Impact of Stressful Life Events on Alcohol Relapse


Maria E. Pagano, PhD

Division of Child Psychiatry, Department of Psychiatry, Case Western Reserve University School of Medicine ( CWRU SOM), Cleveland, OH


Christina M. Delos Reyes, MD

Department of Psychiatry, CWRU SOM, Cleveland, OH



I.        Introduction (slides 3-7)

a.       Stressful life events (SLE) and Relapse (slide 3)

                                                               i.      SLE → relapse1,2,3

                                                             ii.      SLE/relapse link stronger for:

1.      Personality disorder (PD) subtypes

2.      Early versus adult onset alcohol use disorder (AUD)

3.      Particular types of SLE


b.      Prior Research (Slides 4-7)

                                                               i.      Cole et al. 19904,8 (slide 4)

1.      ~ 6,700 male employees

2.      Cumulative SLE → more drinks

                                                             ii.      Cooper et al. 19925,8 (slide 5)

1.      SLE-alcohol link stronger for:

a.       Males

b.      High expectancy of alcohol effects

c.       Avoidant coping with emotions

                                                            iii.      Droomers et al. 19996,8 (slide 6)

1.      Epidemiologic study of ~ 1800 adults

2.      No link between cumulative SLE and number of drinks

                                                           iv.      Methodological limitations of prior research (slide 7)

1.      Retrospective designs

2.      Follow-up periods brief

3.      Alcoholics sampled were treatment seeking (non-random)

4.      Poor measurement of SLE


II.     Collaborative Longitudinal Personality Disorders Study (CLPS) (slides 8-24)

a.       Study aims (slide 8)

                                                               i.      To explore SLE/relapse link

1.      SLE typology explored

a.       Cumulative

b.      Type

c.       Timing

2.      Effects among patient subgroups

a.       PD subtype

b.      AUD chronicity (age of onset)


b.      CLPS overview (slide 10)

                                                               i.      4 sites: Brown, Columbia, Harvard, Yale

                                                             ii.      Sample: 573 PD subjects

1.      86 STPD

2.      175 BPD

3.      158 AVPD

4.      154 OCPD

                                                            iii.      Aim:  to determine the stability and predictors of PD clinical course


c.       Methods: Participants (slide 11)

                                                               i.      32.8 years

                                                             ii.      64% female

                                                            iii.      39% self-referred

                                                           iv.      76% Caucasian

                                                             v.      13 years of education


d.      Methods: Baseline Evaluation (slide 12)

                                                               i.      Rater-administered interview

                                                             ii.      Structured Clinical Interview for DSM-IV Axis 1 Disorders (SCID-I/P)

                                                            iii.      AUD status: history, current

                                                           iv.      Diagnostic Interview for DSM-IV Personality Disorders (DIPD-IV)


e.       Methods: Follow-up (slide 13)

                                                               i.      Follow-up interviews: 6 months, yearly thereafter for 6 years

                                                             ii.      Follow-up Measures:

1.      Longitudinal Interval Follow-Up Evaluation (LIFE)

2.      Axis 1 symptoms tracked weekly

3.      Life Events Assessment (LEA)

4.      Start/end dates of SLE tracked


f.        Methods: Interrater Reliability (slide 14)

                                                               i.      SCID-I/P interrater reliability

1.      Axis I disorders

2.      κ coefficients: 0.57-1.0

                                                             ii.      DIPD interrater reliability

1.      Axis II disorders

2.      κ coefficients: 0.68-0.73

                                                            iii.      Test-retest reliability:  k coefficients 0.69-0.74


g.       Methods: Relapse Defined (slide 15)

                                                               i.      Relapse observed prospectively 

                                                             ii.      Axis 1 symptoms tracked weekly

1.      AUD symptoms meet full DSM-IV criteria for AUD for 2+ weeks

a.       Adults with no AUD history: “new onset” 

b.      Adults with AUD history: “relapse”


h.       Methods: LEA7 (slide 16)

                                                               i.      59 SLE “negative”

                                                             ii.      23 SLE “positive”

                                                            iii.      SLE assessed across 6 domains

1.      Work/school (e.g. Laid off)

2.      Family/living matters (e.g. Miscarriage or still birth)

3.      Love relations (e.g. Spouse/mate died)

4.      Crime/legal matters (e.g. Burglarized)

5.      Financial matters (e.g. Suffered financial loss not related to work)

6.      Health (e.g. Serious injury occurred/worsened)


i.         Methods: Data Analyses (slide 17)

                                                               i.      Analysis of variance, Χ2 tests

1.      Demographic characteristics among subgroups

2.      SLE levels among subgroups

                                                             ii.      Event History Analyses

1.      Kaplan-Meier: time to relapse subgroup comparisons

2.      Cox regressions: time-varying SLE/relapse link examined


j.        Results: Rate SLE Endorsed Over 6 Years (slide 18)

                                                               i.      High prevalence of romance, family SLE

                                                             ii.      Low prevalence of finance, crime SLE

                                                            iii.      90% had 1+ positive SLE and 1+ negative SLE


k.      Table: Predictors of Relapse Over 6 Years (slide 19)

                                                               i.      Negative SLE → relapse overall

                                                             ii.      No positive SLE/relapse link overall

                                                            iii.      Predictors of relapse:

1.      AUD history (H+): 3-fold increase in relapse risk

2.      ASPD+: 2-fold increase in relapse risk

3.      OCPD+: 50% decrease in relapse risk


l.          SLE Impact on Relapse By AUD History (slide 20)

                                                               i.      AUD history (H+):  positive/finance SLE  → relapse

                                                             ii.      No  AUD history (H-): romance SLE  → relapse


m.     SLE Impact on Relapse By OCPD/AUD status (slide 21)

                                                               i.       OCPD+, AUD (H-): romance SLE  → relapse

                                                             ii.      OCPD-, AUD (H-): no SLE/relapse link

                                                            iii.       OCPD+, AUD (H+): no SLE/relapse link

                                                           iv.       OCPD-, AUD (H+): positive/negative/romance/finance SLE → relapse


n.       SLE Impact on Relapse By ASPD/AUD status (slide 22)

                                                               i.      ASPD+, H-:

1.      No  ASPD+/H- subjects

2.      Rare co-occurrence worth noting

                                                             ii.      ASPD-, H-: no SLE/relapse link

                                                            iii.      ASPD+, H+:  finance SLE  → relapse

                                                           iv.      ASPD-, H+: no SLE/relapse link


o.      Summary (slide 23)

                                                               i.      Relapse risk: 2-fold increase for ASPD+, 50% decrease for OCPD+

                                                             ii.      Negative SLE → relapse : all subjects

                                                            iii.      Positive SLE →relapse : AUD H-

                                                           iv.      Romance SLE → relapse : OCPD+/H-

                                                             v.      Finance SLE → relapse : ASPD+/H+


p.      Clinical Implications (slide 24)

                                                               i.      Assessment

1.      Consider PD assessment at intake

2.      Consider SLE/relapse link in light of PD subtype

                                                             ii.      Treatment

1.      Tailor SLE/Relapse prevention strategies for PD/AUD subgroups

2.      Improved use of limited resources




1.      Brady KT and Sonne SC. [1999].The role of stress in alcohol use, alcoholism treatment, and relapse. Alcohol Research and Health; Vol 23: 263-271.


2.      Brown SA, Vik PW, McQuaid JR, Patterson TL et al.  [1990]. Severity of psychological stress and outcome of alcoholism treatment. Journal of Abnormal Psychology; Vol 99: 344-348.


3.      Cerbone FG, Larison CL. [2000]. A bibliographic essay: the relationship between stress and substance use.  Substance Use and Misuse; Vol 35: 757-786.


4.      Cole G, Tucker L, Friedman GM.  [1990]. Relationships among measures of alcohol drinking behavior, life-events, and perceived stress. Psychological Reports; Vol 67: 587-591.


5.      Cooper ML, Russell M, Skinner JB et al.  [1992]. Stress and alcohol use: moderating effects of gender, coping, and alcohol expectancies. Journal of Abnormal Psychology; Vol 101:139-152.


6.      Droomers M, Schrijvers CT, Stronks K et al.  [1999]. Educational differences in excessive alcohol consumption: the role of psychosocial and material stressors. Preventive Medicine; Vol 29:1-10.


7.      Pagano ME, Skodol AE, Stout RL, Shea MT et al. [2004]. Stressful life events as predictors of functioning: findings from the Collaborative Longitudinal Personality Disorders Study. Acta Psychiatrica Scandinavica; Vol 110: 421-429.


8.      Veenstra MY, Lemmens P, Friesema I et al. [2006]. A literature overview of the relationship between life-events and alcohol use in the general population. Alcohol and Alcoholism; Vol 41: 455-463.