3.4 Affective Disorders
Week 2
Depression Part 2
,Sex Differences in Depression
Kendler & Gardner (2014)
Sex Differences in the Pathways to Major Depression: A Study of
Opposite-Sex Twin Pairs
Introduction
● Women have a consistently higher rate of MDD than men
● Etiological pathways for this finding have been explored in previous research (single
risk factors)
○ Marital status/quality
○ Stressful life events
○ Prior anxiety disorders
○ Personality
○ Ruminative propensity
● Because of the genetic and environmental familial factors in MDD, delineating risk
factors that differentiate sexes would be facilitated if there was a design controlling
for sex differences
○ This study → co-twin control design in opposite-sex dizygotic twin pairs (optimal
sample for studying sex differences)
Method
● Sample
○ Started with 9.400 individuals that were twins from Virginia (1940-1974)
○ Final interview 1.000 participants at around 37 years of age
● Outcome variable
○ Participants were asked about the occurrence of 15 DSM-III symptoms for an
MDD in the past year
● Model variables
○ Five developmental periods
■ Childhood → familial risk, low parental warmth, sexual abuse, parental
loss
■ Early adolescence → neuroticism, low self-esteem, early-onset anxiety,
conduct disorder
■ Late adolescence → low educational achievement, lifetime traumas,
nicotine dependence, alcohol and drug use disorders
■ Adulthood → divorce, past history of MDD, low social support
■ Past year → marital satisfaction, distal stressful life events, dependent
proximal and independent proximal stressful life events
● Statistical methods
, ○ Divided into 2 groups so that one member of each pair was present in each
group
○ Genetic and environmental influences were balanced across groups
Results
● General
○ 837/1057 participants had no episodes of MDD in the past year
○ In 12 pairs both members had MDD in the past year
○ Of the 208 with MDD in the past year 62% was female
○ Paths were estimated using statistical software (best fit model)
○ Three levels of analysis on sex differences in individual paths, all outflow paths
from risk variables, total effect of risk variables on liability to MDD
● Individual paths
○ Some individual paths had strong sex differences
○ Stronger in males
■ Childhood sexual abuse to conduct disorder and early onset anxiety
disorder (0.41 vs 0.12)
■ Drug use disorders to distal and dependent proximal stressful life events
(0.16 vs 0)
■ Dependent proximal life events to past year MDD (0.37 vs 0.24)
○ Stronger in females
■ Low parental warmth to early onset anxiety disorders and prior history of
MDD (0.07 vs 0)
■ Low marital satisfaction and social support to past year MDD (0.20 vs 0)
● Risk factors → outflow
○ Low parental warmth, parental loss, lifetime traumas, neuroticism, divorce, social
support, marital satisfaction contribute more strongly to the MDD pathway for
females
○ Low self-esteem, drug use disorder, past history of MDD, distal and dependent
proximal stressful life events contribute more strongly to MDD pathway in
males
● Risk factors → total direct & indirect paths to MDD
○ Compared risk factor severity of sex differences → minimal (<0.02), modest,
moderate, strong sex differences
○ Three variables with modest differences (0.02-0.05)
■ Stronger effect on females (1) → parental warmth
■ Stronger effect on males (2) → childhood sexual abuse & past history of
MDD
○ Four variables with moderate differences (0.05-0.10)
■ Stronger effect on females (2) → neuroticism & divorce
■ Stronger effect on males (2) → conduct disorder & drug use disorder
○ Four variables with strong differences (>0.10)
■ Stronger effect on females (2) → social support & marital satisfaction
, ■ Stronger effect on males (2) → distal & dependent proximal stressful life
events
● Specific classes of stressful life events
○ Three event categories had largest differences in effect size
■ For males most effects → financial problems, work problems, legal
problems
○ For females most effects → relationship problems, serious illnesses in close
social network
Discussion
● Findings of risk factors are consistent with other findings, especially neuroticism
● Support for women being more affected by problems in social network
● Six men risk factors this study found can be divided into 3 (previously established by
research) groups
○ Externalizing psychopathology → drug abuse
○ Prior depressive history → history of MDD
○ Greater sensitivity to specific stressors → childhood sexual abuse, stressful
life events
● Men more likely to be more emotionally involved in occupational & financial
success
● Some findings not consistent with prior research (eg divorce more effects on women)
● These findings not directly comparable to previous studies → complex model
● Findings are congruent with psychoanalytic model by Blatt → cognitive-behavioral,
attachment & interpersonal personal perspective
○ MDD has two forms → anaclictic (deficiencies in caring relationships & unmet
dependency needs) & introjective (inability to meet internal demands for
self-worth & achievement)
○ Males more likely to suffer from introjective & females more likely to suffer from
anaclictic
● Limitations
○ Assumption of causal relationships between predictors & dependent variables
(these variables could have all different types of relationships, probably
reciprocal)
○ Some risk factors were assessed using long term memory and may have been
influenced by recall bias
○ Assumption that multiple independent variables act additively & linearly in
their impact on risk for MDD → unlikely to be true
○ Sample was limited to adult white twins from Virginia
Week 2
Depression Part 2
,Sex Differences in Depression
Kendler & Gardner (2014)
Sex Differences in the Pathways to Major Depression: A Study of
Opposite-Sex Twin Pairs
Introduction
● Women have a consistently higher rate of MDD than men
● Etiological pathways for this finding have been explored in previous research (single
risk factors)
○ Marital status/quality
○ Stressful life events
○ Prior anxiety disorders
○ Personality
○ Ruminative propensity
● Because of the genetic and environmental familial factors in MDD, delineating risk
factors that differentiate sexes would be facilitated if there was a design controlling
for sex differences
○ This study → co-twin control design in opposite-sex dizygotic twin pairs (optimal
sample for studying sex differences)
Method
● Sample
○ Started with 9.400 individuals that were twins from Virginia (1940-1974)
○ Final interview 1.000 participants at around 37 years of age
● Outcome variable
○ Participants were asked about the occurrence of 15 DSM-III symptoms for an
MDD in the past year
● Model variables
○ Five developmental periods
■ Childhood → familial risk, low parental warmth, sexual abuse, parental
loss
■ Early adolescence → neuroticism, low self-esteem, early-onset anxiety,
conduct disorder
■ Late adolescence → low educational achievement, lifetime traumas,
nicotine dependence, alcohol and drug use disorders
■ Adulthood → divorce, past history of MDD, low social support
■ Past year → marital satisfaction, distal stressful life events, dependent
proximal and independent proximal stressful life events
● Statistical methods
, ○ Divided into 2 groups so that one member of each pair was present in each
group
○ Genetic and environmental influences were balanced across groups
Results
● General
○ 837/1057 participants had no episodes of MDD in the past year
○ In 12 pairs both members had MDD in the past year
○ Of the 208 with MDD in the past year 62% was female
○ Paths were estimated using statistical software (best fit model)
○ Three levels of analysis on sex differences in individual paths, all outflow paths
from risk variables, total effect of risk variables on liability to MDD
● Individual paths
○ Some individual paths had strong sex differences
○ Stronger in males
■ Childhood sexual abuse to conduct disorder and early onset anxiety
disorder (0.41 vs 0.12)
■ Drug use disorders to distal and dependent proximal stressful life events
(0.16 vs 0)
■ Dependent proximal life events to past year MDD (0.37 vs 0.24)
○ Stronger in females
■ Low parental warmth to early onset anxiety disorders and prior history of
MDD (0.07 vs 0)
■ Low marital satisfaction and social support to past year MDD (0.20 vs 0)
● Risk factors → outflow
○ Low parental warmth, parental loss, lifetime traumas, neuroticism, divorce, social
support, marital satisfaction contribute more strongly to the MDD pathway for
females
○ Low self-esteem, drug use disorder, past history of MDD, distal and dependent
proximal stressful life events contribute more strongly to MDD pathway in
males
● Risk factors → total direct & indirect paths to MDD
○ Compared risk factor severity of sex differences → minimal (<0.02), modest,
moderate, strong sex differences
○ Three variables with modest differences (0.02-0.05)
■ Stronger effect on females (1) → parental warmth
■ Stronger effect on males (2) → childhood sexual abuse & past history of
MDD
○ Four variables with moderate differences (0.05-0.10)
■ Stronger effect on females (2) → neuroticism & divorce
■ Stronger effect on males (2) → conduct disorder & drug use disorder
○ Four variables with strong differences (>0.10)
■ Stronger effect on females (2) → social support & marital satisfaction
, ■ Stronger effect on males (2) → distal & dependent proximal stressful life
events
● Specific classes of stressful life events
○ Three event categories had largest differences in effect size
■ For males most effects → financial problems, work problems, legal
problems
○ For females most effects → relationship problems, serious illnesses in close
social network
Discussion
● Findings of risk factors are consistent with other findings, especially neuroticism
● Support for women being more affected by problems in social network
● Six men risk factors this study found can be divided into 3 (previously established by
research) groups
○ Externalizing psychopathology → drug abuse
○ Prior depressive history → history of MDD
○ Greater sensitivity to specific stressors → childhood sexual abuse, stressful
life events
● Men more likely to be more emotionally involved in occupational & financial
success
● Some findings not consistent with prior research (eg divorce more effects on women)
● These findings not directly comparable to previous studies → complex model
● Findings are congruent with psychoanalytic model by Blatt → cognitive-behavioral,
attachment & interpersonal personal perspective
○ MDD has two forms → anaclictic (deficiencies in caring relationships & unmet
dependency needs) & introjective (inability to meet internal demands for
self-worth & achievement)
○ Males more likely to suffer from introjective & females more likely to suffer from
anaclictic
● Limitations
○ Assumption of causal relationships between predictors & dependent variables
(these variables could have all different types of relationships, probably
reciprocal)
○ Some risk factors were assessed using long term memory and may have been
influenced by recall bias
○ Assumption that multiple independent variables act additively & linearly in
their impact on risk for MDD → unlikely to be true
○ Sample was limited to adult white twins from Virginia
