aliança terapêutica 2

8/3/2019 aliana teraputica 2

1/15

J Psychother Pract Res, 9:1, Winter 2000 39

Participants Perceptions of Dimensions of theTherapeutic Alliance Over the Course ofTherapy

Alexandra Bachelor, Ph.D.Ramzi Salame, Ph.D.

Received April 22, 1999; revised August 6, 1999; accepted August 19, 1999. From Universite Laval, Quebec, Canada; Dr. Salame is nowat the UNESCO Regional Ofce, Beirut, Lebanon. Address corre-spondence to Dr. Bachelor, Ecole de Psychologie, Universite Laval,Quebec, Canada G1K 7P4.

Copyright 2000 American Psychiatric Association

The course of diverse dimensions of the therapeutic alliance as seen by the therapy participants was examined in two studies. In Study 1, use of the Penn Helping Alliance Method, Therapeutic Alliance Rating System, and Vanderbilt Psychotherapy Process Scale at the 3rd, 10th, and next-to-last therapy sessions revealed signicant increases over time for therapists mean scores on 5 of the 12 subscales. Consistency of participants scores over time varied depending on rater

perspective (therapist or client) and time interval. In Study 2, the Helping Alliance Questionnaire, Working Alliance Inventory, and California Psychotherapy Alliance Scales showed no signicant change in participants average alliance scores on the 9 subscales from the 5th to 10th sessions, and their same-scale ratings were in general consistent. Few differences were observed between therapists and clients average alliance ratings, but the relationship was generally perceived differently within dyads.

(The Journal of Psychotherapy Practice and

Research 2000; 9:3953)

The therapeutic alliance of client and therapist hasbeen the focus of considerable research attentionover the past 15 to 20 years. Bordins1 writings on thegeneralizability across therapy systems of this originallyanalytic concept, together with increasing interest in thealliance as an integrative therapeutic factor,24 stimu-lated intensive research into the role of the alliance intherapy (see Horvath and Symonds5 for a review of thealliance-outcome relationship). More recently, attentionhas turned to the examination of other relevant issues,

including the developmental course of the alliance overtherapy. The stability or variability of the alliance hasimportant clinical as well as empirical implications, in-cluding whether increased focus on the relationship iswarranted at specic time points in therapy andwhether commonly used one-time assessments can beassumed to represent the quality of the alliance through-out therapy.

From a conceptual standpoint, Luborsky6 viewedthe helping alliance as a dynamic, rather than static,phenomenon that is responsive to the changing de-mands of different phases of therapy. The experience of

the therapist as helpful and supportive, one aspect of the alliance, is held to be more evident at the beginning


2/15

Perceptions of Therapeutic Alliance

40 J Psychother Pract Res, 9:1, Winter 2000

of therapy, whereas the sense of shared responsibilityregarding treatment goals, a second aspect, is more typ-ical of later phases of treatment. According to Bordin,7

the working alliancedened in terms of mutuality of goals and tasks supported by the client-therapist bondgrows and deepens with time as inevitable strains ordisruptions are worked through and repaired.

Also of the view that the alliance varies over time,Gelso and Carter8 proposed that its course during ther-apyparticularly in time-limited interventionsin-volves a weakening after an initial development,followed in successful therapy by an increase to earlier,high levels. Along similar lines, Horvath et al.9 positedan initial phase of development for the alliance, held tooccur within the rst ve therapy sessions (and proba-bly peaking during the third session), followed by a sec-ond, more critical phase, during which the therapist

challenges maladaptive patterns, the effect of which isa weakening or rupturing of the alliance that must berepaired if therapy is to continue successfully. On theother hand, Greenberg 10 challenged the high-low-highpattern suggested by these authors, maintaining that insuccessful therapies the alliance either rises or holds a steady value over time. In sum, although consensus hasnot been reached on the specic pattern of develop-ment, several theorists view the alliance, or individualaspects of the alliance, as shifting over the course of therapy.

Relatively few studies have specically investigated

the time course of the alliance. Three recent studiestested the rupture/repair, or U-shaped pattern, pro-posed by Bordin7 and Gelso and Carter8 to characterizethe course of the alliance in therapy. Horvath andMarx,11 examining the working alliance (as assessed bythe Working Alliance Inventory; WAI) at the session-by-session level in 4 clients and their 2 therapists, foundthat therapists combined ratings increased over the rst 5 of 10 therapy sessions and then declined slightly, torecover marginally after the 7th session. Although theseresults suggested a cyclical developmental model, cli-ents alliance ratings showed, to the contrary, a linear

increase over the therapy. Golden and Robbins,12

in a similar study, reported the reverse pattern: the thera-pists working alliance scores gradually increased overthe 12 therapy sessions, whereas his two clients scoreswere lowest during the middle phase (5th8th sessions)of therapyand in one client, scores remained at thislevel on the task and goal alliance subscales. Kivlighanand Shaughnessy,13 in contrast, demonstrated that a di-

rect linear pattern, as opposed to a quadratic (cyclical)pattern, best accounted for both client and therapist working alliance scores over therapy. Sexton et al.,14

although not specically testing the cyclical model of development, reported, in a sequential study that fo-cused on the clients perspective, that the working alli-ancewas largely determined by the rst therapysession.

A high-low-high pattern of alliance development does not appear, then, to be strongly supported by thedata. However, there is some evidence of different de-velopmental patterns over time among therapists andclients alliances.

Some studies, though not focused specically onthe course of the alliance, have employed multiple as-sessments of the client-therapist relationship or its com-ponents and have tested for time differences. Althoughndings remain inconclusive, several results suggest dif-

ferential temporal patterns for specic components of the therapeutic relationship, with positive-toned fea-tures showing greater variation than negative facets. Forexample, using the Therapeutic Alliance Rating Sys-tem, Marziali15 found early session mean ratings forpositive patient and therapist contributions to the alli-ance to be signicantly lower than nal session scores;negative contributions, however, showed no changeover the course of therapy. Similarly, Klee et al.,16 ex-amining positive patient alliance contributions, re-ported an increase in these from early to late sessions.Using a factor-analytic version of the same measure (theCalifornia Therapeutic Alliance Rating System), Mar-mar et al.,17 however, found both positive and negativepatient contributions to the alliance, such as commit-ment and hostility, to increase over therapy, whereastherapist negative behaviors did not vary. An increasefrom early to late sessions in clients positive scores withno change in negative scores was also reported by Lu-borsky et al.18 in more improved patients, using thePenn HelpingAlliance (counting signs) Method. In con-trast, in Hartley and Strupp,19 last-session, comparedwith rst-session, alliance ratings of both positive and

negative patient components (e.g., motivation, resis-tance), as assessed by the Vanderbilt Therapeutic Alli-ance Scale, showed a signicant decrease regardless of outcome status. Still other studies, using these and othermeasures, found no differences in levels of the allianceor its components attributable to phase of therapy.2023

The consistency of the alliance has been examinedin a few studies. Moderate to high correlations between


3/15

Bachelor and Salame


early and late sessions have been reported for the Pennscale (counting signs and rating methods) and the WAIscale,18,22,24 suggesting relative stability of these vari-ables. Other results,25 however, indicated substantialchanges in the Penn characteristics between these timepoints.

Clearly, more work is needed on how the compo-nents of the therapeutic relationship evolve over time.Furthermore, the above results are, in general, based onindependent observer assessments of the therapeutic al-liance, which represent but one perspective on the cli-ent-therapist relationship. Indeed, different evaluativesources (observer, therapist, client) are viewed as pro-viding unique, nonsubstitutable information on therelationship.3,26,27 Several studies have in fact substan-tiated the low convergence of therapists and clientsperceptions of alliance variables,11,12,23,28,29 supporting

separate study of the therapy partners alliances.Given that the therapy participants are privy to a wealth of information at the emotional and empathiclevels30 and may provide the more authentic versionsof the quality of the treatment relationship,31 partici-pants evaluation of the alliance warrants more atten-tion. Empirical data on the time course of the allianceas seen by the therapy participants have primarily in-volved the WAI,2 which focuses on therapist-client agreement.32 Little is known about the course of otheralliance ratings and characteristics. Despite some pre-liminary work,23,3234 there is insufcient evidence toconclude the comparability of different alliance mea-sures and components. The use of multiple measures,in capturing different facets of a construct of interest,may also contribute to a broader understanding of theunderlying construct.35

The goal of the current study, then, was to track thecourse of different facets of the therapeutic allianceacross therapy, from the perspective of the therapy par-ticipants. Two studies are presented that examined par-ticipants perceptions by using different alliancemeasures and assessment times. The rst study exam-

ined the alliance at the 3rd, 10th, and next-to-last ses-sions. The second, later study, although not addressing the developmental pattern of the alliance, allowed forstudy of the briefer time segment from the 5th to 10ththerapy sessions. To provide a more thorough account of the time course of relationship components, percep-tions were assessed both between and within partici-pants.

STUDY 1

Participants and Procedure

Participants were 27 Caucasian therapist-client dy-

ads at a university consultation service serving a largefrancophone community and used as a training facilityfor graduate students in clinical psychology. This sam-ple was drawn from the 47 participant dyads of a pre-vious study, the report of which includes additionaldetails on the constitution of the sample.33 However,none of the analyses effected in the present study havebeen previously reported. The length of therapy rangedfrom 15 to 37 sessions, with a mean of 26 6.44 ses-sions. Therapy length was limited to the duration of theacademic year, ending with the summer semester (al-though clients could continue therapy the following

year with another therapist, if desired). Clients wereseen on a once-a-week basis. These clients were 20women and 7 men, with a mean age of 30.15 7.61years. Means and standard deviations are reported.

Sixty-seven percentof clientswere single, 15% weremarried, and 18% were separated or divorced. Sixty-ve percent had completed junior college or had someuniversity education. Fifty-two percent of clients wereunemployed, 18% were students, and 30% held skilledor semiskilled jobs. Clients overall psychological func-tioning, as rated by trainees supervisors on the GlobalAssessment Scale (GAS),36 fell within the range of moderate symptoms or generally functioning withsome difculty (mean 59.19 11.36) and their diag-noses (n 26) included anxiety-related (38%), person-ality-related (31%), and psychosexual-related (8%)disorders, as well as interpersonal problems (23%).Self-ratings on the Psychiatric Symptom Index (PSI), a measure of self-reported psychiatric status,37 indicatedthe presence of perceived high symptomatology(mean 39.37 19.21). Fourteen clients had had priortherapy with other therapists, mostly at this same con-sultation service. Clients were seen by 17 masters-level

trainees in a rst-year clinical psychology practicum(12women and 5 men; mean age 31.35 6.14 years). Ninetherapists saw 1 client each, six therapists saw 2 clients,and two saw 3 clients. Therapists therapeutic orien-tation, based on their supervisors self-describedapproach, was predominantly humanistic or humanis-tic-existential (70% of therapists); other orientations in-cluded bioenergetic (18%) and psychodynamic (12%).


4/15



Consenting clients, who had been informed at phone registration of an ongoing research project, weremet individually prior to their initial appointment by a research assistant to complete demographic and con-sent forms and the PSI as well as other outcome mea-sures. Alliance ratings were subsequently collected onan individual basis by a research assistant following cli-ents 3rd, 10th, and next-to-last sessions. Therapistscompleted alliance measures at assessment phases cor-responding to those of clients. Because of administrativeslips, cancellations, or other delays, in a few cases client or therapist ratings of adjacent sessions were collected.Trainees supervisors, all clinical psychology facultymembers, provided GAS ratings and diagnoses shortlyafter trainees initial sessions with clients (on the basisof verbal and audiotaped material presented at weeklysupervision). Clients, therapists, and supervisors com-

pleted various outcome measures, including the GASand PSI, generally at 2 weeks following termination.Signicant improvement from intake to terminationwas observed on all measures in the original sample.33

Alliance Measures

The Penn Alliance Rating Method (Penn)22 com-prises ten 10-point Likert-type items, six of which mea-sure the patients experience of receiving help or a helpful attitude from the therapist (HA 1) and four of which measure the patients experience of being in-volved in a joint or team effort with the therapist (HA2). Evidence of the scales internal consistency and pre-dictive validity is reported by the authors.22

The Vanderbilt Psychotherapy Process Scale(VPPS)21 comprises 44 5-point Likert-type items assess-ing seven dimensions of therapist and patient attitudesand behaviors: Patient Exploration (PEXP), Therapist Exploration (TEXP), Patient Participation (PPAR), Pa-tient Hostility (PHOS), Therapist Warmth and Friend-liness (TWFR), Negative Therapist Attitude, andTherapist Directiveness (TDIR). Adequate internal con-

sistencies of individual subscales, and the predictive va-lidity of a broad dimension of patient involvement(comprising two patient subscales) have been demon-strated.21,38 In the current study, the Penn and VPPSitems, originally developed for external observer eval-uation, were reformulated, following Marziali,15 by al-tering pronouns to suit the perspectives of client andtherapist. Unfortunately, the self-report versions (client

and therapist) of the Penn items39 were not available ina French version at the time Study 1 was conducted.

The Therapeutic Alliance Rating System (TARS),15

in patient and therapist versions, comprises 42 items, of which 21 evaluate the therapists positive and negativecontributions to the alliance (TPOS, TNEG, respec-tively) and 21 evaluate the patients positive and nega-tive contributions (PPOS, PNEG, respectively).Marziali15 reported satisfactory internal consistenciesfor the subscales, as well as good discriminant and pre-dictive validity. In the present study, a 5-point Likert-type scale was substituted for the original visual analog format, using the original end points (not at all and a lot).

This study employed the French versions of thethree alliance scales used in Bachelor,33 for which ade-quate reliabilities (coefcient alpha) were obtained

(ranging from 0.68 to 0.87) for the individual subscalesexcepting the VPPS scales for Negative Therapist At-titude, which was deleted from further analysis, and Pa-tient Hostility ( 0.54).

Results

Given the variation in length of treatment acrossclients, preliminary analyses were conducted to ascer-tain the effect of number of sessions on participantsalliance perceptions. No signicant Pearson (Bonfer-roni-adjusted, P 0.00l) correlations were found be-

tween individual client-rated or therapist-rated alliancecharacteristics and number of sessions at any of thethree assessment points. Further, when we divided theclient sample into two groups, with 21 3.23 (n 14)and 31.92 3.40 (n 13) mean therapy sessions, re-spectively, average levels of the individual alliance char-acteristics, either client- or therapist-rated, were not found to differ signicantly between the two groups, re-gardless of assessment time.

Taken together, these results indicate that treatment length did not inuence participants alliance percep-tions. Thus, the two client (and corresponding therapist)

subsamples were pooled in further analyses.Preliminary analyses further showed no signicant

differences between male and female subjects either onthe demographic variables or on the variable of priorconsultation. Clients gender and age were also unre-lated to their alliance ratings at all assessment points. Inaddition, pretherapy GAS and PSI scores, as well asclient diagnoses (categories involving 23% or more of


5/15

Bachelor and Salame


TABLE 1. Means, standard deviations, and comparison of alliance subscales at three assessment times ( N 27)

Scores (Mean SD) F -RatioAllianceDimensions

3rdSession

10thSession

Next-to-Last Session Group Time

GroupTime

Penn Helping Alliance MethodHA 1 Cl 41.81 8.87 44.04 9.44 44.96 7.84 4.47 8.93a 0.058

Th 36.70 7.28 40.67 8.08 41.89 5.89HA 2 Cl 27.37 6.89 28.30 6.80 29.48 5.64 6.43 8.60a 1.09

Th 22.59 6.17 24.70 6.38 27.04 5.21Vanderbilt Psychotherapy Process Scale (VPPS)

PEXP Cl 27.33 4.30 27.74 4.54 27.81 3.45 39.06a 1.15 0.32Th 20.67 5.22 22.04 5.31 21.48 3.72

TEXP Cl 26.26 5.60 28.93 6.23 28.22 5.01 2.95 11.32a 0.38Th 23.81 4.60 26.74 4.24 26.85 4.54

PPAR Cl 26.07 5.20 28.74 4.24 28.67 3.87 2.89 5.94a 0.66Th 25.22 4.60 26.30 4.78 27.07 4.73

PHOS Cl 11.15 2.98 9.74 2.93 9.93 2.38 1.81 7.46a 1.43Th 12.22 3.51 11.30 3.81 9.96 2.83

TWFR Cl 33.67 5.66 35.11 6.40 34.70 5.58 4.69 2.94 0.02Th 30.78 4.59 32.30 5.49 32.04 4.84

TDIR Cl 8.67 3.56 8.63 3.16 10.44 3.66 0.26 6.81a 0.57Th 8.81 3.29 9.63 2.92 10.55 2.90

Therapeutic Alliance Rating Scale (TARS)PPOS Cl 38.81 8.04 42.15 7.91 42.48 7.28 5.34 16.81a 1.84

Th 32.85 8.02 37.96 7.92 40.37 7.43PNEG Cl 18.30 5.71 17.44 3.67 18.11 4.91 1.96 0.03 0.80

Th 19.41 5.81 20.11 5.53 19.22 5.47

TPOS Cl 38.52 7.69 42.81 8.35 41.30 8.21 0.51 13.15a

0.87Th 37.33 5.31 40.56 6.02 41.15 5.89TNEG Cl 12.70 2.84 12.74 2.92 12.96 3.22 51.03a 0.59 0.80

Th 17.48 3.91 18.48 3.35 17.55 3.52

Note: Cl client; Th therapist; HA 1 helping alliance type 1; HA 2 helping alliance type 2; PEXP Patient Exploration;TEXP Therapist Exploration; PPAR Patient Participation; PHOS Patient Hostility; TWFR Therapist Warmth and Friendliness;TDIR Therapist Directiveness; PPOS patient positive; PNEG patient negative; TPOS therapist positive; TNEG therapist negative(contributions to alliance).a P 0.004 (Bonferroni-adjusted).

clients), were unrelated to either therapists or clients3rd-session alliance assessments (allr -and F -valuesnon-signicant at the Bonferroni-adjusted level,P 0.001).

To determine whether alliance ratings differed overtime and between rater perspectives, two-factor analy-ses of variance, with time (3rd vs. 10th and next-to-last sessionsa repeated measure) and group (client vs.therapist) as the independent variables, were per-formed on each of the 12 alliance variables. Observedmeans, standard deviations, and F -ratios are presentedin Table 1.

As can be seen from Table 1, 8 of the 12 alliancedimensions demonstrated signicant time effects (using a Bonferroni-adjusted signicance level of 0.004 tocounter familywise error) in the combined ratings of

clients and therapists: the Penn HA 1 (therapist-offeredhelpfulness and support) and HA 2 (joint collaborationin the work of therapy); the VPPS TEXP (Therapist Ex-ploration) and TDIR (Therapist Directiveness); PPAR (Patient Participation) and PHOS (Patient Hostility);and the TARS PPOS (patient positive contributions)and TPOS (therapist positive contributions). Signicant effects for group were found on the VPPS PEXP (Pa-tient Exploration) and on the TARS TNEG (therapist negative contributions) scales. No group-by-time inter-actions were observed.

Repeated-measures analyses of variance performedseparately on clients and therapists ratings on the eight scales displaying time effects indicated that the ob-served overall time effects were mainly attributable to


6/15



TABLE 2. Pearson same-scale correlations between assessment points ( N 27)

3rd10th Sessions 10thNext-to-Last Sessions 3rdNext-to-Last SessionsScale Therapist Client Therapist Client Therapist Client

HA 1 0.55b 0.45 0.75a 0.71a 0.59a 0.41HA 2 0.47 0.62a 0.54b 0.79a 0.43 0.51PEXP 0.32 0.57a 0.65a 0.71a 0.38 0.64a

TEXP 0.50 0.63a 0.58a 0.50 0.52 0.64a

PPAR 0.60a 0.45 0.41 0.27 0.18 0.39PHOS 0.43 0.70a 0.53 0.30 0.36 0.07TWFR 0.66a 0.55b 0.81a 0.68a 0.68a 0.47TDIR 0.54 0.42 0.51 0.16 0.44 0.22

PPOS 0.59a 0.63a 0.70a 0.54b 0.51 0.31PNEG 0.69a 0.47 0.55b 0.36 0.42 0.51TPOS 0.74a 0.71a 0.59a 0.70a 0.50 0.59a

TNEG 0.53 0.18 0.57a 0.26 0.43 0.42

Note: HA 1 helping alliance type 1; HA 2 helping alliance type 2; PEXP Patient Exploration; TEXP Therapist Exploration;PPAR Patient Participation; PHOS Patient Hostility; TWFR Therapist Warmth and Friendliness; TDIR Therapist Directiveness;PPOS patient positive; PNEG patient negative; TPOS therapist positive; TNEG therapist negative (contributions to alliance).a P 0.002 (Bonferroni-adjusted);bP 0.003.

changes in therapists perceptions. As indicated by post hoc contrasts, therapists mean scores on the ve scalesHA 1, HA 2, TEXP, PPOS, and TPOS showed signi-cant differences between the 3rd and next-to-last ses-sions, and their mean scores on the latter three scales(TEXP, PPOS, TPOS) also differed signicantly be-tween the 3rd and 10th sessions. All differences were inthe direction of increased scores. No signicant effectsfor time were observed for clients ratings of these eight alliance dimensions. Despite observed signicant over-all time effects for the VPPS TDIR, PPAR, and PHOSscales, no signicant effects were found once clientsand therapists scores were examined separately. Withregard to observed main effects for group, Newman-Keuls comparisons indicated that clients gave higherratings than therapists on the VPPS PEXP and lowerratings on the TARS TNEG at each of the 3 assessment

points.As shown in Table 2, Pearson correlations com-puted between therapists 3rd- and 10th-session same-scale ratings on each of the 12 alliance dimensionsrevealed signicant coefcients for ve scales (Bonfer-roni-adjusted level of 0.002): the VPPS PPAR andTWFR and the TARS PPOS, PNEG, and TPOS; thecorrelation for the Penn HA 1 was near-signicant. Be-tween the 10th and next-to-last sessions, seven same-scale correlations were signicant: the Penn HA 1, theVPPS PEXP, TEXP, and TWFR, and the TARS PPOS,

TPOS, and TNEG; the same-scale correlations for HA2 and PNEG were near-signicant. Thus, across the twotime periods, consistency was observed for ratings of TPOS, PPOS, and TWFR, while ratings of HA 1 andPNEG were fairly consistent. Between the 3rd and next-to-last sessions, only two same-scale correlations (HA 1and TWFR) were signicant.

As also shown in Table 2, correlational analyses of clients same-scale ratings on each of the 12 alliancedimensions indicated six signicant Pearson correla-tions between ratings at the 3rd and 10th sessions: HA2, PEXP, TEXP; PHOS, PPOS, and TPOS; the corre-lation for TWFR was near-signicant. Five signicant same-scale correlations were found between 10th andnext-to-last session ratings: HA 1, HA 2, PEXP, TWFR,TPOS; the correlation for PPOS was near-signicant.Across both time periods, consistency of ratings was ob-served for HA 2, PEXP, and TPOS; ratings of TWFR and of PPOS were fairly consistent. Between the 3rdand the next-to-last sessions, only three correlations(PEXP, TEXP, TPOS) computed on same-scale ratingsproved signicant.

STUDY 2

Participants and Procedure

Participants were 30 Caucasian French-speaking therapist-client dyads drawn from three different sites


7/15

Bachelor and Salame


(the same university consultation service as in Study 1,site A,n 8; a community services center, site B,n 10;and private practice, site C, n 12). Clients were 18women and 12 men (mean age 35.10 9.47 years),of whom 40% were single, 30% were married or livedwith a partner, and 30% were separated, divorced, orwidowed. Forty-four percent of clients held skilled orsemiskilled jobs, 23% held professional or minor pro-fessional positions, and the remainder were students orunemployed. Fifty-seven percent held junior college oruniversity diplomas. Diagnoses of the clients (n 29)included anxiety-related (35%), dysthymic-related(10%), and personality-related (17%) disorders, as wellas interpersonal problems (38%). Clients level of psychological functioning (GAS),36 as provided bytheir therapists, fell in the range of moderate symp-toms or generally functioning with some difculty

(mean 59.46 8.37). Clients self-reported symptom-atic status (PSI)37 suggested the presence of perceivedhigh symptomatology (mean 42.34 16.90). Clientswere seen, in general, on a weekly basis, and were intherapy for an average of 18 5.63 sessions (range 929). Half of these clients had previous counseling ex-perience.

Therapists, who were solicited either individuallyor at staff meetings to collaborate in a study on the cli-ent-therapist relationship, were 12 women and 8 men(mean age 35.9 9.77 years), representing 54%, 68%,and 86% of solicited therapists in sites A, B, and C,

respectively. The majority (65%) described their theo-retical orientation as humanistic or humanistic-existen-tial; other approaches included Gestalt, eclectic,bioenergetic, and transpersonal. Fifteen therapists saw1 client, two therapists saw 2 clients, one therapist saw3 clients, and two therapists saw 4 clients each. Threetherapists were licensed psychologists (2.517 yearspost-M.A. experience), one was a licensed socialworker, one a registered nurse (11 and 15 years of coun-seling experience, respectively), 13 were clinical orcounseling psychology practicum students (11 at themasters and 2 at the bachelors level), and two were

screened volunteer helpers at the community servicescenter, with undergraduate university diplomas but noformal counseling training. Prior results indicated nosignicant relationship between therapist experiencelevel and the alliance dimensions currently examinedin a subset of this sample (A. Bachelor, unpublishedreport).

Consenting therapists, who were asked to recruit

from one to three incoming clients (two therapists pro-vided more) to participate, gave clients at their initialtherapy session the research packet (containing a con-sent form assuring anonymity and condentiality of re-sponses, a demographic information sheet, and thePSI), to be returned in a sealed envelope at the follow-ing session. Virtually all solicited clients in sites B andC, and 58% in site A, agreed to participate. At the 5thand 10th therapy sessions, therapists gave participating clients the alliance measures in a sealed packet that con-tained instructions to return the completed question-naires, sealed in the accompanying envelope, at theirnext session. Thirty-ve clients provided data for thetwo assessment times. Of these 35 clients, 5 outliers,who had an average of 55 therapy sessions, were ex-cluded, thus yielding the current sample of 30 clients.Therapists completed the GAS and provided client di-

agnoses following the 2nd therapy session, and theycompleted the alliance measures after the 5th and 10thsessions. Prepost differences in clients self-rated symp-tomatic status (PSI) and in therapists ratings of clientslevel of functioning (GAS)both measures completed,in general, 2 weeks following terminationindicatedsignicant client change (t 3.27, df 27, P 0.01, andt 8.37, df 24, P 0.001, respectively). Both clientsand therapists rated therapy as quite helpful on a 9-point post-therapy index of overall change40 (ranging from extremely harmful to extremely helpful;means: 7.28 1.22 and 7.11 0.87, respectively).

Alliance Measures

The Helping Alliance Questionnaire (HAQ)39 as-sesses two types of helping alliance in 11 items, 8 of which reect perceived helpfulness and support of thetherapist (HA 1) and 3 of which reect the patients col-laboration with the therapist toward the goals of treat-ment (HA 2). Its therapist counterpart, the Therapist Facilitating Behavior Questionnaire, is an 11-item scaleassessing the same two types of alliance. Both scaleswere shown to predict therapy outcome as assessed by

various indices.41

For purposes of statisticalcalculations,the original scoring scheme ( 3 to 3) was transformedinto a positive (1 to 6) scale.

The Working Alliance Inventory (WAI), short form,42 is a 12-item, 7-point, Likert-type measure of thethree alliance dimensions of agreement on tasks(TASK), agreement on goals (GOAL), and develop-ment of bonds (BOND), with parallel forms for client


8/15



and therapist that consist of each of the four highest-loading items from the original3 12-item subscales. Tra-cey and Kokotovic42 report results that support theconstruct validity of the WAI-short form, as well as highinternal consistency estimates for its three subscales.

The California Psychotherapy Alliance Scales(CALPAS)43 (C. Marmar and L. Gaston, unpublished),patient and therapist versions, comprise 24 7-point items assessing four scales: Patient Working Capacity(PWC), Patient Commitment (PC), Working StrategyConsensus (WSC), and Therapist Understanding andInvolvement (TUI). Gaston43 reported satisfactory re-liability ( 0.83) for the total patient scale, as well asevidence of criterion-related validity.

This study used the French translation of the CAL-PAS patient scale (C. Marmar and L. Gaston, unpub-lished) and French translations of the other alliance

scales made by two experienced bilingual clinicians.Es-timates of internal consistencies of the scales, computedon 5th-session ratings of the current sample, were ade-quate, ranging, for therapist ratings from 0.66 to 0.91,and for client ratings from 0.69 to 0.93 with the excep-tion of the HAQ HA 2 and the CALPAS PWC scales(means 0.57 and 0.46, respectively). The latter sub-scale, as well as the other three CALPAS subscales,showed comparatively higher reliabilities in the current study than the original estimates reported.43 The lowestimate for HA 2 could be a result of the limited num-ber of items (three). Results obtained with regard to

these two subscales should thus be viewed with caution.

Results

As in Study 1, because of the range in number of sessions across clients, preliminary analyses were con-ducted to assess the effect of number of sessions onparticipants alliance ratings. No signicant (Bonferroni-adjusted, P 0.003) correlations were observed betweenclients or therapists alliance ratings and number of ses-sions at either assessment point. Further, when we di-vided the client sample into two groups, with a mean of

21.62 4.77 sessions (n 16) and 13.29 2.13 sessions(n 14), respectively, mean client and therapist ratingsof the two groups on the nine alliance characteristicswere not found to differ signicantly, either at the 5thor the 10th therapy session. The two client subsamples(and the corresponding therapist subsamples) were thuscombined in further analyses.

Preliminary analyses also indicated no signicant

differences among the three research sites with regardto clients age, sex, and marital status, as well as initialpsychological status (PSI and GAS scores). Also, maleand female clients did not differwith regard to age, mar-ital status, GAS, or PSI scores. Further, clients gender,age, and diagnoses (categories involving 17% or moreof clients) were unrelated to either therapist- or client-rated alliance dimensions (allr - and F -values nonsignif-icant at the Bonferroni-adjusted level of 0.005).

Two-factor analyses of variance, with one repeatedmeasure (time) and one between-subjects factor (client vs. therapist), conducted on each of the nine alliancevariables and using a Bonferroni-corrected alpha of 0.005, yielded a near-signicant (P 0.006) time effect (combined client and therapist ratings) for the CALPASPWC scale, and signicant group effects for the CAL-PAS TUI and WSC scales. No group-by-time interac-

tions were observed. Scale means, standard deviations,and F -ratios are reported in Table 3.Tests of the difference between means, conducted

separately on clients and therapists PWC scores, in-dicated that the overall near-signicant effect was dueto a difference in therapists ratings between the 5th and10th sessions (t 2.90, df 29, P 0.02). With regard toobserved group effects, inspection of mean scores in-dicated that clients gave higher ratings than therapistson the two variables of TUI and WSC both at the 5thand 10th therapy sessions.

Pearson correlations computed between partici-

pants 5th- and 10th-session same-scale ratings of thenine alliance characteristics revealed that among ther-apists, six of the nine same-scale coefcients proved sig-nicant: the Penn HA 1, the WAI Goal and Bond, andthe CALPAS TUI, PC, and WSC (range 0.510.72, allP 0.005, Bonferroni-adjusted). Same-scale coefcientsfor the Penn HA 2, the WAI Task, and the CALPASPWC were nonsignicant. Among clients, all but one(the CALPAS WSC) of the nine same-scale associationswere signicant (range 0.520.76).

DISCUSSION

The current ndings support, in general, the view that individual characteristics of the therapeutic relationshipevolve differentially over time. The variability or sta-bility of relationship characteristics was specically re-lated to rater perspective (therapist or client) and timephase, as well as data-analytic procedure (between- orwithin-subjects analyses).


9/15

Bachelor and Salame


TABLE 3. Means, standard deviations, and comparison of alliance subscales at two assessment times

Scores (Mean SD) F -RatioAlliance Dimensions 5th Session 10th Session Group Time Group Time

Helping Alliance QuestionnaireHA 1 Cl 41.43 4.88 41.53 5.08 2.45 1.72 1.26

Th 39.00 5.04 40.30 4.89

HA 2 Cl 12.77 2.74 13.30 2.52 0.61 4.83 0.32Th 13.07 2.93 13.97 2.61

Working Alliance InventoryBOND Cl 23.21 4.62 23.76 4.82 0.01 0.00 1.41

Th 22.67 2.62 22.17 2.65TASK Cl 22.69 5.88 22.03 4.86 1.07 0.28 2.99

Th 20.57 4.77 22.80 3.72GOAL Cl 22.24 4.69 22.24 4.44 1.11 0.49 0.49

Th 20.83 4.50 21.48 3.66California Psychotherapy Alliance Scales

TUI Cl 6.49 0.68 6.56 0.59 17.27a 2.77 0.52Th 5.81 0.69 5.98 0.61

PWC Cl 5.38 0.86 5.48 0.84 3.29 7.85b 4.07Th 4.72 1.09 5.34 1.08

PC Cl 6.06 0.88 6.03 0.81 5.88 3.56 4.46Th 5.25 1.30 5.47 0.87

WSC Cl 6.20 0.93 6.17 0.75 12.86a 2.18 3.18Th 5.29 0.98 5.67 0.81

Note: Study 2; above,n 29/30 because of missing data. Cl client; Th therapist; HA 1 helping alliance type 1; HA 2 helping alliancetype 2; TUI Therapist Understanding and Involvement; PWC Patient Working Capacity; PC Patient Commitment; WSC Working Strategy Consensus.a P 0.005 (Bonferroni-adjusted);bP 0.006.

Therapists Alliance Perceptions Across Therapy

In Study 1, therapists average levels of ve posi-tive-toned relationship characteristicstheir helpful-ness and support (HA 1), joint work efforts with clients(HA 2), and exploration of underlying dynamics(TEXP), as well as their own and their clients positiveattitudes and affect (TPOS, PPOS)increased signi-cantly from the near-beginning to the near-conclusionof therapy. The latter three characteristics (TEXP,TPOS, PPOS) also showed signicant increases fromthe 3rd to the 10th therapy session. Contrary to theory,6

the alliance component of helpfulness and support (HA1), and not only the sense of shared responsibility inachieving treatment goals (HA 2), was stronger later intherapy, as judged by therapists. Also, contrary to data drawn from observer ratings,21 exploratory, uncovering

interventions (TEXP) also improved over time in theperspective of therapists. Consistent with Marziali,15

positive but not negative therapist and patient attitudesand affect (TNEG, PNEG) increased over time.

Between the 10th and near-nal sessions, however,no signicant change in therapists average perceptionswas observed. As indicated by the ndings of Study 2,which examined the more specic time segment fromthe 5th to 10th therapy sessions, group changes werenot observed in therapists perceptions after the 5th ses-sion (with one exception, clients working capacity[PWC], e.g., self-disclosure and self-exploration, whichtended to improve).

Results of intra-individual analyses, however, indi-cated differences among therapists in their perceptionsover time of most of the relationship characteristics, in-cluding those that showed improvement in therapists asa group, particularly between the 3rd and 10th and the3rd and next-to-last therapy sessions. (In these two timeintervals, 50% and 83%, respectively, of the variables of Study 1 were not rated consistently.) When these cor-

relational results were considered together with thendings based on mean ratings, over the 3rd to the 10ththerapy sessions, perceptions of their uncovering inter-ventions (TEXP) proved variable among therapists:


10/15



some perceived such efforts to increase, and others judged these interventions to diminish or remain stable.Similarly, 5 of the other 11 therapist-rated alliance char-acteristics of Study 1 displayed variation at the individ-ual level over this time interval. Thus, although sometherapists retained their initial judgments of clients col-laborative efforts (HA 2), self-exploration behaviors(PEXP), and hostile attitudes (PHOS), as well as theirown directive (TDIR) and negative behaviors (TNEG),other therapists viewed these characteristics as decreas-ing and still others saw them as increasing. On the otherhand, the observed mean increases from the 3rd to 10thsession in therapists appraisal of their own as well astheir clients positive attitudes and affects (TPOS,PPOS), proved consistent across therapists. The otherfour relationship characteristicsperceived client par-ticipation (PPAR) and negative attitudes (PNEG) and

their own warmth and friendliness (TWFR) as well ashelpfulness and support (HA 1)were stable (or fairlystable as regards HA 1) across therapists over this timeinterval.

From the 10th session to the near-conclusion of therapy, all therapist-rated characteristics of Study 1 ev-idenced stability, except for perceived directiveness(TDIR) and client hostility (PHOS), which continued touctuate, and perceived client involvement (PPAR),now variable in individual therapists. Stability was alsoevidenced for most alliance characteristics over the 5thto 10th therapy sessions (Study 2). Three exceptions

were perceived collaboration with clients (HA 2), agree-ment on therapeutic tasks (Task), and client working ca-pacity (PWC).

Between the 3rd and near-nal therapy sessions,again taking into account both mean comparisons andintra-individual associations, therapists perceptions forthe most part showed variability, including the dimen-sions that evidenced improvement in therapists as a group (HA 2, TEXP, TPOS, PPOS; the exception wasHA 1 [offered helpfulness and support], which in-creased across therapists). Similarly, the initial assess-ments of the other alliance characteristics of Study 1

(clients self-exploration [PEXP], participation [PPAR],hostility [PHOS], and negative attitudes [PNEG], as wellas their own negativeness [TNEG] and directiveness[TDIR]) increased in some therapists and decreased ordid not change in others; the exception was perceivedwarmth and friendliness (TWFR), which was the onlyalliance component that proved stable across therapistsover this interval.

Clients Alliance Perceptions Across Therapy

With regard to the perceptions that clients enter-tained of the various facets of the therapeutic relation-ship assessed in both studies, the nding of nosignicant change in average levels, regardless of timeperiod, is in line with prior results14,20,2224 and addssupport to the view that the quality of the therapeuticrelationship is established early in therapy.19,22,44 Theresults of intra-individual analyses suggest, however,that this conclusion applies mainly to clients viewed asa group. Similar to the therapists, individual clients re-vised their perceptions of several facets of the therapeu-tic relationship over time.

Over the interval encompassing the 3rd to 10ththerapy sessions, when mean ratings as well as intra-individual associations were considered, clients showed

variability in their perceptions of 5 of the 12 character-istics of Study 1: their therapists helpfulness, support (HA 1), and directiveness (TDIR); their own active en-gagement in therapy (PPAR); and their own as well astherapists negative attitudes (PNEG, TNEG). On theother hand, the following proved stable across clients:initial impressions of their therapists uncovering inter-ventions (TEXP), positive attitudes and affect (TPOS),their own collaboration in the work of therapy (HA 2),self-exploration efforts (PEXP), and both positive andhostile attitudes (PPOS, PHOS). Perceptions of theirtherapists warmth and friendliness (TWFR) also

proved fairly stable over this time interval.From the 10th session to the near-conclusion of therapy, individual clients continued to revise their im-pressions of their therapists directiveness (TDIR), theirown active participation in therapy (PPAR) and theirown as well as their therapists negative attitudes(PNEG, TNEG). Their own hostility (PHOS), as well astheir therapists uncovering work (TEXP), which werestable prior to the 10th session, now also proved vari-able in individual clients. The helpfulness and support of their therapist (HA 1), variable prior to the 10th ses-sion, stabilized among clients. The ve characteristics

of perceived therapist warmth and friendliness (TWFR)and other positive contributions (TPOS), as well as theirown positive attitudes (PPOS), collaboration (HA 2),and self-observation efforts (PEXP), remained stable.Between the 5th and 10th sessions, as assessed in Study2, perceptions of 8 of the 9 alliance facets examined alsoproved stable across clients; the exception was consen-sus on work strategies (WSC).


11/15

Bachelor and Salame


As was observed with therapists, clients assessment of most alliance characteristics at the near-conclusion,as compared with the early relationship, showed vari-ability. Only 3 of the 12 characteristics of Study 1proved stable across clients between the 3rd and near-nal sessions: perceived efforts to explore underlying dynamics both on their part (PEXP) and on the part of therapists (TEXP), and therapists positive attitudes andaffect (TPOS).

Remarks on Participants Alliance PerceptionsOver Time

In sum, although the pattern of variation differedin therapists and clients, participants perceptions of most of the 12 relationship characteristics in Study 1(when both comparisons of means and intra-individualassociations are considered) showed variation over onetime period or another and, for select variables, over alltime phases assessed. Therapists alliance perceptionswere found to shift more prior to the 10th therapy ses-sion (i.e., there were shifts in two-thirds of the relation-ship characteristics during what proved to be the rst third torst half of therapy). After the 10th session, ther-apists perceptions generally stabilized. In contrast,shifts in clients alliances occurred both before and afterthe 10th therapy session, although fewer facets variedfor clients than for therapists. In general, the character-istics that showed variation among individual clientsprior to the 10th session were the ones that continuedto vary subsequently.

Given that alliance perceptions for the most part proved stable from the 5th to the 10th therapy sessionboth in therapists and clients, as observed in Study 2, it may be that many facets of the therapeutic alliance, par-ticularly in the perceptions of therapists, have stabilizedby the 5th session. This conclusion remains tentative,however, because the comparability of alliance mea-sures across the two studies (with the possible exceptionof the conceptually similar Penn scales39 ) remains to beestablished. More work is needed to delineate the dis-tinctive cross-theoretical or cross-method features of thetherapeutic alliance.3,45

Although not an objective of the current report, in-terrelationships among the alliance measures withinsamples were explored. (The comparability of dimen-sions across studies could not be evaluated.) In Study 1,of the 44 interscale coefcients from clients alliance rat-ings at the 3rd, 10th, and next-to-last sessions, 0, 8, and4, respectively, exceeded 0.70. Of the coefcients from

therapists ratings at these three assessment points, 4, 8,and 6, respectively, were greater than 0.70. In Study 2,of the 26 interscale coefcients at the 5th and 10th ses-sions, 15 and 7, respectively, of those based on clientsratings exceeded 0.70, and 18 and 16, respectively, of those based on therapists ratings were larger than 0.70.None of these correlations attained a magnitude of 0.90,the level at which redundancy of constructs can be con-cluded.46

In both therapy partners, variation in the relation-ship was most pronounced between the early and near-nal therapy sessions. This result could reect thevariability observed in the relationship componentsprior to the 10th session and the cumulative effect of small, nonsignicant changes in other variables over the3rd to the 10th and the 10th to near-nal sessions. Neg-ative components of the alliance, such as therapist-per-ceived client hostility (PHOS) and client-perceivednegative client and therapist attitudes (PNEG, TNEG),together with client-perceived active participation(PPAR) and both participants perceptions of therapist directiveness (e.g., offering advice, modeling behavior;TDIR), appeared to be particularly susceptible to indi-vidual perceptual differences, and these componentsmay be more sensitive to contextual factors, such assession content or specic therapeutic work. On theother hand, components such as therapist-perceivedwarmth and friendliness (TWFR) and other positive(e.g., understanding, esteem-enhancing) attitudes andaffect (TPOS), as well as client-perceived efforts to ex-amine and explore feelings and experiences (PEXP),proved stable throughout the therapeutic process andmay reect trait-like characteristics of the participants.

These results, though requiring replication, suggest that single assessments of many facets of the participantsperceptions of the relationship cannot be assumed to berepresentative of their perceptions throughout thecourse of therapy, even though specic perceptions maypersist over longer periods of time. Participants ratings earlyin therapy, in particular, were poor predictors of theirappraisal of the relationship qualities at the near-con-clusion of the therapeutic process. Another implicationof these ndings, if replicated, is that signicant predic-tor alliance relationships established at a specic point in therapy may not generalize to other therapy sessions.Depending on the alliance characteristic investigatedand the measurement point in therapy, more than oneassessment of such relationships may be necessary.

The current results also offered little support for thecyclical model of alliance development. Although nei-


12/15



ther study reported was specically devised to test thismodel, the observed absence (on average ratings) inStudy 1 of signicantly lower scores in positive-tonedand of higher scores in negative-toned variables at the10th compared with the 3rd or near-nal sessions seemsinconsistent with the proposition7,8 that the alliance dis-plays a pattern of strengthening, decay, and repair overthe course of therapy. Similarly, the lack in Study 2 of signicant decreases between the 5th and 10th therapysessions (as indicated bya posteriori t -tests, allP 0.005,Bonferroni-adjusted) in the mean ratings of participantsfor whom the 10th session was near the midpoint of therapy (n 16) seems at variance with a purportedcritical phase9 for the alliance during the middle ther-apy sessions. Although these results may have been in-uenced by the predominantly humanistic therapymodality, which may involve less active use of chal-lenging interventions, they are in line with priordata.10,13,14 However, such a cyclical pattern, or otherpatterns including steady improvement or stability,10

may characterize individual clients and therapists alli-ances. Indeed, the current results suggest the existence,in successful therapy, of different developmental pat-terns of the relationship components and measuresamong therapists and clients.

Although there has been some acknowledgment of such patterns,13,47 individual trajectories of develop-ment have not been the focus of empirical attention inthe study of the time course of the alliance. A clinicallyrelevant avenue of research suggested by the current results could involve the identication of participant subgroups who differentially appraise the alliance overtime (i.e., participants whose perceptions either im-prove, worsen, remain stable, or wax and wane). Suchan investigation could then seek pre-therapy or in-ther-apy factors (e.g., attachment style, specic therapist in-terventions, alliance rupture markers) that inuencecharacteristic developmental patterns. A recent studyalong these lines has been conducted by Piper et al.,29

who reported different patterns of change in the alliancein clients with high, as opposed to low, quality of object relations.

Comparison of Participants Alliance Perceptions

Results of group comparisons indicated that thetherapeutic relationship was perceived similarly bytherapists and clients in terms of average levels of therelationship components. Exceptions were clients self-reection and self-exploration efforts (PEXP), which

were consistently rated higher by clients than by ther-apists across therapy, and, congruent with Marziali,15

therapists negative attitudes (TNEG), which were con-sistently rated lower by clients than by therapists (Study1). Both at the 5th and 10th therapy sessions (Study 2),clients also perceived their therapists understanding and involvement (TUI), as well as the agreement withtheir therapists regarding treatment goals and how toproceed in therapy (WSC), to be signicantly higherthan did therapists. Clients appreciated, it appears, thecommitment of their therapists to understand and helpthem, and they held equally favorable judgments of their own contributions and joint work efforts with ther-apists. Possibly clients entertain a different notion thantherapists of what is entailed in collaborating in thework of therapy. Therapist trainees comparativelymore negative self-evaluations (Study 1) regarding counterproductive attitudes and affect could reect in-creased sensitivity to these attitudes as a result of thesupervision process.

To further explore participants agreement regard-ing the quality of the alliance, post hoc correlations be-tween therapists and clients alliance ratings werecalculated. Results indicated, in general, poor conver-gence of perceptions of the relationship characteristics,regardless of assessment time.

The few exceptions generally reected aspects of the clients involvement and positive contributions at select assessment points. At the 3rd session, client-ther-apist dyads agreed on the strength of the clients activeinvolvement (PPAR); at the 5th session, they agreed onthe clients commitment to therapy (PC);andat the10thsession, they agreed on the therapists directive behav-iors (TDIR); allP 0.001. They were in near agreement (P 0.002) at session 10 regarding the clients positiveattitudes and affects (PPOS).

Thus, although at the group level therapists and cli-ents held generally similar views of the alliance,within particular dyads the therapy partners did not have thesame opinions regarding most features of the therapyrelationship. Consistent with several other stud-ies,11,12,23,29 these results suggest that clients and thera-pists employ different reference bases in appraising thequality of the therapeutic relationship. Although therehas been some speculation on the source of participantsdifferential evaluations (e.g., therapists may evaluate al-liance quality with reference to clients progress, whileclients may evaluate it with reference to extratherapyrelationships13,48 ), this issue awaits further empirical in-vestigation. Because clients alliance perceptions are sig-


13/15

Bachelor and Salame


nicant determinants of therapy outcomesuperior tothose based on other evaluative sources5the observedlow consensus underlines that it is important for thera-pists to attend more closely to their clients viewpointsand regularly check with clients their perceptions of thetherapy relationship.

Limitations

Several limitations to the generalizability of the cur-rent ndings should be noted. First, sample sizes weresmall, and male clientsand therapists, particularly in therst study, were underrepresented. A high proportionof therapists were inexperienced, although therapist level of experience, including lack of experience but presence of an inherently helpful disposition, gener-ally appear to be unrelated to the quality of the thera-

peutic relationship.43,4951

Even highly experiencedtherapists have low-alliance clients in the therapies theyconduct.14 Second, therapists in the two current studieswere predominantly humanistic in orientation, and theresults may not generalize to other approaches. It should be noted, however, that participants ratings of the alliance, as tapped by the CALPAS and WAI scales,have not been found to differ among various therapeu-tic modalities.24,43,52 Third, ndings regarding thecourse of the relationship over therapy were limited tobroad time intervals; the WAI and CALPAS variables,in particular, were examined at only two time points.

More frequent assessment would have allowed for a more sensitive analysis of the developmental course of the therapy relationship and its components. The as-sessments of the alliance near the conclusion of therapymay be confounded with outcome, although this con-cern has not received strong empirical support.9,22,53 Fi-nally, the low observed reliabilities, in the secondsample, of the HAQ HA 2 and the CALPAS PWC

scales warrant caution in interpreting the results ob-tained on these dimensions, and additional research isneeded to establish their content validity.

CONCLUSION

The present ndings, despite their limitations, add tothe understanding of the time course of a broad rangeof relationship characteristics as seen by the therapyparticipants. Research in this area has predominantlyfocused on the three components (or global score) of the Working Alliance Inventory. The current resultspoint to the relevance, in studying the time course of the alliance, of examining the perceptions of each ther-apy partner as well as the individual components of thetherapeutic relationship.

An attempt was made to integrate results from

mean comparisons and correlational analyses in dis-cussing the time course of the alliance. In line with priorobservations,9 variability of perceptions over the timecourse appears greater when individual clients or ther-apists alliance perceptions are monitored, as comparedwith a focus on pooled ratings. As noted by others,29,54

ndings based on aggregate data may mask important individual differences in perceptions of the alliance. Toimprove our understanding of the actual therapist-client relationship, it appears worthwhile to pursue investi-gation of individual differences in perceptions of rela-tionship qualities across the course of therapy, along

with factors that enhance or strain the relationship inparticular clients or therapists, or in participant sub-groups that share similar perceptions.

The authors gratefully acknowledge the assistance of Hans Iverin data analysis. A portion of this paper was presented at the annual meeting of the Society for Psychotherapy ResearchBerkeley, CA, 1992.

REFERENCES1. Bordin E: The generalizability of the psychoanalytic conceptof

the working alliance. Psychotherapy: Theory, Research andPractice 1979; 16:252260

2. Horvath AO, Greenberg L: Development and validation of theWorking Alliance Inventory. Journal of Counseling Psychology1989; 36:223233

3. Horvath AO, Greenberg LS: Introduction, in The Working Al-liance: Theory, Research and Practice, edited by Horvath AO,Greenberg LS. New York, Wiley, 1994, pp 19

4. Stiles W, Shapiro D, Elliott R: Are all psychotherapies equiva-lent? Am Psychol 1986; 41:165180

5. Horvath AO, Symonds BD: Relation between working allianceand outcome in psychotherapy: a meta-analysis. Journal of Counseling Psychology 1991; 38:139149

6. Luborsky L: Helping alliance in psychotherapy, in SuccessfulPsychotherapy, edited by Cleghhorn JL. New York; Brunner/Mazel, 1976, pp 92116

7. Bordin E: A psychodynamic view of counseling psychology.The Counseling Psychologist 1980; 9:6266

8. Gelso CJ, Carter JA: Components of the psychotherapy rela-tionship: their interaction and unfolding duringtreatment.Jour-nal of Counseling Psychology 1994; 41:296306


14/15



9. Horvath AO, Gaston L, Luborsky L: The therapeutic allianceand its measures, in Psychodynamic Treatment and Research,edited by Miller L, Luborsky L, Barber J et al. New York, BasicBooks, 1993, pp 247273

10. Greenberg LS: What is real in the relationship?Comment onGelso and Carter (1994). Journal of Counseling Psychology1994; 41:307310

11. Horvath AO, Marx RW: The development and decay of theworking alliance during time-limited counseling. Canadian Journal of Counseling 1990; 24:240259

12. Golden BR, Robbins SB: The working alliance within time-limited therapy: a case analysis. Professional Psychology: Re-search and Practice 1990; 21:476481

13. Kivlighan D, Shaughnessy P: Analysis of the development of the working alliance using hierarchical linearmodeling. Journalof Counseling Psychology 1995; 42:338349

14. Sexton HC, Hembre K, Kvarme G: The interaction of the al-liance and therapy microprocess: a sequential analysis. J Con-sult Clin Psychol 1996; 64:471480

15. Marziali E: Three viewpoints on the therapeutic alliance.J NervMent Dis 1984; 172:417423

16. Klee MR, Abeles N, Muller RT: Therapeutic alliance: earlyindicators, course, and outcome. Psychotherapy 1990; 27:166174

17. Marmar C, Weiss D, Gaston L: Towards the validation of theCalifornia Therapeutic Alliance Rating System. PsychologicalAssessment 1989; 1:4652

18. Luborsky L, Crits-Christoph P, Alexander L, et al: Two helping alliance methods for predicting outcomes of psychotherapy: a counting signs vs global ratings method. J Nerv Ment Dis 1983;171:480491

19. Hartley D, Strupp H: The therapeutic alliance: its relationshipto outcome in brief psychotherapy, in Empirical Studies of Psy-choanalytic Theories, vol 1, edited by Masling J. Hillsdale, NJ,Lawrence Erlbaum, 1983, pp 137

20. Eaton T, Abeles N, Gutfreund M: Therapeuticalliance and out-come: impact of treatment length and pretreatment symptom-atology. Psychotherapy: Theory, Research and Practice 1988;

25:53654221. Gomes-Schwartz B: Effective ingredients in psychotherapy:prediction of outcome from process variables. J Consult ClinPsychol 1978; 46:10231035

22. Morgan R, Luborsky L, Crits-Christoph P, et al: Predicting theoutcomes of psychotherapy by the Penn Helping Alliance Rat-ing method. Arch Gen Psychiatry 1982; 39:397402

23. Tichenor V, Hill C: A comparison of six measures of working alliance. Psychotherapy: Theory, Research and Practice 1989;26:195199

24. Salvio M, Beutler L, Wood J, et al: The strength of the thera-peutic alliance in three treatments for depression. Psychother-apy Research 1992; 1:3136

25. Crits-Christoph P, Barber J, Kurcias JS: The accuracy of ther-apists interpretations and the development of alliance. Psycho-therapy: Theory, Research and Practice 1993; 3:2535

26. Gurman A: The patients perception of the therapeuticrelation-ship, in Effective Psychotherapy, edited by Gurman A, RazinA. New York, Pergamon, 1977, pp 503543

27. Marmar CR, Horowitz MJ, Weiss DS: The development of thetherapeutic alliance rating system, in The PsychotherapeuticProcess: A Research Handbook, edited by Greenberg LS, Pin-sof WM. New York, Guilford, 1986, pp 367390

28. Mallinckrodt B, Nelson M: Counselor training level and theformation of the psychotherapeutic working alliance. Journal of Counseling Psychology 1991; 38:133138

29. Piper W, Boroto D, Joyce A, et al: Pattern of alliance and out-come in short-term individual psychotherapy. Psychotherapy:Theory, Research and Practice 1995; 32:639647

30. Carter JA: Rening notions through converging paradigms.The Counseling Psychologist 1994; 22:7981

31. Marziali E, Marmar C, Krupnick J: Therapeutic alliance scales:development and relationship to psychotherapy outcome. Am J Psychiatry 1981; 138:361364

32. Hatcher RL, Barends AW: Patients view of the alliance in psy-chotherapy: exploratory factor analysis of three alliance mea-sures. J Consult Clin Psychol 1996; 64:13261336

33. Bachelor A: Comparison and relationshipto outcome ofdiversedimensions of the helping alliance as seen by client and thera-pist. Psychotherapy: Theory, Research and Practice 1991;28:534549

34. Safran J, Wallner K: The relative predictive validity of two ther-apeutic alliance measures in cognitive therapy. PsychologicalAssessment 1991; 3:188195

35. Kazdin A: Preparing and evaluating research reports. Psycho-logical Assessment 1995; 7:228237

36. Endicott J, Spitzer R, Fleiss J, et al: The Global Assessment Scale. Arch Gen Psychiatry 1976; 33:766771

37. Ilfeld F: Further validation of a psychiatric symptom index in a normal population. Psychological Reports 1976; 39:12151228

38. Suh C, Strupp H, OMalley S: The Vanderbilt process mea-sures: the Psychotherapy Process Scale (VPPS) and the Nega-tive Indicators Scale (VNIS), in The PsychotherapeuticProcess:A Research Handbook, edited by Greenberg L, Pinsof W. NewYork, Guilford, 1986, pp 285323

39. Alexander L, Luborsky L: The Penn Helping Alliance Scales,in The Psychotherapeutic Process: A Research Handbook, ed-ited by Greenberg L, Pinsof W. New York, Guilford, 1986, pp325366

40. Green B, Gleser G, Stone W, et al:Relationships amongdiversemeasures of psychotherapy outcome. J Consult Clin Psychol1975; 43:689699

41. Luborsky L, McLellan T, Woody G, et al: Therapist success andits determinants. Arch Gen Psychiatry 1985; 42:602611

42. Tracey T, Kokotovic A: Factor structure of the Working Alli-ance Inventory. Psychological Assessment 1989; 1:207210

43. Gaston L: Reliability and criterion-related validity of the Cali-fornia Psychotherapy Alliance Scales. Psychological Assess-ment 1991; 3:6873

44. Henry WP, Strupp HH: The therapeutic alliance as interper-sonal process, in The Working Alliance: Theory, Research andPractice, edited by Horvath AO, Greenberg LS. New York, Wi-ley, 1994, pp 5184

45. Rice LN, Greenberg LS (eds): Patterns of Change: IntensiveAnalysis of Psychotherapy Process. New York, Guilford, 1984

46. Tabachnick BG, Fidell LS: Using Multivariate Statistics, 2ndedition. New York, Harper and Row, 1989

47. Gaston L, Marmar C: The California Psychotherapy AllianceScales, in The Working Alliance: Theory, Research and Prac-tice, edited by Horvath AO, Greenberg LS. New York, Wiley,1994, pp 85108

48. Tyron GS, Kane AS: Relationship of working alliance to mutualand unilateral termination. Journal of Counseling Psychology1993; 40:3336

49. Dunkle J, Friedlander M: Contribution of therapist experienceand personal characteristics to the working alliance. Journal of Counseling Psychology 1996; 43:456460

50. Gomes-Schwartz B, Schwartz J: Psychotherapy process vari-ables distinguishing the inherently helpful person from the


15/15

Bachelor and Salame


professional psychotherapist. J Consult Clin Psychol 1978;46:196197

51. Strupp H, Hadley S: Specic vs nonspecic factors in psycho-therapy: a controlled study of outcome. Arch Gen Psychiatry1979; 36:11251129

52. Marmar CR, Gaston L, Gallagher D, et al: Therapeuticallianceand outcome in behavioral, cognitive, and brief dynamic psy-chotherapy in late-life depression. J Nerv Ment Dis 1989;17:464472

53. Gaston L, Marmar CR, Thompson LW, et al: Alliance predic-tion of outcome beyond in-treatment symptomatic change aspsychotherapy progresses. Psychotherapy Research 1991;1:104112

54. Henry WP, Strupp HH, Schacht TE, et al: Psychodynamic ap-proaches, in Handbook of Psychotherapy and BehavioralChange, edited by Bergin AE, Gareld SL. New York, Wiley,1994, pp 467508

aliança terapêutica 2

Documents