, Farmington, CT.
Researchers of the specific physiology of the dissociative subtype of PTSD showed a phenotype that appears to have different patterns of emotion dysregulation (Lanius et al., 2010). Although this is not an analysis of the DSM-5 dissociative subtype per se, researchers in one recent study distinguished a phenotype in women with PTSD and early-life trauma that was characterized by dissociative symptomatology and blunted cortisol reactivity (Zaba et al., 2015). Another group of researchers studied cortisol in women with PTSD and childhood sexual abuse and found blunted cortisol reactivity and greater basal levels (Schalinski, Elbert, SteudteSchmiedgen, & Kirschbaum, 2015). An earlier study team distinguished a dissociative disorder phenotype from PTSD that was based on elevated basal cortisol levels and blunted reactivity (Simeon et al., 2007). Together, evidence from these studies indicates that exposure to the early relational trauma that predisposes a person to dissociation and PTSD may affect that individual’s short- and long-term cortisol patterns.
Israel Liberzon, MD, is a professor in the Department of Psychiatry, University of Michigan, Ann Arbor, MI.
Because in utero elevations in hypothalamic– pituitary–adrenal (HPA) axis hormones are
2
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
Heather Rowe, PhD, BSc(Hons), is a senior research fellow in the Jean Hailes Research Unit, Monash University, Melbourne, Victoria, Australia. Hyunhwa Lee, PhD, PMHNP-BC, is an assistant professor in the School of Nursing, University of Nevada, Las Vegas, Las Vegas, NV. Maria Muzik, MD, MS, is an associate professor and Co-Director of the Women and Infants Mental Health Program, Department of Psychiatry, University of Michigan, Ann Arbor, MI.
Methods Design and Setting This prospective, ecological study of diurnal and gestational salivary cortisol profiles was a biobehavioral component of a larger clinical cohort study of the effect of PTSD on childbearing outcomes in a community sample of women in prenatal care. The parent study is the Stress, Trauma, Anxiety, and the Childbearing Year
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
RESEARCH
Seng, J. S. et al.
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
(STACY) Project (NIH R01 NR008767, J. S. Seng, principal investigator). The STACY Project took place in prenatal clinics at three academic health systems in Michigan. Institutional review boards approved the study at all three sites.
Participants In the parent clinical study, investigators recruited a convenience sample designed to be generalizable to community prenatal clinic populations. The women invited to the biobehavioral component were recruited from the parent study sample. For participation in the parent study, eligible women were at least 18 years of age, spoke English, were expecting their first infants, and initiated prenatal care at less than 28 weeks gestation. Maternity clinic nurses invited eligible women to participate in a survey interview about “stressful things that happen to women, emotions, and pregnancy,” and 1,581 women completed the diagnostic interview (Seng, Low, Sperlich, Ronis, & Liberzon, 2009). Of these, 1,049 fit a parent study cohort definition and were enrolled. The parent study cohorts included a lifetime PTSD group; a trauma-exposed, PTSD-negative control group; and a nonexposed control group. For the biobehavioral component, all women enrolled for follow-up were then invited to participate in a single day of salivary cortisol specimen collection for a diurnal component to test Hypothesis A. Women were invited to participate in the longitudinal component to test Hypothesis B if they returned specimen kits; adhered to the timing instructions; and reported no medication use, no smoking, and no endocrine disorders. Enrollment for the longitudinal profile analysis continued until a target sample of 135 was reached.
Variables The primary explanatory variable for the study was the four-level cohort variable: the trauma-exposed control group, the nonexposed control group, and two groups with participants who experienced PTSD (PTSD-only and PTSDD). The participants in the PTSD-only group met lifetime DSM-IV criteria (American Psychiatric Association, 1994), and the participants in the PTSD-D group also indicated presence of depersonalization and/or derealization consistent with the DSM-5 dissociative subtype definition (American Psychiatric Association, 2013). The cortisol diurnal profile included morning, late afternoon, and bedtime cortisol levels. The
281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
longitudinal profile repeated the same diurnal timing at targeted Prenatal Weeks 16, 20, 24, 28, 32, 36, 38, and 40, Postnatal Day 1, and Postnatal Weeks 2, 4, and 6. PTSD is often co-occurrent with other factors known to affect cortisol levels and that also are implicated in preterm birth. These confounding factors include depression, sociodemographic factors, and smoking. In studies of cortisol in pregnancy, it is important to control for gestational age in comparisons of cortisol levels in women at different weeks of gestation, because in descriptive studies, researchers showed that cortisol levels rise across gestation (D’AnnaHernandez, Ross, Natvig, & Laudenslager, 2011). Thus, covariates in the diurnal and longitudinal analyses included depression, smoking, and an index of cumulative sociodemographic risk factors as a proxy for chronic stress. The diurnal profile model also considered gestational age as a confounder.
Data Sources and Measurement Cortisol procedures. Data from our pilot work informed decisions about the protocol for collection of the saliva specimens (Seng, Low, BenAmi, & Liberzon, 2005). We used the minimal protocol needed to define the diurnal curve to maximize participation of a generalizable sample, with the morning (7 a.m. to 9 a.m., on awakening), late afternoon (4 p.m. to 6 p.m., before supper), and evening (10 p.m. to midnight, bedtime) time points. With Salivette tubes (Sarstedt, Newton, New Jersey) we mailed a health checklist to query for acute illness, medication use, and metabolic disorders that we then used to exclude women during the analysis phase. We included an illustrated instruction sheet that was also an adherence checklist to record the time of day and note smoking and eating in the 30 minutes before saliva collection. These procedures were consistent with other community sample studies that used ecologically collected saliva specimens for cortisol (Young & Breslau, 2004). Participants mailed specimen kits to the laboratory in padded, pre-addressed, postage-paid envelopes. We logged and froze these at 80 C until assay. We conducted a radioimmunoassay in triplicate with the use of the Coat-a-Count kit (Diagnostic Products Corporation, Los Angeles, California). Intra- and interassay coefficients of variance were less than 5% and 10%, respectively. We created variables for the cortisol values in natural units (log-transformed).
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
3
RESEARCH
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
Cortisol in Women With PTSD and Dissociative Subtype
Standardized telephone diagnostic interview data. Female interviewers with project-specific training conducted standardized telephone diagnostic interviews, and supervisors audited these interactions for quality across the life of the study. We reimbursed participants $25. We used established measures for psychiatric characteristics, including the Life Stressor Checklist (Wolfe & Kimerling, 1997), which includes queries for 30 types of potentially traumatic events, to determine trauma exposure. The National Women’s Study PTSD Module (NWS-PTSD) was validated in a primarily clinical sample of 528 women during the DSM-IV PTSD Field Trial (Kilpatrick et al., 1994). The NWS-PTSD module was validated against the Structured Clinical Interview for DSM-IV and attained a sensitivity of .99 and specificity of .79 (Resnick, Kilpatrick, Dansky, Saunders, & Best, 1993). In our sample of 395 participants, the internal consistency alpha for the NWS-PTSD was .91, and in the sample of 111 participants it was .93. Psychological dissociation was measured with the eight-item Taxon version of the Dissociative Experience Scales (DES-T; Waller, Putnam, & Carlson, 1996). The developers assessed this shortened version’s discriminant validity as a screening tool and found sensitivity of .80 and specificity of .80 (Waller et al., 1996). We modified the DES-T for telephone use with a response set of never, rarely, sometimes, often, and all the time. Participants were included in the PTSD-D group if they answered sometimes or more frequently to any of the three depersonalization and derealization items. In our sample of 395, the internal consistency was .78, and in the sample of 111 it was .73. We measured diagnosis of depression in the past year with the Composite International Diagnostic Interview, a structured psychiatric diagnostic interview designed to be used by lay interviewers. Its efficacy for diagnosis is supported by extensive field trial data on cross-national reliability and validity (Wittchen, 1994). We used demographic and smoking items from the Centers for Disease Control and Prevention’s Perinatal Risk Assessment and Monitoring Survey (Beck et al., 2002) that have been fielded across the United states since 1987. We summed African American race, being pregnant as a teen, having low education, living in poverty, and living in a zip code with greater rates of crime to obtain a 0 to 5 index as a proxy for sociodemographic risk or
4
stress. We classified a participant as a smoker if she continued to use one or more cigarettes per day after she became aware of her pregnancy. Although only women who we believed to be nonsmokers were enrolled for the longitudinal study, a small number of participants (n ¼ 6) indicated on the specimen procedure checklist that they smoked that day, so we also classified them as smokers. Medical record information. We calculated weeks of gestation on the basis of the expected date of birth determined at the first prenatal visit. Postnatal days and weeks were based on the actual infant birth date. Sample sizes. For Hypothesis A, the single-day, diurnal component, we included every woman who accepted the invitation to participate (N ¼ 449), and, after we eliminated those at gestational age greater than 25 weeks and those with metabolic diseases and medication use, our sample size was 395. For Hypothesis B, the longitudinal component, we estimated sample size (N ¼ 135) on the basis of our pilot data (Seng et al., 2005) to have a power greater than 80% at an alpha of .05 with the use of repeated measures. We retained 111 participants who provided specimens from the prenatal to the postnatal period. Statistical analysis. Preliminary steps in the analysis included examination of the missing data. In the single-day component to test Hypothesis A, there were 80 instances in which the cortisol assay values were less than the 0.003 mg/dl limit of detection of the assay. Because the interval for the undetectable region is small (0 to .003) and the proportion of specimens was small, we imputed the undetectable values with the limit of detection value. We maintained this procedure in the longitudinal study component. The cortisol data included a small proportion of instances in which the assay values were less than the limit of detection (0.003 mg/d), so we substituted this value (Croghan & Egeghy, 2003). Cortisol values followed log-normal distribution, so we examined descriptive data in natural units but used the log-transformed values to test and model the hypotheses. To test Hypothesis A, we used generalized estimating equations (GEEs) to consider the within-individual correlations of repeated-measures cortisol values (Ziegler, Kastner, & Blettner, 1998). We modeled the logtransformed cortisol levels using cohort, time of day, and their interactions. Depression, smoking,
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
RESEARCH
Seng, J. S. et al.
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
cumulative sociodemographic risk index, and gestational age were included as covariates. GEEs were also used to test Hypothesis B. The loess smooth method (Cleveland & Loader, 1996) was used to display smooth curves for cortisol levels at different time points for the four cohorts. The loess smooth plot showed that 34 weeks was a break-point for the trend of cortisol levels. Therefore, we partitioned the data into two sets (before or after 34 weeks) and analyzed them separately. We used GEEs to model the logtransformed cortisol levels by use of cohort, gestational week of specimen collection, and their interactions (see Supplemental Table S1). In addition, depression, smoking during pregnancy, and cumulative sociodemographic risk index were included as covariates. A p value of less than .05 was interpreted to be significant.
Results Sample for Hypothesis A Participants returned cortisol kits at a rate of 43% (449 of the 1,049 women enrolled for followup). In the single-day, diurnal component to test Hypothesis A, we included the 395 participants who remained after we eliminated 2 who had metabolic disorders, 9 who used corticosteroids or psychotropic medications, and 54 who returned specimens collected after 25 completed weeks of gestation. In this sample, the participants’ mean gestational week at the time of the specimen collection was 12.75 (standard deviation ¼ 4.39). This sample of 395 women (see Table 1) had a diverse, relatively low, socioeconomic status profile, and 33% were African American. Nineteen percent had histories of childhood maltreatment. A total of 27% had PTSD: 15% had PTSD-only and 12% had PTSD-D.
Results for Hypothesis A In the single-day diurnal component (see Figure 1), after we adjusted for smoking, depression, cumulative sociodemographic risk index, and gestational age, we did not detect significant effects of group assignment, time of day, or their interactions on the log-transformed cortisol levels. However, we found smoking and gestational age had significant effects on the logtransformed cortisol levels.
Sample for Hypothesis B A smaller subset of participants correctly followed the procedures in the single-day process
505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560
In early pregnancy, cortisol levels in participants in the dissociative subtype group were 8 times greater in the afternoon and 10 times greater at bedtime than those in participants in the nonexposed control group.
and agreed to submit saliva specimen kits up to 12 times (see Table 1). These participants were more likely to be White, to be living in poverty, and to have more than a high school education compared with those in the sample for the singleday component. Childhood maltreatment was experienced by 19% of these participants, and PTSD was present at a greater rate: 30% had PTSD-only and 7% had PTSD-D. On average, they returned a mean of 8.75 (range ¼ 4 to 13) morning and bedtime specimens and 8.50 (range ¼ 3 to 13) afternoon specimens.
Results for Hypothesis B We plotted the longitudinal gestational curves for morning, late afternoon, and bedtime (see Figure 2). For all four cohorts, the loess smooth plot showed that the cortisol levels increased up to 34 weeks gestation and then decreased during the afternoon and bedtime. However, in the morning, the cortisol levels remained constant before 34 weeks gestation and then decreased slightly after that. The PTSD-D cohort had the highest and flattest cortisol curves from early gestation to 6 weeks postpartum. The differences between the PTSD-D and nonexposed cohorts were greatest at afternoon and bedtime. During late gestation, the differences were smaller because all cohorts demonstrated the physiologic gestational cortisol rise. The GEE models showed that participants in the PTSD-D group had significantly greater morning, afternoon, and bedtime cortisol levels than did participants in the nonexposed control group before gestational week 34 in all adjusted models. The PTSD-only group was not significantly different from the other groups. After gestational week 34, adjusted models of between-group comparisons (see Supplemental Table S2) showed no significant differences in morning, afternoon, or bedtime cortisol levels and no significant group membership by week interactions. To quantify the extent of the differences for participants in the PTSD-D group at various weeks, we conducted an additional calculation from the GEEs (see Supplemental Table S3). The
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
5
RESEARCH
561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616
Cortisol in Women With PTSD and Dissociative Subtype
Table 1: Sample Characteristics One Day, n ¼ 395
Longitudinal, n ¼ 111
c2/t
132 (33.4)
23 (20.7)
6.57
.010
Pregnant as a teen (18–20 years old)
62 (15.7)
13 (11.7)
1.09
.297
Living in poverty (income <$15,000)
65 (16.5)
35 (31.5)
12.42
<.001
High school education or less
133 (33.7)
15 (13.5)
17.01
<.001
Residence in zip code with greater rate of crime
122 (30.9)
24 (21.6)
3.62
.060
1.30 1.73
0.99 1.58
1.79
.075
Characteristic, n (%) or Mean SD
p
Sociodemographic risk factors African American race
Cumulative sociodemographic risk index (0–5)
a
Childhood maltreatment history Any sexual, physical, or emotional abuse or neglectb Cumulative types of childhood maltreatment (0–5)
76 (19.2)
21 (18.9)
0.45 0.94
0.46 1.01
Group membership Nonexposed control group
139 (35.2)
Trauma-exposed control group
0.0058
.939
0.094
.926
11.53
.009
34 (30.6)
149 (37.7)
38 (34.2)
PTSD-only group
58 (14.7)
31 (27.9)
PTSD-D group
49 (12.4)
8 (7.2)
Major depression in the past year
39 (9.9)
16 (14.4)
1.84
.174
Smoking during pregnancy
29 (7.3)
6 (5.4)
0.50
.478
Co-occurring conditions
Gestational age
12.75 4.39
Diurnal cortisol levels Morning (peak) cortisol, mg/dl, w8 a.m. (n ¼ 392)
0.41 0.24
Afternoon cortisol, mg/dl, w4 p.m. (n ¼ 387)
0.11 0.12
Bedtime (nadir) cortisol, mg/dl, w10 p.m. (n ¼ 394)
0.09 0.12
Note. PTSD ¼ posttraumatic stress disorder; PTSD-D ¼ dissociative subtype posttraumatic stress disorder. a An index of cumulative sociodemographic risk was created from the sum of the five risk factors for PTSD (listed above). bChild maltreatment history was positive if physical abuse, contact sexual abuse, penetration sexual abuse, emotional abuse or neglect, or physical neglect before age 16 years were reported. An index was created from the sum of these five types of exposure to maltreatment trauma.
print & web 4C=FPO
log-transformed cortisol levels were retransformed by exponentiation. In early pregnancy (Gestational Week 8), the cortisol levels of
Discussion Q4 Figure 1. Log-transformed cortisol levels across 1 day by the four groups.
6
participants in the PTSD-D group were 2 times greater in the morning, 8 times greater in the afternoon, and 10 times greater at bedtime than those of the participants in the nonexposed control group, whereas in late pregnancy (Gestational Week 32), participants in the PTSD-D group had cortisol levels that were less than 2 times greater in the morning and 1.5 times greater levels in the afternoon and at bedtime compared with participants in the nonexposed control group.
With this biobehavioral component of the STACY Project, we have begun to fill a gap in knowledge of the longitudinal pattern of cortisol across the prenatal and postnatal periods for women with PTSD, including specific results related to the
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
RESEARCH
Seng, J. S. et al.
729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
associated with childhood maltreatment, were predictive of lower birth weight and shorter gestation. Dissociation is often present in PTSD as a sequela of childhood maltreatment (Stein et al., 2013) and may be a marker of severity and chronicity (Liotti, 2004), and it may also be a marker of severe stress in pregnancy.
Strengths of this study include the prospective cohort design; the diverse sample, which is large for a biobehavioral study; and the use of established measures to characterize trauma history, PTSD, depression, and dissociation. Furthermore, this was a community sample rather than a psychiatric sample of women who become pregnant; therefore, findings are highly generalizable. Some limitations should also be considered. We attempted to exclude smokers, but 6 of the 111 women disclosed smoking during at least 1 day, and this may have increased cortisol levels; thus, we controlled for smoking. Our specimens were collected at home and sent by mail; therefore, they have greater ecological validity but may have greater error variance than specimens collected in the laboratory. We used lifetime PTSD diagnosis, past-year depression, and dissociation measured as a trait. Current mental health symptom load or diagnoses may show stronger associations.
print & web 4C=FPO
673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728
Figure 2. Smoothed curves through morning, afternoon, and bedtime data points across weeks for the four groups.
dissociative subtype of PTSD. Our main finding was that participants with PTSD-D, and not those with nondissociative PTSD or trauma, had higher and flatter levels of cortisol during early gestation and into the early postpartum period. For participants in the PTSD-D group, the difference was most apparent in early gestation, which is a critical period. An elevated cortisol level has been identified as a risk factor for preterm birth by direct effect on onset of labor and indirect effect on inflammatory processes (Green et al., 2005). The perinatal outcomes of the parent study, published in 2011 (Seng et al., 2011), showed that lifetime PTSD, and even more so, PTSD
Our findings have numerous implications for further research. First, our finding that participants with PTSD-D had greater levels of cortisol than did participants with PTSD-only indicates that in the future, researchers should attend to women with the dissociative subtype and distinguish them from those with PTSD-only. Second, the greatest variation in cortisol levels associated with PTSD-D was in early gestation. Investigators should include specimens from early pregnancy before the physiologic gestational increase in cortisol to observe maximum effect sizes. Third, with regard to the finding that dissociation may lead to increased biological stress, the question arises of whether this is a characteristic of an individual woman. An alternative explanation is that psychosexual aspects of pregnancy act as triggers to dissociative symptoms in survivors of childhood maltreatment in particular and result in persistent, enhanced stress reactivity across gestation that is specific to a woman’s experience of pregnancy (Montgomery, 2013; Seng et al., 2009). Perinatal studies of reactivity to a stressor or trigger (e.g., internal pelvic examination) could help answer this question and indicate
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
7
RESEARCH
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
Cortisol in Women With PTSD and Dissociative Subtype
Cortisol levels in pregnant women with dissociation and posttraumatic stress are consistent with research that shows toxic stress from maltreatment and other adverse childhood experiences.
whether the usual blunted reactivity to stress in pregnancy is observed in this group (Schalinski et al., 2015; Zaba et al., 2015). Fourth, in the perinatal outcomes study of veteran women by Shaw et al. (2014), PTSD in pregnancy and history of military sexual trauma were predictive of shorter gestation. Military sexual trauma has been associated with greater than representative rates of childhood maltreatment trauma (Kelly, Skelton, Patel, & Bradley, 2011). Future research also should be designed to determine whether dissociation in particular or cumulative childhood and adult trauma; psychiatric symptom burden from complex comorbidities, including depression; and the stress of adverse life circumstances that accompany these together exert a toxic stress influence on a woman (Shonkoff et al., 2012).
because the public health cost is so high (Shonkoff, et al., 2012). For perinatal nurses, midwives, obstetricians, and other members of the perinatal care team, focus on childhood maltreatment history and complex posttraumatic sequelae is part of renewed attention to psychosocial care as a means to improve individual and population outcomes (Renfrew et al., 2014). Trauma-informed approaches to care and trauma-specific interventions are being developed for perinatal care (Cuthbert, 2015). Although some women with histories of childhood maltreatment are resilient or recovered by the time they become pregnant, these biological findings indicate that some are very adversely affected psychologically and very stressed during the childbearing year. We can screen and apply a stepped approach to maternity care that includes case-finding and interventions for women with PTSD, posttraumatic depression, and PTSD-D. The stakes are high, the reasons are compelling, and theoretical, qualitative, and service delivery elements are emerging to advance the field in this direction.
Conclusions Finally, studying maternal cortisol levels is a starting point, but other HPA axis components more specific to pregnancy should be considered, including placental corticotrophin-releasing hormone, which was elevated in childhood maltreatment survivors (Moog et al., 2015). Studies of the other stress-response systems that are affected by the HPA axis and associated with childbearing outcomes also warrant study, especially the oxytocin system (Teicher et al., 2003). Our results have clinical practice implications, especially because these biological findings are consistent with reports from clinical and qualitative studies. Since the first narrative reviews and case reports emerged in the early 1990s, researchers have suggested that childhood maltreatment; the sociodemographic disadvantage and risk exposures associated with it, including repeat victimization; and the substantial burden of mental health sequelae all affect women’s childbearing experiences. There is mounting evidence to suggest that these affect a range of childbearing outcomes as well (Sperlich, 2015). The conceptualization of toxic stress is becoming shared across health disciplines
8
We observed very high, perhaps toxic, levels of the stress hormone cortisol in a community sample of pregnant women with the dissociative subtype of PTSD. This complex form of PTSD in adults is strongly associated with history of childhood maltreatment. Knowledge that childhood maltreatment (and other adverse childhood experiences) causes toxic stress is rapidly increasing in public health and pediatrics fields because these experiences have been convincingly linked with disease across the lifespan (Shonkoff et al., 2012). Our data are important because the developmental origin of disease theory (Reynolds, 2013) and research also links high stress and cortisol levels in the fetal environment with early morbidity and mortality (Reynolds et al., 2013). Further research is warranted because PTSD-D–related toxic stress during pregnancy could potentially be mitigated to improve a woman’s perinatal mental health and the intra-uterine environment.
Acknowledgment Funded by NIH grant R01 NR008767 (Seng, principal investigator). The authors acknowledge the long-term contributions of co-investigator Dr. David Ronis.
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896
RESEARCH
Seng, J. S. et al.
897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952
Q1
Supplementary Material Note: To access the supplementary material that accompanies this article, visit the online version of the Journal of Obstetric, Gynecologic, & Neonatal Nursing at http://jognn.org and at https://doi.org/10.1016/j.jogn.2017.10.008.
953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
Lanius, R. A., Vermetten, E., Loewenstein, R. J., Brand, B., Schmahl, C., Bremner, J. D., & Spiegel, D. (2010). Emotion modulation in PTSD: Clinical and neurobiological evidence for a dissociative subtype. American Journal of Psychiatry, 167(6), 640–647. https://doi.org/10.1176/appi.ajp.2009.09081168 Latendresse, G. (2009). The interaction between chronic stress and pregnancy: Preterm birth from a biobehavioral perspective. Journal of Midwifery & Women’s Health, 54(1), 8–17. https://doi. org/10.1016/j.jmwh.2008.08.001
REFERENCES
Q2
Liotti, G. (2004). Trauma, dissociation, and disorganized attachment:
Allen, M. C. (2008). Neurodevelopmental outcomes of preterm infants. Current Opinion in Neurology, 21(2), 123–128. https://doi.org/ 10.1097/WCO.0b013e3282f88bb4 American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Arlington, VA: Author. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: Author.
Three strands of a single braid. Psychotherapy, 41(4), 472–486. https://doi.org/10.1037/0033-3204.41.4.472 Montgomery, E. (2013). Feeling safe: A meta-synthesis of the maternity care needs of women who were sexually abused in childhood. Birth: Issues in Perinatal Care, 40(2), 88–95. https://doi.org/10. 1111/birt.12043 Moog, N. K., Buss, C., Entringer, S., Shahbaba, B., Gillen, D. L.,
Beck, L. F., Morrow, B., Lipscomb, L. E., Johnson, C. H., Gaffield, M.
Hobel, C. J., & Wadhwa, P. D. (2015). Maternal exposure to
E., Rogers, M., & Gilbert, B. C. (2002). Prevalence of selected
childhood trauma is associated during pregnancy with
maternal behaviors and experiences, Pregnancy Risk Assess-
placental–fetal stress physiology. Biological Psychiatry, 79(10),
ment Monitoring System (PRAMS), 1999. Morbidity and Mor-
831–839. https://doi.org/10.1016/j.biopsych.2015.08.032
tality Weekly Report, 51(2), 1–27. Retrieved from https://www.
O’Connor, T. G., Tang, W., Gilchrist, M. A., Moynihan, J. A., Pressman,
cdc.gov/mmwr/preview/mmwrhtmL/ss5102a1.htm
E. K., & Blackmore, E. R. (2014). Diurnal cortisol patterns and
Cleveland, W. S., & Loader, C. (1996). Smoothing by local regression:
psychiatric symptoms in pregnancy: Short-term longitudinal
Principles and methods. In W. Ha¨rdel & M. G. Schimek (Eds.),
study. Biological Psychology, 96, 35–41. https://doi.org/10.
Statistical theory and computational aspects of smoothing (pp. 10–49). New York, NY: Springer.
1016/j.biopsycho.2013.11.002 Renfrew, M. J., Homer, C. S. E., Downe, S., McFadden, A., Muir, N.,
Croghan, C., & Egeghy, P. P. (2003, September 22–24). Methods of
Prentice, T., & ten Hoope-Bender, P. (2014). Midwifery: An
dealing with values below the limit of detection using SAS. Paper
executive summary for The Lancet’s series. Retrieved from
presented at Southeastern SAS User Group, St. Petersburg, FL.
http://www.thelancet.com/pb/assets/raw/Lancet/stories/series/
Cuthbert, C. (2015). Where are we on the journey toward trauma-
midwifery/midwifery_exec_summ.pdf
specific interventions and treatments for the perinatal period?
Resnick, H. S., Kilpatrick, D. G., Dansky, B. S., Saunders, B. E., &
In J. Seng & J. Taylor (Eds.), Trauma informed care in the
Best, C. L. (1993). Prevalence of civilian trauma and post-
perinatal period. Edinburgh: Dunedin Academic Press.
traumatic stress disorder in a representative national sample of
D’Anna-Hernandez, K. L., Ross, R. G., Natvig, C. L., & Laudenslager,
women. Journal of Consulting and Clinical Psychology, 61(6),
M. L. (2011). Hair cortisol levels as a retrospective marker of
984–991. https://doi.org/10.1037/0022-006X.61.6.984
hypothalamic–pituitary axis activity throughout pregnancy:
Reynolds, R. M. (2013). Glucocorticoid excess and the developmental
Comparison to salivary cortisol. Physiology & Behavior, 104(2),
origins of disease: Two decades of testing the hypothesis—
348–353. https://doi.org/10.1016/j.physbeh.2011.02.041
2012 Curt Richter Award Winner. Psychoneuroendocrinology,
Davis, E. P., & Sandman, C. A. (2010). The timing of prenatal exposure
38(1), 1–11. https://doi.org/10.1016/j.psyneuen.2012.08.012
to maternal cortisol and psychosocial stress is associated with
Reynolds, R. M., Labad, J., Buss, C., Ghaemmaghami, P., & Raikkonen,
human infant cognitive development. Child Development, 81,
K. (2013). Transmitting biological effects of stress in utero: Impli-
131–148. https://doi.org/10.1111/j.1467-8624.2009.01385.x
cations for mother and offspring. Psychoneuroendocrinology,
Green, N. S., Damus, K., Simpson, J. L., Iams, J., Reece, E. A., Hobel,
38(9), 1843–1849. https://doi.org/10.1016/j.psyneuen.2013.05.018
C. J., … March Of Dimes Scientific Advisory Committee On
Risnes, K. R., Vatten, L. J., Baker, J. L., Jameson, K., Sovio, U.,
Prematurity. (2005). Research agenda for preterm birth: Rec-
Kajantie, E., … Bracken, M. B. (2011). Birthweight and mortality
ommendations from the March of Dimes. American Journal of
in adulthood: A systematic review and meta-analysis. Interna-
Obstetrics and Gynecology, 193(3, Part 1), 626–635. https://
tional Journal of Epidemiology, 40(3), 647–661. https://doi.org/
doi.org/10.1016/j.ajog.2005.02.106
10.1093/ije/dyq267
Kelly, U. A., Skelton, K., Patel, M., & Bradley, B. (2011). More than
Schalinski, I., Elbert, T., Steudte-Schmiedgen, S., & Kirschbaum, C.
military sexual trauma: Interpersonal violence, PTSD, and
(2015). The cortisol paradox of trauma-related disorders: Lower
mental health in women veterans. Research in Nursing &
phasic responses but higher tonic levels of cortisol are asso-
Health, 34(6), 457–467. https://doi.org/10.1002/nur.20453
ciated with sexual abuse in childhood. PLoS One, 10(8),
Kilpatrick, D. G., Resnick, H. S., Freedy, J. R., Pelcovitz, D., Resick, P.
e0136921. https://doi.org/10.1371/journal.pone.0136921
A., Roth, S., & Van der Kolk, B. (1994). The posttraumatic stress
Schreier, H. M., Bosquet Enlow, M., Ritz, T., Coull, B. A., Gennings, C.,
disorder field trial: Emphasis on criterion a and overall PTSD
Wright, R. O., & Wright, R. J. (2016). Lifetime exposure to
diagnosis. In T. A. Widiger, A. J. Frances, H. A. Pincus, & R.
traumatic and other stressful life events and hair cortisol in a
Ross (Eds.), DSM-IV sourcebook (Vol. 3). Arlington, VA: Amer-
multi-racial/ethnic sample of pregnant women. Stress, 19(1),
ican Psychiatric Publishing.
45–52. https://doi.org/10.3109/10253890.2015.1117447
Lackritz, E. M., Wilson, C. B., Guttmacher, A. E., Howse, J. L., Eng-
Schreier, H. M. C., Enlow, M. B., Ritz, T., Gennings, C., & Wright, R. J.
mann, C. M., Rubens, C. E., … Setting, P. B. R. P. (2013).
(2015). Childhood abuse is associated with increased hair
A solution pathway for preterm birth: Accelerating a priority
cortisol levels among urban pregnant women. Journal of
research agenda. Lancet Global Health, 1(6), E328–E330.
Epidemiology and Community Health, 69(12), 1169–1174.
https://doi.org/10.1016/S2214-109X(13)70120-7
https://doi.org/10.1136/jech-2015-205541
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
9
RESEARCH
Cortisol in Women With PTSD and Dissociative Subtype
1009
Seng, J. S., Low, L. K., Ben-Ami, D., & Liberzon, I. (2005). Cortisol level
1010
and perinatal outcome in pregnant women with posttraumatic
1011 1012
Taylor (Eds.), Trauma informed care in the perinatal period.
1052
Edinburgh, UK: Dunedin Academic Press.
1053
stress disorder: A pilot study. Journal of Midwifery & Women’s
Stein, D. J., Koenen, K. C., Friedman, M. J., Hill, E., McLaughlin, K. A.,
Health, 50(5), 392–398. https://doi.org/10.1016/j.jmwh.2005.
Petukhova, M., … Kessler, R. C. (2013). Dissociation in post-
04.024
traumatic stress disorder: Evidence from the world mental
1055
health surveys. Biological Psychiatry, 73(4), 302–312. https://
1056
doi.org/10.1016/j.biopsych.2012.08.022
1057
1013
Seng, J. S., Low, L. K., Sperlich, M., Ronis, D. L., & Liberzon, I. (2009).
1014
Prevalence, trauma history, and risk for posttraumatic stress
1015
disorder among nulliparous women in maternity care. Obstet-
Teicher, M. H., Andersen, S. L., Polcari, A., Anderson, C. M., Navalta,
rics & Gynecology, 114(4), 839–847. https://doi.org/10.1097/
C. P., & Kim, D. M. (2003). The neurobiological consequences
AOG.0b013e3181b8f8a2
of early stress and childhood maltreatment. Neuroscience and
1016
1054
1058 1059
1017
Seng, J. S., Low, L. K., Sperlich, M., Ronis, D. L., & Liberzon, I. (2011).
1018
Post-traumatic stress disorder, child abuse history, birthweight
1019
and gestational age: A prospective cohort study. British Journal
Waller, N., Putnam, F. W., & Carlson, E. B. (1996). Types of dissociation
1062
1020
of Obstetrics and Gynecology, 118(11), 1329–1339. https://doi.
and dissociative types: A taxometric analysis of dissociative
1063
org/10.1111/j.1471-0528.2011.03071.x
experiences. Psychological Methods, 1(3), 300–321. https://
1021
Seng, J. S., Rauch, S. A., Resnick, H., Reed, C. D., King, A., Low, L. K.,
Biobehavioral Reviews, 27(1–2), 33–44. https://doi.org/10.1016/
1060
S0149-7634(03)00007-1
1061
doi.org/10.1037/1082-989X.1.3.300
1064
1022
… Liberzon, I. (2010). Exploring posttraumatic stress disorder
Wittchen, H. U. (1994). Reliability and validity studies of the WHO—
1065
1023
symptom profile among pregnant women. Journal of Psycho-
Composite International Diagnostic Interview (CIDI): A critical
1066
1024
somatic Obstetrics & Gynecology, 31(3), 176–187. https://doi.
review. Journal of Psychiatric Research, 28(1), 57–84. https://
1067
1025
org/10.3109/0167482X.2010.486453
doi.org/10.1016/0022-3956(94)90036-1
1068
1026
Shannon, M., King, T. L., & Kennedy, H. P. (2007). Allostasis: A theo-
Wolfe, J., & Kimerling, R. (1997). Gender issues in the assess-
1069
retical framework for understanding and evaluating perinatal
ment of posttraumatic stress disorder. In J. P. Wilson & T.
1027
health outcomes. Journal of Obstetric, Gynecologic, & Neonatal
M. Keane (Eds.), Assessing psychological trauma and
1070
1028
Nursing, 36(2), 125–134. https://doi.org/10.1111/j.1552-6909.
PTSD (pp. 192–238). New York, NY: Guilford Press.
1071
1029
2007.00126.x
Yonkers, K. A., Smith, M. V., Forray, A., Epperson, C. N., Costello, D.,
1072
Shaw, J. G., Asch, S. M., Kimerling, R., Frayne, S. M., Shaw, K. A., &
Lin, H., & Belanger, K. (2014). Pregnant women with post-
1073
Phibbs, C. S. (2014). Posttraumatic stress disorder and risk of
traumatic stress disorder and risk of preterm birth. JAMA Psy-
1030 1031 1032
spontaneous preterm birth. Obstetrics & Gynecology, 124(6),
chiatry, 71(8), 897–904. https://doi.org/10.1001/jamapsychiatry.
1111–1119. https://doi.org/10.1097/AOG.0000000000000542
2014.558
1074 1075
1033
Shonkoff, J. P., Garner, A. S., & Committee on Psychosocial Aspects of
Young, E. A., & Breslau, N. (2004). Saliva cortisol in posttraumatic
1076
1034
Child and Family Health, Committee on Early Childhood,
stress disorder: A community epidemiologic study. Biological
1077
1035
Adoption, and Dependent Care, and Section on Developmental
Psychiatry, 56(3), 205–209. https://doi.org/10.1016/j.biopsych.
1078
and Behavioral Pediatrics. (2012). The lifelong effects of early
2004.05.011
1036
childhood adversity and toxic stress. Pediatrics, 129(1), e232–
1037
Zaba, M., Kirmeier, T., Ionescu, I. A., Wollweber, B., Buell, D. R., Gall-
e246. https://doi.org/10.1542/peds.2011-2663
1079
Kleebach, D. J., … Schmidt, U. (2015). Identification and
1080
1038
Simeon, D., Knutelska, M., Yehuda, R., Putnam, F., Schmeidler, J., &
characterization of HPA-axis reactivity endophenotypes in a
1081
1039
Smith, L. M. (2007). Hypothalamic–pituitary–adrenal axis func-
cohort of female PTSD patients. Psychoneuroendocrinology, 55,
1082
1040
tion in dissociative disorders, post-traumatic stress disorder,
102–115. https://doi.org/10.1016/j.psyneuen.2015.02.005
1083
and healthy volunteers. Biological Psychiatry, 61(8), 966–973.
Ziegler, A., Kastner, C., & Blettner, M. (1998). The generalised esti-
1041
https://doi.org/10.1016/j.biopsych.2006.07.030
1042
Sperlich, M. (2015). How does focusing on posttraumatic stress dis-
1043
order shift perinatal mental health paradigms? In J. Seng & J.
mating equations: An annotated bibliography. Biometrical
1084
Journal, 40(2), 115–139. . <115::Aid-Bimj115>3.0.Co;2–6.
1085
https://doi.org/10.1002/(Sici)1521-4036(199806)40:2
1086
1044
1087
1045
1088
1046
1089
1047
1090
1048
1091
1049
1092
1050
1093
1051
1094
10
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
Gestational and Postnatal Cortisol Profiles of Women With Posttraumatic Stress Disorder and the Dissociative Subtype Julia S. Seng, Yang Li, James J. Yang, Anthony P. King, Lisa M. Kane Low, Mickey Sperlich, Heather Rowe, Q3 Hyunhwa Lee, Maria Muzik, Julian D. Ford, and Israel Liberzon
Correspondence Julia S. Seng, PhD, CNM, FAAN, University of Michigan, School of Nursing, 400 N. Ingalls St., Room 3220, Ann Arbor, MI 48109. [email protected]
ABSTRACT
Keywords adverse childhood events cortisol developmental origins of health and disease dissociation dissociative subtype gestation posttraumatic stress disorder toxic stress
Participants: Women expecting their first infants who fit with one of four cohorts: a nonexposed control group, a trauma-exposed control group, a group with PTSD, and a group with the dissociative subtype of PTSD.
Objective: To test the hypothesis that women with posttraumatic stress disorder (PTSD) have greater salivary cortisol levels across the diurnal curve and throughout gestation, birth, and the postpartum period than women who do not have PTSD. Design: Prospective, longitudinal, biobehavioral cohort study. Setting: Prenatal clinics at academic health centers in the Midwest region of the United States.
Methods: In the first half of pregnancy, 395 women provided three salivary cortisol specimens on a single day for diurnal data. A subsample of 111 women provided three salivary cortisol specimens per day, 12 times, from early pregnancy to 6 weeks postpartum for longitudinal data. Trauma history, PTSD, and dissociative symptoms were measured via standardized telephone diagnostic interviews with the use of validated epidemiologic measures. Generalized estimating equations were used to determine group differences. Results: Generalized estimating equations were used to show that women with the dissociative subtype of PTSD had the highest and flattest gestational cortisol level curves. The difference was greatest in early pregnancy, when participants in the dissociative subtype group had cortisol levels 8 times greater in the afternoon and 10 times greater at bedtime than those in the nonexposed control group. Conclusion: Women with the dissociative subtype of PTSD, a complex form associated with a history of childhood maltreatment, may have toxic levels of cortisol that contribute to intergenerational patterns of adverse health outcomes.
JOGNN,
-, -–-;
2017. https://doi.org/10.1016/j.jogn.2017.10.008 Accepted October 2017
Julia S. Seng, PhD, CNM, FAAN, is a professor in the School of Nursing, University of Michigan, Ann Arbor, MI.
(Continued)
The authors report no conflict of interest or relevant financial relationships.
http://jognn.org
P
erinatal outcomes are indicative of the health of a society’s population (Allen, 2008; Risnes et al., 2011). Maternal psychosocial stress has been implicated as one of five pathways to the adverse perinatal outcome of premature birth (Green et al., 2005). It is also associated with lower birth weight (Shannon, King, & Kennedy, 2007) and is a risk factor in the fetus for early lifespan morbidity and mortality (Reynolds, Labad, Buss, Ghaemmaghami, & Raikkonen, 2013). Biological mediators of social determinants of health, including psychosocial stressors, are a priority focus for research to redress preterm birth (Lackritz et al., 2013).
Posttraumatic stress disorder (PTSD) is a severe form of psychological stress that affects 8% of pregnant women (Seng et al., 2010). However, PTSD has only recently been studied in relation to perinatal outcomes. Recent studies in large prenatal clinic samples showed significant associations of PTSD with preterm birth or shorter gestation, especially in current as opposed to lifetime remitted PTSD, PTSD with depression (Seng, Low, Sperlich, Ronis, & Liberzon, 2011; Shaw et al., 2014; Yonkers et al., 2014), and PTSD that is subsequent to childhood maltreatment (Seng et al., 2011) or military sexual trauma (Shaw et al., 2014).
ª 2017 AWHONN, the Association of Women’s Health, Obstetric and Neonatal Nurses. Published by Elsevier Inc. All rights reserved.
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
1
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
RESEARCH
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168
Cortisol in Women With PTSD and Dissociative Subtype
Posttraumatic stress disorder is associated with adverse perinatal outcomes. Elevated cortisol is a plausible mechanism for adverse outcomes and may affect maternal prenatal physiology.
Yang Li, MS, is a doctoral candidate in the School of Nursing, University of Michigan, Ann Arbor, MI. James J. Yang, PhD, is an associate research scientist in the Department of Psychiatry, University of Michigan, Ann Arbor, MI. Anthony P. King, PhD, is a research assistant professor in the Department of Psychiatry, University of Michigan, Ann Arbor, MI. Lisa M. Kane Low, PhD, CNM, FACNM, is an associate professor in the School of Nursing, University of Michigan, Ann Arbor, MI. Mickey Sperlich, PhD, MSW, CPM, is an assistant professor of social work, University at Buffalo (SUNY), Buffalo, NY.
With the release of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013), a diagnosis of PTSD may include a dissociative subtype characterized by altered perception of the environment in the form of derealization or depersonalization. In an epidemiologic analysis of 25,000 adults across 26 nations, 14.4% of those with PTSD fit the dissociative subtype classification (Stein et al., 2013). Dissociation has been associated with more severe PTSD that is diagnosed earlier in the lifespan and is associated with a greater number of lifetime trauma exposures, including history of childhood maltreatment or cumulative adverse childhood experiences (Liotti, 2004; Shonkoff et al., 2012; Stein et al., 2013). Thus, dissociation is an important indicator of severity and chronicity. It also may be a more proximal predictor of shorter gestation than the previously identified risk factor of childhood maltreatment (Seng et al., 2011).
associated with adverse perinatal outcomes (Green et al., 2005), slower infant mental development (Davis & Sandman, 2010), and adverse consequences to lifespan health (Reynolds, 2013), cortisol is an important biomarker for perinatal and developmental research. Data about the effects of PTSD on the HPA axis in pregnant women are very limited. In a recent study on cortisol levels in women at 21 and 34 weeks gestation that was focused on depression but entered trauma history and PTSD as covariates, researchers found no evidence to indicate an effect of PTSD after depression was taken into account (O’Connor et al., 2014). One recent study of gestational cortisol levels in women’s hair showed effects of childhood abuse in one analysis (Schreier, Enlow, Ritz, Gennings, & Wright, 2015) and cumulative trauma in another (Schreier et al., 2016). Depression and PTSD were covariates used to adjust the model in the latter analysis, but the researchers did not report the relative effects. The dissociative subtype of PTSD, a new diagnosis in 2013 (American Psychiatric Association, 2013), requires investigation in perinatal research. Our objective was to contribute to the literature on perinatal psychosocial stress by describing cortisol levels across the diurnal curve for 1 day in the first half of pregnancy and across the longitudinal gestational curve, including time points during pregnancy, birth, and the postnatal period, in women with PTSD, including a subset with the dissociative subtype. We examined four cohorts: a nonexposed control group, a resilient trauma-exposed control group, a group with PTSD (PTSD-only), and a group with the dissociative subtype of PTSD (PTSD-D). Our hypotheses were that (a) women with PTSD have greater diurnal levels of cortisol and (b) these greater levels of cortisol persist across gestation and the postnatal period. We further hypothesized that women with the dissociative subtype would have the greatest levels of diurnal and gestational cortisol.
Julian D. Ford, PhD, ABPP, is a professor in the Department of Psychiatry, University of Connecticut School of Medicine, Farmington, CT.
Researchers of the specific physiology of the dissociative subtype of PTSD showed a phenotype that appears to have different patterns of emotion dysregulation (Lanius et al., 2010). Although this is not an analysis of the DSM-5 dissociative subtype per se, researchers in one recent study distinguished a phenotype in women with PTSD and early-life trauma that was characterized by dissociative symptomatology and blunted cortisol reactivity (Zaba et al., 2015). Another group of researchers studied cortisol in women with PTSD and childhood sexual abuse and found blunted cortisol reactivity and greater basal levels (Schalinski, Elbert, SteudteSchmiedgen, & Kirschbaum, 2015). An earlier study team distinguished a dissociative disorder phenotype from PTSD that was based on elevated basal cortisol levels and blunted reactivity (Simeon et al., 2007). Together, evidence from these studies indicates that exposure to the early relational trauma that predisposes a person to dissociation and PTSD may affect that individual’s short- and long-term cortisol patterns.
Israel Liberzon, MD, is a professor in the Department of Psychiatry, University of Michigan, Ann Arbor, MI.
Because in utero elevations in hypothalamic– pituitary–adrenal (HPA) axis hormones are
2
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
Heather Rowe, PhD, BSc(Hons), is a senior research fellow in the Jean Hailes Research Unit, Monash University, Melbourne, Victoria, Australia. Hyunhwa Lee, PhD, PMHNP-BC, is an assistant professor in the School of Nursing, University of Nevada, Las Vegas, Las Vegas, NV. Maria Muzik, MD, MS, is an associate professor and Co-Director of the Women and Infants Mental Health Program, Department of Psychiatry, University of Michigan, Ann Arbor, MI.
Methods Design and Setting This prospective, ecological study of diurnal and gestational salivary cortisol profiles was a biobehavioral component of a larger clinical cohort study of the effect of PTSD on childbearing outcomes in a community sample of women in prenatal care. The parent study is the Stress, Trauma, Anxiety, and the Childbearing Year
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
RESEARCH
Seng, J. S. et al.
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
(STACY) Project (NIH R01 NR008767, J. S. Seng, principal investigator). The STACY Project took place in prenatal clinics at three academic health systems in Michigan. Institutional review boards approved the study at all three sites.
Participants In the parent clinical study, investigators recruited a convenience sample designed to be generalizable to community prenatal clinic populations. The women invited to the biobehavioral component were recruited from the parent study sample. For participation in the parent study, eligible women were at least 18 years of age, spoke English, were expecting their first infants, and initiated prenatal care at less than 28 weeks gestation. Maternity clinic nurses invited eligible women to participate in a survey interview about “stressful things that happen to women, emotions, and pregnancy,” and 1,581 women completed the diagnostic interview (Seng, Low, Sperlich, Ronis, & Liberzon, 2009). Of these, 1,049 fit a parent study cohort definition and were enrolled. The parent study cohorts included a lifetime PTSD group; a trauma-exposed, PTSD-negative control group; and a nonexposed control group. For the biobehavioral component, all women enrolled for follow-up were then invited to participate in a single day of salivary cortisol specimen collection for a diurnal component to test Hypothesis A. Women were invited to participate in the longitudinal component to test Hypothesis B if they returned specimen kits; adhered to the timing instructions; and reported no medication use, no smoking, and no endocrine disorders. Enrollment for the longitudinal profile analysis continued until a target sample of 135 was reached.
Variables The primary explanatory variable for the study was the four-level cohort variable: the trauma-exposed control group, the nonexposed control group, and two groups with participants who experienced PTSD (PTSD-only and PTSDD). The participants in the PTSD-only group met lifetime DSM-IV criteria (American Psychiatric Association, 1994), and the participants in the PTSD-D group also indicated presence of depersonalization and/or derealization consistent with the DSM-5 dissociative subtype definition (American Psychiatric Association, 2013). The cortisol diurnal profile included morning, late afternoon, and bedtime cortisol levels. The
281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
longitudinal profile repeated the same diurnal timing at targeted Prenatal Weeks 16, 20, 24, 28, 32, 36, 38, and 40, Postnatal Day 1, and Postnatal Weeks 2, 4, and 6. PTSD is often co-occurrent with other factors known to affect cortisol levels and that also are implicated in preterm birth. These confounding factors include depression, sociodemographic factors, and smoking. In studies of cortisol in pregnancy, it is important to control for gestational age in comparisons of cortisol levels in women at different weeks of gestation, because in descriptive studies, researchers showed that cortisol levels rise across gestation (D’AnnaHernandez, Ross, Natvig, & Laudenslager, 2011). Thus, covariates in the diurnal and longitudinal analyses included depression, smoking, and an index of cumulative sociodemographic risk factors as a proxy for chronic stress. The diurnal profile model also considered gestational age as a confounder.
Data Sources and Measurement Cortisol procedures. Data from our pilot work informed decisions about the protocol for collection of the saliva specimens (Seng, Low, BenAmi, & Liberzon, 2005). We used the minimal protocol needed to define the diurnal curve to maximize participation of a generalizable sample, with the morning (7 a.m. to 9 a.m., on awakening), late afternoon (4 p.m. to 6 p.m., before supper), and evening (10 p.m. to midnight, bedtime) time points. With Salivette tubes (Sarstedt, Newton, New Jersey) we mailed a health checklist to query for acute illness, medication use, and metabolic disorders that we then used to exclude women during the analysis phase. We included an illustrated instruction sheet that was also an adherence checklist to record the time of day and note smoking and eating in the 30 minutes before saliva collection. These procedures were consistent with other community sample studies that used ecologically collected saliva specimens for cortisol (Young & Breslau, 2004). Participants mailed specimen kits to the laboratory in padded, pre-addressed, postage-paid envelopes. We logged and froze these at 80 C until assay. We conducted a radioimmunoassay in triplicate with the use of the Coat-a-Count kit (Diagnostic Products Corporation, Los Angeles, California). Intra- and interassay coefficients of variance were less than 5% and 10%, respectively. We created variables for the cortisol values in natural units (log-transformed).
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
3
RESEARCH
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
Cortisol in Women With PTSD and Dissociative Subtype
Standardized telephone diagnostic interview data. Female interviewers with project-specific training conducted standardized telephone diagnostic interviews, and supervisors audited these interactions for quality across the life of the study. We reimbursed participants $25. We used established measures for psychiatric characteristics, including the Life Stressor Checklist (Wolfe & Kimerling, 1997), which includes queries for 30 types of potentially traumatic events, to determine trauma exposure. The National Women’s Study PTSD Module (NWS-PTSD) was validated in a primarily clinical sample of 528 women during the DSM-IV PTSD Field Trial (Kilpatrick et al., 1994). The NWS-PTSD module was validated against the Structured Clinical Interview for DSM-IV and attained a sensitivity of .99 and specificity of .79 (Resnick, Kilpatrick, Dansky, Saunders, & Best, 1993). In our sample of 395 participants, the internal consistency alpha for the NWS-PTSD was .91, and in the sample of 111 participants it was .93. Psychological dissociation was measured with the eight-item Taxon version of the Dissociative Experience Scales (DES-T; Waller, Putnam, & Carlson, 1996). The developers assessed this shortened version’s discriminant validity as a screening tool and found sensitivity of .80 and specificity of .80 (Waller et al., 1996). We modified the DES-T for telephone use with a response set of never, rarely, sometimes, often, and all the time. Participants were included in the PTSD-D group if they answered sometimes or more frequently to any of the three depersonalization and derealization items. In our sample of 395, the internal consistency was .78, and in the sample of 111 it was .73. We measured diagnosis of depression in the past year with the Composite International Diagnostic Interview, a structured psychiatric diagnostic interview designed to be used by lay interviewers. Its efficacy for diagnosis is supported by extensive field trial data on cross-national reliability and validity (Wittchen, 1994). We used demographic and smoking items from the Centers for Disease Control and Prevention’s Perinatal Risk Assessment and Monitoring Survey (Beck et al., 2002) that have been fielded across the United states since 1987. We summed African American race, being pregnant as a teen, having low education, living in poverty, and living in a zip code with greater rates of crime to obtain a 0 to 5 index as a proxy for sociodemographic risk or
4
stress. We classified a participant as a smoker if she continued to use one or more cigarettes per day after she became aware of her pregnancy. Although only women who we believed to be nonsmokers were enrolled for the longitudinal study, a small number of participants (n ¼ 6) indicated on the specimen procedure checklist that they smoked that day, so we also classified them as smokers. Medical record information. We calculated weeks of gestation on the basis of the expected date of birth determined at the first prenatal visit. Postnatal days and weeks were based on the actual infant birth date. Sample sizes. For Hypothesis A, the single-day, diurnal component, we included every woman who accepted the invitation to participate (N ¼ 449), and, after we eliminated those at gestational age greater than 25 weeks and those with metabolic diseases and medication use, our sample size was 395. For Hypothesis B, the longitudinal component, we estimated sample size (N ¼ 135) on the basis of our pilot data (Seng et al., 2005) to have a power greater than 80% at an alpha of .05 with the use of repeated measures. We retained 111 participants who provided specimens from the prenatal to the postnatal period. Statistical analysis. Preliminary steps in the analysis included examination of the missing data. In the single-day component to test Hypothesis A, there were 80 instances in which the cortisol assay values were less than the 0.003 mg/dl limit of detection of the assay. Because the interval for the undetectable region is small (0 to .003) and the proportion of specimens was small, we imputed the undetectable values with the limit of detection value. We maintained this procedure in the longitudinal study component. The cortisol data included a small proportion of instances in which the assay values were less than the limit of detection (0.003 mg/d), so we substituted this value (Croghan & Egeghy, 2003). Cortisol values followed log-normal distribution, so we examined descriptive data in natural units but used the log-transformed values to test and model the hypotheses. To test Hypothesis A, we used generalized estimating equations (GEEs) to consider the within-individual correlations of repeated-measures cortisol values (Ziegler, Kastner, & Blettner, 1998). We modeled the logtransformed cortisol levels using cohort, time of day, and their interactions. Depression, smoking,
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
RESEARCH
Seng, J. S. et al.
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
cumulative sociodemographic risk index, and gestational age were included as covariates. GEEs were also used to test Hypothesis B. The loess smooth method (Cleveland & Loader, 1996) was used to display smooth curves for cortisol levels at different time points for the four cohorts. The loess smooth plot showed that 34 weeks was a break-point for the trend of cortisol levels. Therefore, we partitioned the data into two sets (before or after 34 weeks) and analyzed them separately. We used GEEs to model the logtransformed cortisol levels by use of cohort, gestational week of specimen collection, and their interactions (see Supplemental Table S1). In addition, depression, smoking during pregnancy, and cumulative sociodemographic risk index were included as covariates. A p value of less than .05 was interpreted to be significant.
Results Sample for Hypothesis A Participants returned cortisol kits at a rate of 43% (449 of the 1,049 women enrolled for followup). In the single-day, diurnal component to test Hypothesis A, we included the 395 participants who remained after we eliminated 2 who had metabolic disorders, 9 who used corticosteroids or psychotropic medications, and 54 who returned specimens collected after 25 completed weeks of gestation. In this sample, the participants’ mean gestational week at the time of the specimen collection was 12.75 (standard deviation ¼ 4.39). This sample of 395 women (see Table 1) had a diverse, relatively low, socioeconomic status profile, and 33% were African American. Nineteen percent had histories of childhood maltreatment. A total of 27% had PTSD: 15% had PTSD-only and 12% had PTSD-D.
Results for Hypothesis A In the single-day diurnal component (see Figure 1), after we adjusted for smoking, depression, cumulative sociodemographic risk index, and gestational age, we did not detect significant effects of group assignment, time of day, or their interactions on the log-transformed cortisol levels. However, we found smoking and gestational age had significant effects on the logtransformed cortisol levels.
Sample for Hypothesis B A smaller subset of participants correctly followed the procedures in the single-day process
505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560
In early pregnancy, cortisol levels in participants in the dissociative subtype group were 8 times greater in the afternoon and 10 times greater at bedtime than those in participants in the nonexposed control group.
and agreed to submit saliva specimen kits up to 12 times (see Table 1). These participants were more likely to be White, to be living in poverty, and to have more than a high school education compared with those in the sample for the singleday component. Childhood maltreatment was experienced by 19% of these participants, and PTSD was present at a greater rate: 30% had PTSD-only and 7% had PTSD-D. On average, they returned a mean of 8.75 (range ¼ 4 to 13) morning and bedtime specimens and 8.50 (range ¼ 3 to 13) afternoon specimens.
Results for Hypothesis B We plotted the longitudinal gestational curves for morning, late afternoon, and bedtime (see Figure 2). For all four cohorts, the loess smooth plot showed that the cortisol levels increased up to 34 weeks gestation and then decreased during the afternoon and bedtime. However, in the morning, the cortisol levels remained constant before 34 weeks gestation and then decreased slightly after that. The PTSD-D cohort had the highest and flattest cortisol curves from early gestation to 6 weeks postpartum. The differences between the PTSD-D and nonexposed cohorts were greatest at afternoon and bedtime. During late gestation, the differences were smaller because all cohorts demonstrated the physiologic gestational cortisol rise. The GEE models showed that participants in the PTSD-D group had significantly greater morning, afternoon, and bedtime cortisol levels than did participants in the nonexposed control group before gestational week 34 in all adjusted models. The PTSD-only group was not significantly different from the other groups. After gestational week 34, adjusted models of between-group comparisons (see Supplemental Table S2) showed no significant differences in morning, afternoon, or bedtime cortisol levels and no significant group membership by week interactions. To quantify the extent of the differences for participants in the PTSD-D group at various weeks, we conducted an additional calculation from the GEEs (see Supplemental Table S3). The
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
5
RESEARCH
561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616
Cortisol in Women With PTSD and Dissociative Subtype
Table 1: Sample Characteristics One Day, n ¼ 395
Longitudinal, n ¼ 111
c2/t
132 (33.4)
23 (20.7)
6.57
.010
Pregnant as a teen (18–20 years old)
62 (15.7)
13 (11.7)
1.09
.297
Living in poverty (income <$15,000)
65 (16.5)
35 (31.5)
12.42
<.001
High school education or less
133 (33.7)
15 (13.5)
17.01
<.001
Residence in zip code with greater rate of crime
122 (30.9)
24 (21.6)
3.62
.060
1.30 1.73
0.99 1.58
1.79
.075
Characteristic, n (%) or Mean SD
p
Sociodemographic risk factors African American race
Cumulative sociodemographic risk index (0–5)
a
Childhood maltreatment history Any sexual, physical, or emotional abuse or neglectb Cumulative types of childhood maltreatment (0–5)
76 (19.2)
21 (18.9)
0.45 0.94
0.46 1.01
Group membership Nonexposed control group
139 (35.2)
Trauma-exposed control group
0.0058
.939
0.094
.926
11.53
.009
34 (30.6)
149 (37.7)
38 (34.2)
PTSD-only group
58 (14.7)
31 (27.9)
PTSD-D group
49 (12.4)
8 (7.2)
Major depression in the past year
39 (9.9)
16 (14.4)
1.84
.174
Smoking during pregnancy
29 (7.3)
6 (5.4)
0.50
.478
Co-occurring conditions
Gestational age
12.75 4.39
Diurnal cortisol levels Morning (peak) cortisol, mg/dl, w8 a.m. (n ¼ 392)
0.41 0.24
Afternoon cortisol, mg/dl, w4 p.m. (n ¼ 387)
0.11 0.12
Bedtime (nadir) cortisol, mg/dl, w10 p.m. (n ¼ 394)
0.09 0.12
Note. PTSD ¼ posttraumatic stress disorder; PTSD-D ¼ dissociative subtype posttraumatic stress disorder. a An index of cumulative sociodemographic risk was created from the sum of the five risk factors for PTSD (listed above). bChild maltreatment history was positive if physical abuse, contact sexual abuse, penetration sexual abuse, emotional abuse or neglect, or physical neglect before age 16 years were reported. An index was created from the sum of these five types of exposure to maltreatment trauma.
print & web 4C=FPO
log-transformed cortisol levels were retransformed by exponentiation. In early pregnancy (Gestational Week 8), the cortisol levels of
Discussion Q4 Figure 1. Log-transformed cortisol levels across 1 day by the four groups.
6
participants in the PTSD-D group were 2 times greater in the morning, 8 times greater in the afternoon, and 10 times greater at bedtime than those of the participants in the nonexposed control group, whereas in late pregnancy (Gestational Week 32), participants in the PTSD-D group had cortisol levels that were less than 2 times greater in the morning and 1.5 times greater levels in the afternoon and at bedtime compared with participants in the nonexposed control group.
With this biobehavioral component of the STACY Project, we have begun to fill a gap in knowledge of the longitudinal pattern of cortisol across the prenatal and postnatal periods for women with PTSD, including specific results related to the
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
RESEARCH
Seng, J. S. et al.
729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
associated with childhood maltreatment, were predictive of lower birth weight and shorter gestation. Dissociation is often present in PTSD as a sequela of childhood maltreatment (Stein et al., 2013) and may be a marker of severity and chronicity (Liotti, 2004), and it may also be a marker of severe stress in pregnancy.
Strengths of this study include the prospective cohort design; the diverse sample, which is large for a biobehavioral study; and the use of established measures to characterize trauma history, PTSD, depression, and dissociation. Furthermore, this was a community sample rather than a psychiatric sample of women who become pregnant; therefore, findings are highly generalizable. Some limitations should also be considered. We attempted to exclude smokers, but 6 of the 111 women disclosed smoking during at least 1 day, and this may have increased cortisol levels; thus, we controlled for smoking. Our specimens were collected at home and sent by mail; therefore, they have greater ecological validity but may have greater error variance than specimens collected in the laboratory. We used lifetime PTSD diagnosis, past-year depression, and dissociation measured as a trait. Current mental health symptom load or diagnoses may show stronger associations.
print & web 4C=FPO
673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728
Figure 2. Smoothed curves through morning, afternoon, and bedtime data points across weeks for the four groups.
dissociative subtype of PTSD. Our main finding was that participants with PTSD-D, and not those with nondissociative PTSD or trauma, had higher and flatter levels of cortisol during early gestation and into the early postpartum period. For participants in the PTSD-D group, the difference was most apparent in early gestation, which is a critical period. An elevated cortisol level has been identified as a risk factor for preterm birth by direct effect on onset of labor and indirect effect on inflammatory processes (Green et al., 2005). The perinatal outcomes of the parent study, published in 2011 (Seng et al., 2011), showed that lifetime PTSD, and even more so, PTSD
Our findings have numerous implications for further research. First, our finding that participants with PTSD-D had greater levels of cortisol than did participants with PTSD-only indicates that in the future, researchers should attend to women with the dissociative subtype and distinguish them from those with PTSD-only. Second, the greatest variation in cortisol levels associated with PTSD-D was in early gestation. Investigators should include specimens from early pregnancy before the physiologic gestational increase in cortisol to observe maximum effect sizes. Third, with regard to the finding that dissociation may lead to increased biological stress, the question arises of whether this is a characteristic of an individual woman. An alternative explanation is that psychosexual aspects of pregnancy act as triggers to dissociative symptoms in survivors of childhood maltreatment in particular and result in persistent, enhanced stress reactivity across gestation that is specific to a woman’s experience of pregnancy (Montgomery, 2013; Seng et al., 2009). Perinatal studies of reactivity to a stressor or trigger (e.g., internal pelvic examination) could help answer this question and indicate
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
7
RESEARCH
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
Cortisol in Women With PTSD and Dissociative Subtype
Cortisol levels in pregnant women with dissociation and posttraumatic stress are consistent with research that shows toxic stress from maltreatment and other adverse childhood experiences.
whether the usual blunted reactivity to stress in pregnancy is observed in this group (Schalinski et al., 2015; Zaba et al., 2015). Fourth, in the perinatal outcomes study of veteran women by Shaw et al. (2014), PTSD in pregnancy and history of military sexual trauma were predictive of shorter gestation. Military sexual trauma has been associated with greater than representative rates of childhood maltreatment trauma (Kelly, Skelton, Patel, & Bradley, 2011). Future research also should be designed to determine whether dissociation in particular or cumulative childhood and adult trauma; psychiatric symptom burden from complex comorbidities, including depression; and the stress of adverse life circumstances that accompany these together exert a toxic stress influence on a woman (Shonkoff et al., 2012).
because the public health cost is so high (Shonkoff, et al., 2012). For perinatal nurses, midwives, obstetricians, and other members of the perinatal care team, focus on childhood maltreatment history and complex posttraumatic sequelae is part of renewed attention to psychosocial care as a means to improve individual and population outcomes (Renfrew et al., 2014). Trauma-informed approaches to care and trauma-specific interventions are being developed for perinatal care (Cuthbert, 2015). Although some women with histories of childhood maltreatment are resilient or recovered by the time they become pregnant, these biological findings indicate that some are very adversely affected psychologically and very stressed during the childbearing year. We can screen and apply a stepped approach to maternity care that includes case-finding and interventions for women with PTSD, posttraumatic depression, and PTSD-D. The stakes are high, the reasons are compelling, and theoretical, qualitative, and service delivery elements are emerging to advance the field in this direction.
Conclusions Finally, studying maternal cortisol levels is a starting point, but other HPA axis components more specific to pregnancy should be considered, including placental corticotrophin-releasing hormone, which was elevated in childhood maltreatment survivors (Moog et al., 2015). Studies of the other stress-response systems that are affected by the HPA axis and associated with childbearing outcomes also warrant study, especially the oxytocin system (Teicher et al., 2003). Our results have clinical practice implications, especially because these biological findings are consistent with reports from clinical and qualitative studies. Since the first narrative reviews and case reports emerged in the early 1990s, researchers have suggested that childhood maltreatment; the sociodemographic disadvantage and risk exposures associated with it, including repeat victimization; and the substantial burden of mental health sequelae all affect women’s childbearing experiences. There is mounting evidence to suggest that these affect a range of childbearing outcomes as well (Sperlich, 2015). The conceptualization of toxic stress is becoming shared across health disciplines
8
We observed very high, perhaps toxic, levels of the stress hormone cortisol in a community sample of pregnant women with the dissociative subtype of PTSD. This complex form of PTSD in adults is strongly associated with history of childhood maltreatment. Knowledge that childhood maltreatment (and other adverse childhood experiences) causes toxic stress is rapidly increasing in public health and pediatrics fields because these experiences have been convincingly linked with disease across the lifespan (Shonkoff et al., 2012). Our data are important because the developmental origin of disease theory (Reynolds, 2013) and research also links high stress and cortisol levels in the fetal environment with early morbidity and mortality (Reynolds et al., 2013). Further research is warranted because PTSD-D–related toxic stress during pregnancy could potentially be mitigated to improve a woman’s perinatal mental health and the intra-uterine environment.
Acknowledgment Funded by NIH grant R01 NR008767 (Seng, principal investigator). The authors acknowledge the long-term contributions of co-investigator Dr. David Ronis.
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org
841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896
RESEARCH
Seng, J. S. et al.
897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952
Q1
Supplementary Material Note: To access the supplementary material that accompanies this article, visit the online version of the Journal of Obstetric, Gynecologic, & Neonatal Nursing at http://jognn.org and at https://doi.org/10.1016/j.jogn.2017.10.008.
953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
Lanius, R. A., Vermetten, E., Loewenstein, R. J., Brand, B., Schmahl, C., Bremner, J. D., & Spiegel, D. (2010). Emotion modulation in PTSD: Clinical and neurobiological evidence for a dissociative subtype. American Journal of Psychiatry, 167(6), 640–647. https://doi.org/10.1176/appi.ajp.2009.09081168 Latendresse, G. (2009). The interaction between chronic stress and pregnancy: Preterm birth from a biobehavioral perspective. Journal of Midwifery & Women’s Health, 54(1), 8–17. https://doi. org/10.1016/j.jmwh.2008.08.001
REFERENCES
Q2
Liotti, G. (2004). Trauma, dissociation, and disorganized attachment:
Allen, M. C. (2008). Neurodevelopmental outcomes of preterm infants. Current Opinion in Neurology, 21(2), 123–128. https://doi.org/ 10.1097/WCO.0b013e3282f88bb4 American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Arlington, VA: Author. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: Author.
Three strands of a single braid. Psychotherapy, 41(4), 472–486. https://doi.org/10.1037/0033-3204.41.4.472 Montgomery, E. (2013). Feeling safe: A meta-synthesis of the maternity care needs of women who were sexually abused in childhood. Birth: Issues in Perinatal Care, 40(2), 88–95. https://doi.org/10. 1111/birt.12043 Moog, N. K., Buss, C., Entringer, S., Shahbaba, B., Gillen, D. L.,
Beck, L. F., Morrow, B., Lipscomb, L. E., Johnson, C. H., Gaffield, M.
Hobel, C. J., & Wadhwa, P. D. (2015). Maternal exposure to
E., Rogers, M., & Gilbert, B. C. (2002). Prevalence of selected
childhood trauma is associated during pregnancy with
maternal behaviors and experiences, Pregnancy Risk Assess-
placental–fetal stress physiology. Biological Psychiatry, 79(10),
ment Monitoring System (PRAMS), 1999. Morbidity and Mor-
831–839. https://doi.org/10.1016/j.biopsych.2015.08.032
tality Weekly Report, 51(2), 1–27. Retrieved from https://www.
O’Connor, T. G., Tang, W., Gilchrist, M. A., Moynihan, J. A., Pressman,
cdc.gov/mmwr/preview/mmwrhtmL/ss5102a1.htm
E. K., & Blackmore, E. R. (2014). Diurnal cortisol patterns and
Cleveland, W. S., & Loader, C. (1996). Smoothing by local regression:
psychiatric symptoms in pregnancy: Short-term longitudinal
Principles and methods. In W. Ha¨rdel & M. G. Schimek (Eds.),
study. Biological Psychology, 96, 35–41. https://doi.org/10.
Statistical theory and computational aspects of smoothing (pp. 10–49). New York, NY: Springer.
1016/j.biopsycho.2013.11.002 Renfrew, M. J., Homer, C. S. E., Downe, S., McFadden, A., Muir, N.,
Croghan, C., & Egeghy, P. P. (2003, September 22–24). Methods of
Prentice, T., & ten Hoope-Bender, P. (2014). Midwifery: An
dealing with values below the limit of detection using SAS. Paper
executive summary for The Lancet’s series. Retrieved from
presented at Southeastern SAS User Group, St. Petersburg, FL.
http://www.thelancet.com/pb/assets/raw/Lancet/stories/series/
Cuthbert, C. (2015). Where are we on the journey toward trauma-
midwifery/midwifery_exec_summ.pdf
specific interventions and treatments for the perinatal period?
Resnick, H. S., Kilpatrick, D. G., Dansky, B. S., Saunders, B. E., &
In J. Seng & J. Taylor (Eds.), Trauma informed care in the
Best, C. L. (1993). Prevalence of civilian trauma and post-
perinatal period. Edinburgh: Dunedin Academic Press.
traumatic stress disorder in a representative national sample of
D’Anna-Hernandez, K. L., Ross, R. G., Natvig, C. L., & Laudenslager,
women. Journal of Consulting and Clinical Psychology, 61(6),
M. L. (2011). Hair cortisol levels as a retrospective marker of
984–991. https://doi.org/10.1037/0022-006X.61.6.984
hypothalamic–pituitary axis activity throughout pregnancy:
Reynolds, R. M. (2013). Glucocorticoid excess and the developmental
Comparison to salivary cortisol. Physiology & Behavior, 104(2),
origins of disease: Two decades of testing the hypothesis—
348–353. https://doi.org/10.1016/j.physbeh.2011.02.041
2012 Curt Richter Award Winner. Psychoneuroendocrinology,
Davis, E. P., & Sandman, C. A. (2010). The timing of prenatal exposure
38(1), 1–11. https://doi.org/10.1016/j.psyneuen.2012.08.012
to maternal cortisol and psychosocial stress is associated with
Reynolds, R. M., Labad, J., Buss, C., Ghaemmaghami, P., & Raikkonen,
human infant cognitive development. Child Development, 81,
K. (2013). Transmitting biological effects of stress in utero: Impli-
131–148. https://doi.org/10.1111/j.1467-8624.2009.01385.x
cations for mother and offspring. Psychoneuroendocrinology,
Green, N. S., Damus, K., Simpson, J. L., Iams, J., Reece, E. A., Hobel,
38(9), 1843–1849. https://doi.org/10.1016/j.psyneuen.2013.05.018
C. J., … March Of Dimes Scientific Advisory Committee On
Risnes, K. R., Vatten, L. J., Baker, J. L., Jameson, K., Sovio, U.,
Prematurity. (2005). Research agenda for preterm birth: Rec-
Kajantie, E., … Bracken, M. B. (2011). Birthweight and mortality
ommendations from the March of Dimes. American Journal of
in adulthood: A systematic review and meta-analysis. Interna-
Obstetrics and Gynecology, 193(3, Part 1), 626–635. https://
tional Journal of Epidemiology, 40(3), 647–661. https://doi.org/
doi.org/10.1016/j.ajog.2005.02.106
10.1093/ije/dyq267
Kelly, U. A., Skelton, K., Patel, M., & Bradley, B. (2011). More than
Schalinski, I., Elbert, T., Steudte-Schmiedgen, S., & Kirschbaum, C.
military sexual trauma: Interpersonal violence, PTSD, and
(2015). The cortisol paradox of trauma-related disorders: Lower
mental health in women veterans. Research in Nursing &
phasic responses but higher tonic levels of cortisol are asso-
Health, 34(6), 457–467. https://doi.org/10.1002/nur.20453
ciated with sexual abuse in childhood. PLoS One, 10(8),
Kilpatrick, D. G., Resnick, H. S., Freedy, J. R., Pelcovitz, D., Resick, P.
e0136921. https://doi.org/10.1371/journal.pone.0136921
A., Roth, S., & Van der Kolk, B. (1994). The posttraumatic stress
Schreier, H. M., Bosquet Enlow, M., Ritz, T., Coull, B. A., Gennings, C.,
disorder field trial: Emphasis on criterion a and overall PTSD
Wright, R. O., & Wright, R. J. (2016). Lifetime exposure to
diagnosis. In T. A. Widiger, A. J. Frances, H. A. Pincus, & R.
traumatic and other stressful life events and hair cortisol in a
Ross (Eds.), DSM-IV sourcebook (Vol. 3). Arlington, VA: Amer-
multi-racial/ethnic sample of pregnant women. Stress, 19(1),
ican Psychiatric Publishing.
45–52. https://doi.org/10.3109/10253890.2015.1117447
Lackritz, E. M., Wilson, C. B., Guttmacher, A. E., Howse, J. L., Eng-
Schreier, H. M. C., Enlow, M. B., Ritz, T., Gennings, C., & Wright, R. J.
mann, C. M., Rubens, C. E., … Setting, P. B. R. P. (2013).
(2015). Childhood abuse is associated with increased hair
A solution pathway for preterm birth: Accelerating a priority
cortisol levels among urban pregnant women. Journal of
research agenda. Lancet Global Health, 1(6), E328–E330.
Epidemiology and Community Health, 69(12), 1169–1174.
https://doi.org/10.1016/S2214-109X(13)70120-7
https://doi.org/10.1136/jech-2015-205541
JOGNN 2017; Vol. -, Issue -
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
9
RESEARCH
Cortisol in Women With PTSD and Dissociative Subtype
1009
Seng, J. S., Low, L. K., Ben-Ami, D., & Liberzon, I. (2005). Cortisol level
1010
and perinatal outcome in pregnant women with posttraumatic
1011 1012
Taylor (Eds.), Trauma informed care in the perinatal period.
1052
Edinburgh, UK: Dunedin Academic Press.
1053
stress disorder: A pilot study. Journal of Midwifery & Women’s
Stein, D. J., Koenen, K. C., Friedman, M. J., Hill, E., McLaughlin, K. A.,
Health, 50(5), 392–398. https://doi.org/10.1016/j.jmwh.2005.
Petukhova, M., … Kessler, R. C. (2013). Dissociation in post-
04.024
traumatic stress disorder: Evidence from the world mental
1055
health surveys. Biological Psychiatry, 73(4), 302–312. https://
1056
doi.org/10.1016/j.biopsych.2012.08.022
1057
1013
Seng, J. S., Low, L. K., Sperlich, M., Ronis, D. L., & Liberzon, I. (2009).
1014
Prevalence, trauma history, and risk for posttraumatic stress
1015
disorder among nulliparous women in maternity care. Obstet-
Teicher, M. H., Andersen, S. L., Polcari, A., Anderson, C. M., Navalta,
rics & Gynecology, 114(4), 839–847. https://doi.org/10.1097/
C. P., & Kim, D. M. (2003). The neurobiological consequences
AOG.0b013e3181b8f8a2
of early stress and childhood maltreatment. Neuroscience and
1016
1054
1058 1059
1017
Seng, J. S., Low, L. K., Sperlich, M., Ronis, D. L., & Liberzon, I. (2011).
1018
Post-traumatic stress disorder, child abuse history, birthweight
1019
and gestational age: A prospective cohort study. British Journal
Waller, N., Putnam, F. W., & Carlson, E. B. (1996). Types of dissociation
1062
1020
of Obstetrics and Gynecology, 118(11), 1329–1339. https://doi.
and dissociative types: A taxometric analysis of dissociative
1063
org/10.1111/j.1471-0528.2011.03071.x
experiences. Psychological Methods, 1(3), 300–321. https://
1021
Seng, J. S., Rauch, S. A., Resnick, H., Reed, C. D., King, A., Low, L. K.,
Biobehavioral Reviews, 27(1–2), 33–44. https://doi.org/10.1016/
1060
S0149-7634(03)00007-1
1061
doi.org/10.1037/1082-989X.1.3.300
1064
1022
… Liberzon, I. (2010). Exploring posttraumatic stress disorder
Wittchen, H. U. (1994). Reliability and validity studies of the WHO—
1065
1023
symptom profile among pregnant women. Journal of Psycho-
Composite International Diagnostic Interview (CIDI): A critical
1066
1024
somatic Obstetrics & Gynecology, 31(3), 176–187. https://doi.
review. Journal of Psychiatric Research, 28(1), 57–84. https://
1067
1025
org/10.3109/0167482X.2010.486453
doi.org/10.1016/0022-3956(94)90036-1
1068
1026
Shannon, M., King, T. L., & Kennedy, H. P. (2007). Allostasis: A theo-
Wolfe, J., & Kimerling, R. (1997). Gender issues in the assess-
1069
retical framework for understanding and evaluating perinatal
ment of posttraumatic stress disorder. In J. P. Wilson & T.
1027
health outcomes. Journal of Obstetric, Gynecologic, & Neonatal
M. Keane (Eds.), Assessing psychological trauma and
1070
1028
Nursing, 36(2), 125–134. https://doi.org/10.1111/j.1552-6909.
PTSD (pp. 192–238). New York, NY: Guilford Press.
1071
1029
2007.00126.x
Yonkers, K. A., Smith, M. V., Forray, A., Epperson, C. N., Costello, D.,
1072
Shaw, J. G., Asch, S. M., Kimerling, R., Frayne, S. M., Shaw, K. A., &
Lin, H., & Belanger, K. (2014). Pregnant women with post-
1073
Phibbs, C. S. (2014). Posttraumatic stress disorder and risk of
traumatic stress disorder and risk of preterm birth. JAMA Psy-
1030 1031 1032
spontaneous preterm birth. Obstetrics & Gynecology, 124(6),
chiatry, 71(8), 897–904. https://doi.org/10.1001/jamapsychiatry.
1111–1119. https://doi.org/10.1097/AOG.0000000000000542
2014.558
1074 1075
1033
Shonkoff, J. P., Garner, A. S., & Committee on Psychosocial Aspects of
Young, E. A., & Breslau, N. (2004). Saliva cortisol in posttraumatic
1076
1034
Child and Family Health, Committee on Early Childhood,
stress disorder: A community epidemiologic study. Biological
1077
1035
Adoption, and Dependent Care, and Section on Developmental
Psychiatry, 56(3), 205–209. https://doi.org/10.1016/j.biopsych.
1078
and Behavioral Pediatrics. (2012). The lifelong effects of early
2004.05.011
1036
childhood adversity and toxic stress. Pediatrics, 129(1), e232–
1037
Zaba, M., Kirmeier, T., Ionescu, I. A., Wollweber, B., Buell, D. R., Gall-
e246. https://doi.org/10.1542/peds.2011-2663
1079
Kleebach, D. J., … Schmidt, U. (2015). Identification and
1080
1038
Simeon, D., Knutelska, M., Yehuda, R., Putnam, F., Schmeidler, J., &
characterization of HPA-axis reactivity endophenotypes in a
1081
1039
Smith, L. M. (2007). Hypothalamic–pituitary–adrenal axis func-
cohort of female PTSD patients. Psychoneuroendocrinology, 55,
1082
1040
tion in dissociative disorders, post-traumatic stress disorder,
102–115. https://doi.org/10.1016/j.psyneuen.2015.02.005
1083
and healthy volunteers. Biological Psychiatry, 61(8), 966–973.
Ziegler, A., Kastner, C., & Blettner, M. (1998). The generalised esti-
1041
https://doi.org/10.1016/j.biopsych.2006.07.030
1042
Sperlich, M. (2015). How does focusing on posttraumatic stress dis-
1043
order shift perinatal mental health paradigms? In J. Seng & J.
mating equations: An annotated bibliography. Biometrical
1084
Journal, 40(2), 115–139. . <115::Aid-Bimj115>3.0.Co;2–6.
1085
https://doi.org/10.1002/(Sici)1521-4036(199806)40:2
1086
1044
1087
1045
1088
1046
1089
1047
1090
1048
1091
1049
1092
1050
1093
1051
1094
10
JOGNN, -, -–-; 2017. https://doi.org/10.1016/j.jogn.2017.10.008
FLA 5.5.0 DTD JOGN288_proof 22 November 2017 1:46 pm ce
http://jognn.org