- Email: [email protected]
Trauma History Is Associated With Psychological Distress and Somatic Symptoms in Homebound Older Adults
Trauma History Is Associated With Psychological Distress and Somatic Symptoms in Homebound Older Adults Andrew J. Petkus, M.A., Amber M. Gum, Ph.D., Bellinda King-Kallimanis, M.S., Julie Loebach Wetherell, Ph.D.
Objectives: To examine the physical health, anxiety, and depressive symptoms of homebound older adults in relation to a history of trauma exposure. Design: Secondary analysis of cross-sectional data. Setting: Participants were recruited by in-home aging services case managers. Participants: The sample consisted of 136 older adults receiving in-home services. Measurements: Participants completed a research battery that included the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID), Modified Mini-Mental Status Examination, and Brief Symptom Inventory-18 (BSI-18). Case managers provided standard assessments of health and functional status. Data were analyzed using t tests, 2 tests, and logistic regression. Results: Approximately half (49%) of the participants reported experiencing a traumatic event. When compared with participants who did not report a history of trauma, trauma-exposed participants were on average 3 years younger, had one more chronic medical condition, were taking two more nonpsychotropic medications, and were more likely to report poor self-rated health. Furthermore, trauma-exposed participants were more likely to experience clinically significant somatic and anxiety symptoms on the BSI-18 and were more likely to meet criteria for a SCID anxiety disorder diagnosis. After controlling for physical health and cognitive functioning, trauma exposure was still associated with psychiatric disorders and higher levels of somatic and anxiety symptoms. Conclusions: Trauma-exposed participants were younger but experienced worse physical and mental health than those without a trauma history. This suggests that trauma exposure across the lifespan could have a long-term negative impact on homebound older adults’ health and quality of life. (Am J Geriatr Psychiatry 2009; 17:810 –818) Key Words: Homebound elderly, trauma, health, anxiety, depression
Received November 24, 2008; revised April 3, 2009; accepted May 7, 2009. From the San Diego State University/University of California, San Diego Joint Doctoral Program in Clinical Psychology, San Diego, CA (AJP); Department of Aging and Mental Health, Louis de la Parte Florida Mental Health Institute, University of South Florida, Tampa, FL (AMG); Department of Medical Psychology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands (BK-K); and Department of Psychiatry, University of California, San Diego, CA (JLW). Send correspondence and reprint requests to Andrew J. Petkus, M.A., 3350 La Jolla Village Drive, Building 13, Room 118, San Diego, CA 92161. e-mail: [email protected] © 2009 American Association for Geriatric Psychiatry
810
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al.
A
pproximately two-thirds of individuals in the United States will experience at least one traumatic event in their lifetime.1 A traumatic event is an event in which one is exposed to actual or threatened death, serious injury, or threat to the physical integrity of self or others.2 Exposure to a traumatic event can lead to psychological difficulties, the most prototypical of which is posttraumatic stress disorder (PTSD). The PTSD is characterized by fear, reexperiencing the traumatic event, avoidance, emotional numbing, and increased arousal.2 Studies have also shown a relationship between trauma and other outcomes such as depression3– 6 and other anxiety disorders.7 Research also suggests that the impact of trauma may be long lasting; exposure to childhood trauma8 and to traumatic event at any time during the lifespan6,9 have been associated with poorer mental health in older adulthood. A substantial body of literature documents the association between trauma and physical health. Women with a history of sexual assault,10,11 physical or sexual abuse,12 or childhood emotional abuse7 report more somatic complaints than women with no trauma history. In primary care samples, male trauma survivors have increased risk of diabetes and arthritis, whereas female survivors are at greater risk for cancer and digestive disorders.13 These adverse health effects may persist into older adulthood. Older adults who had experienced any traumatic event throughout their lives have reported poorer self-rated health as well as increased disability and more chronic health conditions.8,14 Furthermore, PTSD and depressive symptoms later in life may further lead to negative health perceptions in the elderly.14 Traumas can have long-lasting effects on how individuals perceive themselves, their coping abilities, and the world.15 Although older adults may have experienced traumatic events years or even decades ago and have coped successfully in midlife, age-related declines in physical and cognitive functioning may re-evoke distressing reactions to past traumas in old age.16 One study found that 10% of older adults experience flashbacks, nightmares, or intrusive memories of traumatic events that occurred decades ago.17 Moreover, new stressful events may amplify these symptoms, putting individuals facing current stressors at risk for experiencing sequelae from past traumatic events.16 Studies suggest that 10%–30% of older adults are facing substantial disability and chronic illness that
Am J Geriatr Psychiatry 17:9, September 2009
render them homebound.18,19 This population is at high risk for depression and other forms of psychological distress19 as well as for nursing home placement and institutionalized care. Research with homebound elders, therefore, is important because improved understanding of this population may lead to more effective interventions and strategies to postpone institutionalization. Because of their current significant stressors like chronic illness, disability, and cognitive decline, homebound elders represent one group of older adults who may be particularly vulnerable to the long-term effects of prior traumas. Because of the high prevalence of disability and medical conditions in this population, one might speculate that the physical health and cognitive functioning of homebound elders may be more relevant to these elders’ emotional functioning than traumas that occurred years or decades ago. Because of the chronic effects of trauma, however, an alternate hypothesis is that past traumas may still negatively impact the psychological experiences of homebound older adults independently of current health-related stressors. Despite the potential importance of trauma in a population at risk for adverse health consequences and need for institutional care, to our knowledge, no research has examined relationships of psychological trauma with health outcomes among homebound older adults. The purpose of this study was to investigate the relationship between history of trauma exposure and the physical health and psychological distress of homebound elders. It was hypothesized that participants who reported exposure to a trauma at any point in life would be in worse physical health and exhibit higher levels of distress than participants who were not exposed to a trauma. It was further hypothesized that trauma exposure would be associated with psychological distress even after controlling for physical health and cognitive functioning.
METHODS Data Source and Sample Data for these analyses were drawn from a larger project, investigating the mental health needs of
811
Trauma Exposure and Homebound Elders homebound older adults. A brief summary of the study design, sampling procedures, measures, and other procedures are described later; a more detailed description appears in another article.20 A sample of 142 older adults receiving in-home social services through aging service agencies provided data through in-person interviews; data were also extracted from standardized assessments administered by the case managers. Individuals were eligible to participate if they were aged 60 years or older, spoke English, and did not have a known diagnosis of dementia per the case manager. In total, 231 individuals agreed to be contacted by the research team and 142 individuals completed the interview, resulting in a response rate of 61.5%. Of those who did not complete the interview, 70 refused primarily because of health problems, 11 were unable to participate because of cognitive impairment, 6 were unable to be contacted after numerous attempts, and 2 died before scheduling interviews. For the questions related to trauma history, four participants refused and data for two participants were missing, resulting in a final sample of N ⫽ 136.
Measures Diagnosis of Depressive and Anxiety Disorders. The Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID)21 was administered to assess for current depressive and anxiety disorders. The following disorders were assessed: major depressive disorder, dysthymic disorder, depressive disorder not otherwise specified, depressive disorder because of medical condition or substance, panic disorder, agoraphobia, specific phobia, social phobia, obsessive-compulsive disorder, PTSD, generalized anxiety disorder, anxiety disorder not otherwise specified, and anxiety because of general medical condition or substance. All interviewers had at least a bachelor’s degree and were trained and supervised by a licensed clinical psychologist (AG). All interviews were audiotaped (unless refused by the participant) and a random 20% of the interviews were rated by AG to assess interrater reliability. Kappas ranged from 0.61 to 1.00 across disorders representing high interrater reliability.22
812
Trauma History. Trauma history was derived from the PTSD section of the anxiety disorders module of the SCID. Participants were asked: “Sometimes things happen to people that are extremely upsetting—things like being in a life-threatening situation like a major disaster, very serious accident, or fire; being physically assaulted or raped; seeing another person killed or dead or badly hurt, or hearing about something horrible that has happened to someone you are close to. At any time during your life, have any of these kinds of things happened to you?” Participants who responded “yes” were asked to briefly describe the nature of each event, the year of each event, and their age at the time of each event. Green’s23 criteria for traumatic events were used to determine what constitutes a traumatic event. Using these criteria, a traumatic event was defined as an event in which the individual experienced one or more of the following: threat to one’s life or bodily integrity; severe physical harm or injury; receipt of intentional injury/harm; exposure to the grotesque; violent/sudden loss of a loved one; witnessing or learning of violence to a loved one; learning of exposure to a noxious agent; or causing death or severe harm to another. The first author and the second author independently coded each event as traumatic or not based on these criteria and interrater reliability was assessed. A kappa value of 0.95 was achieved, representing high interrater reliability.22 In cases of disagreement, the first author and the second author discussed discrepancies and the final decision was reached by consensus. Indicators of Physical Health. Health variables included the following: a) self-rated health; b) number of prescription nonpsychotropic medications; c) number of chronic medical conditions (cancer, diabetes, emphysema, heart problems, liver problems, stroke, Parkinson disease, allergies, asthma, bladder/kidney problems, high blood pressure, gall bladder problems, or thyroid problems); and d) functional impairment as measured by a combined total of number of limitations in activities of daily living (ADL; bathing or showering, dressing, eating, getting in or out of bed or chairs, walking, and using the toilet) and instrumental activities of daily living (IADL; using the telephone, shopping, preparing food, housekeeping, laundry, transportation, taking medications, and handling finances). Self-rated health was assessed by asking participants to rate
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al. their health using a 5-point scale (ranging from “1poor” to “5-excellent”). Number of prescribed medications was measured by asking the participant to provide a list of all medications that were being taken (values for sample ranged from 0 to 20). Number of chronic medical conditions (values for sample ranged from 1 to 15) and ADL and IADL limitations (values for sample ranged from 2 to 14) were derived from the case manager’s assessment nearest in time to the research interview. Modified Mini-Mental Status Examination. The 3MS24 is a measure of cognitive functioning and is a modification of the Mini-Mental Status Examination.25 Scores ranged from 0 to 100, with higher scores representing better functioning. The 3MS has strong validity and reliability.24 Cronbach’s ␣ in this sample was 0.84. Brief Symptom Inventory-18. The Brief Symptom Inventory-18 (BSI-18)26 was administered to measure somatic, depressive, and anxiety symptoms. Each item assesses distress during the past week on a 5-point scale ranging from “0-not at all” to “4-extremely.” The total score ranges from 0 to 72 and the Somatic, Depression, and Anxiety subscale scores range from 0 to 24; higher scores represent more distress. The BSI-18 manual contains normative data for men and women and suggests adopting a t score of 63 as a cut score to represent caseness on each respective subscale. In this sample, Cronbach’s ␣ values were 0.86 depression, 0.81 anxiety, 0.71 somatic, and 0.89 for the total score.
regression model included the following covariates: age 60 –93 years, sex (0 ⫽ man, 1 ⫽ woman), number of chronic medical conditions (range: 1–15), combined number of ADL and IADL limitations (range: 2–14), number of nonpsychotropic medications (range: 0 –20), 3MS score (range: 64 –100), and history of trauma exposure (0 ⫽ absent, 1 ⫽ present).
RESULTS Frequency and Age of Traumatic Events Figure 1 depicts the proportion of participants reporting each category of traumatic event. Nearly half of the sample (N ⫽ 67, 49%) reported experiencing at least one traumatic event. The most common trauma experienced was the sudden unexpected loss of a loved one (N ⫽ 35, 27.7%). Table 1 provides information regarding participants’ age at the time of the traumatic event and number of events experienced. Participants most frequently experienced traumatic events between the ages of 19 and 59 years and were least likely to report a trauma at the age of 60 years or older (N ⫽ 14, 10.3%). Among the participants who reported experiencing a trauma, most reported experiencing only one traumatic event (N ⫽ 40, 58.7%).
Comparisons by Trauma Data Analysis Descriptive statistics for all variables were computed. To assess the first hypothesis, comparisons between those with and without a history of trauma were conducted for all other variables (demographics, depressive and anxiety symptoms, diagnoses, indicators of physical health, and 3MS) using t tests for continuous variables and 2 tests for dichotomous variables. The BSI-18 data were significantly skewed; therefore, these data were dichotomized using the caseness rule of a t score of 63 proposed by the authors of the scale as a cutoff to represent clinically significant distress.26 Furthermore, logistic regression analysis was used to examine relationships between covariates and BSI-18 caseness (yes/no) and presence of SCID disorder (yes/no). For each outcome, the final
Am J Geriatr Psychiatry 17:9, September 2009
Table 1 provides descriptive information comparing those who did (N ⫽ 67) and did not (N ⫽ 69) experience a trauma. On average, participants experiencing a trauma were significantly younger and more likely to report being in poor health. They also had more medical conditions and were taking more prescribed nonpsychotropic medications. A significantly greater proportion of trauma-exposed participants were identified by the BSI-18 as currently exhibiting clinically significant anxiety and somatic symptoms. Trauma-exposed participants were also more likely to meet criteria for an anxiety disorder or any Axis I disorder overall. The two groups did not differ with regard to depressive symptoms, cognitive performance, functional impairment, education, sex, and marital status.
813
Trauma Exposure and Homebound Elders
FIGURE 1.
Frequency of Traumatic Events Reported (N ⴝ 136)
Logistic Regression of Overall Distress, Anxiety, Depressive, and Somatic Symptoms Table 2 displays results from logistic regression, investigating current distress and psychiatric disorder by exposure to a trauma, after controlling for covariates. Across all regression analyses, residuals were examined and one outlier was identified. Analyses were rerun with this participant excluded and the results did not change; therefore, we included the outlier in our analyses. For BSI-18 somatic caseness, the final multivariate logistic regression model was statistically significant, (2(7) ⫽ 31.4, p ⬍0.01), and the model was found to accurately classify 72.8% of participants. In the multivariate model, experiencing a trauma (odds ratio [OR]: 2.23, 95% confidence interval [CI]: 1.01– 4.93, z (1) ⫽ 1.98, p ⬍0.05) and number of chronic medical conditions (OR: 1.36, 95% CI: 1.11–1.67, z (1) ⫽ 2.90, p ⬍0.01) were associated with higher levels of somatic symptoms. For BSI-18 depressive caseness, the multivariate model was not statistically significant (2(7) ⫽ 8.87, p ⫽ 0.26). Furthermore, no individual covariates were significant at the p ⬍0.05 level.
814
For BSI-18 anxiety caseness, the multivariate model was statistically significant (2(7) ⫽ 14.6, p ⬍0.05), and the model accurately classified 84.6% of participants. Trauma exposure (OR: 3.39, 95% CI: 1.02–11.26, z (1) ⫽ 2.00, p ⬍0.05) and poor cognitive performance (OR: 0.92, 95% CI: 0.86 – 0.98, z (1) ⫽ ⫺2.49, p ⬍0.02) were associated with clinically significant anxiety. The multivariate model with the outcome of meeting criteria for any Axis I disorder was statistically significant (2(7) ⫽ 15.5, p ⬍0.03), and the model accurately classified 76.1% of participants. Participants were more likely to meet criteria for an Axis I disorder if they had experienced a trauma (OR: 2.69, 95% CI: 1.05– 6.87, z (1) ⫽ 2.07, p ⬍0.04) or had more chronic medical conditions (OR: 1.26, 95% CI: 1.02–1.54, z (1) ⫽ 2.16, p ⬍0.04). The multivariate model with the outcome of meeting criteria for a depressive disorder was statistically significant (2(7) ⫽ 14.6, p ⬍0.05), and the model accurately classified 88.2% of participants. The number of physical conditions was a significant covariate (OR: 1.32, 95% CI: 1.03–1.69, z (1) ⫽ 2.20, p ⬍0.03), with more medical conditions being associated with greater like-
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al.
TABLE 1.
Sample Characteristics (N ⴝ 136)
Variable Demographics Age, years Education, years Sex Male Female Race White Minority Marital status Married Separated/divorced Never married Widow/widower Age at trauma 0–18 19–59 60⫹ Number of traumas 1 2 3 Mental health Current SCID disorder Any Depressive Anxiety PTSD BSI-18 caseness Somatic Depression Anxiety 3MS Physical health Self-rated health Poor Fair Good Very good/excellent Chronic physical conditions ADL ⫹ IADL limitations Nonpsychotropic medications
No Trauma (N ⴝ 69) N (%) or M (SD)
Trauma (N ⴝ 67) N (%) or M (SD)
t or 2 (df)
p
76.46 (8.23) 12.00 (3.08)
73.24 (8.19) 12.64 (2.34)
2.29 (134) ⫺1.37 (134)
0.024a 0.175
14 (20.3) 55 (79.7)
14 (20.9) 53 (79.1)
0.01 (1)
0.930
50 (72.5) 19 (27.5)
51 (76.1) 16 (23.9)
0.24 (1)
0.626
11 (15.9) 17 (24.6) 6 (8.7) 35 (50.7)
9 (13.4) 26 (38.8) 4 (6.0) 28 (41.8)
3.23 (3)
0.356
6.66 (1) 2.76 (1) 5.77 (1)
0.010b 0.097 0.016a
9.52 (1) 0.09 (1) 5.27 (1) ⫺0.75 (134)
0.002b 0.762 0.022a 0.453
— — —
14 (20.9) 36 (53.7) 14 (20.9)
— — —
40 (59.7) 22 (32.8) 5 (7.4)
9 (13.2) 5 (7.2) 4 (5.9) 0 (0.0)
21 (31.8) 11 (16.4) 13 (19.7) 3 (4.5)
22 (31.9) 12 (17.4) 4 (5.9) 87.52 (9.1)
39 (58.2) 13 (19.4) 14 (20.9) 88.63 (8.2)
17 (24.6) 23 (33.3) 22 (31.9) 7 (9.1) 4.56 (2.40) 7.72 (3.04) 6.26 (3.98)
32 (47.8) 25 (37.3) 9 (13.4) 1 (1.5) 5.61 (2.77) 7.73 (2.86) 8.80 (4.78)
16.10 (4)
⫺2.36 (134) ⫺0.01 (134) ⫺3.38 (134)
0.003b
0.020a 0.989 0.001b
Notes: Missing data: age of trauma: 12 from ages 0 to 18 years, 9 from ages 19 to 59 years, 10 from ages 60⫹. a p ⬍0.05. b p ⱕ0.01.
lihood of having a depressive disorder. Trauma exposure was not significant (OR: 2.05, 95% CI: 0.60 –7.02, z (1) ⫽ 1.14, p ⫽ 0.26). The multivariate model with the outcome of meeting criteria for a SCID anxiety disorder was statistically significant (2(7) ⫽ 16.7, p ⬍0.02), and the model accurately classified 86.6% of participants. Trauma was not significant (OR: 2.94, 95% CI: 0.84 –10.38, z (1) ⫽ 1.68, p ⫽ 0.10); participants were more likely to meet criteria
Am J Geriatr Psychiatry 17:9, September 2009
for an anxiety disorder if they were younger (OR: 0.92, 95% CI: 0.85– 0.99, z (1) ⫽ ⫺2.20, p ⬍0.03).
CONCLUSIONS Exposure to a traumatic event at any point during one’s lifetime was common for this sample of home-
815
Trauma Exposure and Homebound Elders
TABLE 2.
Trauma History as a Covariate of Mental Health Problems After Controlling for Demographics, Health, and Cognitive Performance
BSI-18 caseness Somatic Depression Anxiety Psychiatric disorder Any Depression Anxiety
Odds Ratio
Standard Error
95% Confidence Interval
2.23 0.90 3.39
0.90 0.45 2.08
1.01 0.34 1.02
2.69 2.05 2.95
1.29 1.29 1.89
1.05 0.60 0.84
z (df)
p
4.93 2.37 11.26
1.98 (1) (⫺1) 2.00 (1)
0.049a 0.836 0.046a
6.87 7.02 10.38
2.08 (1) 1.14 (1) 1.68 (1)
0.039a 0.255 0.092
Notes: The Wald test was performed for all hypothesis tests of individual odds ratio parameters above. a p ⬍0.05.
bound older adults, with 49% of the sample reporting a trauma history. This proportion is less than the prevalence of trauma exposure (68%) found in epidemiologic samples of older adults;1 whether this represents a difference between homebound and active elderly or is a byproduct of insufficiently sensitive methods used to detect trauma in this study is unknown. Although they were younger, trauma-exposed participants appeared to be in worse physical health: they had more chronic medical conditions and somatic symptoms, were more likely to report poor health, and were taking more nonpsychotropic medications, when compared with participants who had not experienced a trauma. In addition, traumaexposed participants were more likely to meet criteria for an anxiety or any psychiatric disorder and to report clinically significant levels of anxiety symptoms. After controlling for physical health and cognitive functioning, trauma exposure was still associated with psychiatric disorders and somatic and anxiety symptoms. These findings are largely consistent with the study hypotheses, except that significant relationships were not found for depressive symptoms or when depressive or anxiety psychiatric disorders were examined separately. Trauma exposure was associated with poor health as measured by medical conditions, medications, self-reported health, and somatic symptoms. This finding is consistent with the other studies of trauma in older adults.8 Notably, trauma-exposed participants were in worse health, however, they were on average 3 years younger than participants who had not experienced a trauma. This suggests that trauma exposure may contribute to an accelerated process of physical decline and becoming homebound. Further-
816
more, the relationships of trauma were significant across a variety of health and mental health outcomes despite the fact that most traumas occurred ⬎10 years earlier. This provides additional support for the proposition that the effects of traumas can persist during long periods. Biopsychosocial changes after trauma exposure may explain the relationship between trauma and declining health. Trauma exposure can lead to psychological changes such as PTSD, depression, and increased stress. These psychological changes have been associated with biological changes such as increased activity of the hypothalamic-pituitary adrenal axis and attentional changes such as increased sensitivity to somatic symptoms.27 Thus, trauma exposure earlier in life may produce biopsychosocial changes that in turn may increase allostatic load and lead to poor health.27 Trauma was also associated with psychiatric disorders and anxiety symptoms. These relationships remained present after controlling for physical health and cognitive functioning: in fact, trauma exposure was the strongest covariate for each outcome. The stresses associated with being homebound may evoke feelings of helplessness and not being in control, thus re-evoking the experience of their prior trauma. No statistically significant association between depression and trauma exposure was found. This finding is not consistent with other studies of trauma and depression because other studies have reported that trauma exposure was associated with increased depressive symptoms.9 This finding, however, is consistent with a study documenting higher levels of anxiety, but not depression, in older Holocaust sur-
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al. vivors.28 Although depressive disorders were more than twice as prevalent among those with a trauma history as among those without a trauma history (16.4% versus 7.2%) in this study, the difference was not statistically significant. Having a depressive disorder was associated with number of chronic physical conditions, consistent with a large body of research.29 The insensitive measurement and retrospective recall of traumatic events are important limitations of this study, which could have resulted in an underestimate of the prevalence of traumas. Checklists of traumatic events may be more reliable than openended questions in determining trauma exposure.30 Inconsistencies in reporting past traumatic events have been documented when using the open-ended question from the SCID PTSD module to assess for past trauma exposure.31 Although some studies report that the retrospective recall of trauma can be reliable,32 participants who are suffering from physical illness or psychological distress may potentially exhibit a recall bias for traumatic events. Both insensitive measurement of past trauma and potential recall biases may have synergistically contributed to the relatively low frequency of trauma exposure found in this sample and affected our results. Thus, an alternate explanation of our findings is that participants in worse physical health and experiencing more anxiety were potentially more likely to recall traumatic events than those who were doing well. These limitations highlight the need for longitudinal studies that investigate the relationships between trauma and well-being prospectively over time. Additional limitations include not having a continuous measure of PTSD symptomatology. Research has suggested that PTSD symptoms may mediate the relationship between trauma and poor health.33 Fi-
nally, although the sample size was sufficient to find significant relationships on a number of variables, we were unable to examine the impact of multiple traumas. Experiencing cumulative traumatic events has been shown to be associated with worse outcomes than experiencing a single traumatic event.6 The primary strength of this study is that it is the first to examine trauma exposure in a homebound population of older adults. These findings also have several clinical implications. Existing empirically supported trauma-focused interventions such as exposure therapy, cognitive behavior therapy, and cognitive processing therapy may be appropriate for homebound older adults with distress related to trauma history, if appropriately adapted for this population. Identification of moderators of the relationship between trauma and physical and mental health need to be identified to help detect homebound elders at elevated risk for the effects of trauma for whom such interventions are indicated. Identification of mediators of this relationship is important as isolating mechanisms may sharpen the focus of treatment. Finally, preventive interventions should be explored; trauma-focused interventions delivered earlier in life may prevent, or at least postpone, physical decline and becoming homebound. In conclusion, these findings support the need for further research on the impact of trauma on the health and psychological functioning of homebound older adults.
The authors thank the older adults who participated in this study, the aging service agencies that collaborated with us, and Murray Stein M.D. for his contributions. This study was supported by a University of South Florida New Researcher grant awarded to Amber M. Gum, Ph.D.
References 1. Norris FH: Epidemiology of trauma: frequency and impact of different potentially traumatic events on different demographic groups. J Consult Clin Psychol 1992; 60:409 – 418 2. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders: DSM–IV, Fourth Edition. Washington, DC, American Psychiatric Association, 1994 3. Lincoln KD, Chatters LM, Taylor RJ: Social support, traumatic events, and depressive symptoms among African Americans. J Marriage Fam 2005; 67:754 –766 4. McQuaid JR, Pedrelli P, McCahill ME, et al: Reported trauma, post-traumatic stress disorder and major depression among primary care patients. Psychol Med 2001; 31:1249 –1257
Am J Geriatr Psychiatry 17:9, September 2009
5. Tanskanen A, Hintikka J, Honkalampi K, et al: Impact of multiple traumatic experiences on the persistence of depressive symptoms—a population-based study. Nord J Psychiatry 2004; 58: 459 – 464 6. Turner RJ, Lloyd DA: Lifetime traumas and mental health: the significance of cumulative adversity. J Health Soc Behav 1995; 36:360 –376 7. Spertus IL, Yehuda R, Wong CM, et al: Childhood emotional abuse and neglect as predictors of psychological and physical symptoms in women presenting to a primary care practice. Child Abuse Negl 2003; 27:1247–1258 8. Krause N, Shaw BA, Cairney J: A descriptive epidemiology of
817
Trauma Exposure and Homebound Elders lifetime trauma and the physical health status of older adults. Psychol Aging 2004; 19:637– 648 9. Krause N: Lifetime trauma, emotional support, and life satisfaction among older adults. Gerontologist 2004; 44:615– 623 10. Stein MB, Lang AJ, Laffaye C, et al: Relationship of sexual assault history to somatic symptoms and health anxiety in women. Gen Hosp Psychiatry 2004; 26:178 –183 11. Jamieson DJ, Steege JF: The association of sexual abuse with pelvic pain complaints in a primary care population. Am J Obstet Gynecol 1997; 177:1408 –1412 12. McCauley J, Kern DE, Kolodner K, et al: Clinical characteristics of women with a history of childhood abuse: unhealed wounds. JAMA 1997; 277:1362–1368 13. Norman SB, Means-Christensen AJ, Craske MG, et al: Associations between psychological trauma and physical illness in primary care. J Trauma Stress 2006; 19:461– 470 14. Rauch SA, Morales KH, Zubritsky C, et al: Posttraumatic stress, depression, and health among older adults in primary care. Am J Geriatr Psychiatry 2006; 14:316 –324 15. Van der Kolk BA, McFarlane AC, Weisæth L: Traumatic Stress: the Effects of Overwhelming Experience on Mind, Body, and Society. New York, Guilford, 1996 16. Grazaiano R: Trauma and aging. J Gerontol Soc Work 2003; 40:3–21 17. Creamers M, Parslow R: Trauma exposure and posttraumatic stress disorder in the elderly: a community prevalence study. Am J Geriatr Psychiatry 2008; 16:853– 856 18. Ganguli M, Fox A, Gilby J, et al: Characteristics of rural homebound older adults: a community-based study. J Am Geriatr Soc 1996; 44:363–370 19. Bruce ML, McAvay GJ, Raue PJ, et al: Major depression in elderly home health care patients. Am J Psychiatry 2002; 159:1367–1374 20. Gum AM, Petkus AJ, McDougal SJ, et al: Behavioral health needs and problem recognition by older adults receiving home-based aging services. Int J Geriatr Psychiatry 2009; 24:400 – 408 21. First MB, Spitzer RL, Miriam G, et al: Structured Clinical Interview for the DSM–IV–TR Axis I Disorders, Research Version, Patient
818
Edition With Psychotic Screen. New York, Biometrics Research, New York State Psychiatric Institute, 2002 22. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 1977; 33:159 –174 23. Green BL: Defining trauma: terminology and generic stressor dimensions. J Appl Soc Psychol 1990; 20:1632–1642 24. Teng EL, Chui HC: Manual for the Administration and Scoring of the Modified Mini-Mental State (3MS) Test. Los Angeles, CA, University of Southern California School of Medicine, 1991 25. Folstein MF, Folstein SE, McHugh PR: ‘Mini-Mental State’: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189 –198 26. Derogatis L: Brief Symptom Inventory (BSI) 18: Administration, Scoring, and Procedures Manual. Minneapolis, MN, National Computer Systems, 2000 27. Green BL, Kimerling R, Schnurr PP: Trauma, posttraumatic stress disorder, and health status, in Trauma and Health: Physical Health Consequences of Exposure to Extreme Stress. Edited by Schnurr PP, Green BL. Washington, DC, American Psychological Association, 2004, pp 13– 42 28. Kohn R, Levav I, Sharon A: Psychopathology among holocaust survivors sixty years thereafter. Am J Geriatr Psychiatry 2009; 17:A96 29. Blazer DG: Depression in late life: review and commentary. J Gerontol Series A Biol Sci Med Sci 2003; 58:249 –265 30. Franklin CL, Sheeran T, Zimmerman M: Screening for trauma histories, posttraumatic stress disorder (PTSD), and subthreshold PTSD in psychiatric outpatients. Psychol Assess 2002; 14:467– 471 31. Hepp U, Gamma A, Milos G, et al: Inconsistency in reporting potentially traumatic events. Br J Psychiatry 2006; 188:278 –283 32. Krinsley KE, Gallagher JG, Weathers FW, et al: Consistency of retrospective reporting about exposure to traumatic events. J Trauma Stress 2003; 16:399 – 409 33. Schnurr PP, Green BL: Understanding relationships among trauma, post-traumatic stress disorder, and health outcomes. Adv Mind Body Med 2004; 20:18 –29
Am J Geriatr Psychiatry 17:9, September 2009
Objectives: To examine the physical health, anxiety, and depressive symptoms of homebound older adults in relation to a history of trauma exposure. Design: Secondary analysis of cross-sectional data. Setting: Participants were recruited by in-home aging services case managers. Participants: The sample consisted of 136 older adults receiving in-home services. Measurements: Participants completed a research battery that included the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID), Modified Mini-Mental Status Examination, and Brief Symptom Inventory-18 (BSI-18). Case managers provided standard assessments of health and functional status. Data were analyzed using t tests, 2 tests, and logistic regression. Results: Approximately half (49%) of the participants reported experiencing a traumatic event. When compared with participants who did not report a history of trauma, trauma-exposed participants were on average 3 years younger, had one more chronic medical condition, were taking two more nonpsychotropic medications, and were more likely to report poor self-rated health. Furthermore, trauma-exposed participants were more likely to experience clinically significant somatic and anxiety symptoms on the BSI-18 and were more likely to meet criteria for a SCID anxiety disorder diagnosis. After controlling for physical health and cognitive functioning, trauma exposure was still associated with psychiatric disorders and higher levels of somatic and anxiety symptoms. Conclusions: Trauma-exposed participants were younger but experienced worse physical and mental health than those without a trauma history. This suggests that trauma exposure across the lifespan could have a long-term negative impact on homebound older adults’ health and quality of life. (Am J Geriatr Psychiatry 2009; 17:810 –818) Key Words: Homebound elderly, trauma, health, anxiety, depression
Received November 24, 2008; revised April 3, 2009; accepted May 7, 2009. From the San Diego State University/University of California, San Diego Joint Doctoral Program in Clinical Psychology, San Diego, CA (AJP); Department of Aging and Mental Health, Louis de la Parte Florida Mental Health Institute, University of South Florida, Tampa, FL (AMG); Department of Medical Psychology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands (BK-K); and Department of Psychiatry, University of California, San Diego, CA (JLW). Send correspondence and reprint requests to Andrew J. Petkus, M.A., 3350 La Jolla Village Drive, Building 13, Room 118, San Diego, CA 92161. e-mail: [email protected] © 2009 American Association for Geriatric Psychiatry
810
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al.
A
pproximately two-thirds of individuals in the United States will experience at least one traumatic event in their lifetime.1 A traumatic event is an event in which one is exposed to actual or threatened death, serious injury, or threat to the physical integrity of self or others.2 Exposure to a traumatic event can lead to psychological difficulties, the most prototypical of which is posttraumatic stress disorder (PTSD). The PTSD is characterized by fear, reexperiencing the traumatic event, avoidance, emotional numbing, and increased arousal.2 Studies have also shown a relationship between trauma and other outcomes such as depression3– 6 and other anxiety disorders.7 Research also suggests that the impact of trauma may be long lasting; exposure to childhood trauma8 and to traumatic event at any time during the lifespan6,9 have been associated with poorer mental health in older adulthood. A substantial body of literature documents the association between trauma and physical health. Women with a history of sexual assault,10,11 physical or sexual abuse,12 or childhood emotional abuse7 report more somatic complaints than women with no trauma history. In primary care samples, male trauma survivors have increased risk of diabetes and arthritis, whereas female survivors are at greater risk for cancer and digestive disorders.13 These adverse health effects may persist into older adulthood. Older adults who had experienced any traumatic event throughout their lives have reported poorer self-rated health as well as increased disability and more chronic health conditions.8,14 Furthermore, PTSD and depressive symptoms later in life may further lead to negative health perceptions in the elderly.14 Traumas can have long-lasting effects on how individuals perceive themselves, their coping abilities, and the world.15 Although older adults may have experienced traumatic events years or even decades ago and have coped successfully in midlife, age-related declines in physical and cognitive functioning may re-evoke distressing reactions to past traumas in old age.16 One study found that 10% of older adults experience flashbacks, nightmares, or intrusive memories of traumatic events that occurred decades ago.17 Moreover, new stressful events may amplify these symptoms, putting individuals facing current stressors at risk for experiencing sequelae from past traumatic events.16 Studies suggest that 10%–30% of older adults are facing substantial disability and chronic illness that
Am J Geriatr Psychiatry 17:9, September 2009
render them homebound.18,19 This population is at high risk for depression and other forms of psychological distress19 as well as for nursing home placement and institutionalized care. Research with homebound elders, therefore, is important because improved understanding of this population may lead to more effective interventions and strategies to postpone institutionalization. Because of their current significant stressors like chronic illness, disability, and cognitive decline, homebound elders represent one group of older adults who may be particularly vulnerable to the long-term effects of prior traumas. Because of the high prevalence of disability and medical conditions in this population, one might speculate that the physical health and cognitive functioning of homebound elders may be more relevant to these elders’ emotional functioning than traumas that occurred years or decades ago. Because of the chronic effects of trauma, however, an alternate hypothesis is that past traumas may still negatively impact the psychological experiences of homebound older adults independently of current health-related stressors. Despite the potential importance of trauma in a population at risk for adverse health consequences and need for institutional care, to our knowledge, no research has examined relationships of psychological trauma with health outcomes among homebound older adults. The purpose of this study was to investigate the relationship between history of trauma exposure and the physical health and psychological distress of homebound elders. It was hypothesized that participants who reported exposure to a trauma at any point in life would be in worse physical health and exhibit higher levels of distress than participants who were not exposed to a trauma. It was further hypothesized that trauma exposure would be associated with psychological distress even after controlling for physical health and cognitive functioning.
METHODS Data Source and Sample Data for these analyses were drawn from a larger project, investigating the mental health needs of
811
Trauma Exposure and Homebound Elders homebound older adults. A brief summary of the study design, sampling procedures, measures, and other procedures are described later; a more detailed description appears in another article.20 A sample of 142 older adults receiving in-home social services through aging service agencies provided data through in-person interviews; data were also extracted from standardized assessments administered by the case managers. Individuals were eligible to participate if they were aged 60 years or older, spoke English, and did not have a known diagnosis of dementia per the case manager. In total, 231 individuals agreed to be contacted by the research team and 142 individuals completed the interview, resulting in a response rate of 61.5%. Of those who did not complete the interview, 70 refused primarily because of health problems, 11 were unable to participate because of cognitive impairment, 6 were unable to be contacted after numerous attempts, and 2 died before scheduling interviews. For the questions related to trauma history, four participants refused and data for two participants were missing, resulting in a final sample of N ⫽ 136.
Measures Diagnosis of Depressive and Anxiety Disorders. The Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID)21 was administered to assess for current depressive and anxiety disorders. The following disorders were assessed: major depressive disorder, dysthymic disorder, depressive disorder not otherwise specified, depressive disorder because of medical condition or substance, panic disorder, agoraphobia, specific phobia, social phobia, obsessive-compulsive disorder, PTSD, generalized anxiety disorder, anxiety disorder not otherwise specified, and anxiety because of general medical condition or substance. All interviewers had at least a bachelor’s degree and were trained and supervised by a licensed clinical psychologist (AG). All interviews were audiotaped (unless refused by the participant) and a random 20% of the interviews were rated by AG to assess interrater reliability. Kappas ranged from 0.61 to 1.00 across disorders representing high interrater reliability.22
812
Trauma History. Trauma history was derived from the PTSD section of the anxiety disorders module of the SCID. Participants were asked: “Sometimes things happen to people that are extremely upsetting—things like being in a life-threatening situation like a major disaster, very serious accident, or fire; being physically assaulted or raped; seeing another person killed or dead or badly hurt, or hearing about something horrible that has happened to someone you are close to. At any time during your life, have any of these kinds of things happened to you?” Participants who responded “yes” were asked to briefly describe the nature of each event, the year of each event, and their age at the time of each event. Green’s23 criteria for traumatic events were used to determine what constitutes a traumatic event. Using these criteria, a traumatic event was defined as an event in which the individual experienced one or more of the following: threat to one’s life or bodily integrity; severe physical harm or injury; receipt of intentional injury/harm; exposure to the grotesque; violent/sudden loss of a loved one; witnessing or learning of violence to a loved one; learning of exposure to a noxious agent; or causing death or severe harm to another. The first author and the second author independently coded each event as traumatic or not based on these criteria and interrater reliability was assessed. A kappa value of 0.95 was achieved, representing high interrater reliability.22 In cases of disagreement, the first author and the second author discussed discrepancies and the final decision was reached by consensus. Indicators of Physical Health. Health variables included the following: a) self-rated health; b) number of prescription nonpsychotropic medications; c) number of chronic medical conditions (cancer, diabetes, emphysema, heart problems, liver problems, stroke, Parkinson disease, allergies, asthma, bladder/kidney problems, high blood pressure, gall bladder problems, or thyroid problems); and d) functional impairment as measured by a combined total of number of limitations in activities of daily living (ADL; bathing or showering, dressing, eating, getting in or out of bed or chairs, walking, and using the toilet) and instrumental activities of daily living (IADL; using the telephone, shopping, preparing food, housekeeping, laundry, transportation, taking medications, and handling finances). Self-rated health was assessed by asking participants to rate
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al. their health using a 5-point scale (ranging from “1poor” to “5-excellent”). Number of prescribed medications was measured by asking the participant to provide a list of all medications that were being taken (values for sample ranged from 0 to 20). Number of chronic medical conditions (values for sample ranged from 1 to 15) and ADL and IADL limitations (values for sample ranged from 2 to 14) were derived from the case manager’s assessment nearest in time to the research interview. Modified Mini-Mental Status Examination. The 3MS24 is a measure of cognitive functioning and is a modification of the Mini-Mental Status Examination.25 Scores ranged from 0 to 100, with higher scores representing better functioning. The 3MS has strong validity and reliability.24 Cronbach’s ␣ in this sample was 0.84. Brief Symptom Inventory-18. The Brief Symptom Inventory-18 (BSI-18)26 was administered to measure somatic, depressive, and anxiety symptoms. Each item assesses distress during the past week on a 5-point scale ranging from “0-not at all” to “4-extremely.” The total score ranges from 0 to 72 and the Somatic, Depression, and Anxiety subscale scores range from 0 to 24; higher scores represent more distress. The BSI-18 manual contains normative data for men and women and suggests adopting a t score of 63 as a cut score to represent caseness on each respective subscale. In this sample, Cronbach’s ␣ values were 0.86 depression, 0.81 anxiety, 0.71 somatic, and 0.89 for the total score.
regression model included the following covariates: age 60 –93 years, sex (0 ⫽ man, 1 ⫽ woman), number of chronic medical conditions (range: 1–15), combined number of ADL and IADL limitations (range: 2–14), number of nonpsychotropic medications (range: 0 –20), 3MS score (range: 64 –100), and history of trauma exposure (0 ⫽ absent, 1 ⫽ present).
RESULTS Frequency and Age of Traumatic Events Figure 1 depicts the proportion of participants reporting each category of traumatic event. Nearly half of the sample (N ⫽ 67, 49%) reported experiencing at least one traumatic event. The most common trauma experienced was the sudden unexpected loss of a loved one (N ⫽ 35, 27.7%). Table 1 provides information regarding participants’ age at the time of the traumatic event and number of events experienced. Participants most frequently experienced traumatic events between the ages of 19 and 59 years and were least likely to report a trauma at the age of 60 years or older (N ⫽ 14, 10.3%). Among the participants who reported experiencing a trauma, most reported experiencing only one traumatic event (N ⫽ 40, 58.7%).
Comparisons by Trauma Data Analysis Descriptive statistics for all variables were computed. To assess the first hypothesis, comparisons between those with and without a history of trauma were conducted for all other variables (demographics, depressive and anxiety symptoms, diagnoses, indicators of physical health, and 3MS) using t tests for continuous variables and 2 tests for dichotomous variables. The BSI-18 data were significantly skewed; therefore, these data were dichotomized using the caseness rule of a t score of 63 proposed by the authors of the scale as a cutoff to represent clinically significant distress.26 Furthermore, logistic regression analysis was used to examine relationships between covariates and BSI-18 caseness (yes/no) and presence of SCID disorder (yes/no). For each outcome, the final
Am J Geriatr Psychiatry 17:9, September 2009
Table 1 provides descriptive information comparing those who did (N ⫽ 67) and did not (N ⫽ 69) experience a trauma. On average, participants experiencing a trauma were significantly younger and more likely to report being in poor health. They also had more medical conditions and were taking more prescribed nonpsychotropic medications. A significantly greater proportion of trauma-exposed participants were identified by the BSI-18 as currently exhibiting clinically significant anxiety and somatic symptoms. Trauma-exposed participants were also more likely to meet criteria for an anxiety disorder or any Axis I disorder overall. The two groups did not differ with regard to depressive symptoms, cognitive performance, functional impairment, education, sex, and marital status.
813
Trauma Exposure and Homebound Elders
FIGURE 1.
Frequency of Traumatic Events Reported (N ⴝ 136)
Logistic Regression of Overall Distress, Anxiety, Depressive, and Somatic Symptoms Table 2 displays results from logistic regression, investigating current distress and psychiatric disorder by exposure to a trauma, after controlling for covariates. Across all regression analyses, residuals were examined and one outlier was identified. Analyses were rerun with this participant excluded and the results did not change; therefore, we included the outlier in our analyses. For BSI-18 somatic caseness, the final multivariate logistic regression model was statistically significant, (2(7) ⫽ 31.4, p ⬍0.01), and the model was found to accurately classify 72.8% of participants. In the multivariate model, experiencing a trauma (odds ratio [OR]: 2.23, 95% confidence interval [CI]: 1.01– 4.93, z (1) ⫽ 1.98, p ⬍0.05) and number of chronic medical conditions (OR: 1.36, 95% CI: 1.11–1.67, z (1) ⫽ 2.90, p ⬍0.01) were associated with higher levels of somatic symptoms. For BSI-18 depressive caseness, the multivariate model was not statistically significant (2(7) ⫽ 8.87, p ⫽ 0.26). Furthermore, no individual covariates were significant at the p ⬍0.05 level.
814
For BSI-18 anxiety caseness, the multivariate model was statistically significant (2(7) ⫽ 14.6, p ⬍0.05), and the model accurately classified 84.6% of participants. Trauma exposure (OR: 3.39, 95% CI: 1.02–11.26, z (1) ⫽ 2.00, p ⬍0.05) and poor cognitive performance (OR: 0.92, 95% CI: 0.86 – 0.98, z (1) ⫽ ⫺2.49, p ⬍0.02) were associated with clinically significant anxiety. The multivariate model with the outcome of meeting criteria for any Axis I disorder was statistically significant (2(7) ⫽ 15.5, p ⬍0.03), and the model accurately classified 76.1% of participants. Participants were more likely to meet criteria for an Axis I disorder if they had experienced a trauma (OR: 2.69, 95% CI: 1.05– 6.87, z (1) ⫽ 2.07, p ⬍0.04) or had more chronic medical conditions (OR: 1.26, 95% CI: 1.02–1.54, z (1) ⫽ 2.16, p ⬍0.04). The multivariate model with the outcome of meeting criteria for a depressive disorder was statistically significant (2(7) ⫽ 14.6, p ⬍0.05), and the model accurately classified 88.2% of participants. The number of physical conditions was a significant covariate (OR: 1.32, 95% CI: 1.03–1.69, z (1) ⫽ 2.20, p ⬍0.03), with more medical conditions being associated with greater like-
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al.
TABLE 1.
Sample Characteristics (N ⴝ 136)
Variable Demographics Age, years Education, years Sex Male Female Race White Minority Marital status Married Separated/divorced Never married Widow/widower Age at trauma 0–18 19–59 60⫹ Number of traumas 1 2 3 Mental health Current SCID disorder Any Depressive Anxiety PTSD BSI-18 caseness Somatic Depression Anxiety 3MS Physical health Self-rated health Poor Fair Good Very good/excellent Chronic physical conditions ADL ⫹ IADL limitations Nonpsychotropic medications
No Trauma (N ⴝ 69) N (%) or M (SD)
Trauma (N ⴝ 67) N (%) or M (SD)
t or 2 (df)
p
76.46 (8.23) 12.00 (3.08)
73.24 (8.19) 12.64 (2.34)
2.29 (134) ⫺1.37 (134)
0.024a 0.175
14 (20.3) 55 (79.7)
14 (20.9) 53 (79.1)
0.01 (1)
0.930
50 (72.5) 19 (27.5)
51 (76.1) 16 (23.9)
0.24 (1)
0.626
11 (15.9) 17 (24.6) 6 (8.7) 35 (50.7)
9 (13.4) 26 (38.8) 4 (6.0) 28 (41.8)
3.23 (3)
0.356
6.66 (1) 2.76 (1) 5.77 (1)
0.010b 0.097 0.016a
9.52 (1) 0.09 (1) 5.27 (1) ⫺0.75 (134)
0.002b 0.762 0.022a 0.453
— — —
14 (20.9) 36 (53.7) 14 (20.9)
— — —
40 (59.7) 22 (32.8) 5 (7.4)
9 (13.2) 5 (7.2) 4 (5.9) 0 (0.0)
21 (31.8) 11 (16.4) 13 (19.7) 3 (4.5)
22 (31.9) 12 (17.4) 4 (5.9) 87.52 (9.1)
39 (58.2) 13 (19.4) 14 (20.9) 88.63 (8.2)
17 (24.6) 23 (33.3) 22 (31.9) 7 (9.1) 4.56 (2.40) 7.72 (3.04) 6.26 (3.98)
32 (47.8) 25 (37.3) 9 (13.4) 1 (1.5) 5.61 (2.77) 7.73 (2.86) 8.80 (4.78)
16.10 (4)
⫺2.36 (134) ⫺0.01 (134) ⫺3.38 (134)
0.003b
0.020a 0.989 0.001b
Notes: Missing data: age of trauma: 12 from ages 0 to 18 years, 9 from ages 19 to 59 years, 10 from ages 60⫹. a p ⬍0.05. b p ⱕ0.01.
lihood of having a depressive disorder. Trauma exposure was not significant (OR: 2.05, 95% CI: 0.60 –7.02, z (1) ⫽ 1.14, p ⫽ 0.26). The multivariate model with the outcome of meeting criteria for a SCID anxiety disorder was statistically significant (2(7) ⫽ 16.7, p ⬍0.02), and the model accurately classified 86.6% of participants. Trauma was not significant (OR: 2.94, 95% CI: 0.84 –10.38, z (1) ⫽ 1.68, p ⫽ 0.10); participants were more likely to meet criteria
Am J Geriatr Psychiatry 17:9, September 2009
for an anxiety disorder if they were younger (OR: 0.92, 95% CI: 0.85– 0.99, z (1) ⫽ ⫺2.20, p ⬍0.03).
CONCLUSIONS Exposure to a traumatic event at any point during one’s lifetime was common for this sample of home-
815
Trauma Exposure and Homebound Elders
TABLE 2.
Trauma History as a Covariate of Mental Health Problems After Controlling for Demographics, Health, and Cognitive Performance
BSI-18 caseness Somatic Depression Anxiety Psychiatric disorder Any Depression Anxiety
Odds Ratio
Standard Error
95% Confidence Interval
2.23 0.90 3.39
0.90 0.45 2.08
1.01 0.34 1.02
2.69 2.05 2.95
1.29 1.29 1.89
1.05 0.60 0.84
z (df)
p
4.93 2.37 11.26
1.98 (1) (⫺1) 2.00 (1)
0.049a 0.836 0.046a
6.87 7.02 10.38
2.08 (1) 1.14 (1) 1.68 (1)
0.039a 0.255 0.092
Notes: The Wald test was performed for all hypothesis tests of individual odds ratio parameters above. a p ⬍0.05.
bound older adults, with 49% of the sample reporting a trauma history. This proportion is less than the prevalence of trauma exposure (68%) found in epidemiologic samples of older adults;1 whether this represents a difference between homebound and active elderly or is a byproduct of insufficiently sensitive methods used to detect trauma in this study is unknown. Although they were younger, trauma-exposed participants appeared to be in worse physical health: they had more chronic medical conditions and somatic symptoms, were more likely to report poor health, and were taking more nonpsychotropic medications, when compared with participants who had not experienced a trauma. In addition, traumaexposed participants were more likely to meet criteria for an anxiety or any psychiatric disorder and to report clinically significant levels of anxiety symptoms. After controlling for physical health and cognitive functioning, trauma exposure was still associated with psychiatric disorders and somatic and anxiety symptoms. These findings are largely consistent with the study hypotheses, except that significant relationships were not found for depressive symptoms or when depressive or anxiety psychiatric disorders were examined separately. Trauma exposure was associated with poor health as measured by medical conditions, medications, self-reported health, and somatic symptoms. This finding is consistent with the other studies of trauma in older adults.8 Notably, trauma-exposed participants were in worse health, however, they were on average 3 years younger than participants who had not experienced a trauma. This suggests that trauma exposure may contribute to an accelerated process of physical decline and becoming homebound. Further-
816
more, the relationships of trauma were significant across a variety of health and mental health outcomes despite the fact that most traumas occurred ⬎10 years earlier. This provides additional support for the proposition that the effects of traumas can persist during long periods. Biopsychosocial changes after trauma exposure may explain the relationship between trauma and declining health. Trauma exposure can lead to psychological changes such as PTSD, depression, and increased stress. These psychological changes have been associated with biological changes such as increased activity of the hypothalamic-pituitary adrenal axis and attentional changes such as increased sensitivity to somatic symptoms.27 Thus, trauma exposure earlier in life may produce biopsychosocial changes that in turn may increase allostatic load and lead to poor health.27 Trauma was also associated with psychiatric disorders and anxiety symptoms. These relationships remained present after controlling for physical health and cognitive functioning: in fact, trauma exposure was the strongest covariate for each outcome. The stresses associated with being homebound may evoke feelings of helplessness and not being in control, thus re-evoking the experience of their prior trauma. No statistically significant association between depression and trauma exposure was found. This finding is not consistent with other studies of trauma and depression because other studies have reported that trauma exposure was associated with increased depressive symptoms.9 This finding, however, is consistent with a study documenting higher levels of anxiety, but not depression, in older Holocaust sur-
Am J Geriatr Psychiatry 17:9, September 2009
Petkus et al. vivors.28 Although depressive disorders were more than twice as prevalent among those with a trauma history as among those without a trauma history (16.4% versus 7.2%) in this study, the difference was not statistically significant. Having a depressive disorder was associated with number of chronic physical conditions, consistent with a large body of research.29 The insensitive measurement and retrospective recall of traumatic events are important limitations of this study, which could have resulted in an underestimate of the prevalence of traumas. Checklists of traumatic events may be more reliable than openended questions in determining trauma exposure.30 Inconsistencies in reporting past traumatic events have been documented when using the open-ended question from the SCID PTSD module to assess for past trauma exposure.31 Although some studies report that the retrospective recall of trauma can be reliable,32 participants who are suffering from physical illness or psychological distress may potentially exhibit a recall bias for traumatic events. Both insensitive measurement of past trauma and potential recall biases may have synergistically contributed to the relatively low frequency of trauma exposure found in this sample and affected our results. Thus, an alternate explanation of our findings is that participants in worse physical health and experiencing more anxiety were potentially more likely to recall traumatic events than those who were doing well. These limitations highlight the need for longitudinal studies that investigate the relationships between trauma and well-being prospectively over time. Additional limitations include not having a continuous measure of PTSD symptomatology. Research has suggested that PTSD symptoms may mediate the relationship between trauma and poor health.33 Fi-
nally, although the sample size was sufficient to find significant relationships on a number of variables, we were unable to examine the impact of multiple traumas. Experiencing cumulative traumatic events has been shown to be associated with worse outcomes than experiencing a single traumatic event.6 The primary strength of this study is that it is the first to examine trauma exposure in a homebound population of older adults. These findings also have several clinical implications. Existing empirically supported trauma-focused interventions such as exposure therapy, cognitive behavior therapy, and cognitive processing therapy may be appropriate for homebound older adults with distress related to trauma history, if appropriately adapted for this population. Identification of moderators of the relationship between trauma and physical and mental health need to be identified to help detect homebound elders at elevated risk for the effects of trauma for whom such interventions are indicated. Identification of mediators of this relationship is important as isolating mechanisms may sharpen the focus of treatment. Finally, preventive interventions should be explored; trauma-focused interventions delivered earlier in life may prevent, or at least postpone, physical decline and becoming homebound. In conclusion, these findings support the need for further research on the impact of trauma on the health and psychological functioning of homebound older adults.
The authors thank the older adults who participated in this study, the aging service agencies that collaborated with us, and Murray Stein M.D. for his contributions. This study was supported by a University of South Florida New Researcher grant awarded to Amber M. Gum, Ph.D.
References 1. Norris FH: Epidemiology of trauma: frequency and impact of different potentially traumatic events on different demographic groups. J Consult Clin Psychol 1992; 60:409 – 418 2. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders: DSM–IV, Fourth Edition. Washington, DC, American Psychiatric Association, 1994 3. Lincoln KD, Chatters LM, Taylor RJ: Social support, traumatic events, and depressive symptoms among African Americans. J Marriage Fam 2005; 67:754 –766 4. McQuaid JR, Pedrelli P, McCahill ME, et al: Reported trauma, post-traumatic stress disorder and major depression among primary care patients. Psychol Med 2001; 31:1249 –1257
Am J Geriatr Psychiatry 17:9, September 2009
5. Tanskanen A, Hintikka J, Honkalampi K, et al: Impact of multiple traumatic experiences on the persistence of depressive symptoms—a population-based study. Nord J Psychiatry 2004; 58: 459 – 464 6. Turner RJ, Lloyd DA: Lifetime traumas and mental health: the significance of cumulative adversity. J Health Soc Behav 1995; 36:360 –376 7. Spertus IL, Yehuda R, Wong CM, et al: Childhood emotional abuse and neglect as predictors of psychological and physical symptoms in women presenting to a primary care practice. Child Abuse Negl 2003; 27:1247–1258 8. Krause N, Shaw BA, Cairney J: A descriptive epidemiology of
817
Trauma Exposure and Homebound Elders lifetime trauma and the physical health status of older adults. Psychol Aging 2004; 19:637– 648 9. Krause N: Lifetime trauma, emotional support, and life satisfaction among older adults. Gerontologist 2004; 44:615– 623 10. Stein MB, Lang AJ, Laffaye C, et al: Relationship of sexual assault history to somatic symptoms and health anxiety in women. Gen Hosp Psychiatry 2004; 26:178 –183 11. Jamieson DJ, Steege JF: The association of sexual abuse with pelvic pain complaints in a primary care population. Am J Obstet Gynecol 1997; 177:1408 –1412 12. McCauley J, Kern DE, Kolodner K, et al: Clinical characteristics of women with a history of childhood abuse: unhealed wounds. JAMA 1997; 277:1362–1368 13. Norman SB, Means-Christensen AJ, Craske MG, et al: Associations between psychological trauma and physical illness in primary care. J Trauma Stress 2006; 19:461– 470 14. Rauch SA, Morales KH, Zubritsky C, et al: Posttraumatic stress, depression, and health among older adults in primary care. Am J Geriatr Psychiatry 2006; 14:316 –324 15. Van der Kolk BA, McFarlane AC, Weisæth L: Traumatic Stress: the Effects of Overwhelming Experience on Mind, Body, and Society. New York, Guilford, 1996 16. Grazaiano R: Trauma and aging. J Gerontol Soc Work 2003; 40:3–21 17. Creamers M, Parslow R: Trauma exposure and posttraumatic stress disorder in the elderly: a community prevalence study. Am J Geriatr Psychiatry 2008; 16:853– 856 18. Ganguli M, Fox A, Gilby J, et al: Characteristics of rural homebound older adults: a community-based study. J Am Geriatr Soc 1996; 44:363–370 19. Bruce ML, McAvay GJ, Raue PJ, et al: Major depression in elderly home health care patients. Am J Psychiatry 2002; 159:1367–1374 20. Gum AM, Petkus AJ, McDougal SJ, et al: Behavioral health needs and problem recognition by older adults receiving home-based aging services. Int J Geriatr Psychiatry 2009; 24:400 – 408 21. First MB, Spitzer RL, Miriam G, et al: Structured Clinical Interview for the DSM–IV–TR Axis I Disorders, Research Version, Patient
818
Edition With Psychotic Screen. New York, Biometrics Research, New York State Psychiatric Institute, 2002 22. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 1977; 33:159 –174 23. Green BL: Defining trauma: terminology and generic stressor dimensions. J Appl Soc Psychol 1990; 20:1632–1642 24. Teng EL, Chui HC: Manual for the Administration and Scoring of the Modified Mini-Mental State (3MS) Test. Los Angeles, CA, University of Southern California School of Medicine, 1991 25. Folstein MF, Folstein SE, McHugh PR: ‘Mini-Mental State’: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189 –198 26. Derogatis L: Brief Symptom Inventory (BSI) 18: Administration, Scoring, and Procedures Manual. Minneapolis, MN, National Computer Systems, 2000 27. Green BL, Kimerling R, Schnurr PP: Trauma, posttraumatic stress disorder, and health status, in Trauma and Health: Physical Health Consequences of Exposure to Extreme Stress. Edited by Schnurr PP, Green BL. Washington, DC, American Psychological Association, 2004, pp 13– 42 28. Kohn R, Levav I, Sharon A: Psychopathology among holocaust survivors sixty years thereafter. Am J Geriatr Psychiatry 2009; 17:A96 29. Blazer DG: Depression in late life: review and commentary. J Gerontol Series A Biol Sci Med Sci 2003; 58:249 –265 30. Franklin CL, Sheeran T, Zimmerman M: Screening for trauma histories, posttraumatic stress disorder (PTSD), and subthreshold PTSD in psychiatric outpatients. Psychol Assess 2002; 14:467– 471 31. Hepp U, Gamma A, Milos G, et al: Inconsistency in reporting potentially traumatic events. Br J Psychiatry 2006; 188:278 –283 32. Krinsley KE, Gallagher JG, Weathers FW, et al: Consistency of retrospective reporting about exposure to traumatic events. J Trauma Stress 2003; 16:399 – 409 33. Schnurr PP, Green BL: Understanding relationships among trauma, post-traumatic stress disorder, and health outcomes. Adv Mind Body Med 2004; 20:18 –29
Am J Geriatr Psychiatry 17:9, September 2009