Association Between Maternal Depressogenic Cognitive Style During Pregnancy and Offspring Cognitive Style 18 Years Later.

0.55 MB·English

by  Pearson, Rebecca M.

Article
Association Between Maternal Depressogenic Cognitive
Style During Pregnancy and Offspring Cognitive
Style 18 Years Later
Rebecca M. Pearson, Ph.D. Objective: Understanding the origins of offspring’s depressogenic cognitive style
negative cognitive style could provide atage18.
Charles Fernyhough, Ph.D a means to prevent adult depression. Results: A positive association was ob-
Cognitive style is an important target for
served between maternal and offspring
intervention because although it is not
Richard Bentall, Ph.D. cognitive styles: a one-standard-deviation
possible to remove the stress and adver-
increasein maternal depressogenic cogni-
sitiesinpeople’slives,itmaybepossibleto
Jonathan Evans, M.D. tive style score during pregnancy was
modify interpretation of such adversities significantly associated with a mean in-
through cognitive style. Children may de-
crease of 0.1 standard deviations in off-
Jon Heron, Ph.D.
velop a negative cognitive style through
springdepressogeniccognitivestylescoreat
modelingthestyleoftheirmothers.How-
age18.Thiseffectremainedafteradjusting
Carol Joinson, Ph.D. ever,findingshavebeeninconsistentonthe
formaternalandoffspringdepressionand
association.Theauthorstestedthehypoth-
explained21%oftheassociationbetween
Alan L. Stein, F.R.C.Psych. esis that there is an independent associa- maternalandoffspringdepression.
tion between maternal and offspring
Conclusions: Although the mechanisms
Glyn Lewis, Ph.D. depressogeniccognitivestyle. remaintobeelucidated,thefindingsare
Method: Datafromover4,000mothers consistent with the idea that a mother’s
andchildrenfromtheAvonLongitudinal cognitivestyle(irrespectiveofherdepres-
Study of Parents and Children cohort sion status) influences that of her child.
study in the United Kingdom were used Thissuggeststhatinterventionstoimprove
to investigate the association between amother’scognitivestylecouldhelppre-
maternal depressogenic cognitive style vent her offspring from developing de-
before the offspring’s birth and the pressionduringadulthood.
(AmJPsychiatry2013;170:434–441)
D
epression during adulthood is one of the leading article,however,wefocusondepressogeniccognitivestyle
causesoftheglobalburdenofdisease(1),makingresearch defined as the tendency to attribute negative events to
into prevention a priority. While it is rarely possible to causes that are internal (caused by self), global (have an
preventtheoccurrenceoftheadverseeventsandcircum- impact on a wide domain of activities), and stable (are
stancesthatarethoughttoincreasetheriskofdepression,it difficulttochange).Cognitivestyleisimportantbecauseit
maybepossibletopreventdevelopmentofdepressogenic represents the application of cognitive biases to the self
interpretationsofadverseevents,whichmayinturnreduce andtheinterpretationofadverseevents.Indeed,cognitive
theriskofdevelopingdepression.Understandingtheorigins style is one of the therapeutic targets of cognitive
of depressogenic cognitive style is therefore important in behavioraltherapy,andpreviousresearchhasshownthat
strategiestopreventdepression. this style is associated with both concurrent and future
Beck’s cognitive theory of depression (2) suggests that episodesofdepression(5).
thepresenceofnegativebeliefsystemsregardingtheself, Twinstudiessuggestthattheremaybebothgeneticand
the world, and the future underlie the development of environmentalinfluencesondepressogeniccognitivestyle
depression. Beck’s theory is consistent with research on (6).However,thereportedgeneticcorrelationsforcogni-
attributionalstyles,whichsuggeststhattheapplicationof tivestylearesmall,andsmallerthanthosefordepression
such depressogenic beliefs to one’s interpretation of or anxiety disorders themselves (7). Therefore, cognitive
negative life events is associated with depression (3). stylemaydeveloplargelythroughexperience.Oneprocess
RecentresearchhasextendedBeck’stheorytomorebasic bywhichnegativecognitivestylemaydevelopisthrough
cognitive biases. For example, attentional bias toward theexplanationsofeventsprovidedbycaregivers,partic-
negative stimuli is also associated with vulnerability to ularly mothers. Several studies have suggested that
depression(4).Suchbiasesmayrepresentalowercogni- negativeandcriticalmaternalfeedbackisassociatedwith
tive level of depressogenic beliefs: the beliefs we hold depressogenic offspring cognitive styles (8–10). Mothers’
direct the information we bias and vice versa. In this owncognitivestylemayalsoprovideamodelforthechild
434
ajp.psychiatryonline.org AmJPsychiatry170:4,April2013
PEARSON,FERNYHOUGH,BENTALL,ETAL.
toimitate(11,12);basedonobservationsoftheirmother’s in southwest England who had an estimated date of delivery
inferencesaboutherselfandcircumstances,childrenwill between April 1, 1991, and December 31, 1992, were invited to
takepart.Thechildrenof15,247pregnancieswererecruited(20).
developtheirowncognitivestyle.
The representative nature of the original ALSPAC sample has
However, there are also theoretical reasons for ques-
beeninvestigatedbycomparisonwiththe1991NationalCensus
tioning whether there is any direct association between dataofresidentsinthecountyofAvon(20).Ethicalapprovalfor
maternalandoffspringcognitivestyles.First,itispossible the study was obtained from the ALSPAC Law and Ethics
that a mother’s negative beliefs about herself are not Committeeand theLocalResearchEthicsCommittees.Written
sufficiently expressed in her behavior to be modeled. consent was obtained from all participants after they received
a description of the study. Detailed information has been
Second,itispossiblethatthechildmayusethemother’s
collected on the cohort since early pregnancy (20; see also
attributions to reach diametrically opposite conclusions. http://www.alspac.bris.ac.uk). In this study, we used data from
Forexample,ifthemothermakesanexternalattribution thesubsampleofALSPACsingletonoffspringwhoattendedthe
and blames the child for her misfortunes, the child may researchclinicatage18.
learntoblamehim-orherself(8).
Measures
Althoughsomestudiesreportcorrelationsbetweenma-
Offspring cognitive style. The Cognitive Style Questionnaire–
ternal and offspring cognitive style (8, 13, 14), at least as
Short Form (CSQ-SF), which was developed from the original
manystudiesfailtofindevidenceforanyassociation(10,15,
Cognitive Style Questionnaire (21) and has been shown to be
16). This inconsistency may reflect a failure to detect real reliableandvalidforuseinadolescents(22),wasadministeredto
associations because the studies were limited to cross- theALSPACchildrenatthefourthTeenFocusClinic(meanage:
sectionaldesigns(14,16)orhadsmallsamplesizes(,400) 17 years, 10 months). The CSQ-SF presents eight hypothetical
(8–10, 13). Previous studies have also measured mother’s events relating to failures in academic, employment, and in-
terpersonal relationships. For each event, participants are in-
cognitivestyleaftertheindexchildwasborn,thusfailingto
structed to vividly imagine themselves in that situation and
eliminatethepossibilityofreversecausality. thinkcarefullyaboutthereasonfortheevent.Next,usingaLikert
In addition, previous studies have limited their investi- scale of agreement from 1–5, participants rate the extent to
gationstowithinchildhoodandearlyadolescence(9,10,13). whichthisreasonwascausedbyinternalversusexternalfactors
(themselves or others), specific versus global factors (has an
Itis,however,importanttounderstandtheantecedentsof
impactonallareasorthisspecificsituation),andstableversus
cognitivestylesthatarecarriedintoadulthood,wherethe
unstable factors (will persist or not) and the extent to which it
main public health burdens of depression are manifest. reflects theirself-worth(meansthey are flawed)(Figure 1). For
Cognitivestylesmayalsobeopentochanginginfluences eachscenario,twoitemsrelatetoeachofthefourdimensions,
(10), and as the child develops, the mother may become resultingin64items,withtotalscoresrangingfrom64to320.A
lessinfluentialincomparisontopeers(17). fifth dimension relating to negative consequences was omitted
from the CSQ-SF. Higher scores indicate a more negative style.
Finally, it is unclear whether the association between Total scores showed a normal distribution, with a Cronbach’s
maternal and offspring cognitive styles is independent of alpha(measuringinternalconsistency)of0.88,whichiscompa-
maternaldepression.Maternalcognitivestyleandmaternal rabletopreviousstudies(22).Aprincipal-componentsanalysisof
depression are strongly related (18), and maternal de- thescoresforthefourdimensionsindicatedthatasinglefactor
with an eigenvalue of 3.25 explained 65% of the variance. All
pression is also a risk factor for offspring depressogenic
subscalescoresloadedonthisfactor(loadingsrangedfrom0.37
cognitivestyle(10).Partofthisassociationmaybeexplained
to0.85),whichiscomparabletopreviousstudies(22).Weused
bythecognitivestyleofdepressedmothers.However,the CSQ-SFtotalscoreastheprimaryoutcomevariable.
effect of maternal depression could also operate through Atotalof4,693adolescentscompletedatleastoneitemonthe
mechanisms that are unrelated to cognitive style. For questionnaire, and 3,387 completed all 64 items. For those who
completedatleast75%oftheitems(equivalenttoatleastsixofthe
example,maternaldepressionmayactasadversityforthe
eightscenarios),missingitemswereimputedfromtheindividuals’
child or have an impact on the child’s cognitive style
medianscoresacrossallotheritemsinthesubscalefromwhich
throughitsassociationwithlesssensitiveparenting(19). itemsweremissing.Resultsarereportedfortheimputedscores.
In this study, we sought to improve on previous in- Allanalyseswererepeatedwithcompleteitemscoresonly,andthe
vestigations by using data from over 4,000 families from resultswerecomparable(dataavailableonrequest).
the Avon Longitudinal Study of Parents and Children Maternal cognitive style during pregnancy. A broader mea-
(ALSPAC)cohortintheUnitedKingdomtoinvestigatethe sure of intrapersonal sensitivity (23) given to mothers as part
association between maternal cognitive style measured ofaquestionnairemailedoutat18weeksofpregnancywasused
to derive a measure of maternal cognitive style (18). Six items
during pregnancy and offspring cognitive style 18 years
relatingtonegativecognitionsoutlinedinBeck’scognitivetheory
later and how any such association was related to the
ofdepressionwereselectedonatheoreticalbasisinaprevious
associationbetweenmaternalandoffspringdepression. study (18). These items originated from the Dysfunctional
AttitudesScale,whichhasbeenvalidatedagainstareviewer-led
Method measureofcognitivedistortionsusinghypotheticalstories(24).
Selected items were judged to measure negative cognitions
according toBeck’s theory;two authors(G.L.and J.E.) selected
Participants
theseitemsindependently.Itemswereexcludediftheyincluded
The sample comprised participants from ALSPAC. All preg- wordsthatwerelikelytobeconfoundedbycurrentmood,such
nant women residing in the former Avon Health Authority as “worry” or “feel.” The internal consistency of the items was
435
AmJPsychiatry170:4,April2013 ajp.psychiatryonline.org
MATERNALANDOFFSPRINGCOGNITIVESTYLES
FIGURE1.OffspringandMaternalCognitiveStyleItems
Offspring Cognitive Style Items Relation to Depressogenic Thinking
Example Scenario: Imagine you go to a party and people are not interested in you.
Think carefully about the reason for people not being interested in you.
The reason is my fault Internal
The reason affects all areas of my life Global
The reason will cause the same negative event in the future Stable
The reason means there is something wrong with me as a person Self-worth
Maternal Cognitive Style Items Relation to Depressogenic Thinking
I avoid saying what I think for fear of being rejected Global and self-worth
If others knew the real me, they would not like me Internal and self-worth
If other people knew what I am really like, they would think less of me Internal and self-worth
I always expect criticism Global, stable, and self-worth
I don’t like people to really know me Global, stable, and internal
My value as a person depends enormously on what others think of me Global and self-worth
high(Cronbach’salpha=0.77).Eachitemwasratedforagreement depressive ideas, poor concentration, sleep problems, and fa-
ona4-pointLikertscale,withpossiblescoresrangingfrom0to3. tiguetoaccountforsubthresholdsymptoms.
Summed scores could range from 0 to 18, with higher scores
reflecting more negative cognitions. A total of 12,175 mothers Statistical Analysis
completed the six items, with a mean score of 4.6 (SD=3.4). Wefirstexaminedtherelationshipbetweenmaternalcognitive
Higher scores for mothers who were not currently depressed styleandoffspringcognitivestyleinaseriesoflinearregression
werepredictiveoffutureepisodesofdepressionyearsafterbirth models. The exposurevariable was maternalnegative cognitive
(18).Thenegativecognitionsendorsedbythemothersrelateto style, treated as a continuous score. The outcome variable was
the dimensions on the CSQ-SF. For example, to agree with the offspringtotal CSQ-SF score.We examined aunivariate model,
statement “I always expect criticism” reflects low self-worth, thenintroducedpotential confoundingvariables(maternaland
which, as indicated by the endorsement of “always,” is both offspring depression) separately into the model to investigate
aglobalandastableattribution.Asthismeasurewastakenasan their impact on the main association. Although later maternal
index of maternal cognitive style, it is referred to as such depression measures were available, they were not included in
hereafter(seeFigure1). ourmodelsbecauseofthestrongcorrelationamongtimingsof
maternaldepression.However,werepeatedallanalysesreported
Confounding variables. Variables previously shown to be as- here with antenatal replaced by postnatal timing of depression
sociatedwithmaternalandoffspringcognitivestyleswerecon- and obtained comparable results (data available on request).
trolled for (10). Maternal characteristics were derived from Therewerenolatermeasuresofmaternalcognitivestyle.
questionnaires given during pregnancy. These included age, We examined a final multivariate model that included all
education level, social class, (ranked from 1 to 5, with lower confoundingvariables,andthenwerepeatedthefinalmultivar-
scores indicating higher status), parity, smoking during preg- iate analysis excluding mothers who exceeded thresholds for
nancy,anddepressionmeasuredusingtheEdinburghPostnatal depression at the time they completed the cognitive style
Depression Scale (25) at 18 weeks of pregnancy on the same measure. Analyses were conducted using Stata, version 12
questionnaire from which the measure of maternal cognitive (StataCorp,CollegeStation,Tex.).
style was derived. The Edinburgh Postnatal Depression Scale
Mediation analyses. To investigate whether the association
is a 10-item self-report questionnaire specifically designed to
between maternal and offspring cognitive style mediated any
screenforperinataldepression;scores.12havebeenshownto
association between maternal and offspring depression, we
have a high sensitivity (81.1%) and specificity (95.7%) in
investigated the association between maternal and offspring
predicting major depressive disorder (26). We also included
depression before and after including cognitive styles and
offspringgenderandoffspringdepression,whichwasmeasured
performedamediationanalysisusingMplus.
with the Clinical Interview Schedule–Revised (27) at the same
researchclinicastheCSQ-SF. Missing data. In a sensitivity analysis, we used multiple im-
putations to account for missing data. Given the availability of
Offspringdepression.TheClinicalInterviewSchedule–Revised information on sociodemographic variables, we assumed that
is a self-administered, computerized interview that establishes missingness is dependent on observed data and used multiple
theseverityofsymptomsthatconstituteanxietyanddepression chained equations using all study variables and sociodemo-
disorders using algorithms based on ICD-10 criteria (27). The graphic indicators to impute 30 data sets. We then repeated
version we used derives an ICD-10 diagnosis as well as symp- analysesacrosstheimputeddatasetscombiningestimatesusing
tomseverityscoresfordepression,depressivethoughts,anxiety, Rubin’s rules (28). Given the difficulty in imputing missing
panic, phobia, sleep, concentration, and fatigue, on a scale of outcomes (28), we initially imputed only up to the sample for
0–4.Weusedabinaryvariable(depressed,notdepressed);cases which complete cognitive style (CSQ-SF) outcome data were
werethosewithaprimarydiagnosisofmild,moderate,orsevere available (N=3,845). However, given the association between
depression using ICD-10 criteria. We also used a depressive CSQ-SF and mood, we used five earlier measures of mood to
symptom score derived by summing scores for depression, predict CSQ-SF and repeated the imputation model up to
436
ajp.psychiatryonline.org AmJPsychiatry170:4,April2013
PEARSON,FERNYHOUGH,BENTALL,ETAL.
TABLE1.CharacteristicsofCompleteCaseSampleandOverallAvonLongitudinalStudyof ParentsandChildren(ALSPAC)
Sample
CompleteCaseSample
Characteristic (N=2,528) ALSPACSamplea p
N % TotalN N %
Maternaleducation 8,827 ,0.001
Universitydegree 572 23 982 11
Alevel(noncompulsory,afterage16) 771 31 1,957 22
Olevel/vocation(compulsory,beforeage16) 1,036 41 4,345 49
NoneorCertificateofSecondaryEducation 149 6 1,543 17
SocialClass 7,280 ,0.001
1(higherprofessional) 228 9 351 5
2(professional) 968 38 2,116 29
3(skilled) 1,135 45 3,828 53
4(unskilled) 169 7 795 11
5(unemployed) 27 1 187 3
Pregnancyandmaternaldata 10,072
First-bornchild 1,286 51 4,374 43 ,0.001
Femalechild 1,411 56 7,977 46 ,0.001
Mean SD TotalN Mean SD
Mother’sage(years) 29.7 4.3 27.6 5.0 ,0.001
Mother’sdepressionscoreduringpregnancy 6.1 4.4 7.2 4.9 ,0.001
Maternalcognitivestyle 5.0 3.7 5.0 3.5 0.758
Childdepressionandcognitivestyledata 1,669
ChildCSQscore 162 20 161 20 0.794
N % N %
Childdepressiondiagnosisatage18,based 181 7 148 9 0.044
ontheClinicalInterviewSchedule–Revised
aForthecoresingletonALSPACsamplenotinthecompletecasesample,N=11,089.Percentagesarebasedontheproportionofthosewith
dataonthesevariables,asindicatedinthe“TotalN”column.
a starting sample with at least one measure of mood or tem- TABLE2.MeanScoreontheCognitiveStyleQuestionnaire–
perament as well as exposure (N=10,322). This model should Short Form Among Offspring at Age 18, by Maternal
be interpreted with caution, however, as our ability to predict CognitiveStyleScoreDuringPregnancya
CSQ-SFscoreswaslimited. Score
MaternalCognitiveStyleTertile Mean SD 95%CI
Results
Lowesttertile(N=1,076) 161 20.0 160–162
Atotal of3,845offspringprovideduseabledataonthe Middletertile(N=877) 162 20.0 161–164
CSQ-SFatage18.Fromthissample,3,320oftheirmothers Highesttertile(N=813) 164 20.2 163–165
had provided maternal cognitive style data during preg- a Maternalcognitivestylescoresaregroupedintotertilesrepresent-
inglowscores(0–2),moderatescores(3–5),andhighscores(6–18).
nancy.Thissamplewasfurtherreducedto2,528mothers
The table demonstrates a dose-response relationship between
andchildrenforwhomcompletedataforallconfounding maternal cognitive style and offspring cognitive style score,
variableswereavailable.Demographiccharacteristics for indicating that mothers with higher cognitive style scores have
offspringwithhigherscores.
thesamplewithcompletedataascomparedwiththerest
ofALSPACaresummarizedinTable1.
cognitive style. For a one-standard-deviation increase in
AssociationsWith Offspring CognitiveStyle
maternal depression score, offspring cognitive style score
Meanoffspringcognitivestyle(CSQ-SF)scores,bylevel increased by 1.28 points (95% CI=0.40–2.17; p=0.005).
ofmaternalcognitivestyle,arepresentedinTable2.Linear However, the increase fell to 0.17 (95% CI=–0.002 to 0.36;
regressionanalyses provided evidencefor a positiveasso- p=0.09)afteradjustmentformaternalcognitivestyle.
ciationbetweenmaternalandoffspringcognitivestyle.This
association held in the sample for whom complete data Associations WithOffspring Depression
and Mediating RoleofCognitive Styles
wereavailableforconfoundingvariablesafterweincluded
adjustments for maternal and offspring depression and Cognitive style scores for depressed offspring were 14
additional confounding variables, excluding mothers who points (0.8 standard deviations) higher (95% CI=12–18;
were depressed at baseline, and imputing missing data p,0.001)thanthosefortheirnondepressedpeers.There
(Table 3). There was also evidence for a univariate wasalsoanassociationbetweenmaternalcognitivestyle
association between maternal depression and offspring andoffspringdepression(oddsratio=1.18,95%CI=1.01–1.40;
437
AmJPsychiatry170:4,April2013 ajp.psychiatryonline.org
MATERNALANDOFFSPRINGCOGNITIVESTYLES
TABLE 3.Mean Increase in Cognitive Style Questionnaire–Short Form (CSQ-SF) Score Among Offspring for Each 6-Point
IncreaseinMaternalCognitiveStyleScorea
IncreaseinOffspringCSQ-SFScore
Model b 95%CI p
Model1a:UnivariateassociationwithchildCSQ-SFscore,allavailable 2.55 1.38–3.72 ,0.001
data(N=3,320)
Model1b:UnivariateassociationwithchildCSQ-SFscore,complete 2.86 1.51–4.22 ,0.001
cases(N=2,528)
Model2:Adjustedforconcurrentadolescentdepression 2.59 1.26–3.90 ,0.001
Model3:Adjustedformother’santenataldepression 2.45 0.79–4.10 0.004
Model4:Alladjustmentsb 1.96 0.53–3.40 0.007
Model5:Asinmodel4plusexclusionofthosewhosemotherswho 1.81 0.22–3.40 0.025
weredepressedatbaseline(N=2,209)
Model6:Asinmodel4,withimputationsformissingdataforthose 2.03 0.79–3.25 ,0.001
whohadCSQ-SFscores(N=3,845)
Model7:Asinmodel4,withimputationsformissingdataandCSQ-SF 2.40 0.90–3.86 0.002
scores(N=10,322)
a Models 1 and 2 show results for univariate linear regression analyses; models 2 and 3 show results after adjustments for offspring and
maternaldepression;model4showsresultsafteradjustmentforconfoundingvariables;andmodels6and7showresultsaftermissingdata
imputation. b indicates the increase in CSQ-SF score for each 6-point increase in maternal cognitive style score, which represents the
approximatedifferencebetweenthelowestandhighesttertiles.Models2–4wereconductedforthoseinthesampleforwhomcompletedata
wereavailable(N=2,528).Model5wasrestrictedtothoseforwhomcompletedatawereavailableandwhosemothershaddepressionscores
belowthresholdatbaseline.
bAdjustments were made for adolescent depression continuous score and diagnosis indicator, maternal antenatal depression total score,
maternalage,maternaleducation,socialclass,maternalhistoryofdepression,childgender,andparityofmotherwithindexchild.
p=0.049),whichwasreducedafteradjustmentforoffspring Strengths andLimitations
cognitivestyle(oddsratio=1.10,95%CI=0.93–1.31;p=0.248). Strengthsofthestudyincludethelargesamplesize,the
Finally, there was an association between maternal de- long-term follow-up spanning the life of offspring from
pression and offspring depression. For a one-standard- before birth into adulthood, and the availability of con-
deviationincreaseinmaternaldepressionscore,theodds founding variables, particularly concurrent measures of
of offspring depression were increased by 23% (odds bothmaternalandoffspringdepression.
ratio=1.23, 95% CI=1.04–1.44; p=0.013). This associa- Alimitationofthestudyisthelosstofollow-upfromthe
tion was reduced after maternal and offspring cognitive original sample. Young adults who attended the clinic at
styleswere included inthe model (odds ratio=1.15, 95% age 18 were less likely to come from poorer families.
CI=0.96–1.39;p=0.138).AsshowninFigure2,mediation However,sensitivityanalysesaccountingformissingdata
analysis provides evidence that 21% of the association provided no evidence that missing data introduced bias.
between maternal and offspring depression was medi- Another limitation is that the measurement of maternal
atedbymaternalandoffspringcognitivestyles. and offspring cognitive style differed. The maternal
measure related to negative cognitive style in general,
whereas the offspring measure involved attribution of
cognitive styles to hypothetical events. According to
Discussion
cognitive theories, the two measures should reflect the
An increase of approximately one standard deviation same underlying constructs. For example, if a mother
in maternal cognitive style score during pregnancy was alwaysexpectscriticism,itseemsreasonabletoinferthat
associatedwithanincreaseofapproximately0.1standard this outlook would be applied to the attributions of
deviationsinoffspringcognitivestylescoreatage18.The negative events. Nonetheless, there was likely to have
small effect size may explain previous inconsistent find- beenconsiderablemeasurementerrorinmeasuringcog-
ings because it suggests that this association may have nitivestyles,forthematernalmeasureinparticular;some
beenmissedinsmallerstudieslackingsufficientstatistic- subjects with negative cognitive styles may not have
al power. Our analyses provide evidence that maternal reportedthisinthefewquestionsavailable,inwhichcase
andoffspringcognitivestylesexplained21%oftheassocia- their score on the measure would not accurately reflect
tionbetweenmaternalandoffspringdepression.Itisstrik- theircognitivestyle.Thiswillhaveaddednoisetothedata,
ing that the association between maternal and offspring likelyleadingtoanunderestimateofanyassociation.
cognitive styles should remain despite the possible re- Therearealsoseveralquestionsthatourfindingscannot
duction in maternal influence as the child ages and the answer. For example, the stability of maternal cognitive
potential for the mother’s cognitive style to change stylethroughoutthechild’slifewasunclear,whichlimits
throughthechild’slife. our understanding as to when it influences the child.
438
ajp.psychiatryonline.org AmJPsychiatry170:4,April2013
PEARSON,FERNYHOUGH,BENTALL,ETAL.
FIGURE 2.Theoretical Pathways Linking Maternal Depression and Maternal Cognitive Style With Offspring Cognitive Style
andDepressiona
(Unadjusted association between
maternal and offspring depression,
0.088, p=0.014)
Offspring depression
Maternal depression 0.070
diagnosis at age 18
in pregnancy (total score) (p=0.048)
(yes/no)
0.450 0.024 0.332
(p<0.001) (p=0.317) (p<0.001)
Offspring cognitive
Maternal cognitive style 0.073
style at age 18
in pregnancy (total score) (p=0.002)
(total score)
aAnalysisbasedonparticipantsforwhomcompletedatawereavailable(N=2,528).Inthediagram,thearrowsrepresentregressionsandthus
associations between variables. All regression path coefficients are standardized, so all effects sizes are directly comparable. The total
association between maternal depression and offspring depression at age 18 was 0.088 (95% CI=0.017–0.159, p=0.014). This total effect
comprisesalldirectandindirectpathwaysfrommaternaltooffspringdepression.Thiswouldcomprisethedirectassociationbeforecognitive
stylesareconsideredinthemodel.Incontrast,oncematernalandchildcognitivestylesareconsidered,thedirectassociationdropsto0.070
(p=0.048).Thetotalindirecteffectthroughmotherandoffspringcognitivestyleswas0.019(95%CI=0.017–0.020,p=0.010),ascalculatedin
Mplusfromtheproductoftheindirectpathwaysinthefigure.Thus,21%(0.019/0.088)oftheoriginaltotalassociationbetweenmaternaland
childdepressionwasmediatedbymotherandchildcognitivestyles.Themodelfitindicesweregood;x2=0.140,df=1,p=0.708,rootmean
squareerrorofapproximation=0.0,95%CI=0.00–0.038,comparativefitindex=1.0;standarderrorsandthusconfidenceintervalsofdirect
andindirecteffectswereestimatedusingbootstrapping.Theestimatorwasweightedleastsquares.
Furthermore, the lack of a parenting measure limits our is initially brought about, the important point is that
understandingastohowmaternalcognitivestyleinfluences maternal depression and cognitive style were strongly
thechild.Finally,althoughpreviousstudieshavereported correlated in pregnancy and that the association with
limitedheritabilityofdepressogeniccognitivestyles(6,29), cognitivestylespartiallyexplainedtheeffectsofmaternal
unmeasuredsharedgeneticfactorsmayalsoexplainpartof depression. Transmission of cognitive styles is therefore
theassociation. one potential pathway from maternal to offspring de-
pressionthatismodifiableandwouldprovideatargetfor
Mechanisms
prevention. Furthermore, maternal cognitive style was
One explanation for the association between mother associated with offspring cognitive style irrespective of
and offspring cognitive styles is that negative cognitive maternal depression. Thus, maternal cognitive style is
styleis amarker for depression andthereforetheresults a potential target for prevention of offspring depression
reflecttheestablishedassociationbetweenmaternaland evenifthemotherisnotdepressedatthatpoint.
offspringdepression(12,30).However,theassociationre-
Interventions
mainedafteradjustments for maternal andoffspringde-
pression,indicatingthatthisisnotalikelyexplanation.In There are two possible strategies that might prevent
contrast, maternal and child cognitive styles mediated the transmission of cognitive styles. First, maternal
asignificantpartoftheassociationbetweenmaternaland cognitivestylecouldbemodified.Forexample,cognitive
offspring depression. Because maternal depression and therapy is designed to modify cognitive styles, and
maternal cognitive style were measured on the same evidencesuggeststhatithasalastingpositiveinfluence
occasion,wecannotdeterminethedirectionofcausality. on cognitive style (31). As described above, it is also
Previous evidence suggests that cognitive style leads to possible that maternal depression leads to negative
depression(18);however,depressedstatescouldalsolead cognitive styles, and thus in depressed mothers anti-
tonegativecognitivestyle.Whicheverwaytheassociation depressant treatment could prevent development of a
439
AmJPsychiatry170:4,April2013 ajp.psychiatryonline.org
MATERNALANDOFFSPRINGCOGNITIVESTYLES
PatientPerspectives
(cid:129) A 19-year-old female patient presented to psychiatric (cid:129) The following exchange occurred in a face-to-face
serviceswithdepression.Shebegantreatmentwithcognitive- interactionbetweenamotherandheryounginfant.The
behavioral therapy. During discussions of her conditional infant appeared to be sleepy, and the mother initially
beliefswithhertherapist,thepatientdescribedmemoriesof commentedonthisfact.However,afewmomentslater,
her mother’s reactions and interpretations of events in her withoutanychangeintheinfant’sbehavior,themother
andhersister’slives:“Ifwecamesecondinclass,mumwould said,“Ithinkyou’reangryatme.”Aplausibleexplanation
say, “It’s a shame you never come first in anything.” If my isthatthemotherwasinappropriatelyattributingtoher
sister or I made any mistakes, mum would get upset and childnegativeandcriticalemotionsexpectedfromothers.
thinkshehadfailedasaparent.Shewouldsaythatwewere Ifachildgrowsupwithfrequentexposuretonegativeand
uselessandwouldneverbesuccessful.Ibegantothinkthat critical parental evaluations, these will arguably contrib-
whateverIdid,inwhateverareaofmylife,Iwouldneverbe utetotheformationofanegativemodelofself.(Example
goodenough,everythingIdidjustcausedupset.”(Example provided by J. Cassidy, personal communication, July
drawnfromG.L.’sclinicalexperience.) 2012.)
negativecognitivestyle.Second,ratherthanfocusingon the findings suggest that modifying maternal cognitive
changing the mother’s cognitive style, interventions style irrespective of the current presence of maternal
could focus on preventing transmission by changing depression could help prevent the development of
whatamothersaysandhowshebehaves.Topreventthe a depressogenic cognitive style in offspring. Given the
imitationofcognitivestyles,interventionscouldworkon association between cognitive style and clinical levels of
preventingmothersfromprovidingexamplesofnegative depression in this and other studies (35), these findings
cognitivestyle.Forexample,commentingthatthechild’s provideanotherpossibleroutetopreventingdepression.
act was bad rather than that the child is bad would be
modifying a global and personal attribution that is
specific and external. Negative maternal cognitive style Received May 23, 2012; revisions received Aug. 1 and Aug. 22,
2012; accepted Aug. 30, 2012 (doi: 10.1176/appi.ajp.2012.
may also result in less specific forms of negative par- 12050673).FromtheAcademicUnitofPsychiatry,OakfieldHouse,
enting,whichmayalsoneedtobetargeted(32).Forex- OakfieldGrove,UniversityofBristol,Bristol,U.K.;theDepartmentof
Psychology,DurhamUniversity,Durham,U.K.;theInstituteofMental
ample, mother-child interventions such as video
and Behavioural Sciences, University of Liverpool, Liverpool, U.K.;
feedback have been shown to modify mothers’ expres- and the Department of Psychiatry, Oxford University, Oxford, U.K.
sion of negative language and behavior and increase AddresscorrespondencetoDr.Pearson([email protected].
uk).
responsiveness (33). Finally, a mother may apply her
Dr. Bentall has received a speaking honorarium from Lilly. The
negative beliefs to her interpretation of child behavior: other authors report no financial relationships with commercial
amotherwhoalwaysfeelscriticizedmayperceivethat— interests.
and act as though—her child is critical of her (see the The UK Medical Research Council (Grant 74882), the Wellcome
Trust (Grant 076467), and the University of Bristol provide core
second patient perspective). Children who are told that support for the Avon Longitudinal Study of Parents and Children
they hold negative and critical thoughts may come (ALSPAC).ThisstudywassupportedbyaWellcomegrantheldbyDr.
Lewis.
to believe that they are negative people. In this case,
Theauthorsaregratefultothefamilieswhotookpartinthestudy,
interventions could work on improving the mother’s the midwives for help in recruiting them, and the whole ALSPAC
awareness of the child’s mental processes, or mind- team,whichincludesinterviewers,computerandlaboratorytechni-
mindedness(34).Forexample,changingamother’sfocus cians, clerical workers, research scientists, volunteers, managers,
receptionists, and nurses. Patient Perspectives were provided in
toherchild’sratherthanherownmentalstatesmayhelp collaborationwithJ.CassidyandE.Meins.
heractappropriatelytothechild’sneeds.
Conclusions References
To our knowledge, this is the first study to provide 1. WorldHealthOrganization:TheGlobalBurdenofDisease:2004
evidenceforaninfluence,persistingintoearlyadulthood, Update.Geneva,WorldHealthOrganization,2008
2. Beck AT: Cognition, affect, and psychopathology. Arch Gen
of a mother’s cognitive style on her offspring’s cognitive Psychiatry1971;24:495–500
style.Theeffectsizeisrelativelysmall,butgiventhedose- 3. AlloyLB,AbramsonLY,MetalskyGI,HartlageS:Thehopeless-
response relationship, the 18-year longitudinal time ness theory of depression: attributional aspects. Br J Clin Psy-
chol1988;27:5–21
frame, the low-risk sample, and potential measurement
4. Joiner TE Jr, Wagner KD: Parental, child-centered attributions
error resulting in underestimation of any associations,
and outcome: a meta-analytic review with conceptual and
suchaneffectislikelytobemeaningful.Themechanisms methodologicalimplications.JAbnormChildPsychol1996;24:
underlying this effect remain to be established, although 37–52
440
ajp.psychiatryonline.org AmJPsychiatry170:4,April2013
PEARSON,FERNYHOUGH,BENTALL,ETAL.
5. PrinsteinMJ,AikinsJW:Cognitivemoderatorsofthelongitudi- 22. MeinsE,McCarthy-JonesS,FernyhoughC,LewisG,BentallRP,
nal association between peer rejection and adolescent de- Alloy LB: Assessing negative cognitive style: development and
pressivesymptoms.JAbnormChildPsychol2004;32:147–158 validationofashort-formversionoftheCognitiveStyleQues-
6. LauJY,RijsdijkF,EleyTC:Ithink,thereforeIam:atwinstudyof tionnaire.PersIndividDif2012;52:581–585
attributional style in adolescents. J Child Psychol Psychiatry 23. Boyce P, Parker G: Development of a scale to measure in-
2006;47:696–703 terpersonalsensitivity.AustNZJPsychiatry1989;23:341–351
7. ZavosHM,RijsdijkFV,GregoryAM,EleyTC:Geneticinfluences 24. Weissman AN, Beck AT: Development and validation of the
onthecognitivebiasesassociatedwithanxietyanddepression DysfunctionalAttitudesScale:apreliminaryinvestigation.Pro-
symptomsinadolescents.JAffectDisord2010;124:45–53 ceedingsofthe62ndAnnualMeetingoftheAmericanEduca-
8. AlloyLB,AbramsonLY,TashmanNA,BerrebbiDS,HoganME, tionalResearchAssociation.Toronto,March1978
WhitehouseWG,CrossfieldAG,MoroccoA:Developmentalori- 25. Cox JL, Holden JM, Sagovsky R: Detection of postnatal de-
ginsofcognitive vulnerability todepression:parenting,cogni- pression: development of the 10-item Edinburgh Postnatal
tive,andinferentialfeedbackstylesoftheparentsofindividuals DepressionScale.BrJPsychiatry1987;150:782–786
athighandlowcognitiveriskfordepression.CognitTherRes 26. MurrayL,CarothersAD:ThevalidationoftheEdinburghPost-
2001;25:397–423 natalDepressionScaleonacommunitysample.BrJPsychiatry
9. GarberJ,FlynnC:PredictorsofDepressiveCognitionsinYoung 1990;157:288–290
Adolescents.CognitTherRes2001;25:353–376 27. LewisG,PelosiAJ,ArayaR,DunnG:Measuringpsychiatricdis-
10. MezulisA,FunasakiK,HydeJS:Negativecognitivestyletrajec- orderinthecommunity:astandardizedassessmentforuseby
tories in the transition to adolescence. J Clin Child Adolesc layinterviewers.PsycholMed1992;22:465–486
Psychol2011;40:318–331 28. White IR, Royston P, Wood AM: Multiple imputation using
11. BanduraA,HustonAC:Identificationasaprocessofincidental chainedequations:issuesandguidanceforpractice.StatMed
learning.JAbnormSocPsychol1961;63:311–318 2011;30:377–399
12. Goodman SH, Gotlib IH: Risk for psychopathology in the chil- 29. FarmerA,HarrisT,RedmanK,MahmoodA,SadlerS,McGuffin
dren of depressed mothers: a developmental model for un- P: The Cardiff Depression Study: a sib-pair study of dysfunc-
derstanding mechanisms of transmission. Psychol Rev 1999; tional attitudes in depressed probands and healthy control
106:458–490 subjects.PsycholMed2001;31:627–633
13. SeligmanME,PetersonC,KaslowNJ,TanenbaumRL,AlloyLB, 30. Goodman SH, Rouse MH, Connell AM, Broth MR, Hall CM,
Abramson LY: Attributional style and depressive symptoms Heyward D: Maternal depression and child psychopathology:
amongchildren.JAbnormPsychol1984;93:235–238 a meta-analytic review. Clin Child Fam Psychol Rev 2011; 14:
14. StarkKD,SchmidtKL,JoinerTEJr:Cognitivetriad:relationship 1–27
todepressivesymptoms,parents’cognitivetriad,andperceived 31. JarrettRB,Vittengl JR,DoyleK,ClarkLA:Changesincognitive
parentalmessages.JAbnormChildPsychol1996;24:615–631 content during and following cognitive therapy for recurrent
15. OliverJM,BergerLS:Depression,parent-offspringrelationships, depression: substantial and enduring, but not predictive of
andcognitivevulnerability.JSocBehavPers1992;7:415–429 changeindepressivesymptoms.JConsultClinPsychol2007;75:
16. Kaslow NJ,Rehm LP,Pollack SL,SiegelAW: Attributional style 432–446
andself-controlbehaviorindepressedandnondepressedchil- 32. Stein A, Craske MG,Lehtonen A, Harvey A,Savage-McGlynn E,
dren and their parents. J Abnorm Child Psychol 1988; 16: Davies B, Goodwin J, Murray L, Cortina-Borja M, Counsell N:
163–175 Maternalcognitionsandmother-infantinteractioninpostnatal
17. Stevens EA, Prinstein MJ: Peer contagion of depressogenic at- depressionandgeneralizedanxietydisorder.JAbnormPsychol
tributional styles among adolescents: a longitudinal study. J (Epubaheadofprint,Jan30,2012)
AbnormChildPsychol2005;33:25–37 33. Stein A, Woolley H, Senior R, Hertzmann L, Lovel M, Lee J,
18. EvansJ,HeronJ,LewisG,ArayaR,WolkeD;ALSPACstudyteam: CooperS,WheatcroftR,ChallacombeF,PatelP,Nicol-Harper
Negativeself-schemasandtheonsetofdepressioninwomen: R, Menzes P, Schmidt A, Juszczak E, Fairburn CG: Treating
longitudinalstudy.BrJPsychiatry2005;186:302–307 disturbances in the relationship between mothers with bu-
19. Lovejoy MC, Graczyk PA, O’Hare E, Neuman G: Maternal de- limic eating disorders and their infants: a randomized, con-
pression and parenting behavior: ameta-analytic review. Clin trolled trial of video feedback. Am J Psychiatry 2006; 163:
PsycholRev2000;20:561–592 899–906
20. BoydA,GoldingJ,MacleodJ,LawlorDA,FraserA,HendersonJ, 34. Meins E, Fernyhough C, Fradley E, Tuckey M: Rethinking ma-
Molloy L, Ness A, Ring S, Davey Smith G: Cohort Profile: the ternal sensitivity: mothers’ comments on infants’ mental pro-
“Childrenofthe90s”:theindexoffspringoftheAvonLongitu- cesses predict security of attachment at 12 months. J Child
dinal Study of Parents and Children. Int J Epidemiol (Epub PsycholPsychiatry2001;42:637–648
aheadofprint,Apr16,2012) 35. MezulisAH,AbramsonLY,HydeJS,HankinBL:Isthereauni-
21. HaeffelGJ,GibbBE,MetalskyGI,AlloyLB,AbramsonLY,Hankin versalpositivitybiasinattributions?Ameta-analyticreviewof
BL,JoinerTEJr,SwendsenJD:Measuringcognitivevulnerability individual, developmental, and cultural differences in the
to depression: development and validation of the Cognitive self-serving attributional bias. Psychol Bull 2004; 130:
StyleQuestionnaire.ClinPsycholRev2008;28:824–836 711–747
441
AmJPsychiatry170:4,April2013 ajp.psychiatryonline.org