Abstract

Objectives.

This study examined the longitudinal association between a prior history of sexual assault (SA), typically in youth, and decreasing executive functioning (EF) in old age and whether the apolipoprotein (APOE) ϵ4 allele modifies this relationship.

Method.

In this longitudinal study, 846 community-dwelling older adults at baseline completed questions about SA history and two tests of EF. Over the 10 years following this baseline visit, participants completed up to 3 follow-up cognitive assessments. Mixed-effects models first examined the longitudinal association between SA and EF performance. Last, preplanned analyses examined whether the APOE ϵ4 allele modified this association.

Results.

A single SA exposure was not associated with EF declines. Repeated SA exposure was associated with steeper declines in both EF measures. For Trails B, there was a significant interaction between any SA exposure and the APOE ϵ4 allele, such that having either repeated or isolated SA as well as APOE ϵ4 was associated with faster decline.

Discussion.

SA exposure earlier in life may increase risk for declines in EF 50–60 years later in old age, particularly in the context of the APOE ϵ4 allele. These results generally support a diathesis-stress model of decreased cognitive reserve.

Approximately 66% of the U.S. population will experience a traumatic event (i.e., an event in which one is exposed to actual or threatened death, serious injury, or threat to the physical integrity of self or other; American Psychiatric Association, 1994) during their lifetime (Norris, 1992). Approximately 33% of females and 15% of males report a history of experiencing any form of sexual assault (SA) during their lifetime (Siegel, Sorenson, Golding, Burnam, & Stein, 1987; Sorenson, Stein, Siegel, Golding, & Burnam, 1987). SA is the traumatic event most commonly associated with posttraumatic stress disorder (PTSD) in civilian females (Norris, 1992).

Biological responses to stress such as SA may affect the prefrontal cortex (PFC) and disrupt PFC-mediated abilities, such as executive functioning (EF; Holmes & Wellman, 2009; Liu & Aghajanian, 2008; Radley et al., 2004; Wellman, 2001). Thus, the combination of age-related declines and prior SA may make older adult SA survivors particularly vulnerable to greater declines in EF. Only a few studies have examined the long-term effects of trauma exposure on cognitive functioning in older adults. These studies suggest that deficits in memory (Golier, Harvey, Legge, & Yehuda, 2006) and EF (Hart et al., 2008) may be associated with PTSD symptoms. Additionally, other research has found that greater stress during childhood is associated with worse cognitive functioning in older adulthood (Fors, Lennartsson, & Lundberg, 2009). Research on this topic, however, has mainly focused on specific populations such as war veterans and Holocaust survivors. Although SA is one of the most common traumatic events experienced by women, to our knowledge no studies have investigated a history of SA in relation to cognitive functioning in later adulthood.

The apolipoprotein (APOE) ϵ4 isoform allele is the most common genotype associated with increased risk for declines in EF and cognitive functioning in later life (Bookheimer & Burggren, 2009; Small, Rosnick, Fratiglioni, & Backman, 2004). The interaction of this allele with prior stress is important, as studies indicate that carriers of the APOE ϵ4 allele may be more vulnerable to the adverse effects of stress, potentially including SA (Lee et al., 2008). Research in younger adults from families with a history of bipolar disorder support this hypothesis; greater severity of childhood sexual trauma combined with the presence of an APOE ϵ4 allele was associated with poorer cognitive functioning, including EF (Savitz, van der Merwe, Stein, Solms, & Ramesar, 2007).

The present study is an exploratory examination into the association of SA history and APOE ϵ4 with EF in old age. The first aim was to examine the association between SA history and EF by comparing individuals with no SA exposure, a single SA exposure, or repeated SA exposure. We hypothesize that a dose–response association will exist such that those with a history of repeated SA exposure will have worse EF over time. The second aim was to investigate whether the APOE ϵ4 allele modifies the association between SA history and EF. We hypothesize that individuals with both a history of SA and the APOE ϵ4 allele will demonstrate greater decline in cognitive functioning over time when compared with individuals with SA only, APOE ϵ4 only, or neither. To our knowledge, this is the first study to examine EF change over time in older adults in relation to history of SA and the APOE ϵ4 allele.

METHOD

Participants

All participants were from the Rancho Bernardo Study of Healthy Aging (Criqui, Barrett-Connor, & Austin, 1978). In 1972, all adults aged 30 and older living in a southern California community were invited to participate in a study of heart disease risk factors. Of the potential 6,155 participants aged 30–79, 5,052 were enrolled, representing an 82% response rate. Refusing to participate in the study was not related to sex or age. Beginning in 1992, all surviving community-dwelling members of the original study were asked to participate in a follow-up study investigating osteoarthritis and diabetes, which included measures of EF and queried past SA exposure. In total, 1,383 participants completed this assessment, which served as the baseline assessment for the present analyses. Previously, in a visit between 1988 and 1992, 876 (61.2%) of the 1,383 participants completed an assessment, which included genetic testing for the APOE allele. Following this baseline visit, participants completed up to three follow-up cognitive function assessments. Time intervals between these assessments varied, but on average participants completed 2.5 assessments with approximately three years between assessments (see Table 1). Of these 876 participants, 30 were missing data on key variables (9 were missing education data and 21 were missing Beck Depression Inventory scores at baseline and all follow-up assessments), leaving 846 participants for the present analysis.

Table 1.

Sample Characteristics at Baseline (N = 846)

Variable M (SD) or N (%) 
Age (years) 74.13 (8.97) 
Caucasian 842 (99.5) 
Education  
    Less than high school 26 (3.1) 
    High school 229 (27.1) 
    Some college education 591 (69.9) 
Female 508 (60.0) 
Taking anxiolytic medication 55 (6.5) 
Taking antidepressant medication 24 (2.8) 
Current smoker 63 (7.4) 
Body mass index 25.1 (3.9) 
Experienced any sexual assault 82 (9.7) 
Experienced repeated sexual assault 17 (2.0) 
Age at first sexual assault (years) 16.7 (11.0) 
Time since sexual assault (years) 55.1 (11.9) 
Has at least one APOE ϵ4 allele 202 (23.9) 
Experienced sexual assault and has APOE ϵ4 allele 21 (2.5) 
Experienced sexual assault and does not have APOE e4 allele 61 (7.2) 
Did not experience sexual assault and has APOE e4 allele 181 (21.4) 
Did not experience sexual assault and does not have APOE e4 allele 583 (68.9) 
BDI 5.44 (4.42) 
Alcohol abuse (2 or more on CAGE score) 76 (9.0) 
MMSE 27.92 (2.18) 
MMSE < 27 at baseline 132 (15.6) 
Trails B (s) 132.36 (60.52) 
Verbal fluency (number of words) 18.22 (5.04) 
Time from baseline to visit 2 (years) 3.76 (0.82) 
Time from baseline to visit 3 (years) 6.43 (0.79) 
Time from baseline to visit 4 (years) 10.02 (0.77) 
Number of assessments  
    One 846 (100.0) 
    Two 588 (69.5) 
    Three 411 (48.6) 
    Four 227 (26.8) 
Variable M (SD) or N (%) 
Age (years) 74.13 (8.97) 
Caucasian 842 (99.5) 
Education  
    Less than high school 26 (3.1) 
    High school 229 (27.1) 
    Some college education 591 (69.9) 
Female 508 (60.0) 
Taking anxiolytic medication 55 (6.5) 
Taking antidepressant medication 24 (2.8) 
Current smoker 63 (7.4) 
Body mass index 25.1 (3.9) 
Experienced any sexual assault 82 (9.7) 
Experienced repeated sexual assault 17 (2.0) 
Age at first sexual assault (years) 16.7 (11.0) 
Time since sexual assault (years) 55.1 (11.9) 
Has at least one APOE ϵ4 allele 202 (23.9) 
Experienced sexual assault and has APOE ϵ4 allele 21 (2.5) 
Experienced sexual assault and does not have APOE e4 allele 61 (7.2) 
Did not experience sexual assault and has APOE e4 allele 181 (21.4) 
Did not experience sexual assault and does not have APOE e4 allele 583 (68.9) 
BDI 5.44 (4.42) 
Alcohol abuse (2 or more on CAGE score) 76 (9.0) 
MMSE 27.92 (2.18) 
MMSE < 27 at baseline 132 (15.6) 
Trails B (s) 132.36 (60.52) 
Verbal fluency (number of words) 18.22 (5.04) 
Time from baseline to visit 2 (years) 3.76 (0.82) 
Time from baseline to visit 3 (years) 6.43 (0.79) 
Time from baseline to visit 4 (years) 10.02 (0.77) 
Number of assessments  
    One 846 (100.0) 
    Two 588 (69.5) 
    Three 411 (48.6) 
    Four 227 (26.8) 

Note. Missing data: age when sexual assault occurred, 6; baseline BDI score, 7; baseline Trails B, 14; MMSE, 21; baseline verbal fluency, 1. BDI = Beck Depression Inventory, 1. MMSE = Mini Mental State Examination. APOE = apolipoprotein.

During the follow-up period, 350 (41.4%) of these elderly participants died; most of the others who were missing assessments were no longer living in the community due to moving away or institutionalization. All study procedures were approved by the University of California, San Diego Institutional Review Board, and all participants provided written informed consent.

Outcome: Executive Functioning

To test EF, the Trail Making Test, part B (Trails B) from the Halstead–Reitan Neuropsychological Test Battery (Reitan, 1958) was administered by specially trained personnel at each visit. The Trails B test measures visual tracking, attention, cognitive flexibility, and psychomotor speed. Scoring is the amount of time spent to complete the test, to a maximum of 300 s. A higher score represents poorer test performance. Trails A, which measures psychomotor speed, was not administered.

Tests of fluency are an additional method to assess executive functioning (Cummings & Miller, 2007). Thus, a verbal fluency test (Borkowski, Benton, & Spreen, 1967) was administered at each visit. Participants were asked to name as many animals as they could in 60 s. The outcome variable is the total number of animals named. Repetitions (naming the same animal more than once), variants (referring to the same animal by a different name), and intrusions (replying with a nonanimal) are not counted in this score. To minimize learning effect, some visits used vegetables and fruits instead of animals.

Sexual Assault History

History of SA was assessed by asking participants standard questions that were adapted from the questions used in the Los Angeles Epidemiologic Catchment Area Study (Sorenson et al., 1987). Participants were asked the following: “In your lifetime, has anyone ever tried to pressure or force you to have unwanted sexual contact? (Sexual contact: touching your sexual parts, you touching their sexual parts, or intercourse).” Participants who had responded “yes” answered a question about whether this was an isolated or repeated event, how old they were when this first happened, and if they received counseling following the event. The use of this method to assess past SA history has been used in past SA research (Stein & Barrett-Connor, 2000).

APOE Genotyping

Genotype information was collected during the visit from 1988 to 1992 using standard techniques. Sequanna Therapeutics extracted genomic DNA from whole blood cells. Gel electrophoresis following polymerase change reaction amplification around the diagnostic polymorphic sites was conducted to determine APOE genotype. See Siest and colleagues (1995) for additional details regarding the APOE genotyping. There were too few homozygotes for APOE ϵ4 allele for separate analysis, so the genetic information was dichotomized into two categories: those who had at least one and those who did not have any APOE ϵ4 alleles.

Other Covariates

At the baseline visit, sex, race, education, and date of birth were recorded. Age in years at baseline and all follow-up assessments were computed from the participant’s date of birth. Current use of anxiety and depression medication was assessed at the baseline visit by a questionnaire and validated by a nurse who examined pills and prescriptions brought to the clinic for that purpose. Baseline body mass index (BMI) was calculated by specially trained nurses who measured height and weight of participants wearing clothing and no shoes. Participants self-reported their current smoking status. Participants were administered the Mini Mental Status Examination (MMSE) at every assessment as a screening test for possible dementia. Depressive symptoms were assessed at every assessment with the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961). Cronbach’s alpha for the BDI at baseline was .75, representing good internal consistency. Current alcohol abuse was assessed by the CAGE alcohol screener (Ewing, 1984). Scores were dichotomized into abuse (no/yes) using the cutoff score of two or greater suggesting alcohol abuse (Dhalla & Kopec, 2007).

Statistical Analysis

Descriptive statistics were computed for all variables and covariates. The primary outcome was EF measured by Trails B and verbal fluency tests. Mixed-effects regression models were conducted to examine the longitudinal association between SA and EF and the interaction between SA and APOE ϵ4 with EF. We chose mixed-effect regression models instead of a cross-sectional linear regression approach in order to increase power by using the total number of assessment points for each participant. This approach allowed us to examine changes over time. In these models, time was represented as age of the participant at each assessment. This approach to modeling time has been described and used in other longitudinal investigations (Caselli et al., 2009; Cohen, Cohen, West, & Aiken, 2003). The results were modeled using “age” versus “duration in the study” because this is an observational study and modeling time using age is more informative. Participants entered the study at different ages and were assessed at approximately equal time points from their original assessment. Duration in study is less meaningful and harder to interpret than participant age. Thus, we chose to model time with age. Separate models were analyzed for each outcome. Past research has identified the following to be associated with cognitive functioning and were thus considered as covariates in all models: education, taking antidepressant medication at baseline (0 = no, 1 = yes), taking anxiolytic medication at baseline (0 = no, 1 = yes), BDI score (McClintock, Husain, Greer, & Cullum, 2010), smoking status (0 = not currently a smoker, 1 = currently a smoker; Ernst, Heishman, Spurgeon, & London, 2001), alcohol abuse (0 = no, 1 = yes), and BMI (Gunstad et al., 2007). Sex was included as a covariate to account for the fact that SA exposure occurred less frequently in men. In order to control for global cognitive functioning and depression, MMSE and BDI scores were included in each model and treated as continuous, time-varying covariates. All continuous covariates (BDI, BMI, and MMSE) were centered on the mean. Due to the limited number of participants with SA, we did not consider interactions between other covariates that may be age-related, such as MMSE, and time.

Other longitudinal investigations of cognitive functioning have suggested that a quadratic association exists in which cognitive functioning decreases little from year to year in younger old age, whereas decreases tend to be greater from year to year with increasing age (Caselli et al., 2009). Descriptive plots of participants’ Trails B and verbal fluency test scores by age were examined. Drawing upon past research as well as the scatter plot, it was clear that the association between Trails B performance and age was quadratic, whereas that between verbal fluency performance and age was linear. Thus, for models examining Trails B as an outcome, both linear and quadratic age effects were modeled as fixed effects as well as random effects to represent time. For models examining verbal fluency as an outcome, the linear age effect was modeled as a fixed effect as well as a random effect to represent time. The likelihood ratio test was used to assess the significance of random time effect.

For Aim 1, SA exposure was broken into one trichotomous ordinal variable: no SA exposure = 0, isolated SA exposure = 1, and repeated SA exposure = 2. Although the SA variable is ordinal, we included it as a categorical variable in the models because the experience of no SA, an isolated SA, and repeated SAs are qualitatively different. Main effects of all the covariates and the two-way interactions of the SA variable and age were considered in these models. For Aim 2, the genetic analyses, all two-way and three-way interactions between age, SA, and APOE were considered in each model. Main effects of all the covariates were kept in the final model, but the interaction was removed from the final model if p > .05. Models were also rerun with the covariates that were not significant at p < .05 removed in a stepwise fashion. The results from these models did not differ when covariates were removed, so the reported models contain all covariates. All analyses included all participants who had at least one EF evaluation. To examine differential loss to follow-up, all final models were rerun with individuals who had at least two, three, or four assessments. The results from these models did not differ; thus, the results reported are based on all individuals with at least one EF evaluation time point. Additionally, analyses were repeated after excluding the 132 participants who had a baseline MMSE score of 26 or lower, suggesting possible dementia; the results did not differ. Statistical analyses were conducted using SPSS version 17.0 and STATA version 9.2.

RESULTS

Baseline characteristics of the participants are presented in Table 1. They were on average 74.1 (SD = 8.9) years old, 60% female, and 99.5% Caucasian. Almost 70% had at least some college education. About one in ten reported an SA (9.7%); on average, participants were 16.7 (SD = 11.0) years old when the first SA occurred. On average, the sexual assault occurred 55.1 (SD = 11.9) years prior to the baseline assessment. Of those who reported an SA, 13 (15.9%) were male and 69 (84.1%) were female. Almost one quarter of participants had at least one APOE ϵ4 allele (23.9%). The APOE ϵ4 allele distributions were in Hardy–Weinberg equilibrium.

The results of the analysis comparing the association between no SA exposure, isolated SA, and repeated SA exposure and EF are presented in Table 2. For Trails B, both the repeated SA by age squared interaction (β = −0.08, SE = 0.12, p = .49) and the isolated SA by age squared interaction (β = 0.05, SE = 0.04, p = .23) were not significant. After removal of the nonsignificant age squared by SA interaction, the repeated SA by age interaction was significant (β = 2.88, SE = 1.11, p = .02), suggesting that a history of repeated SA was associated with significantly greater decline in Trails B performance over time (see Figure 1). The isolated SA by age interaction was not significant (β = −0.03, SE = 0.53, p = .96). For verbal fluency, the isolated SA by age interaction (β = 0.03, SE = 0.05, p = .58) was not significant. The SA by age interaction was removed, and the model was rerun. The repeated SA by age interaction was significant (β = −0.23, SE = 0.10, p = .02), suggesting that repeated SA was associated with significantly greater decline in verbal fluency over time (see Figure 1).

Table 2.

Final Mixed Effects Regression Models Exploring the Association Between Isolated or Repeated Sexual Assault Exposure and Performance on Trails B (N = 834) and Verbal Fluency (N = 846)

Variable Trails B Verbal fluency 
β (SEp Value β (SEp Value 
Age 3.11(0.19) <.01 −0.17(0.02) <.01 
Age quadratic 0.11(0.01) <.01   
Isolated sexual assault −4.74 (5.96) .43 0.91 (0.52) .08 
Repeated sexual assault 36.18(11.71) <.01 −0.74 (0.99) .45 
Repeated sexual assault × Age 2.83(1.11) .02 −0.23(0.10) .02 
Variable Trails B Verbal fluency 
β (SEp Value β (SEp Value 
Age 3.11(0.19) <.01 −0.17(0.02) <.01 
Age quadratic 0.11(0.01) <.01   
Isolated sexual assault −4.74 (5.96) .43 0.91 (0.52) .08 
Repeated sexual assault 36.18(11.71) <.01 −0.74 (0.99) .45 
Repeated sexual assault × Age 2.83(1.11) .02 −0.23(0.10) .02 

Note. Bolded terms are significant at the p < .05 level. All above models control for sex, education, antianxiety and depression medication usage, global cognitive functioning, and depressive symptoms.

Figure 1.

Predicted performance on Trails B and verbal fluency over time for individuals with no history of sexual assault, isolated sexual assault exposure, and repeated sexual assault exposure.

Figure 1.

Predicted performance on Trails B and verbal fluency over time for individuals with no history of sexual assault, isolated sexual assault exposure, and repeated sexual assault exposure.

The analyses examining the effect of the APOE ϵ4 on the association between SA and EF were conducted next. For Trails B, after controlling for covariates, the three way interaction between SA, APOE ϵ4, and quadratic age was statistically significant (β = −0.30, SE = 0.10, p= .01). Figure 2 shows that those who had experienced an SA and had the APOE ϵ4 allele had a concave downward slope (i.e., experienced greater change earlier), whereas participants in the other groups had a convex slope (minimal changes early in old age and more rapid increase later in old age). For verbal fluency, the three-way interaction between SA, APOE ϵ4, and age was not statistically significant (β = 0.02, SE = 0.10, p = .83). After removing the three-way interaction as well as the 2 two-way interactions (APOE ϵ4 by age and SA by age) from the model, the SA by APOE ϵ4 interaction was stronger but not statistically significant (β = 1.86, SE = 1.06, p = .08).

Figure 2.

Predicted performance on Trails B over time for individuals with and without APOE ϵ4 and history of sexual assault.

Figure 2.

Predicted performance on Trails B over time for individuals with and without APOE ϵ4 and history of sexual assault.

DISCUSSION

This study is an initial exploratory investigation of a history of sexual assault in youth as a stressor, both on its own and in conjunction with the APOE ϵ4 allele, in relation to decreasing EF in older community-dwelling adults. Contrary to our first hypothesis, experiencing a single SA in youth was by itself not associated with greater decline in EF in late life. A significant dose–response association was present, however, such that the participants who reported repeated SA had greater declines on both the Trails B and verbal fluency tests when compared with participants who reported no SA exposure or isolated SA exposure. Furthermore, participants with both the APOE ϵ4 allele and a history of either isolated or repeated SA experienced a greater decline in old age on the Trails B test. The APOE ϵ4 allele in conjunction with a history of SA was not associated with significantly greater decline in verbal fluency.

Processing speed is another cognitive domain that is captured on the Trails B test. Typically the Trails A score is subtracted from Trails B to partial out the effects of processing speed. These participants did not complete Trails A; thus, these findings suggest that individuals with history of repeated SA exposure or any SA exposure along with the APOE ϵ4 allele may be at a higher risk for declines in EF or processing speed.

The authors are not aware of any research that has reported the longitudinal association of SA with EF late in life. The finding that a single SA experience was not associated with poorer EF is consistent with one study in younger adults (Jenkins, Langlais, Delis, & Cohen, 2000) but inconsistent with two others (Navalta, Polcari, Webster, Boghossian, & Teicher, 2006; Stein, Kennedy, & Twamley, 2002). The finding that repeated exposure to a traumatic event may be associated with worse outcomes is consistent with other research investigating the long term effects of trauma (Stein & Barrett-Connor, 2000; Turner & Lloyd, 1995). The finding that individuals with the APOE ϵ4 allele and a positive SA history (either repeated or isolated) showed steeper declines in EF or processing speed at a younger age is consistent with the results of a cross-sectional study in which individuals from families with a history of bipolar disorder who experienced more childhood trauma and had the APOE ϵ4 allele had a higher rate of cognitive decline (Savitz et al., 2007).

The present study has several limitations. First, although the cohort included more than 800 well-characterized older adults who were assessed as many as four times, the significant three-way interaction between quadratic age, SA and APOE ϵ4 should be interpreted with caution due to the fact that only 21 participants had experienced an SA and also had the APOE ϵ4 allele. It should be noted that sensitivity analyses including only participants with four assessment points support the robustness of these findings. However, these findings should be considered preliminary, and further research with larger samples is needed to replicate and confirm these results. Second, although the study is prospective, as with any observational study, the causal mechanisms are unclear. Third, the sample was Caucasian and relatively well educated, potentially limiting the generalizability of the results. Fourth, no measure of PTSD symptoms was available following SA, most of which occurred many years previously. It is unclear whether deficits in EF are associated with SA trauma exposure alone or the result of developing PTSD.

Although SA history was assessed using a standard question common to most epidemiological research (Siegel et al., 1987; Sorenson et al., 1987), the use of a more detailed history of SA may be more informative than the standard simple questions used in this investigation (Franklin, Sheeran, & Zimmerman, 2002). Additionally, because of the wording of the question (“tried to pressure or force you”), events that are coded as SAs could include both unwanted advances that were not perceived as traumatic as well as more violent episodes. Alternatively, the argument that any unwanted sexual advance, especially in youth, is traumatic, salient, and has enduring effect can be made. Our data supports this hypothesis that any unwanted sexual advance is salient by the fact that participants recalled these events that occurred on average more than fifty years ago suggests that these events were indeed valid, salient, and unforgotten. We were also unable to control for other factors that can increase the risk of cognitive decline, particularly in the presence of the APOE e4 allele, such as history of head trauma (Ariza et al., 2006) or family dysfunction (Dong, Anda, Dube, Giles, & Felitti, 2003). Finally, although the two neuropsychological tests used to examine EF are well validated, a more extensive neuropsychological test battery that includes other cognitive domains such as working memory would be useful in future research.

The primary strength of this exploratory investigation is that it is the first to examine the association of a common traumatic stressor, alone and in conjunction with an established genetic risk factor, with EF longitudinally in community-dwelling older adults. Our findings generally support a diathesis-stress model of decreased cognitive reserve in later life, in which a single risk factor (such as an isolated exposure to trauma) may not have a noticeable effect, whereas multiple risk factors (such as multiple exposures to trauma or one exposure to trauma in the presence of genetic vulnerability) acting together can produce long-term negative effects on cognition later in life. They further highlight the need for additional research to investigate early sexual trauma exposure and late life cognitive decline. Other longitudinal studies are needed to confirm these results and investigate potential causal mechanisms, which could lead to preventive interventions.

The findings also have important clinical implications. Research has suggested that deficits in EF are common in individuals with depressive or anxiety disorders later in life. Furthermore, experiencing SA, particularly early in life, may increase risk for depression or anxiety. Thus, neuropsychological and psychological assessments in older adults should include questions about past trauma history, as this may play a role in current cognitive functioning.

FUNDING

The Rancho Bernardo Study was funded by the National Institutes of Health/National Institute on Aging grants AG07181 and AG028507 and the National Institute of Diabetes and Digestive and Kidney Diseases, grant DK31801.

References

American Psychiatric Association
Diagnostic and statistical manual of mental disorders: DSM-IV
 , 
1994
4th ed.
Washington, DC
Author
Ariza
M
Pueyo
R
Matarin Mdel
M
Junque
C
Mataro
M
Clemente
I
Sahuquillo
J
Influence of APOE polymorphism on cognitive and behavioural outcome in moderate and severe traumatic brain injury
Journal of Neurology, Neurosurgury, & Psychiatry
 , 
2006
, vol. 
77
 (pg. 
1191
-
1193
doi:10.1136/jnnp.2005.085167
Beck
A
Ward
C
Mendelson
M
Mock
J
Erbaugh
J
An inventory for measuring depression
Archives of General Psychiatry
 , 
1961
, vol. 
44
 (pg. 
561
-
571
doi:10.1001/archpsyc.1961.01710120031004
Bookheimer
S
Burggren
A
APOE-4 genotype and neurophysiological vulnerability to Alzheimer's and cognitive aging
Annual Reviews in Clinical Psychology
 , 
2009
, vol. 
5
 (pg. 
343
-
362
doi:10.1146/annurev.clinpsy.032408.153625
Borkowski
J
Benton
A
Spreen
O
Word fluency and brain damage
Neuropsychologia
 , 
1967
, vol. 
5
 (pg. 
135
-
140
doi:10.1016/0028-3932(67)90015-2
Caselli
RJ
Dueck
AC
Osborne
D
Sabbagh
MN
Connor
DJ
Ahern
GL
Reiman
EM
Longitudinal modeling of age-related memory decline and the APOE epsilon4 effect
New England Journal of Medicine
 , 
2009
, vol. 
361
 (pg. 
255
-
263
doi:10.1056/NEJMoa0809437
Cohen
J
Cohen
P
West
S
Aiken
L
Applied multiple regression/correlation analysis for the behavioral sciences
 , 
2003
3rd ed.
Mahwah, NJ
Lawrence Erlbaum Associates
Criqui
MH
Barrett-Connor
E
Austin
M
Differences between respondents and non-respondents in a population-based cardiovascular disease study
American Journal of Epidemiology
 , 
1978
, vol. 
108
 (pg. 
367
-
372
)
Cummings
JL
Miller
B
Miller
B
Conceptual and clinical aspects of the frontal lobe
The human frontal lobes: Functions and disorders
 , 
2007
New York
Guilford Press
Dhalla
S
Kopec
JA
The CAGE questionnaire for alcohol misuse: A review of reliability and validity studies
Clinical and Investigative Medicine
 , 
2007
, vol. 
30
 (pg. 
33
-
41
)
Dong
M
Anda
RF
Dube
SR
Giles
WH
Felitti
VJ
The relationship of exposure to childhood sexual abuse to other forms of abuse, neglect, and household dysfunction during childhood
Child Abuse and Neglect
 , 
2003
, vol. 
27
 (pg. 
625
-
639
doi:10.1016/S0145-2134(03)00105-4
Ernst
M
Heishman
SJ
Spurgeon
L
London
ED
Smoking history and nicotine effects on cognitive performance
Neuropsychopharmacology
 , 
2001
, vol. 
25
 (pg. 
313
-
319
doi:10.1016/S0893-133X(01)00257-3
Ewing
J
Detecting alcoholism: The CAGE questionnaire
Journal of the American Medical Association
 , 
1984
, vol. 
252
 (pg. 
1905
-
1907
doi:10.1001/jama.252.14.1905
Fors
S
Lennartsson
C
Lundberg
O
Childhood living conditions, socioeconomic position in adulthood, and cognition in later life: exploring the associations
The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences
 , 
2009
, vol. 
64
 (pg. 
750
-
757
doi:10.1093/geronb/gbp029
Franklin
CL
Sheeran
T
Zimmerman
M
Screening for trauma histories, posttraumatic stress disorder (PTSD), and subthreshold PTSD in psychiatric outpatients
Psychological Assessment
 , 
2002
, vol. 
14
 (pg. 
467
-
471
doi:10.1037//1040-3590.14.4.467
Golier
JA
Harvey
PD
Legge
J
Yehuda
R
Memory performance in older trauma survivors: Implications for the longitudinal course of PTSD
Annals of the New York Academy of Sciences
 , 
2006
, vol. 
1071
 (pg. 
54
-
66
doi:10.1196/annals.1364.006
Gunstad
J
Paul
RH
Cohen
RA
Tate
DF
Spitznagel
MB
Gordon
E
Elevated body mass index is associated with executive dysfunction in otherwise healthy adults
Comprehensive Psychiatry
 , 
2007
, vol. 
48
 (pg. 
57
-
61
doi:10.1016/j.comppsych.2006.05.001
Hart
J
Jr.
Kimbrell
T
Fauver
P
Cherry
BJ
Pitcock
J
Booe
LQ
Freeman
TW
Cognitive dysfunctions associated with PTSD: Evidence from World War II prisoners of war
Journal of Neuropsychiatry and Clinical Neurosciences
 , 
2008
, vol. 
20
 (pg. 
309
-
316
doi:10.1176/appi.neuropsych.20.3.309
Holmes
A
Wellman
CL
Stress-induced prefrontal reorganization and executive dysfunction in rodents
Neuroscience and Biobehavioral Reviews
 , 
2009
, vol. 
33
 (pg. 
773
-
783
doi:10.1016/j.neubiorev.2008.11.005
Jenkins
MA
Langlais
PJ
Delis
DA
Cohen
RA
Attentional dysfunction associated with posttraumatic stress disorder among rape survivors
Clinical Neuropsychologist
 , 
2000
, vol. 
14
 (pg. 
7
-
12
doi:10.1076/1385-4046(200002)14:11-8;FT007
Lee
BK
Glass
TA
Wand
GS
McAtee
MJ
Bandeen-Roche
K
Bolla
KI
Schwartz
BS
Apolipoprotein e genotype, cortisol, and cognitive function in community-dwelling older adults
American Journal of Psychiatry
 , 
2008
, vol. 
165
 (pg. 
1456
-
1464
doi:10.1176/appi.ajp.2008.07091532
Liu
RJ
Aghajanian
GK
Stress blunts serotonin- and hypocretin-evoked EPSCs in prefrontal cortex: role of corticosterone-mediated apical dendritic atrophy
Proceedings of the National Academy of Sciences of the United States of America
 , 
2008
, vol. 
105
 (pg. 
359
-
364
doi:10.1073/pnas.0706679105
McClintock
SM
Husain
MM
Greer
TL
Cullum
CM
Association between depression severity and neurocognitive function in major depressive disorder: a review and synthesis
Neuropsychology
 , 
2010
, vol. 
24
 (pg. 
9
-
34
doi:10.1037/a0017336
Navalta
CP
Polcari
A
Webster
DM
Boghossian
A
Teicher
MH
Effects of childhood sexual abuse on neuropsychological and cognitive function in college women
Journal of Neuropsychiatry and Clinical Neurosciences
 , 
2006
, vol. 
18
 (pg. 
45
-
53
doi:10.1176/appi.neuropsych.18.1.45
Norris
FH
Epidemiology of trauma: Frequency and impact of different potentially traumatic events on different demographic groups
Journal of Consulting and Clinical Psychology
 , 
1992
, vol. 
60
 (pg. 
409
-
418
doi:10.1037//0022-006X.60.3.409
Radley
JJ
Sisti
HM
Hao
J
Rocher
AB
McCall
T
Hof
PR
Morrison
JH
Chronic behavioral stress induces apical dendritic reorganization in pyramidal neurons of the medial prefrontal cortex
Neuroscience
 , 
2004
, vol. 
125
 (pg. 
1
-
6
doi:10.1016/j.neuroscience.2004.01.006
Reitan
R
Validity of the trail-making test as an indicator of organic brain disease
Perceptual and Motor Skills
 , 
1958
, vol. 
8
 (pg. 
271
-
276
)
Savitz
J
van der Merwe
L
Stein
DJ
Solms
M
Ramesar
R
Genotype and childhood sexual trauma moderate neurocognitive performance: A possible role for brain-derived neurotrophic factor and apolipoprotein E variants
Biological Psychiatry
 , 
2007
, vol. 
62
 (pg. 
391
-
399
doi:10.1016/j.biopsych.2006.10.017
Siegel
JM
Sorenson
SB
Golding
JM
Burnam
MA
Stein
JA
The prevalence of childhood sexual assault. The Los Angeles Epidemiologic Catchment Area Project
American Journal of Epidemiology
 , 
1987
, vol. 
126
 (pg. 
1141
-
1153
)
Siest
G
Pillot
T
Regis-Bailly
A
Leininger-Muller
B
Steinmetz
J
Galteau
MM
Visvikis
S
Apolipoprotein E: An important gene and protein to follow in laboratory medicine
Clinical Chemistry
 , 
1995
, vol. 
41
 (pg. 
1068
-
1086
)
Small
BJ
Rosnick
CB
Fratiglioni
L
Backman
L
Apolipoprotein E and cognitive performance: A meta-analysis
Psychology and Aging
 , 
2004
, vol. 
19
 (pg. 
592
-
600
doi:10.1037/0882-7974.19.4.592
Sorenson
SB
Stein
JA
Siegel
JM
Golding
JM
Burnam
MA
The prevalence of adult sexual assault. The Los Angeles Epidemiologic Catchment Area Project
American Journal of Epidemiology
 , 
1987
, vol. 
126
 (pg. 
1154
-
1164
)
Stein
MB
Barrett-Connor
E
Sexual assault and physical health: Findings from a population-based study of older adults
Psychosomatic Medicine
 , 
2000
, vol. 
62
 (pg. 
838
-
843
)
Stein
MB
Kennedy
CM
Twamley
EW
Neuropsychological function in female victims of intimate partner violence with and without posttraumatic stress disorder
Biological Psychiatry
 , 
2002
, vol. 
52
 (pg. 
1079
-
1088
doi:10.1016/S0006-3223(02)01414-2
Turner
RJ
Lloyd
DA
Lifetime traumas and mental health: The significance of cumulative adversity
Journal of Health and Social Behavior
 , 
1995
, vol. 
36
 (pg. 
360
-
376
doi:10.2307/2137325
Wellman
CL
Dendritic reorganization in pyramidal neurons in medial prefrontal cortex after chronic corticosterone administration
Journal of Neurobiology
 , 
2001
, vol. 
49
 (pg. 
245
-
253
doi:10.1002/neu.1079

Author notes

Decision Editor: Dr Robert West