loss, trauma, and resilience - rehpa loss, and trauma_by george bonanno pavi 25... · driven by the...

Loss, Trauma, and Resilience:

From Heterogeneity to Flexibility

George A. Bonanno, Ph.D.

Bad things happen . . .

Frequency of stressors across 4 years of college - weekly on-line survey - 50 events from 4 categories

. . . frequently

Lalande & Bonanno (2011) Psychological Trauma

Driven by the poignancy and cost of these events, both clinical and scientific inquiry has focused primarily on psychological damage 1.Psychopathology 2.Average impact of the event itself

But . . . the picture is incomplete

Two Common Approaches

Bonanno (2004) American Psychologist; Bonanno, Westphal, & Mancini (2011) Annual Review Clinical Psychology

1. Focus on extreme: psychopathology • chronic grief and depression (10%-65%) • Posttraumatic Stress Disorder (PTSD) (5%-90%)

• Limitations – Emphasis on pathology can result in sampling bias

The limits of diagnoses and the problem of averages


1. Focus on extreme: psychopathology • chronic grief and depression (10%-65%) • Posttraumatic Stress Disorder (PTSD) (5%-90%)

• Limitations – Emphasis on pathology exacerbates sampling bias – Uninformative about the underlying distribution

The limits of diagnoses and the problem of averages


PTE

1 year 2 years

Hea

lth

p

atho

logy

PTE

Chronic Psychopathology (PTSD, MDD, etc.)

1 year 2 years

Hea

lth

p

atho

logy

Non-psychopathology?

?

?

?

1. Focus on extreme: psychopathology – chronic grief and depression (10%-65%) – Posttraumatic Stress Disorder (PTSD) (5%-90%)

2. Focus on average: impact of the event • Compare groups exposed vs. non-exposed • Compare across different types of events

The Limits of diagnoses and the problem of averages

Bonanno (2004) AP; Bonanno, Westphal, & Mancini (2011) ARCP

PTE

1 year 2 years

Hea

lth

p

atho

logy

Average response

Chronic Psychopathology

?

?

?

• Averages are potentially misleading – often mistaken for mode – may be driven by extreme groups

• The problem with averages is that they . . . • “average out the interesting effects”

– (Bloembergen & Zewail, 1984, Journal of Physical Chemistry)

• “often do not describe reality at all” – (Banholzer, 2010, Research Technology Management)

• “are misleading” – (Parkinson, 1989, Climate Dynamics; B. F. Skinner, Journal of General

Psychology, 1936; Nissan & Shahmoon, 1993, Journal of Economics and Finance; Gaut et al., 2007, Nature Reviews, Genetics; Boyd, 1984 and Haigh, 1994; Mathematical Gazette; Day, 2008, Business Law Brief; Gregson et al, 1993, Biological Psychiatry . . . . . . . )

• can mask changes in the majority – (Casselli & Cunningham, 2009; Oxford Economic Papers)

• Averages are curious statistics that are meant to make

poor people happy. Averages tell us you will eat at least one chicken a year. But if you can’t afford a chicken, don’t worry; someone else will eat two

– paraphrased from La Statistica by Trilussa*

• "I really don't trust statistics much. A man with his head

in a hot oven and his feet in a freezer has statistically an average body temperature ”

– Charles Bukowski

* Paraphrased from archaic Italian by Antonio Malgaroli

The problem with averages

PTE

1 year 2 years

Hea

lth

p

atho

logy

Average response

Psychopathology

? Super-coper? Pathological?

“resilient” “resiliency” “resilience” in titles of social science journals

freq

uenc

y

A broader approach: 20 years of mapping heterogeneity

• Part I: Identifying trajectories of outcome

�Bonanno et al., (1995) JPSP; Bonanno et al. (2002) JPSP; Bonanno (2004) American Psychologist

Bonanno (2004) American Psychologist

PTE

Chronic 5-30%

1 year 2 years

Hea

lth

p

atho

logy

modal response Recovery 15-25%

Delayed 0-15%?

Resilience 55-65%

A broader approach: Mapping heterogeneity

• part I: Identifying trajectories of outcome – Limitations:

• Unsophisticated (“by hand” or primitive algorithm) • Theory driven • Not sensitive to parameters of heterogeneity

Changing Lives of Older Couples (CLOC): A prospective study

• 1,532 married individuals from Detroit area • 205 lost a spouse during the 7-year course of

the study, – interviewed prior to bereavement (on average 3

years pre-loss), – Interviewed at least twice after bereavement (6 and

18 months post-loss).

Bonanno, Wortman et al. (2002). JPSP

-1

-0.5

0

0.5

1

1.5

2

2.5D

epre

ssio

n (C

ESD

)

Bonanno et al. (2002). JPSP

3 yrs pre-loss

6 mo. post-loss 18 mo. post-loss

Resilience

Depressed-Improved

Recovery

Chronic grief

Chronic depression

9% 17%

11% 12% 45%

-1

-0.5

0

0.5

1

1.5

2

2.5D

epre

ssio

n (C

ESD

)

3 yrs pre-loss


Resilience

Depressed-Improved

Recovery

Chronic grief

Chronic depression

9% 17%

11% 12% 45%


-1

-0.5

0

0.5

1

1.5

2

2.5D

epre

ssio

n (C

ESD

)

3 yrs pre-loss


Resilience

Relief

Recovery

Chronic grief

Chronic depression

9% 17%

11% 12% 45%

56%


Boerner, Wortman, & Bonanno (2005). J Gerontology

Depressed-improved

A broader approach: Mapping heterogeneity

• part I: Identifying trajectories of outcome – Limitations:

• Unsophisticated (“by hand” or primitive algorithm) • Theory driven • Not sensitive to parameters of heterogeneity

• part II: Latent trajectory modeling

Posttraumatic stress

Posttraumatic stress/ grief

Normality and Homogeneity

- assumption that data should be normal - most conventional statistical approaches require normality

Posttraumatic stress Posttraumatic stress/grief

PTSD/CG

resilience

Positive skew with arbitrary cut-offs

Posttraumatic stress Posttraumatic stress/grief

Positive skew with arbitrary cut-offs

?

“Le Petite Prince” Antoine de Saint-Exupéry

15

Heterogeneity: HLM, Latent Class Growth Analysis (LCGA), Latent Growth Mixture Modeling (LGMM)

?

Posttraumatic stress/grief

Heterogeneity Latent Growth Mixture Modeling (LGMM): unique distributions: trajectories with random effects

Freq

uenc

y

Parameters of heterogeneity over time

Clark et al. (2008)

German Panel Data Nationally representative sampling of German Households followed 19 years (1984-2003) N = 16,795 DV = life satisfaction EVENTS: unemployment layoff marriage divorce death of spouse birth of child Clark, Diener et al. (2008) The Economic Journal

Clark et al. (2008) Clark, Diener et al. (2008)

Clark et al. (2008) Mancini, Bonanno, & Clark (2011) J Individual Diffs

bereavement Widowhood (4 latent growth trajectories)

58.7%

21.0%

Years before and after loss

4.4% 15.9%

Clark et al. (2008)

Divorce (3 latent growth trajectories)

Years before and after divorce Mancini, Bonanno, & Clark (2011) J Individual Diffs

9.1% 71.8%

19.1%

PARENTHOOD

PARENTHOOD

Increasing 4.3%

Galatzer-Levy, Murzursky, Mancini, & Bonanno (2011). Journal of Family Psychology

Stable high 84.2%

Decreasing 7.2% Stable low 4.2%

Galatzer-Levy & Bonanno (2011), Soc. Sci. & Med.

�Re-analyzing the CLOC data using LGMM

Traumatic injury (US) • 330 men and women • Single-incident traumatic injury (motor vehicle

crash, fall, gun-shot) • Taken to level 1 trauma center • required emergency surgery • PTSD and depression

– Hospitalization – 1 month – 3 months – 6 months

DeRoon-Cassini, Mancini, Rusch, & Bonanno (2010) Rehabilitation Psychology

DeRoon-Cassini, Mancini, Rusch, & Bonanno (2010), Rehab Psych

21.8%

60.5%

(0-5

1)

Spinal Cord Injury

• 233 SCI patients recruited from spinal cord centers in England, Switzerland, Sweden, Germany, Austria, and Ireland.

• Data collected soon after injury and at 3 months, 12 months, and 24 months

Bonanno, Kennedy, Galatzer-Levy, Lude, & Elfstrom (2012), Rehabilitation Psychology

Bonanno, et al. (2012), Rehab Psych

Depression at hospitalization, 3, 12, and 24 months post-injury

Population norm 3.5 Haug et al (2004)

Bonanno et al., (2012), Rehab Psych

Anxiety at hospitalization, 3, 12, and 24 months post-injury

Population norm = 3.9 Haug et al (2004)

2008 NIU mass shooting (n = 660)

Orcutt, Bonanno, Hanna, Miron (2014) Journal of Traumatic Stress

Pre- 27 days 6 mo. 12 mo. 18 mo. 24 mo. 30 mo.

The Psychological Cost of War

• The Millennium Cohort Study (Tyler Smith et al.) – Prospective, with pre- and post-deployment data – Large pool (Ongoing enrollment targets 140,000;

77,047 enrolled in initial panel, 30% deployed) – Confidential/anonymous

Single deployers

Bonanno et al., 2012, Brit. J Psychiatry

Bonanno & Geraci (2013)

Flexibility During and After Combat Deployment

Prospective study of US army soldiers deployed in Afghanistan

81.5%

7.3%

11.2%

base deployment 6 mo. 12 mo. 18 mo.

Bereavement Bonanno et al. (1995, 1999)

bereavement Bonanno et al. (2002, 2004)

bereavement Galatzer-Levy & Bonanno (2012)

bereavement Mancini et al. (2011)

Terrorist attack (9/11) Bonanno et al. (2005)

Terrorist attack (9/11) Bonanno et al. (2006, 2007)

SARS (bio-disaster) Bonanno et al. (2008)

Traumatic injury deRoon-Cassini et al. (2010)

Breast cancer surgery Lam et al. (2010)

Mass shooting Orcutt et al. (2013)

Job loss Galatzer-Levy et al. (2010)

divorce Mancini et al. (2011)

Birth of a child Galatzer-Levy et al. (2011)

Combat deployment Bonanno et al. (2012)

Spinal cord lesion Bonanno et al. (2012)

Resilient Chronic

Bereavement 53% 14% Bonanno et al. (1995, 1999)

bereavement 56% 17% Bonanno et al. (2002, 2004)

bereavement 66% 14% Galatzer-Levy & Bonanno (2012)

bereavement 59% 21% Mancini et al. (2011)

Terrorist attack (9/11) * 35% 29% Bonanno et al. (2005)

Terrorist attack (9/11) 56% 6% Bonanno et al. (2006, 2007)

SARS (bio-disaster) * 35% 42% Bonanno et al. (2008)

Traumatic injury 61% 21% deRoon-Cassini et al. (2010)

Breast cancer surgery 66% 15% Lam et al. (2010); Burton et al (2014)

Mass shooting 62% 8% Orcutt et al. (2013)

Job loss 69% 4% Galatzer-Levy et al. (2010)

divorce 72% 19% Mancini et al. (2011)

Birth of a child 84% 7% Galatzer-Levy et al. (2011)

Combat deployment 83% 7% Bonanno et al. (2012)

Spinal cord lesion 53% 12% Bonanno et al. (2012)

Resilient Chronic

Resilience and positive adjustment • Absence of symptoms and distress • Subjective well-being and life satisfaction • Level of mental health and functioning

– Less cortisol dysregulation (diurnal profile)(Ong et al., 2011)

• Positive adjustment as rated confidentially by close friends-relatives – Bereaved partners (Bonanno et al., 2005)

– High-exposure survivors of 9/11 (Bonanno et al., 2005).

• Positive emotion and experiences – Positive body image after cancer surgery (Lam et al., 2012)

– Comfort from positive memories of deceased (Bonanno et al., 2004)

A broader approach: Mapping individual differences

• phase I: Identifying trajectories of outcome • phase II: Latent trajectory modeling • phase III: Predictors

Why are most people resilient but others not?

Predictors of resilient outcomes?

Multiple, unique predictors with small effect sizes

Multiple, unique predictors with small effect sizes • Pre-event context

– Demographic factors (older, male, greater education) – Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice, shared cultural norms) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-

efficacy, perceived control, etc). – Genetic disposition (G X E)

• Proximal exposure – Witnessing death, serious injury to others – Objective danger to self – Extent of loss (death, loss of property)

• The aftermath (distal exposure) – distal exposure (loss of economic, personal, or health resources) – Reduced search for meaning, worry, rumination – Reduced ongoing stress – Coping and appraisal – Positive emotion and regulatory flexibility

Bonanno et al., (2010). PSPI; Bonanno et al., (2011), Annual Review of Clinical Psych


– Demographic factors (age, gender, education) – Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-



• The aftermath (distal exposure) – loss of resources (economic, personal, or health) – search for meaning, worry, rumination – ongoing stress – Coping and appraisal – Positive emotion and regulatory flexibility


Representative sample of New Yorkers first 6 months after 9/11 (N = 2752)

Bonanno, Galea et al. (2006, 2007) Psychological Science; JCCP

Representative sample of New Yorkers first 6 months after 9/11 (N = 2752)


– Demographic factors (age, gender, education) – Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-






– Demographic factors (older, male, greater education) – Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-



• The aftermath (distal exposure) – loss of resources (economic, personal, or health) – search for meaning, worry, rumination – ongoing stress – Coping and appraisal – Positive emotion and regulatoryflexiblity


* Resources * – Conservation of Resources (COR) theory (Hobfoll,

1998, 2002)

– Resources help buffer the impact of adversity • Material: income, financial reserves • Work: employment, possibility of livelihood • Energy (health): absence of disease, health insurance,

availability of care • Interpersonal: Social support, affinity groups


– Demographic factors (older, male, greater education) – Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-





Optimism

• Expectation of positive events in the future when there is no evidence to support those expectations (Scheier & Carver, 1985; Carver et al., 2010)

• Trait optimism: – General health and well-being: proactive health-

related coping (Carver et al., 2010, Clinical Psychology Review).

– Neural mechanism: imagining positive future events associated with amygdala and rostral anterior cingulate; greater in trait optimism (Sharot et al. 2007, Nature)

Optimism and depression before and after a heart attack

Galatzer-Levy & Bonanno (2014)

63.8%

17.6%

9.9%

8.7%

increased mortality

Baseline optimism: “In future, likely that I will . . .”

+ “live past 75” + “leave inheritance -“see double-digit inflation

Controlling for income

heart attack


– Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-





Bonanno,Kennedy, Galatzer-Levy, Lunde, & Elfstrom (2012), Rehab Psych

Depression, Anxiety and Spinal Cord Lesion

Population norm 3.5 Haug et al (2004)

Challenge (vs. threat) appraisals (adjectives)

Regulatory Flexibility

• Theories of coping and emotion regulation emphasize their dynamic nature (Cole, Lazarus, Folkman, Gross et al…..)

– Person-situation interaction – “fit” with shifting nature of situational demands

• in practice, we tend toward static categorization of strategies as adaptive (e.g., reappraisal, finding meaning) or maladaptive (e.g., avoidance, suppression)

• “Fallacy of uniform efficacy” (Bonanno & Burton, 2013)

Bonanno, 2012, Memory; Bonanno et al. 2004, Psych Science; Bonanno & Burton, 2013, Perspectives

Costs and benefits in nature

• There is no such thing as a perfect adaptation • Every adaptation in nature has both benefits and

costs • the peacock’s colorful tail

Costs and benefits in nature

• There is no such thing as a perfect adaptation • Every adaptation in nature has both benefits and

cost • the Cheetah’s speed

Context sensitivity Repertoire Feedback

Evaluate demands and opportunities

Select regulatory strategy

Monitor and modify as needed

Maintain strategy

Cease strategy

Adjust strategy

Select a new strategy

stre

ssor

Re-evaluate demands and opportunities

Bonanno & Burton (2013), Perspectives on Psychological Science

What is happening? What do I need to do? What am I able to do? Is it working?

Context sensitivity Repertoire Feedback

Evaluate demands and opportunities

Select regulatory strategy

Monitor and modify as needed

Maintain strategy

Cease strategy

Adjust strategy

Select a new strategy

S

tres

sor

Re-evaluate demands and opportunities

INDIVIDUAL DIFFERENCES

Coifman & Bonanno (2010) JAP, Gupta & Bonanno (2011), JAP; Burton et al., (2012) Depress &Anxiety

Emotion Context Sensitivity • Emotions are functional . . evolved as solutions to

specific threats and opportunities • The functions of emotions are “context bound” (Cole et al.,

1994)

• Match between contextual demands and specific functions of an emotion facilitates adaptation

sadness

match adaptive benefits

• Attention inward • Facilitates reframing • Evokes

sympathy/care from others

Emotion functions

Emotion Context Sensitivity • Emotions are functional . . evolved as solutions to

specific threats and opportunities • The functions of emotions are “context bound” (Cole et al.,

1994)

• Match between contextual demands and the specific functions of an emotion facilitates adaptation

Emotion functions

mismatch maladaptive

anger

• prepares to physically defense the self

• signals to warn off others

Emotion Context Insensitivity and Psychopathology

• Depression (MDD) – Report more sadness overall, but do not vary sadness

across contexts (e.g., sad versus other films) – Context sensitivity in former depressed predicted less

likelihood of relapse

Rottenberg et al., 2002, 2005 , Emotion, JAP

Emotion Context Insensitivity and Psychopathology

• Complicated Grief (CG) – Less context sensitivity across interview and film tasks – Evidence of context sensitivity early in bereavement

predicted reduced depressive symptoms later in bereavement

– Context insensitivity limited primarily to expression

Coifman & Bonanno, 2010, JAP Bullock & Bonanno, 2012, JEP Diminich & Bonanno, 2014, JAP

Expressive Flexibility

1

2

3

4

5

6

7

Enhancement Suppression control

Expressive regulation condition

Affe

ct ra

tings

• View evocative photos on computer, rate emotion • “Person in other room will try to guess your emotion” • Three within-subjects conditions

Bonanno et al., 2004, Psych Science

1

2

3

4

5

6

7

Enhancement Suppression control

expressed emotion (observer ratings)

Expressive regulation condition

Affe

ct ra

tings

Expressive enhancement ability enhancement – no monitor (control)

Suppression ability No monitor (control) -- suppression

expressive flexibility (EF) and adjustment

• NYC freshmen beginning college just days before the September 11th terrorist attack.

9/11

W1 distress

2003 2001 2002

W2 distress

EF task Expression ability * Suppression ability *

Flexibility score*

Difference (polarity) ns

Bonanno et al., 2004, Psych Science

Estim

ated

Mar

gina

l Mea

ns4.5

4.0

3.5

3.0

2.5

2.0

1.5

Estimated Marginal Means of MEASURE_

Married asymptomatic Complicated bereaved grief

Expression condition

Control condition

Suppression condition

Expressive Flexibility and Complicated Grief

Cod

ed e

xpre

ssio

n of

em

otio

n

Gupta & Bonanno (2011) Journal of Abnormal Psychology


– Demographic factors (older, male, greater education) – Preparation and prior exposure – Economic resources (employment, income). – Beliefs (acceptance of death, justice, shared cultural norms) – Social resources (support, social network) – Personality (trait self-enhancement, optimism, hardiness, coping self-



• The aftermath (distal exposure) – distal exposure (loss of economic, personal, or health resources) – Reduced search for meaning, worry, rumination – Reduced ongoing stress – Coping and appraisal – Positive emotion and regulatory flexibility


THANK YOU!

loss, trauma, and resilience - rehpa loss, and trauma_by george bonanno pavi 25... · driven by the...

Documents