53 Graduate Student Journal of Psychology Spring 2024 - Vol. 22 Copyright 2024 by the Department of Counseling and Clinical Psychology Teachers College, Columbia University The Effects of Psychosocial and Traumatic Stressors on MCI Diagnosis Katherine Goulden Department of Psychiatry and Behavioral Sciences, Stanford University There is no single cause of mild cognitive im- pairment (MCI), a neurodegenerative condition defined as “clinically significant memory impair- ment that does not meet the criteria for dementia” (Petersen, 2011, p. 2). MCI is an amnestic disorder representing an intermediate stage between norma- tive aging and Alzheimer's dementia. Though all MCI demonstrate neuropsychological impairment, diagnosis can be broken into subtypes amnestic or non-amnestic (Rountree et al., 2007) for more sta- ble estimates of prevalence and rates of returning to normal cognitive functioning (Jak et al., 2009). As age increases, so does the risk of developing MCI. A lifetime’s worth of factors and exposures can affect the risk of developing this unhealthy form of cognitive decline. Apolipoprotein genotype (Tang et al., 2023), education (Tervo et al., 2004), and general cognitive ability during adulthood (Corbo et al., 2023) have been associated with an increased risk of develop- ing MCI, but neither the exact causes of MCI nor the influence of a lifetime of stressors is completely un- derstood. There is no known cure for MCI, thus the identification of modifiable risk factors is important for prevention and earlier detection of those at risk. Although research on MCI risk factors is extensive, the identification of the role of stress is incomplete and has never been investigated in a sample of twins. Exposure to stress in adulthood and midlife has been shown by previous studies to increase the likeli- hood of an MCI diagnosis on the neurobiological level (Kritikos et al., 2023; Song et al., 2020). Stress occurs during a threat and when environmental demands exceed adaptive capacity, with threat associated exter- nal and internal stimuli eliciting the reactions defined as stressors. Potential stressors encountered during adulthood include a long list of adverse psychosocial and physical forces. A psychosocial factor is defined by Hemingway and Marmot (1999) as phenomena that are “potentially related to the social environment and to pathophysiologic changes…Psychosocial factors may act alone or combine in clusters and may exert effects at different stages of the life course” (p. 2). The stressor can be a discrete event such as the death of a spouse, or a prolonged exposure, such as racial discrimination ex- perienced by minority Americans (Turner et al., 2017). In the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-4), Criterion A1, traumatic events are defined as “an event that involves actual or threatened death or serious injury, or other threat to one’s personal integrity” and includes “learning about the unexpected or violent death, serious harm, or threat of death or injury experienced by a family mem- ber or other close associate” (First et al., 2004, p. 14). Posttraumatic stress disorder (PTSD) is a disorder that can be present at any age. It is triggered by either witnessing or experiencing an event that presents a threat to one’s safety. Symptoms may include flash- backs, nightmares, severe anxiety, and uncontrollable Background and Objective: Toxic stress exposure can have effects across the lifespan. Studies of civilians and veterans suggest a connection between psychosocial and traumatic stressor exposure in adulthood and a di- agnosis of dementia later in life. The objective of this study was to investigate the impact of psychosocial and traumatic stressors on rates of MCI (Mild Cognitive Impairment) diagnosis in Vietnam Era Twin Study of Aging participants. Methods: 1,237 twin participants from the VETSA study were aged 61.72 ± 2.44 years at the time of data collection. Traumatic stress was measured by clinical interviewing, with psychosocial stressors quantified by self-report measures. Neuropsychological assessment determined MCI diagnosis. Previously con- ducted genotyping determined ApoE genotype. Mixed model analysis was used to determine effects on MCI diagnosis. Results: Our results from the mixed model analysis did not find a significant relationship between psychosocial and traumatic stress exposure and MCI diagnosis. PTSD diagnosis, measured by the DIS-III-R, collected for the Harvard Drug Study in 1996 (F = 0.249, p = 0.618) does not have a significant effect on MCI diagnosis. Life stress exposure, measured by Holmes and Rahe (1967), (F = 0.249, p = 0.618) does not have a significant effect on MCI diagnosis. Significant associations were determined using the Type III fixed effects. As- sociations were considered statistically significant at p < 0.05, two-tailed. Implications: Few subjects in Wave 2 of VETSA had MCI (n = 147), due in part to the age of the participants at the time (Mean 61.72 ± (2.44 years)). This led to a lack of power in our analysis. Future studies should examine all available VETSA data. Keywords: veterans, mild cognitive impairment, psychosocial stress, traumatic stress 54 GOULDEN thoughts about the event (Hathaway et al., 2010). It is also associated with cognitive impairments unique to the condition in the domains of verbal learning, speed of information processing, attention, work- ing memory, and verbal memory (Scott et al., 2015). A neurobiological pathway has been proposed to describe how PTSD influences dementia, with trau- matic exposure triggering persistent over-activation of the hypothalamic-pituitary-adrenal (HPA) axis and the adrenergic system (Leister & Menke, 2020). The literature on the impact of psychosocial and traumatic stress on MCI outcomes has been mixed. A study by Wang et al., (2018) of non-military el- ders of both sexes found a significant dose-depen- dent relationship between PTSD symptom sever- ity and developing dementia later in life. A study by Peavy et al., (2009) also found that chronic stress and stressful life events led to general memory decline in cognitively normal (CN) individuals as well as those diagnosed with MCI. As for conversion from CN to MCI, Peavy et al., (2012) did not find an association between stressful experiences and change to MCI. These findings could result from the variability of ways that stress and PTSD were operationalized. Although research with elderly veteran partici- pants varies from that of elderly civilian participants, most VETSA participants were not exposed to combat (Kremen et al., 2013). However, veterans are exposed to factors that are unique to military service (Siben- er, 2014). In a study of 181,093 elderly veterans by Yaffe et al. (2010), the 7-year cumulative incident de- mentia rate amongst veterans with PTSD was 10.6% while those without PTSD had a rate of 6.6%. While Yaffe et al. (2010) did not specify the veterans’ era, in another study greater incidence of MCI was ob- served specifically in Vietnam Era veterans who had also been diagnosed with PTSD (Weiner et al., 2017). Previous studies have indicated that psychosocial factors such as racism (Moon et al., 2019), workplace adversity (Nabe-Nielsen, et al., 2019), and divorce (Eriksson, 2015) are risk factors for dementia. But in a meta-analysis of 24 longitudinal studies examining cat- egories of toxic psychosocial and trauma-related stress, Bougea et al. (2022) found suggestive, yet non-robust evidence that psychosocial and traumatic types of stress are associated with increased risk of dementia in later life. As VETSA participants have aged, the focus of the multi-institutional VETSA study has shifted from a focus on substance use to early identification of risk for MCI and Alzheimer’s disease (AD). VETSA par- ticipants are all part of the Vietnam Era Twin Regis- try. VETSA selection criteria were (1) being in one’s fifties at the time of recruitment and (2) that both twins in a pair had to be willing to participate in the baseline assessment (Kremen et al., 2013). A narrow subject age range of participants enhances VETSA’s ability to examine within-individual differences and change over time. Another key aspect of the study de- sign was an extensive neuropsychological test battery. The study is also unique in that we have cognitive as- sessment scores from participants when they were in- ducted into the military at the age of 17 to 25 years old. The present study was designed to investigate the influence of psychosocial and traumatic stressors on MCI diagnosis. It is hypothesized that participants with exposure to psychosocial and traumatic stress- ors would be more likely to have developed MCI. Methods Participants VETSA participants were recruited from the Harvard Drug Study (Tsuang et al., 2001). 1,237 twins participated (349 monozygotic pairs, 265 dizy- gotic pairs, and 9 unpaired). Attrition-replacement participants were included as a subset of the Wave 2 participants. The attrition-replacement participants are twin pairs from the Vietnam Era Twin Registry in the same age range as the returning Wave 2 partic- ipants. At Wave 2 of data collection, the mean age of participants was 61.72 ± (2.44 years). All VETSA par- ticipants were in the military sometime between 1965 and 1975. The majority did not see combat or serve in Vietnam. The sample was entirely male and 95.4% (n = 712) white. Black Americans made up 4% (n = 30), Hispanics represented 0.3% (n = 2) and 0.3% were missing information on race (n = 2). The average life- time education was 12.4 (± 1.3 years) (see Appendix). Instrumentation Holmes and Rahe Stress Scale Life stressors were measured using the Holmes and Rahe Social Readjustment Rating Scale. The Holmes and Rahe Social Readjustment Rating Scale is a 100-item questionnaire (α = .8458), com- posed of 43 life events. An individual's total score measures the amount of stress the individual has ex- perienced in the past year. The tool has been exten- 55 STRESSORS ON MCI DIAGNOSIS sively studied, and its reliability and validity are well established. Cronbach's alpha for various populations ranges from 0.82 to 0.90 (Holmes & Rahe, 1967). DIS-III-R Subjects were interviewed using the Diagnostic Interview Schedule Version III Revised (DIS-III-R), a structured interview employed in epidemiological research. Interviews were performed over the tele- phone by the Institute for Survey Research at Temple University. Responses to the DIS-III-R were used to diagnose psychiatric disorders according to the revised third edition of the Diagnostic and Statistical Manu- al of Mental Disorders (Coopers and Michels, 1988). PTSD Checklist (PCL-R) The PTSD Checklist (PCL) is a self-report rat- ing scale for assessing posttraumatic stress disorder. It consists of 17 items (α = 0.87) which correspond to the DSM-III symptoms of PTSD. Examinees are instructed to indicate how much they have been bothered by each symptom in the past month using a 5-point (1-5) scale. The anchors for the severity rat- ings range from "Not at all" to "Extremely." The PCL can be used as a continuous measure of PTSD symp- tom severity by summing scores across the 17 items. AFQT The Armed Forces Qualification Test (AFQT) is a 50-minute paper-and-pencil test consisting of 100 multiple-choice items (α = .88) that was administered just prior to military induction (Bayroff & Anderson, 1963). The items equally represent the four domains of vocabulary, arithmetic word problems, knowledge of tools and mechanical or electrical equipment, and spatial visualization, which involves matching folded and unfolded box patterns (Uhlaner & Bolanovich, 1952). Originally intended as a measure of military trainability, further research has found the AFQT to be a highly g-loaded test (Orme et al., 2001), with g be- ing the construct of general intelligence (Humphreys, 1979). VETSA investigators received permission from the United States Department of Defense to re-ad- minister a version of the AFQT that is similar to the AFQT versions that had been administered to VETSA subjects just prior to their induction into the military (1965–1975). This version has been used in previous research (Grafman et al., 1988). Scores from the time of induction are also available to VETSA investigators. Genotyping ApoE genotype is integral in AD and MCI re- search as it accounts for as much as 50% of the attrib- utable risk for AD in many populations (Ashford, 2004). As per Lyons et al. (2013), ApoE genotyping was conducted for the 1,237 VETSA participants at either the Boston University or University of Cal- ifornia, San Diego site. ApoE genotype was deter- mined using previously described conditions (Emi et al., 1988; Hixson & Vernier, 1990). Due to the low rate of participants that possessed e4 allele, ho- mozygous and heterozygous carriers were grouped. Jak-Bondi MCI This study uses VETSA data that utilized the Jak-Bondi (Jak, et al., 2009) operationalization of MCI. Conservative criteria were used in VETSA, such that it requires impairment on two measures with- in a domain, with impairment identified as 1.5 SD below normative data (Jak et al., 2009). According to these standardized criteria, individuals in VETSA were classified as normal if, at most, performance on one measure within one or two cognitive domains fell more than 1.5 SD below age-appropriate norms. Procedure Mixed modelling was used to test the asso- ciation between demographic factors, psychoso- cial and traumatic variables, and MCI diagnosis. All analyses were conducted in SPSS 29.0.0.0. All measures were assessed at the individual level. Be- cause our sample consisted of twins, we used a linear mixed modelling approach to account for the clus- tering of twins within families by including a family ID variable as a random effect. Separate analyses were performed for each measure. Standardized scores were used for all outcome measures and for AFQT, educa- tion, and performance on neuropsychological tests. Model 1 tested whether age at testing date for Wave 2 (61.72 ± 2.44 years) had an effect on MCI out- come. Model 2 tested the association of intelligence, as measured by AFQT performance upon military induction (collected between 1965–1975) with MCI outcome. Model 3 tested the association of ApoE e4 allele status on MCI outcome. Model 4 examined the correlation between years of education on MCI out- come. Model 5 tested the association of race on MCI outcome. Model 6 tested the association of ethnicity (Hispanic/Non-Hispanic) on MCI outcome. Model 7 tested the association of PTSD diagnosis at Harvard Drug Study (Tsuang, et al., 2001) data collection date (1996) as measured by the DIS-III-R on MCI out- 56 GOULDEN come. Model 8 tested the association of psychosocial stressors over the last two years before the Wave 2 test- ing date on MCI outcome. Model 9 tested the asso- ciation of PTSD symptoms at Harvard Drug Study (Tsuang, et al., 2001) data collection date (1996), as measured by the DIS-III-R collection on MCI out- come. Significant associations were determined using the type III fixed effects. Associations were consid- ered statistically significant at p < 0.05, two-tailed. Results Table 1 shows that Age (F = 3.879, p = 0.05) did not have a significant effect on MCI diagnosis, al- though it exhibited a trend towards a predictive effect. Intelligence, measured by the AFQT taken between ages 18-25, was found to have no significant impact on MCI diagnosis (F = 3.523, p = 0.061). The ApoE e4 allele(s) status (F = .236, p = 0.628) was also found to have no significant impact on MCI diagnosis. Ed- ucation, however, (F = 4.667, p = 0.031) was found to have a significant effect on MCI diagnosis. Nei- ther Race (F = 2.195, p = 0.139) , nor Ethnicity (F = 2.252, p = 0.134) had a significant effect on MCI diagnosis. PTSD diagnosis, as measured by the DIS- III-R (Coopers and Michel, 1988), collected for the Harvard Drug Study in 1996 (F = 0.249, p = 0.618) did not have a significant effect on MCI diagnosis. Life stress exposure, as measured by the Holmes and Rahe (1967), (F = 0.249, p = 0.618) also did not have a significant impact on MCI diagnosis. Lastly, Table 1 shows that PTSD symptoms (F = 0.006, p = 0.936) did not have a significant impact on MCI diagnosis. Discussion Our findings in a mixed model analysis of VETSA Wave 2 did not support the hypothesis. No significant associations between the stress factors we examined and MCI diagnosis in Wave 2 was observed. A signifi- cant relationship between education and MCI was ob- served, but is not surprising, as epidemiological studies consistently report that a high level of education is as- sociated with a reduced risk of cognitive impairment (Anttila et al., 2002; Fratiglioni & Wang, 2007; Ngan- du et al., 2007). The results of these studies indicate that education might reflect the extent of early cogni- tive stimulation of the brain which may influence glob- al cognitive abilities. The average lifetime education of the participants was 12.4 (± 1.3 years; see Appendix A). Previous studies investigated the relationship be- tween psychosocial and traumatic stressor activities with cognitive decline. We hypothesize that our results may vary from those obtained by Yaffe et al. (2010) be- cause their study tracked health records over a period of seven years, while our analysis looked at new MCI diagnoses over a shorter time span between each VET- SA data collection point. Further, their subjects were more racially diverse and included women. Our study only investigated the onset of MCI, while their study included all types of dementias, including end-stage Alzheimer’s Disease. Most importantly, the mean baseline age of their veteran subjects was 68.8 years. The study by Weiner et al. (2017) primarily aimed to establish the relationship between traumatic brain injury, PTSD, and Alzheimer’s Disease biomarkers. Their study’s conceptualization of PTSD may have been more relevant to finding the connection be- tween traumatic exposure and cognitive impairment, as it accounted for current and lifetime PTSD in- stead of PTSD status at one point in 1996. Another feature of their study that contributed to their find- ing was that their subjects had a mean age of 67.8. As for Peavy et al. (2009), their operationalization of MCI was the less stringent Peterson et al. (1999) criteria. In one study by Oltra-Cucarella et al. (2018) criteria for MCI misclassified 24% of the sample com- pared to the more conservative Jak-Bondi MCI criteria (Jak et al., 2009) used for the current study. Stress was also measured as a cortisol rating and from responses to the Life Events and Difficulties Schedule (Brown & Harris, 1978), which quantified events over the partic- ipants’ entire adult life instead of within the last two years before the study visit. Stressful events were also self-reported every six months, at which time cortisol was measured. This data was collected for two to three years. Aside from this difference in operationalization of toxic stress, the mean age of participants was 78.8 years old, making their participants much older than ours. Wang et al. (2018) studied non-military elders and found a significant dose-dependent relation- ship between PTSD symptom severity and devel- oping dementia later in life. Though the average age of their participants was 55.44 years, PTSD severity was indicated by the frequency of psychiatric clin- ic visits for PTSD. This operationalization of PTSD is not through clinically supported or uniform diag- nosis criteria. As for psychosocial and trauma-related 57 STRESSORS ON MCI DIAGNOSIS stress, Bougea et al. (2022) may have found results that differed from ours because they used a wide va- riety of conceptualizations of psychosocial stress. For dementia diagnosis, one included study used self-re- port responses to quantify dementia. All partici- pant data examined was for people 65 years or older. One of the main limitations of the current study was the small number of people that met the diagnos- tic criteria for MCI, which in total was only 147 after adjusting for age, education, and practice effects. This small number resulted in a lack of statistical power. Ad- ditional issues with the analysis may be from our oper- ationalization of psychosocial and traumatic stressors. The measure of psychosocial stress used, the Holmes and Rahe (1967), only pertained to the two years of the participant’s life prior to test administration. A measure that accounts for the entirety of adulthood, from 18 years old onwards, would more accurately quantify the total adult stress burden. As for using the PTSD diagnosis conferred in 1996, the literature sup- ports that just one incidence of PTSD permanently alters the brain (Hendrickson & Raskind, 2016). The mechanism is that trauma causes permanent neuronal changes that harm learning, habituation, and stimulus discrimination. Some of these neuronal changes that have a continuing impact do not even depend on ac- tual exposure to reminders of the trauma for expres- sion (Van der Kolk, 2003). Even so, a diagnosis back in 1996 may not be as relevant to cognitive status during the data collection period of VETSA Wave 2, which occurred in the 2010’s (between 2009 and 2014). Our insignificant findings are still relevant to MCI risk factor research. Our limitations highlight the im- portance of subject selection and support existing re- search on the typical age of onset for MCI (Howieson et al., 2008). Future analyses should be conducted with VETSA study data that covers a longer period of the twins’ lives. Future analyses should also include data collected when the participants were older. These two suggestions should be followed so that a longer period of toxic stress exposure can be quantified and so that the subjects will be older, giving more time for MCI to develop. Higher rates of MCI can be expected to emerge as the participants age. The results did not indicate that toxic psychosocial stress exposure and/or traumatic stress had an impact on rates of MCI diagno- sis within a subject pool of veterans from the VETSA study. Monitoring the health of veterans as they age is important for addressing the epidemic of cognitive im- pairment in the general population. Even those with primarily non-combatant roles are at risk for adverse health outcomes due to their service. Veterans have made great sacrifices and the scientific community must ensure that their unique healthcare needs are met. References Anttila, T., Helkala, E. L., Kivipelto, M., Hallikainen, M., Alhainen, K., Heinonen, H., Mannermaa., A., Tuomilehto., J., Soininen, H., & Nissinen, A. (2002). Midlife income, occupation, APOE status, and dementia: A population-based study. Neu- rology, 59(6), 887-893. https://doi.org/10.1212/ WNL.59.6.887 Ashford, J. W. (2004). APOE genotype effects on Alz- heimer’s disease onset and epidemiology. Journal of Molecular Neuroscience, 23, 157-165. https:// doi.org/10.1385/JMN:23:3:157 Bougea, A., Anagnostouli, M., Angelopoulou, E., Spanou, I., & Chrousos, G. (2022). Psycho- social and trauma-related stress and risk of de- mentia: A meta-analytic systematic review of longitudinal studies. Journal of Geriatric Psychi- atry and Neurology, 35(1), 24–37. https://doi. org/10.1177/0891988720973759 Brown, G., Harris, T., Social Origins of Depres- sion: A Study of Psychiatric Disorders in Wom- en. London, Tavistock, 1978. https://doi. org/10.4324/9780203714911 Cohen S., Kessler R.C., Gordon U.L., Strategies for measuring stress in studies of psychiatric and physical disorder. In: Cohen S, Kessler RC, Gordon UL, eds. Measuring Stress: A Guide for Health and Social Scientists. New York, NY: Oxford University Press; 1995:3-26 https://doi. org/10.1016/j.jad.2016.08.013 Coopers, A. M., & Michels, R. (1988). Diagnostic and statistical manual of mental disorders, revised (DSM-III-R). American journal of Psychiatry, 145(10), 1300-1301. Corbo, I., Marselli, G., Di Ciero, V., & Casagrande, M. (2023). The protective role of cognitive reserve in mild cognitive impairment: A systematic review. Journal of Clinical Medicine, 12(5), 1759. https:// doi.org/10.3390/jcm12051759 Eriksson-Sörman, D., & Umeå universitet. Institu- tionen för psykologi. (2015). The influence of 58 GOULDEN social relationships and leisure activity on adult cognitive functioning and risk of dementia longi- tudinal population-based studies. Department of Psychology, Umeå University. https://urn.kb.se/ resolve?urn=urn:nbn:se:umu:diva-101840 First, M. B., France, A., & Pincus, H. A. (2004). DSM- IV-TR Guidebook. American Psychiatric Publish- ing, Inc. Fratiglioni, L., & Wang, H. X. (2007). Brain reserve hy- pothesis in dementia. Journal of Alzheimer's Dis- ease, 12(1), 11-22. doi: 10.3233/jad-2007-12103 Gatz, M., Mortimer, J. A., Fratiglioni, L., Johansson, B., Berg, S., Reynolds, C. A., & Pedersen, N. L. (2006). Potentially modifiable risk factors for de- mentia in identical twins. Alzheimer’s & Demen- tia, 2(2), 110–117. https://doi.org/10.1016/j. jalz.2006.01.002 Hathaway, L. M., Boals, A., & Banks, J. B. (2010). PTSD symptoms and dominant emotional response to a traumatic event: An examination of DSM-IV Cri- terion A2. Anxiety, Stress & Coping, 23(1), 119-126. https://doi.org/10.1080/10615800902818771 Hendrickson, R. C., & Raskind, M. A. (2016). Nor- adrenergic dysregulation in the pathophysiology of PTSD. Experimental Neurology, 284, 181-195. https://doi.org/10.1016/j.expneurol.2016.05.014 Hemingway, H., & Marmot, M. (1999). Evidence based cardiology: Psychosocial factors in the aeti- ology and prognosis of coronary heart disease. Sys- tematic review of prospective cohort studies. BMJ (Clinical research ed.), 318(7196), 1460–1467. https://doi.org/10.1136/bmj.318.7196.1460 Holmes, T. H., & Rahe, R. H. (1967). The social readjustment rating scale. Journal of psychoso- matic research, 11(2), 213–218. https://doi. org/10.1016/0022-3999(67)90010-4 Howieson, D. B., Carlson, N. E., Moore, M. M., Was- serman, D., Abendroth, C. D., Payne-Murphy, J., & Kaye, J. A. (2008). Trajectory of mild cog- nitive impairment onset. Journal of the Interna- tional Neuropsychological Society, 14(2), 192-198. https://doi.org/10.1017/S1355617708080375 Humphreys, L. G. (1979). The construct of general intelligence. Intelligence, 3(2), 105-120. https:// doi.org/10.1016/0160-2896(79)90009-6 Jak, A. J., Bondi, M. W., Delano-Wood, L., Wierenga, C., Corey-Bloom, J., Salmon, D. P., & Delis, D. C. (2009). Quantification of five neuropsycholog- ical approaches to defining mild cognitive impair- ment. The American Journal of Geriatric Psychia- try : Official Journal of the American Association for Geriatric Psychiatry, 17(5), 368–375. https:// doi.org/10.1097/JGP.0b013e31819431d5 Kremen, W. S., Franz, C. E., & Lyons, M. J. (2013). VETSA: The Vietnam era twin study of aging. Twin Research and Human Genetics, 16(1), 399- 402. https://doi.org/10.1017/thg.2012.86 Kritikos, M., Diminich, E. D., Meliker, J., Mielke, M., Bennett, D. A., Finch, C. E., ... & Luft, B. J. (2023). Plasma amyloid beta 40/42, phosphorylat- ed tau 181, and neurofilament light are associated with cognitive impairment and neuropathological changes among World Trade Center responders: A prospective cohort study of exposures and cog- nitive aging at midlife. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 15(1). https://doi.org/10.1002/dad2.12409 Leistner, C., & Menke, A. (2020). Hypothalam- ic–pituitary–adrenal axis and stress. Handbook of Clinical Neurology, 175, 55-64. https://doi. org/10.1016/B978-0-444-64123-6.00004-7 Lyons, M. J., Genderson, M., Grant, M. D., Logue, M., Zink, T., McKenzie, R., Franz, C. E., Paniz- zon, M., Lohr, J. B., Jerskey, B., & Kremen, W. S. (2013). Gene‐environment interaction of ApoE genotype and combat exposure on PTSD. Amer- ican Journal of Medical Genetics Part B: Neuro- psychiatric Genetics, 162(7), 762-769. https://doi. org/10.1002/ajmg.b.32154 Moon, H., Badana, A. N. S., Hwang, S. Y., Sears, J. S., & Haley, W. E. (2019). Dementia Prevalence in Older Adults: Variation by Race/Ethnicity and Immigrant Status. The American Journal of Geriatric Psychiatry, 27(3), 241–250. https://doi. org/10.1016/j.jagp.2018.11.003 Nabe-Nielsen, K., Hansen, Å. M., Ishtiak-Ahmed, K., Grynderup, M. B., Gyntelberg, F., Islamoska, S., Mortensen, E. L., Phung, T. K. T., Rod, N. H., Waldemar, G., Westendorp, R. G. J., & Garde, A. H. (2019). Night shift work, long working hours and dementia: A longitudinal study of the Danish Work Environment Cohort Study. BMJ Open, 9(5). https://doi.org/10.1136/bmjopen-2018-027027 Ngandu, T., von Strauss, E., Helkala, E., L., Winblad, B., Nissinen, A., Tuomilehto, J., H. Soininen, & Kivipelto, M. (2007). Education and demen- 59 STRESSORS ON MCI DIAGNOSIS tia: What lies behind the association? Neurology, 69(14), 1442-1450. https://doi.org/10.1212/01. wnl.0000277456.29440.16 Oltra-Cucarella, J., Sánchez-SanSegundo, M., Lipnic- ki, D. M., Sachdev, P. S., Crawford, J. D., Perez-Vi- cente, J. A., Ferrer-Cascales, R. (2018). Using the base rate of low scores helps to identify progres- sion from amnestic MCI to AD. Journal of the American Geriatrics Society, 66(7), 1360–1366. https://doi.org/10.1111/jgs.15412 Orme, D. R., Brehm, W., & Ree, M. J. (2001). Armed Forces Qualification Test as a measure of premorbid intelligence. Military Psychology, 13(4), 187-197. https://doi.org/10.1207/S15327876MP1304_1 Peavy, G. M., Salmon, D. P., Jacobson, M. W., Hervey, A., Gamst, A. C., Wolfson, T., Patterson, T. L., Goldman, S., Mills, P. J., Khandrika, S., & Galas- ko, D. (2009). Effects of chronic stress on memory decline in cognitively normal and mildly impaired older adults. The American Journal of Psychiatry, 166(12), 1384–1391. https://doi.org/10.1176/ appi.ajp.2009.09040461 Petersen, R. C. (2011). Mild cognitive impairment. New England Journal of Medicine, 364(23), 2227- 2234. https://doi.org/10.1056/NEJMcp0910237 Petersen, R. C., Smith, G. E., Waring, S. C., Ivnik, R. J., Tangalos, E. G., & Kokmen, E. (1999). Mild cognitive impairment: Clinical characterization and outcome. Archives of Neurology, 56(3), 303– 308. https://doi.org/10.1001/archneur.56.3.303 Portet, F., Ousset, P. J., Visser, P. J., Frisoni, G. B., Nobili, F., Scheltens, P., Vellas, B., Touchon, J., & MCI Working Group of the European Con- sortium on Alzheimer's Disease (EADC) (2006). Mild cognitive impairment (MCI) in medical practice: A critical review of the concept and new diagnostic procedure. Report of the MCI Work- ing Group of the European Consortium on Alz- heimer's Disease. Journal of Neurology, Neurosur- gery, and Psychiatry, 77(6), 714–718. https://doi. org/10.1136/jnnp.2005.085332 Rountree, S. D., Waring, S. C., Chan, W. C., Lupo, P. J., Darby, E. J., & Doody, R. S. (2007). Impor- tance of subtle amnestic and non-amnestic defi- cits in mild cognitive impairment: Prognosis and conversion to dementia. Dementia and Geriatric Cognitive Disorders, 24(6), 476-482. https://doi. org/10.1159/000110800 Sattler, C., Toro, P., Schönknecht, P., & Schröder, J. (2012). Cognitive activity, education and socio- economic status as preventive factors for mild cognitive impairment and Alzheimer's disease. Psychiatry Research, 196(1), 90-95. https://doi. org/10.1016/j.psychres.2011.11.012 Scott, J. C., Matt, G. E., Wrocklage, K. M., Crnich, C., Jordan, J., Southwick, S. M., & Schweinsburg, B. C. (2015). A quantitative meta-analysis of neu- rocognitive functioning in posttraumatic stress disorder. Psychological Bulletin, 141(1), 105–140. https://doi.org/10.1037/a0038039 Sibener, L., Zaganjor, I., Snyder, H. M., Bain, L. J., Egge, R.; Carrillo, M. C. (2014). Alzheimer’s Dis- ease prevalence, costs, and prevention for military personnel and veterans. Alzheimer’s and Dementia, 10, (3). https://doi.org/10.1016/j.jalz.2014.04.011 Song, H., Sieurin, J., Wirdefeldt, K., Pedersen, N. L., Almqvist, C., Larsson, H., Valdimarsdóttir, U. A.; Fang, F. (2020). Association of stress-related disor- ders with subsequent neurodegenerative diseases. JAMA Neurology,77(6), 700–709. https://doi. org/10.1001/jamaneurol.2020.0117 Tang, Z., Zhu, Y., & Wang, Z. (2023). Characterizing Alzheimer's Disease Biomarker Cascade Through Non-linear Mixed Effect Models. Pre-print. Tervo, S., Kivipelto, M., Hänninen, T., Vanhanen, M., Hallikainen, M., Mannermaa, A., & Soininen, H. (2004). Incidence and risk factors for mild cogni- tive impairment: a population-based three-year fol- low-up study of cognitively healthy elderly subjects. Dementia and Geriatric Cognitive Disorders, 17(3), 196-203. https://doi.org/10.1159/000076356 Turner, A. D., James, B. D., Capuano, A. W., Aggarwal, N. T., & Barnes, L. L. (2017). Perceived stress and cognitive decline in different cognitive domains in a cohort of older African Americans. The Ameri- can Journal of Geriatric Psychiatry, 25(1), 25-34. https://doi.org/10.1016/j.jagp.2016.10.003 Tsuang M.T., Bar J.L., Harley R.M., & Lyons M.J. (2001). The Harvard twin study of substance abuse: What we have learned. Harvard Review Psychiatry, 9(6), 267– 279. https://doi.org/10.1080/10673220127912 Uhlaner, J. E., & Bolanovich, D. J. (1952). De- velopment of Armed Forces Qualification Test and Predecessor Army Screening Tests, 1946- 1950, 1–62 Van der Kolk, B. A. (2003). Psycho- biology of posttraumatic stress disorder. Textbook 60 of biological psychiatry, 319-344. doi: 10.21236/ AD0000191 Veitch, D., E Friedl, K., & W Weiner, M. (2013). Mil- itary risk factors for cognitive decline, dementia and Alzheimer’s disease. Current Alzheimer Re- search, 10(9), 907-930. https://doi.org/10.2174/1 5672050113109990142 Wang, T. Y., Wei, H. T., Liou, Y. J., Su, T. P., Bai, Y. M., Tsai, S. J., Yang, A. C., Chen, T. J., Tsai, C. F., & Chen, M. H. (2018). Risk for developing demen- tia among patients with posttraumatic stress dis- order: A nationwide longitudinal study. Journal of Affective Disorders, 205, 306–310. https://doi. org/10.1016/j.jad.2016.08.013 Weiner, M. W., Harvey, D., Hayes, J., Landau, S. M., Aisen, P. S., Petersen, R. C., Tosun, D., Veitch, D. P., Jack, C. R., Decarli, C., Saykin, A. J., Grafman, J., & Neylan, T. C. (2017). Effects of traumatic brain injury and posttraumatic stress disorder on development of Alzheimer’s disease in Vietnam veterans using the Alzheimer’s Disease neuroim- aging initiative: Preliminary report. Clinical Inter- ventions, 3, 177–188. https://doi.org/10.1016/j. trci.2017.02.005 Yaffe, K., Vittinghoff, E., Lindquist, K., Barnes, D., Covinsky, K. E., Neylan, T., Marmar, C. (2010). Posttraumatic stress disorder and risk of dementia among US veterans. Archives of General Psychi- atry, 67(6), 608-613. https://doi.org/10.1001/ archgenpsychiatry.2010.61 GOULDEN 61 STRESSORS ON MCI DIAGNOSIS Table 1 Mixed Model Tests of Fixed Effects on MCI Diagnosis Note. Dependent Variable: MCI diagnosis at VETSA Wave 2. *at ages 17–25 years, as measured by the AFQT upon military induction between 1965–1975, collected for the Harvard Drug Study **no e4 allele vs. homo- or heterozygous for e4 *** as measured by the DIS-III-R, collected for the Harvard Drug Study in 1996 **** measured over the last two years from testing date by the Holmes and Rahe (1967). 62 GOULDEN Appendix A Note. Subject demographic data from Lyons et al. (2013).