The Marriage Checkup:

An Indicated Preventive Intervention for Treatment-Avoidant

Couples at Risk for Marital Deterioration
James v. Cordova, Clark University
Rogina L. Scott, Marina Dorian, Shilagh Mirgain, Daniel Yaeger,Alison Groot
University of Illinois at Urbana-Champaign

Prior to dissolution, it is likely that couples that become se-

verely distressed first pass through an at-risk stage in which

they experience early symptoms of marital deterioration
but have not yet suffered irreversible damage to their marriage. It is during this "at-risk" stage when couples might
benefit most from early intervention. In response to this
need we have developed an indicated intervention program
called the Marriage Checkup (MC) based on the principles
of motivational interviewing. The current randomized
study provides preliminary evidence for the attractiveness,
tolerability, efficacy, and mechanisms of change of the MC.

MARITAL DETERIORATION is one of the leading

causes of human suffering. Relationship difficulties
are among the most common reasons that people
seek psychological services ("Mental Health," 1995).
Even so, the majority of people suffering from relationship difficulties do not seek the help of mental health professionals and those who do seek help
most frequently see clergy or physicians (Doherty,
Lester, & Leigh, 1986; Veroff, Douvan, & Kulka,
1981). It has been estimated that at any one time
20% of all marriages in the U.S. are significantly
distressed (Beach, Arias, & O'Leary, 1987). Given
an estimate of 56 million married-couple family
households in 2000 (Fields & Casper, 2001), that is
approximately 11.2 million marriages that may be
in serious jeopardy of dissolution at any one time.
In addition to the suffering inherent in marital deterioration and divorce, these processes have been associated with a number of other sources of human
suffering. For example, it has been estimated that
the risk of experiencing a major depressive episode
Address correspondence to James V. Cordova, Department of
Psychology, Clark University, 950 Main Street, Worcester, MA
01610; e-mail: [email protected].
BEHAVIORTHERAPY36, 301-309, 2005
005-7894/0510301-030951.00/0

Copyright2005 by Associationfor Advancementof BehaviorTherapy

All rights for reproductionin any form reserved.

is somewhere between 10 and 25 times greater for

those experiencing significant relationship distress
(Weissman, 1987; Whisman, 2001).
It seems likely that, prior to reaching the dissolution stage, couples that become severely distressed
and eventually dissolve their marriages first pass
through an at-risk stage. During this stage, they experience early symptoms of marital deterioration
but have not yet suffered irreversible damage to
their marriage (Cordova, Warren, & Gee, 2001).
Couples in this at-risk stage are unlikely to seek
conventional, tertiary marital therapy, either because they have not yet become distressed enough
to see the need or because the time, expense, or
stigma of therapy present too great a barrier. Such
couples are also not likely to seek premarital or
newlywed interventions because they are in established marriages and do not perceive themselves as preparing for married life. It is during this
at-risk stage, however, that couples might benefit
most from early intervention. Such indicated earlyintervention programs have the potential to fill a
niche that lies between the inoculation against marital distress provided by prevention programs and
the intensive treatment of severe distress provided
by couple therapy.
Indicated early intervention with at-risk couples
has several goals, each with attendant challenges.
The first goal is to reach populations that are at risk
for relationship deterioration. However, involving
such couples in an intervention program presents
unique challenges. Whereas couples seeking marital therapy and premarital education are motivated
to pursue these interventions, either as a result of
their distress or because of their desire to start their
married lives on the right foot, at-risk couples in established marriages are motivated by neither. Such
couples are unlikely to perceive themselves as distressed enough to seek marital therapy. They may
also be suspicious of therapy or may not think of it
as a viable or desirable option for economic, time,
or social reasons. Any successful intervention must
overcome these barriers to attract at-risk couples.

302

CORDOVA

The second goal is efficient assessment of risk

potential, meaning that brief and effective means
for identifying the demonstrated predictors and
correlates of marital deterioration must be constructed. The attendant challenge involves bridging
the gap between the available empirical literature
concerning predictors and correlates of marital deterioration with couples presenting from within the
community of laypersons.
The third goal is to effectively promote marital
health in the short run, meaning that the intervention should work to immediately improve the relationship satisfaction of participating couples. The
challenge for such interventions is that they must be
brief in order to be attractive to at-risk couples, and
yet they must also be sufficiently powerful to stimulate quick relationship improvement. This goal is
important because a quick boost in marital health
and emotional closeness may be a necessary part of
motivating partners to work collaboratively toward
stable marital health.
In response to both the need for early intervention
with at-risk couples and the challenges presented by
that need, we have developed an intervention program called the Marriage Checkup (MC; Cordova
et al., 2001). The MC is a brief, two-session assessment and feedback intervention utilizing Miller
and Rollnick's (2002) motivational interviewing
strategies and Jacobson and Christensen's (1998)
acceptance-promotion strategies. Research on the
MC to date has demonstrated that this format is effective at attracting couples considered at risk for
ongoing marital deterioration but who are otherwise not seeking relationship treatment. In addition,
research has demonstrated that the MC is easily tolerated (97% completion rate) and safe for use with
at-risk couples (Cordova et al., 2001). Longitudinal
follow-up demonstrated that (a) relationship distress
remained significantly improved 2 years following
the intervention; (b) receiving a treatment recommendation as part of the MC predicted subsequent
treatment seeking for wives; and (c) couples' affective
tone following the MC predicted later marital satisfaction (Gee, Scott, Castellani, & Cordova, 2002).
The previous studies were uncontrolled, however,
so observed improvements could not be confidently
attributed to participation in the MC. Therefore, it
remains to be demonstrated that the MC is an efficacious indicated intervention for promoting the relationship health of participant couples. In addition,
previous research has not addressed the mechanisms
by which the MC is theorized to promote relationship health.
Theoretically, the MC should improve relationship satisfaction and stability by increasing couples'
motivation to pursue maritally healthy habits and

ET

AL.

by increasing intimacy and acceptance of common

differences. Specifically, the MC is expected to facilitate couples' progress through the stages of change
(Prochaska & DiClemente, 1984), from stages in
which they are less motivated to work on improving
their marriages to stages in which they are more motivated to pursue and maintain marital health. According to theory (Miller & Rollnick, 2002), motivational feedback facilitates movement through
several successive stages of change (Prochaska &
DiClemente, 1984). The first is a precontemplative
stage, in which partners suffering from problem areas
in their relationship do not recognize these areas as
problematic or subject to change. The second is a
contemplation stage in which partners recognize
that they have relationship problems but are ambivalent about what, if anything, to do about those
problems. The third stage is a determination stage in
which partners recognize their relationship problems, are determined to address those problems, but
may not know what to do. The fourth stage is an action stage, in which partners recognize their problems and are taking specific steps to address them.
At this stage, efforts to change may or may not be
effective. The fifth stage is a maintenance stage, in
which changes have been made, and partners work
to maintain those changes. The sixth stage is either
an escape stage, in which the problems are resolved,
or a relapse stage, in which the problems recur, and
the couple moves back into one of the former stages.
In addition, the MC is designed to improve intimacy by facilitating partners' expressions of emotional vulnerability (Cordova & Scott, 2001). The
MC is also designed to facilitate greater acceptance
of common differences by highlighting the softer
emotions and understandable reasons associated
with partners' behavior (Cordova, Jacobson, &
Christensen, 1998). Theoretically, these in turn facilitate partners' motivation to work collaboratively toward greater marital health.
Several hypotheses were tested in the current
study.
First, it was hypothesized that the MC would attract couples that could be categorized as at risk
for ongoing relationship deterioration.
Second, it was hypothesized that participants
would tolerate the intervention well and would
neither refuse to participate nor drop out of treatment in substantial numbers.
Third, it was hypothesized that couples participating in the MC would report increases in relationship satisfaction and that a no-treatment control
group would not show comparable improvement.
Fourth, it was hypothesized that MC couples
would demonstrate increases in intimacy and acceptance, as well as increases in motivation to im-

THE

MARRIAGE

prove the quality of the relationship, and that notreatment control couples would not report such
increases.
Fifth, it was hypothesized that intimacy, acceptance, and increased motivation would mediate the
effect of treatment on relationship satisfaction.

Method
PARTICIPANTS

The study involved 74 couples responding to newspaper advertisements. Because we were interested
in non-tertiary-level couples, only couples with no
previous history of couple therapy were included in
the study. All couples, prior to being randomly assigned, were asked if they would be willing to continue in the study even if assigned to the control
condition. Those couples that agreed were randomized to either the M C or no-treatment control
group. Couples assigned to the control group were
thanked for their willingness to contribute to the
project and informed that they would be paid $50
for their participation. 1
The sample was 92.5% White. Husbands' mean
age was 37.6 years (SD = 12.3), and wives' mean age
was 35.7 years (SD = 11.9). Couples were married
on average 9.8 years (SD = 10.45), and had an average of 16 years of education for both husbands
and wives. Couples had an average of 1.1 children
(SD = 1.1).
PROCEDURES

Treatment couples were mailed questionnaires and

returned them at their assessment session. Details
of the M C procedure are provided below. Following feedback, couples were given another battery
of questionnaires to complete. Control group couples received the same battery of pretreatment
questionnaires and returned those questionnaires
by mail. Control couples subsequently received the
same battery of questionnaires following an interval designed to match that of the M C couples, and
again returned those questionnaires by mail. Control couples were paid $50 for their participation.
Treatment couples did not receive monetary compensation for participating in the MC.
MEASURES

The Marital Satisfaction Inventory-Revised

(MSI-R; Snyder, 997). The MSI-R is a 150-item,
13-subscale self-report measure of marital satisfac1As noted later, this procedure was found to be problematic.
Pretreatment differences in Areas of Change and Precontemplation
were found and may have resulted from participants' knowledge of
their group assignment. Control couples appear to have muted
their desire for change and treatment couples appear to have felt
fleer to report greater desire for change.

CHECKUP

303

tion. It measures the amount of marital distress

along 11 relationship dimensions (e.g., global distress, sexual dissatisfaction, affective communication). Cronbach's alpha derived from a combined
sample of 2,040 individuals in the general population and 100 individuals in marital therapy ranged
from .70 (Dissatisfaction with Children Scale) to
.93 (Global Distress Scale) with a mean coefficient
of .82. Test-retest reliability over a 6-week interval
in a sample of 210 individuals in the general population ranged from .74 (Global Distress Scale) to
.88 (Role Orientation Scale) with a mean coefficient of .79 (Snyder, 1997). In the current sample,
the Cronbach's alpha for our composite MSI-R
variable was .92.

The Intimate Safety Questionnaire (ISQ; Cordova, Gee, Warren, & McDonald, 2004). The ISQ
is a 13-item self-report scale designed to measure intimate safety as defined by Cordova and Scott (2001).
The ISQ measures degree of comfort being vulnerable with an intimate partner across a range of relationship domains. Items include "When I need to
cry, I go to my partner, .... I feel uncomfortable talking to my partner about our sexual relationship,"
"I feel comfortable telling my partner things I
would not tell anybody else," and "It's hard to
apologize to my partner." Respondents rated each
statement on a 5-point scale (0 = never, 4 = always).
Internal reliability has been found to be adequate,
with alphas of .84 and .92 for men and women, respectively, and test-retest reliabilities of r = .89 and
r = .91 for men and women, respectively. In the current sample, coefficient alpha was .88 and .91 for
men and women respectively. The ISQ has been
found to be significantly correlated with all of the
subscales of the Personal Assessment of Intimacy in
Relationships Questionnaire (PAIR; Schaefer & O1son, 1981) and was found to be particularly highly
correlated with the emotional intimacy subscale of
the PAIR (r = - . 8 2 and r = - . 8 0 for wives and husbands, respectively). In addition, the ISQ has been
found to be significantly correlated with the Global
Distress Scale of the Marital Satisfaction Inventory
(Snyder, 1979) and the Marital Status Inventory
(Weiss & Cerreto, 1980), providing support for its
construct validity. We use the ISQ as our measure of
intimacy in this study because it is a theory-driven
questionnaire most consistent with our theory of
change with regard to the MC. Additional details regarding the ISQ can be obtained from the first author.

The Couples Stages of Change Questionnaire

(C-SCQ; Dorian & Cordova, 2001). The degree to
which partners were in each of the four stages of
change was measured using the 32-item C-SCQ.
The C-SCQ was adapted from the original 32-item
Stages of Change Questionnaire (McConnaughy,

304

CORDOVA

Prochaska, & Velicer, 1983) to reflect issues regarding partners' marital relationship. Four scales of eight
items each measured four of the theoretical stages
of change. Each partner rated how strongly he or
she agreed or disagreed with each item on a 5-point
Likert scale (1 = strongly disagree, 5 = strongly
agree). Sample items from each stage include (1)
precontemplative: "As far as I'm concerned, I don't
have any problems in my marriage that need changing"; (2) contemplative: "I think my marriage might
be ready for some improvement"; (3) action: "I am
doing something about the issues in my marriage
that have been bothering me"; and (4) maintenance: "I'm working hard to prevent the reoccurrence of problems we've already worked out in our
marriage." Internal consistency for the measure as a
whole was high (o~ = .90 and .82 for husbands and
wives, respectively). Internal consistency for the
four subscale varied from ot = .82 to a = .90, except
for wives' precontemplation subscale, which produced a relatively low alpha (or = .64).

The Areas of Change Questionnaire (A CQ; Weiss,

Hops, &Patterson, 1973). The ACQ is a 34-item
scale asking spouses to rate how much change
they want from their partners across a range of
content areas. Sample items include, "I want my
partner to participate in decisions about spending
money, .... I want my partner to spend time keeping
the house clean," and "I want my partner to have
meals ready on time." Questions are answered on a
scale from - 3 (much less) to +3 (much more),
with ratings of 0 reflecting no desire for change.
Internal consistency was adequate, with alphas of
.82 and .85 for wives and husbands, respectively.
In the current study the ACQ is used to assess
level of overall acceptance of the partner, with
higher change scores reflecting less acceptance. Although change and acceptance need not necessarily be regarded as opposite ends of the same continuum, the desire for less change should be an
adequate representation of greater acceptance in
most cases.
THE

MARRIAGE

CHECKUP

The MC is a two-session intervention. The MC was

advertised as an informational health service available to all married couples interested in learning
more about the health of their marriage. Advertisements for the MC deliberately made no mention of
marital distress or relationship problems in order to
attract couples that may have been at risk but that
may not have perceived themselves as distressed or
suffering from any particular problems. The MC is
deliberately designed to attract both distressed and
nondistressed couples, because some at-risk couples may still self-evaluate as nondistressed (17%

ET

AL.

of our current sample). In order to attract couples

suspicious of marital therapy, advertisements specifically noted that the MC was not marital therapy. The MC was brief and offered free of charge
in order to diminish economic and time barriers.
Finally, it was made clear that couples would be
provided with information and that it would be up
to them what, if anything, to do with that information. This was done in order to assure couples that
were ambivalent about seeking help that the MC
required nothing of them beyond the assessment
and feedback sessions. The specific wording of the
MC advertisement was as follows:
The University Couples Research Program is
offering for a limited time a free Marriage
Checkup (MC) for couples who would like to
find out more about the health of their marriage and whether their relationship is suffering from any common problems. The MC is
not part of any treatment program. Rather it is
an informational marital health service. Consultation is completely confidential. Objective
personal feedback of results will be provided.
It is up to the couple to determine what, if anything, to do with the feedback received.
The MC assesses for variables that the literature
has found are associated with marital distress and/
or deterioration, including marital satisfaction, domestic violence, intimacy, commitment, and communication style. Completion of questionnaires required
approximately 2 hours. The assessment session involved an interview about the history of the relationship (Buehlman, Gottman, & Katz, 1992).
Two problem-solving interactions were also used in
order to assure that (a) the two most problematic issues were discussed, and (b) both the husband and
wife chose an issue of particular significance. Finally, a postinteraction interview probed for the
understandable reasons and soft emotions associated with the problems discussed by the partners as
a means of fostering greater intimacy and acceptance
(Christensen, Jacobson, & Babcock, 1995). The
entire in-lab assessment session required approximately 2 hours.
The feedback session was provided approximately
2 weeks following the assessment session. The format used for the feedback session was a modification of that proposed by Worthington et al. (1995).
The same format was followed for all written and
face-to-face feedback; however, the content of the
feedback was individualized for each couple. Feedback was provided based on a motivational interviewing model (Miller & Rollnick, 2002) in which
participants were provided with objective information stemming from the research literature about

THE

MARRIAGE

the strengths and risk factors detected in their marriage. Feedback began with an overview of the
couple's early history together. The overview was
designed to (a) highlight the characteristics that
originally attracted the partners to each other; (b)
highlight the partners' shared history; and (c) begin
the session with a positive emotional tone.
The next section of the feedback reviewed the partners' strengths as a couple. Indications from the questionnaires, relationship history, or problem-solving
interaction of particular strengths--such as high intimacy, we-ness, and effective c o m m u n i c a t i o n - were emphasized. An attempt was made to make
the strengths section at least as long as the weaknesses section in order to draw partners' attention
to the positive qualities of their relationship.
The next section presented the partners with
their scores on the questionnaires. Therapists discussed each set of scores with the partners and solicited their feedback regarding the accuracy of the
general interpretation.
For the next section, entitled "areas for potential
improvement," two of the partners' most problematic issues were presented. Problematic behavior or
interpersonal patterns were discussed in relation to
the relevant empirical and therapeutic literature in
an attempt to educate partners about the potential
negative long-term interpersonal consequences. Provided in this way, such information is thought to
foster motivation to actively address the target
issues by developing discrepancies between those
problematic issues and partners' valuing of the longterm health of their relationship. Next, partners
were presented with a menu of suggestions for how
they might actively cope with the presented issues.
In addition, partners were encouraged to share their
own ideas for how best to address or cope with the
issues at hand. It was emphasized that partners
were free to choose which, if any, course of action
best appealed to them.
The entire feedback session generally required 2
hours. Overall, participants invest between 5 to 6
hours in the MC. In practice, outside the context of
a research project, it is possible to shorten the length
of the M C considerably by limiting the number of
assessment instruments and limiting the number
of questions asked in the interview.

Results
PROTOCOL

ADHERENCE

Therapists included four of the authors (J.C., R.S.,

M.D., and S.M.) and three advanced clinical doctoral students, all of w h o m were trained and supervised by the first author. In order to assess therapist
adherence to the M C protocol, an adherence scale

305

CHECKUP

was developed. Nineteen codes reflected therapist

behavior during the assessment and feedback sessions of the MC. Four undergraduate students
served as coders, and one of the authors (D.Y.)
served as the coding trainer and supervisor. Weekly
meetings were held to discuss ratings and agree on
consensus ratings. Ten of the 39 treatment couple
tapes (26%) were randomly selected and those 10
tapes were rated by each of the four coders. Each
behavior was rated on a 5-point scale of therapist
adherence ranging from 1 (not at all) to 5 (extensively). Intraclass correlations ranged from .31 to
1.00, with an average of .79. Because intraclass
correlations for some items did not adequately reflect the degree of consensus between raters, percent agreement within one level of the scale was
also calculated across all raters. Percent agreement
ranged from .72 to 1.0. The average adherence rating for most of the codes was at the upper end of
the 5-point scale, indicating that therapists were
able to adhere to the M C manual. 2
PROPORTION
ATTRACTED

OF AT-RISK
TO THE

COUPLES

MC

The first hypothesis was that a substantial proportion of the couples self-referring for the M C would
belong to the theoretically proposed group of
couples at risk for marital deterioration. The assumption being tested was that such treatmentavoidant at-risk couples exist, and that they will
volunteer to participate in an informational checkup.
In order to attract couples that might be at risk, hut
not yet self-evaluating as distressed, participation
was not limited to distressed partners; thus, there
was a possibility that the recruited sample would
consist entirely of low-risk couples. We operationally defined at-risk couples as those in which at
least one partner scored in the moderately to severely distressed range on the Global Distress Scale
(T scores above 50) or scored in the severely distressed range on any of the other satisfaction relevant subscales of the MSI-R (T scores above 60).
This definition allowed us to include as at-risk
those partners who were either (a) severely globally
distressed; (b) moderately globally distressed; or (c)
globally satisfied but severely distressed in at least
one area of their relationship. Individuals who did
not meet any of these criteria were classified as
nondistressed.
Across the sample, 77% of the couples that selfreferred to the M C (n = 57) were classified as at-

2Interested readers can contact the first author for a table of the
adherence codes, mean ratings, intraclass correlations, and percent
agreement within one level.

306

CORDOVA

TABLE I Group Comparison of At-Rislq Therapy-Seeking,

and Community Couples
At Risk
GDS
AFC
PSC
AGG
Ti-O
FIN
SEX

57.5 (8.1)
55.4 (7.5)
55.1 (7.3)
49.6 (7.4)
54.0 (9.0)
52.6 (7.6)
52.8 (8.6)

Therapy
64.9 (6.9)
61.3 (7.7)
62.5 (7.8)
56.9 (10.3)
60.1 (8.2)
57.4 (10.6)
56.3 (10.3)

Community
47.7
47.6
47.3
49.8
49.1
50.3
49.4

(7.8)
(8.7)
(9.4)
(9.3)
(8.9)
(8.9)
(9.6)

p<

5.81/-8.12
4.69/-6.40
5.93/-6.39
5.16/0.20
4.25/-3.50
3.27/-0.97
2.27/-2.49

.001/.001
.001/.001
.001/.001
.001/ns
.001/.001
.01/.06
.05/.05

Note. Therapy and communitydata from Snyder(1997). Numbers before the

slashare for the comparisonbetweenat-riskand therapycouple~Numbersafterthe slasharefor the comparisonbetweenat-riskand community
couples.Alldegreesof freedomwere 155for comparisonsto the therapy
couples and 209 for comparisonsto the communitycouple~GDS =
Global DistressScale;AFC= AffectiveCommunication;PSC = ProblemSolving Communication;AGG = Aggression;TI'O = Time Together;
FIN = DisagreementAbout Finances;SEX = SexualDissatisfaction.

risk and 23% (n = 17) as nondistressed. In addition, of the at-risk couples, 56% (n = 32) did not
meet conventional severe distress criteria as measured by the GDS (T > 60) and 26% (n = 15) had
at least one partner that met conventional nondistress criteria as measured by the GDS (T < 50).
Thus, a substantial percentage of those couples
presenting for participation in the MC met our criteria for being considered at risk for continued
relationship deterioration.
To begin exploring the validity of our operational
definition of at-risk couples, we compared the means
for our at-risk sample to those reported by Snyder
(1997) for a sample of 100 therapy couples and 154
community couples. Results generally fit the prediction that in terms of deterioration of relationship
functioning, at-risk couples would be between average community couples and treatment-seeking couples (see Table 1). T tests revealed that the mean GDS
score for our at-risk sample was significantly smaller
than the mean couple GDS score in Snyder's therapy
sample and significantly larger than the mean couple
GDS score in Snyder's community sample. The same
pattern held for Affective Communication (AFC),
Problem-Solving Communication (PSC), Time Together (TTO), Sexual Dissatisfaction (SEX), and Disagreement About Finances (FIN). Although significantly different from Snyder's therapy couple group,
our sample was not significantly different from Snyder's community sample for AGG.
TREATMENT

ET

AL.

MC, 44 (37%) declined to participate. Of those,

20 stated that they were no longer interested (e.g.,
some had changed their minds since calling, others
had found that their spouse was not interested), 14
reported scheduling difficulties or that they did not
have enough time, 8 were moving or lived out of
state, 1 reported family problems, and 1 provided
no reason for refusal. All 39 couples assigned to
the MC condition completed the protocol. Of the
35 control couples, 32 returned the postassessment
questionnaires.
EFFICACY

OF THE

M,C

The third hypothesis was that the MC would be efficacious at quickly lowering relationship distress
from pre- to postintervention and that a no-treatment control condition would not be equally efficacious. The main outcome variable used was a broad
measure of relationship distress derived from the
MSI-R. In order to use the questionnaire as a broad
measure of relationship distress reflecting our
working definition of "at-risk," as well as to minimize measurement error, intercorrelations between
the subscales of the measure were examined; those
subscales that correlated above r = .70 were averaged together to create a composite relationship
distress score. Among the subscales included in this
composite score were the GDS, the Affective Communication Scale, the Problem-Solving Communication Scale and the Time Together Scale. The correlation between the couples' composite score and
the couples' GDS was r(74) = 0.92.
With regard to gender, paired t-tests on the demographic variables age, education, and income revealed only one difference: husbands were significantly older than wives, t(73) -- -4.02, p < .001.
Similar analyses on the dependent variables revealed
that husbands scored higher on the Aggression
Scale of the MSI-R, t(73) = - 2 . 9 7 , p < .01, and
that wives scored higher on the Areas of Change
Scale, t(72) = 2.70, p < .01, and the Contemplation Scale of the Stages of Change Inventory, t(69) =
3.43, p < .001. None of these differences resulted
in notable effects in later analyses.
Because gender effects were minimal, husbands'
and wives' data were combined into couple summary scores to minimize measurement error and to
simplify reporting of the results. The average correlation for husbands' and wives' preintervention
scores was r = .66. All reported analyses were conducted on those couple scores. 3 Analyses of mean

TOLERANCE

The second hypothesis was that the MC would be

easily tolerated by participant couples, resulting in
low refusal to participate and dropout rates. Of the
120 couples that called expressing interest in the

3Analyses of the data for husbands and wives separately did not
result in substantially different results from those reported for the
couple as the unit of analysis. These results are available from the
first author.

THE

MARRIAGE

TABLE 2 Primary Outcome Analyses Comparing MC With

Control on Relationship Distress, Intimacy, Acceptance, and
Motivation to Change
MC
Measure

M (SD)

Control
n

M (SD)

F(dO, p

39 54.1 (8.1) 35 51.3 (7.8)

39 51.7 (8.3) 35 50.3 (7.4) F(I, 72) = 4.24*

Intimacy
Pre
Post

38
38

Acceptance
Pre
37
Post
37
Action
Pre
Post

37
37

.qp2

.06

3.1 (0.4)
2.8 (0.7) F(I, 67) -- 21.6"** .24

19.2 (9.8) 29 12.3 (5.6)

15.0 (8.3) 29 1.8 (6.5) F ( I , 6 4 ) = 9 . 4 0 " *
3.3 (0.7) 27
3.7 (0.5) 27

//

.47"**

307

Change in Intimacy
~ 3

o.

Treatment x Time

Distress
Pre
Post

3.0 (0.5) 31
3.1 (0.5) 31

CHECKUP

3.3 (0.6)
3.4 (0.6) F(I,62) = 9.62**

.13

.13

Note. Acceptance = Lower scoresindicate greater partner acceptance.~lp2 =

partial eta squaredis a measureof effect sizeequalingthe proportion
of the effect + error variancethat is attributableto the effect.
* p < .05;** p < .0 I; *** p ( .00 I.

differences between the treatment and control groups

on the demographic variables revealed that the two
groups differed on wives' age and that there was a
trend toward a difference in husbands' age, t(72) =
-2.22, p < .05, and t(72) = -1.96, p -- .053, respectively, with wives and husbands in the control
group being on average 6 years older than the
wives and husbands in the treatment group. Further analyses revealed, however, that neither wives'
nor husbands' age correlated with any of the dependent variables, and therefore there was no need
to use these variables as covariates.
A series of mixed-design ANOVAs with relationship distress, intimacy, acceptance, and motivation
to change serving as dependent measures were conducted to directly compare groups on degree of
pre- to postintervention change. Treatment (MC
vs. control) served as a between-subjects factor and
Time (pre vs. postintervention) served as a withinsubjects f a c t o r . 4 In these analyses the effect of interest is the Treatment x Time interaction term. A significant interaction indicates that the degree of
change in pre- versus posttreatment scores for a dependent measure varied according to treatment condition. These analyses revealed significant Treatment Time interactions for relationship distress,
intimacy, acceptance, and motivation to take direct
action to improve the quality of the relationship
4Analyses including gender as an additional repeated-measures
variable reveal no significant interactions with gender. Therefore,
we continue presenting analyses with couple as the unit of analysis.

Group Membership
, Changein Marital Distress
Direct:,
-.25*
Hediated:
-.08
FIGURE I Change in intimacy as mediator of the association
between treatment group and change in marital satisfaction. Betas
are presented next to the appropriate pathways. * p < .05;
* * * p < .00 I.

(see Table 2). Inspection of the pre- to postintervention means suggested that the degree of positive
change was greater in the MC condition than in the
control condition on each of these variables, providing preliminary evidence in favor of the MC relative to the control condition, s
MEDIATION

OF THE TREATMENT

EFFECT

The fourth hypothesis concerned mechanisms of

change. We hypothesized that changes in intimacy,
motivation to change, and acceptance would each
mediate the association between participation in
treatment and changes in relationship distress. Pre/
post change scores were calculated for relationship
distress, intimacy, areas of change (our acceptance
measure), precontemplation and action. Correlations were calculated between these change scores
and treatment group membership (0 = no treatment,
1 = MC) as a convenient way to assess whether the
assumptions for conducting a mediation regression
analyses were met (Baron & Kenny, 1986). Correlations supported testing change in intimacy as the
sole mediator. Regression analyses were conducted
following Baron and Kenny, regressing the potential mediator (change in intimacy) on the independent variable (treatment group), regressing the dependent variable (change in relationship distress)
on the independent variable, and regressing the dependent variable on both the independent variable
and on the mediator. Results revealed that the association between treatment group and change in
relationship distress was no longer significant after
accounting for the effects of change in intimacy (see
Figure 1). The results suggest that changes in intimacy mediate the association between treatment and
changes in relationship distress.
Sin order to assess whether treatment effects were greater for
distressed versus nondistressed couples as defined by conventional
scores on the GDS, we compared the pre to posttreatment change
scores for distressed versus nondistressed couples and found no significant differences in terms of amount of change between the distressed and nondistressed group.

308

CORDOVA

Discussion
AND

TOLERABILITY

MC

The data presented here suggest that a population of

couples at risk for marital deterioration does exist
and that such couples will participate in the MC.
Comparisons to treatment-seeking and community
samples support the assumption that at-risk couples
are more distressed than community couples, but
less distressed than couples that are actively seeking therapy. Theoretically, at-risk couples are in a
stage between healthy relationship satisfaction and
marital distress severe enough to motivate therapy
or divorce seeking. This is one of the first studies to
suggest that at-risk couples in established marriages
exist in the population and that they can be successfully recruited into a targeted intervention (Cordova
et al., 2001). Additionally, these data demonstrate
that the format of an intervention like the MC is
easily tolerated by participants, resulting in very
high participation rates and no dropout. Thus, it
appears that the MC has the potential to reach and
deliver services effectively to a population of atrisk couples that are unlikely to otherwise seek or
receive early intervention.
THE

EFFICACY

BRIEF

AL.

MECHANISMS

THE ATTRACTIVENESS
OF THE

ET

OF THE

M C AS A

INTERVENTION

The data suggest that the MC may effectively provide a quick boost to the relationship satisfaction
of otherwise treatment-avoidant couples. This quick
boost may ultimately prove to be an important
component of an intervention designed to prevent
future relationship deterioration in at-risk couples,
by contributing to partners' motivation to work collaboratively toward stable marital health. In this
study, participation in the MC appeared to quickly
promote broad improvements in marital health, including a general sense of improved relationship
satisfaction, feelings of deeper intimacy, a greater
acceptance of partners for each other, and an increase in motivation to actively attend to the quality of the relationship.
Compared to MC couples, control couples actually
reported a decrease in marital intimacy over time. On
the one hand, this decrease may reflect the continuing decline in relationship health that theoretically
characterizes at-risk couples. On the other hand,
this decrease may reflect a negative reaction to confronting relationship distress outside the context of
an active intervention like the MC. This potentially
iatrogenic effect should be taken into account in
future research in this area, as assessment without
intervention may cause undue stress on already atrisk relationships.

OF CHANGE

Mediation analyses supported the intimacy theory

of change. Specifically, intimacy theory suggests
that events that increase opportunities for engaging
in interpersonally vulnerable behavior can set in motion those processes that develop and sustain more
stable intimate partnerships and deeper feelings of intimate safety that in turn contribute to partners' relationship satisfaction (Cordova & Scott, 2001). The
MC is designed to foster those intimacy processes
by uncovering interpersonal vulnerabilities at the
heart of partners' most pressing issues and helping
couples to develop a more compassionate understanding of each other and the relationship as a
whole.
Neither changes in acceptance nor motivation to
change mediated the treatment effect. Future research will explore the potential of both of these
processes as mechanisms of change given their centrality to the premises of the MC and the modest
sample size of the current study.
FUTURE

DIRECTIONS

Clearly this study provides only preliminary evidence for the efficacy of the MC as an indicated preventative intervention. Work remains to be done to
determine how to maximize the potential strengths
of this type of intervention approach and minimize
the potential weaknesses. However, at this early
stage, it appears to be a fruitful avenue to pursue as
this early evidence suggests this approach may be
providing marital health benefits that might otherwise be absent from the lives of at-risk couples. Couples attracted to participate in the MC pilot studies
had not ever previously engaged in any form of help
seeking for their relationship and were not, at the
time of first contact, interested in any form of tertiary treatment. Given that, the number of positive
relationship changes initiated by participation in the
MC likely benefited the overall health of these marriages more than no intervention at all.
Our plans for future studies include the addition
of an attention control condition to actively control
for demand and placebo effects and to more directly
test the motivational, intimacy-facilitating, and
acceptance-promoting change mechanisms. In addition, future studies will include greater scaffolding
of couples' motivation to take action to improve
their marital health. For example, couples that indicate during their feedback session that they might be
interested in pursuing a recommendation for marital therapy would receive assistance from the consultant in making the initial phone calls to recommended therapists before leaving the office. Finally,
future studies will include a booster checkup to

THE

MARRIAGE

further assess the quality of the relationship, to follow up on initial recommendations, and to reinforce efforts toward greater marital health.
LIMITATIONS

One limitation of this study is the lack of ethnic diversity, limiting its generalizability. In addition, the
sample size was modest, limiting the study's potential to detect smaller effects and to fully test all relevant mechanisms of change. We also introduced at
least one group difference prior to the first assessment by informing couples of their group assignment following their screening, rather than following their completion of the first assessment. Finally,
longitudinal follow-up will be required to test the
durability of the treatment effect versus relapse.
References
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research:
Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.
Beach, S. R. H., Arias, I., & O'Leary, K. D. (1987). The relationship of marital satisfaction and social support to
depressive symptomology. Journal of Psychopathology and
Behavioral Assessment, 8, 305-316.
Buehlman, K., Gottman, J. M., & Katz, L. (1992). How a
couple views their past predicts their future: Predicting
divorce from an oral history interview. Journal of Family
Psychology, 5, 295-318.
Christensen, A., Jacobson, N. S., & Babcock, J. C. (1995).
Integrative behavioral couple therapy. In N. S. Jacobson &
A. S. Gurman (Eds.), Clinical handbook of couples therapy
(pp. 31-64). New York: The Guilford Press.
Cordova, J. V., Gee, C. G., Warren, L. Z., & McDonald, R. P.
(2004). Intimate safety: Measuring the private experience
of intimacy in men and women. Manuscript in preparation.
Cordova, J. V., Jacobson, N. S., & Christensen, A. (1998).
Acceptance versus change interventions in behavioral couples therapy: Impact on couples' in-session communication.
Journal of Marital and Family Therapy, 24, 437-455.
Cordova, J. V., & Scott, R. (2001). Intimacy: A behavioral
interpretation. The Bebavior Analyst, 24, 75-86.
Cordova, J. V., Warren, L. Z., & Gee, C. B. (2001). Motivational
interviewing with couples: An intervention for at-risk couples.
Journal of Marital and Family Therapy, 27, 315-326.
Doherty, W. J., Lester, M. E., & Leigh, G. K. (1986). Marriage
Encounter weekends: Couples who win and couples who
lose. Journal of Marital and Family Therapy, 12, 49-61.
Dorian, M., & Cordova, J. V. (2001). The couples stages of
change questionnaire. Unpublished questionnaire.
Fields, J., & Casper, L. M. (2001). America's families and living arrangements: March 2000. Current Population
Reports, P20-537. Washington, DC: U.S. Census Bureau.
Gee, C. B., Scott, R. L., Castellani, A. M., & Cordova, J. V.
(2002). Predicting 2-year marital satisfaction from part-

CHECKUP

309

ners' reaction to a marriage checkup. Journal of Marital

and Family Therapy, 28, 399-408.
Heyman, R. E., Sayers, S. L., & Bellack, A. S. (1994). Global
marital satisfaction versus marital adjustment: An empirical
comparison of three measures. Journal of Family Psychology, 8, 432-446.
Jacobson, N. S., & Christensen, A. (1998). Acceptance and

change in couple therapy: A therapist's guide to transforming relationships. New York: W. W. Norton.
Margolin, G., Talovic, S., & Weinstein, C. D. (1983). Areas of
change questionnaire: A practical approach to marital
assessment. Journal of Consulting and Clinical Psychology,

51,944-955.
McConnaughy, E. A., Prochaska, J. O., & Velicer, W. E (1983).
Stages of change in psychotherapy: Measurement and
sample profiles. Psychotherapy: Theory, Research & Practice, 20, 368-375.
Mental health: Does therapy help? (1995, November). Consumer Reports, pp. 734-739.
Miller, W. R., & Rollnick, S. (2002). Motivational Interviewing: Preparingpeople for change. New York: The Guilford
Press.
Prochaska, J. O., & DiClemente, C. C. (1984). The transtheo-

retical approach: Crossing traditional boundaries of therapy. Homewood, IL: Dow Jones/Irwin.
Schaefer, M. T., & Olson, D. H. (1981). Assessing intimacy:
The PAIR inventory. Journal of Marital and Family Therapy, 7, 47-60.
Snyder, D. K. (1979). Multidimensional assessment of marital
satisfaction. Journal of Marriage and the Family, 41, 813823.
Snyder, D. K. (1997). Marital satisfaction inventory, revised.
Los Angeles: Western Psychological Services.
Veroff, J., Douvan, E., & Kulka, R. A. (1981). The innerAmerican: A self-portrait from 1957 to 1976. New York: Basic.
Weiss, R. L., & Cerreto, M. C. (1980). The Marital Status
Inventory: Development of a measure of dissolution potential. The American Journal of Family Therapy, 8, 80-86.
Weiss, R. L., Hops, H., & Patterson, G. R. (1973). A framework for conceptualizing marital conflict, a technology for
altering it, and some data for evaluating it. In L. D. Handy
& E. L. Mash (Eds.), Behavior change: Methodology, concepts and practice (pp. 309-342). Champaign, IL: Research
Press.
Weissman, M. M. (1987). Advances in psychiatric epidemiology: Rates and risks for major depression. American Journal of Public Healtb, 77, 445-451.
Whisman, M. A. (2001). The association between depression
and marital satisfaction. In S. R. H. Beach (Ed.), Marital

and family processes in depression: A scientific foundation

for clinicalpractice (pp. 3-24). Washington, DC: American
Psychological Association.
Worthington, E. L., McCullough, M. E., Shortz, J. L., Mindes,
E. J., Sandage, S. J., & Chartrand, J. M. (1995). Can couples assessment and feedback improve relationships? Assessment as a brief relationship enrichment procedure. Journal

of Counseling Psychology, 42,466-475.

RECEIVED: March 15, 2004
ACCEPTED: December 17, 2004