INTRODUCTION

Although child and adolescent mental health problemsare widespread in rural Australia,1 specialist services arenot readily available in many country areas. This meansthat children and adolescents from rural communities areoften unable to access timely specialist help. Similarly,primary care providers such as school counsellors, generalpractitioners and community health nurses have no easyway of seeking a second opinion regarding child and adol-escent mental health issues. To meet the needs of regional

Australia better, some child and adolescent mental healthteams provide prompt assessment followed by brieftherapy.2–6 Given the distances and travelling times in thecountry, optimising a few specialist sessions in this way isoften preferable for families and professionals alike.

Although brief mental health therapy presents aneconomical and convenient alternative to longer and moreintense treatments, there are only a few evaluations of itseffectiveness with children’s mental health problems.These early studies are encouraging and suggest that shortcourses of therapy may be as effective as extended treat-ment.5,6 Standardised programs for specific presentationsare also showing promising results, for example an eight-session treatment has been developed for childhooddepressive symptoms among school students.7 There isalso some evidence that even minimal specialist input canassist children and their families. In a recent follow-upstudy of a metropolitan child and adolescent mental

Aust. J. Rural Health (2000) 8, 161–166

Correspondence: Dr Stephen Allison, Child and AdolescentMental Health Service, Flinders Medical Centre, Bedford Park,South Australia 5042, Australia. Email: stephen.allison@flinders.edu.au

Accepted for publication March 2000.

BRIEF THERAPY FOR CHILDREN’SMENTAL HEALTH PROBLEMS:OUTCOMES IN A RURAL SETTING

1Faculty of Health Sciences, Flinders University of South Australia, Adelaide, South Australia and2Southern Child and Adolescent Mental Health Service, Adelaide, South Australia, Australia

ABSTRACT: In rural areas where specialist services are not readily accessible, there are practical advantages toproviding short courses of therapy for children’s mental health problems. Fifty-five parents from a rural area in SouthAustralia completed questionnaires before and 3 months after receiving brief, family-orientated therapy from a childand adolescent mental health service. Clinically, these families were assessed as being suitable for brief therapybecause the children had moderately severe emotional and behavioural problems and the families were not severelydysfunctional. The screening process sought to exclude more serious and complex presentations so those families werenot included in the current study. Questionnaires measuring the severity of children’s mental health problems (ChildBehavior Checklist) and the degree of family dysfunction (Family Assessment Device general functioning subscale)suggested that, on average, the sample met these entry criteria. The follow-up evaluation after brief therapy foundsignificant overall improvement in the Child Behavior Checklist scores, however some children continued to showsubstantial difficulties. Parents’ satisfaction with therapy was related to these scores after treatment. The resultssuggest that brief therapy was appropriate for most of the families and that routine follow-up could help determinethose who might benefit from further assistance.

KEY WORDS: brief therapy, child and adolescent mental health, follow-up study, outcome evaluation.

Stephen Allison,1 Leigh Roeger,2 Vikki Dadds2 and Graham Martin1

Original Article

162 AUSTRALIAN JOURNAL OF RURAL HEALTH

health service in Australia, most parents expressed satis-faction with brief outpatient treatment.8 Sixty-five percentwere ‘mostly or very satisfied’ with even a single session,and this increased to 76% with two or three sessions and90% with four or more sessions.8 Despite this recent work,brief therapy evaluation has generally lagged behind prac-tice, especially in country areas. The current study soughtto investigate whether rural children were achieving satis-factory outcomes from brief mental health intervention.

The first issue of interest was whether brief therapywas offered to an appropriate group of children andfamilies. The clinical assessment process aimed to selectfamilies that could benefit over a relatively short period;generally these were families without serious dysfunctionwhere the children had moderately severe problems. Thecurrent study used standardised questionnaires to mea-sure whether children and families satisfied these entrycriteria and whether there had been improvement over thecourse of the treatment. It was also considered likely thatparents’ satisfaction would be higher when brief therapywas used appropriately for less severe mental health prob-lems. Hence, the study was interested in whether therewas a significant relationship between the severity of thechildren’s presenting problems and the parents’ satisfac-tion with the outcome of brief therapy.

The second issue investigated was whether parents’opinions of the professional qualities of the therapy teamrelated to the outcomes at follow-up. In mental healthstudies, a positive professional relationship has beentermed the ‘therapeutic alliance’ and the quality of thealliance is a major determinant of outcomes.9 In the con-text of child and adolescent mental health, parents ofteninitiate therapy, their permission is required for treatmentwhen children are younger than 14 years and their co-operation is usually crucial, especially in brief, family-orientated treatment. Parental satisfaction with theprofessional qualities of their mental heath therapistmight therefore be an important determinant of treatmentoutcome. There are few published studies of satisfactionwith rural child and adolescent mental health services inAustralia, but evaluations of other rural health serviceslend support to the idea that personal factors are criticalfor the acceptance of services in country communities.Professional relationships built on qualities of trust, sup-port, courtesy and communication have been found to beimportant to patients attending country hospitals10 orselecting a general practice.11 The current study sought toinvestigate whether these qualities in the professionalrelationship also related to therapy outcomes, in particularwhether parent acceptance of the professionals was linkedwith the outcomes achieved for their children.

METHOD

The present study was conducted over a 12-month periodas part of a service-wide evaluation of the Southern Childand Adolescent Mental Health Service (CAMHS). TheCAMHS operates an ‘open door’ policy and providesassistance to a wide range of children and families withmild to severe mental health problems across the southernmetropolitan and country area of South Australia. TheCAMHS has a family therapy orientation, but children arealso seen individually for part or whole sessions. Familysessions usually last 1.5 hours and individual child ses-sions run for approximately 1 hour. Approximately one-third of children attending CAMHS have severe or verycomplex problems requiring extended treatment. Anotherone-third attend only for assessment and do not receiveongoing treatment, while the remaining one-third receivebrief therapy. Brief therapy in the present study is definedas treatment (not only an assessment) that was completedwithin 1 year. For CAMHS ‘brief therapy’ clients, thistypically involves six sessions of between 1 and 1.5 hourswith a clinician.

The data were collected from the CAMHS rural teamlocated in the southern country region of South Australia.The CAMHS rural team comprises two social workers, onepsychiatric nurse, one psychologist and a psychiatrist whovisits for 2 days per month. The rural team provides spe-cialist services such as family therapy, play therapy, inter-agency consultation with schools and child protection,psychological testing and psychiatric assessment asrequired. During the study period, the rural team offeredan assessment for new referrals within 1–3 weeks of theirfirst telephone contact. The screening process used byCAMHS has been described in more detail in an earlierpaper2 but usually the first session consisted of a 1.5hours semistructured family interview designed to assessthe severity of the presenting problems, the degree ofimpairment and any associated family dysfunction. At theconclusion of this interview, the clinician discussed withthe family and the child the best approach for addressingtheir concerns and a plan of action was developed.

Prior to the screening session, the parents of consec-utive clients were invited to participate in the study.Parents who consented to participate completed question-naires relating to their child’s mental health problems, thelevel of their family functioning and finally at the conclu-sion of treatment, their satisfaction with the services thatthey had received. Parents were chosen to complete thequestionnaires in preference to children because parentsprovide more valid and reliable assessments of children’smental health problems and family functioning. Altern-

BRIEF THERAPY AND CHILDREN’S MENTAL HEALTH: S. ALLISON ET AL. 163

ative methods of direct assessment of children (such asobservation of children’s behaviour at home or school) orusing structured individual interviews were not employedin the present study because of the high costs inherent inthese approaches.

The study materials (information sheets and question-naires) were approved by the Clinical Research EthicsCommittee of Flinders Medical Centre. The study wascarried out by members of an independent evaluationteam who did not provide direct client services. Parentswere informed that any information they provided wouldbe kept confidential from CAMHS therapists and thattheir participation was voluntary and would not affect theprovision of any CAMHS services. Parents were contactedby mail before their first appointment (time 1) and again3 months after their last appointment (time 2). At bothtime 1 and time 2, standard mail-survey techniques12

were used that included an initial mail out, reminder cardand repeat questionnaire for non-respondents. Thisresulted in a return rate of 62% at time 2.

At both times, the questionnaire included measures ofchildren’s mental health problems (Child Behavior Check-list; CBCL) and family functioning (Family AssessmentDevice general functioning subscale; FAD). The CBCL isa widely used measure that allows comparisons to bemade with results from Australian and international stud-ies.13,14 The checklist has 112 questions for parents andthese can be used to derive a ‘total problem score’.13 TheFAD general functioning subscale is also a frequentlyused questionnaire with comparison data available.15,16 Itis a 12-item rating scale of the overall level of familydysfunction (with higher scores indicating greater dys-function).

At time 2 (at the 3-month follow-up) parents wereasked to indicate whether the child’s main presentingproblem was ‘better’, ‘unchanged’ or ‘worse’. They alsocompleted rating scales for their satisfaction with thequality of care (four items), the outcomes achieved fromtherapy (four items) and overall satisfaction (one item).Each aspect was rated on a five-point Likert scale(delighted, satisfied, mixed, dissatisfied, terrible). Theaim of the extreme anchor points of ‘delighted’ and‘terrible’ was to give a more normal distribution fromresponses.17 The questionnaire data were analysed withSPSS 9.0 for Windows (SSPS Inc., Chicago, USA) usingStudent’s paired t-tests and Spearman correlations.

RESULTS

Overall, 55 parents participated in the study and provideddata at both data-collection times. Almost all the respon-

dents were mothers or stepmothers (98%) but one naturalfather was also included. The average age of the childrenwas 9 years (range 4–16 years), and boys (58%, n = 32)outnumbered girls (42%, n = 23) in the sample.

Overall, the majority of parents were satisfied withbrief therapy, being either ‘delighted’ (33%, n = 17) or‘mostly satisfied’ (39%, n = 20). The remainder rating theservice they received as ‘mixed’ (22%, n = 11), ‘mostlydissatisfied’ (2%, n = 1) or ‘terrible’ (4%, n = 2). In addi-tion to their overall satisfaction, parents also rated theirsatisfaction with more specific aspects of therapy on thesame five-point Likert scale. In Table 1, parents’ ratings ofsatisfaction are shown from the most to least favourable.From this ranking it is clear that nearly all parents weresatisfied with the qualities of the team members (manner,understanding, thoroughness and competence, which arethe highest ranked items). The majority of parents werealso satisfied with the outcomes of therapy (amount ofhelp received, problems relieved, helping child to copeand preventing relapse), but as shown in Table 1, theseitems all ranked lower than the items about the mentalhealth team’s professional qualities.

Of the 55 parents who gave a global rating of theirchild’s improvement in therapy, 82% (n = 45) reportedthat the presenting problems were better, with 79% ofthese parents attributing the improvement to the brieftherapy. Of the remaining parents, 14% (n = 8) rated thepresenting problems as unchanged and 4% (n = 2) ratedthe problems as worse.

Before therapy (time 1), the mean CBCL total problemscore was 59.9 (SD = 29.4) for boys and 42.9 (SD = 16.5)for girls, which indicated that their problems were in themoderate range.13,14 At time 2 the mean CBCL total prob-lem score for boys was 51.4 (SD = 16.5) and 32.5 for girls(SD = 26.1). Paired t-tests showed that these differences

TABLE 1: Parents’ satisfaction with the professional qualities

of the mental health team and the outcomes of brief therapy

(items are ranked by percentage satisfaction; n = 51)

Delighted Mostly Mixed Mostly Terrible

satisfied dissatisfied

Manner 59 35 2 2 2

Understanding 49 35 12 2 2

Thoroughness 43 31 22 2 2

Competence 39 43 14 4 0

Help received 33 39 22 6 0

Problems relieved 26 37 29 4 4

Child coping 20 45 27 4 4

Preventing relapse 20 33 33 8 6

164 AUSTRALIAN JOURNAL OF RURAL HEALTH

were statistically significant for boys (t = 2.1, P < 0.05)and girls (t = 2.6, P < 0.01). Some children still had highscores at follow-up, with 27% (n = 14) scoring in the clin-ical range using a ‘cut-point’ of 60. At time 1 the meanscore for FAD general functioning was 1.9 (SD = 0.4) forboth boys’ and girls’ families, which indicated that onaverage these families were not severely dysfunctional.16

The mean score was marginally lower at time 2 (1.8;SD = 0.04); this change was statistically significant(t = 2.1, P < 0.05).

Using Spearman correlation, there was a significantpositive relationship between the change in CBCL totalproblem score between times 1 and 2 (T1-T2 CBCL) andthe satisfaction item that asked specifically about problemrelief (Table 2). There were significant inverse correla-tions between time 2 total problem score (T2 CBCL) andthe items that assessed problems relieved, child coping,preventing relapse and overall satisfaction. This indicatesthat the burden of problems remaining after therapy wasassociated with lower parent satisfaction in these areas.The same pattern was evident for parental reported familyfunctioning at time 2 (T2 FAD), with significant inversecorrelations with these satisfaction items; parents ratingtheir families as more dysfunctional were less satisfiedwith these outcomes of therapy. Interestingly, the samepattern was present for the total problem scores at time 1(T1 CBCL), indicating that higher intake scores were asso-ciated with lower satisfaction 3 months after brief therapy.

DISCUSSION

The current study suggests that brief therapy for child andadolescent mental health problems can be acceptable torural parents. As found in previous evaluations of countryservices, the human dimension is important in rural com-munities.10,11 Parents who brought their children to theservice and then participated in the family-orientatedtherapy program had a high regard for the professional

qualities of the team. Parents generally found team mem-bers to be competent, understanding and thorough. Thesequalities would assist in providing the basis for the trust-ing therapeutic relationships with parents that are neededfor successful outcomes with their children.

After brief therapy, most parents indicated that fromtheir point of view the main presenting problem was betterand they usually cited brief therapy as the reason for theimprovement. Overall, there was also a significant reduc-tion in children’s total mental health problems during theperiod of therapy and follow-up. For some children andfamilies with more severe problems, substantial difficul-ties remained even after significant improvement. It isnotable that at follow-up, problem scores were still in theclinical range for 27% of children. These findings suggestthat some families did not receive enough assistance froma brief course of therapy. Future protocols could includeroutine telephone contact with parents to assess whetherimprovements have been maintained after brief therapyand whether children could benefit from more treatment.

Parents whose children had severe mental health prob-lems tended to be more disappointed with brief therapy. Inparticular, children with more severe problems at intakewere rated by their parents as receiving less problem relieffrom brief therapy and the parents felt they were lessequipped to prevent relapse. At follow-up, more severechildhood problems and poorer family functioning wereboth related to lower parental satisfaction with therapy. Incontrast, the relationship between the change in total prob-lems scores and satisfaction with outcomes held for onlythe specific item that measured problems relieved. Thissuggests that the quantum of problems remaining aftertherapy was salient to parent satisfaction ratings. Parentsof children with severe problems should be given realisticinformation about the likely outcome of brief therapy andprovided with the opportunity to pursue more intensiveforms of treatment if this is indicated.

TABLE 2: Correlations between parental satisfaction with outcomes and parents’ ratings of family functioning and children’s mental

health problems at time 1 and time 2 (n = 51)

T1 FAD T2 FAD T1 CBCL T2 CBCL T1 CBCL- T2 CBCL

Help received –0.07 –0.33* –0.37** –0.29 0.08

Problems relieved –0.10 –0.42** –0.38** –0.49** 0.36*

Child coping –0.00 –0.30* –0.33* –0.36* 0.10

Preventing relapse –0.09 –0.42** –0.52** –0.46** 0.11

Overall satisfaction –0.13 –0.46** –0.36* –0.36** 0.17

*P < 0.05, **P < 0.01. T1, time 1 (before the first appointment); T2, time 2 (3 months after the last appointment); FAD, Family

Assessment Device general functioning subscale; CBCL, Child Behavior Checklist.

BRIEF THERAPY AND CHILDREN’S MENTAL HEALTH: S. ALLISON ET AL. 165

Parents’ satisfaction with the personal manner, under-standing, competence and thoroughness of the cliniciantreating their child was not associated with their ratings ofimprovement in children’s mental health problems. It islikely that a ‘ceiling effect’ (with most parents being satis-fied with relationships) limited the chances of finding theexpected association with therapeutic improvement.Equally, the role of the therapeutic alliance in family-orientated therapy may be more complex than in individ-ual psychotherapy.9 For example, the relationship betweenthe professional and the child (which was not measured inthe current study) and the relationships with parents, mayeach have an effect on outcomes.

The main limitation of the present study was the lackof a control group to determine whether mental healthgains were due to therapy. The improvements found in thecurrent study could have occurred for a wide variety ofreasons other than therapy and might just represent therate of spontaneous recovery among children with mentalhealth problems. It is also possible that parent reportsmay not give an accurate measure of improvement butsimply show ‘regression to the mean’ with parentsacknowledging fewer problems when completing the ques-tionnaires for a second time. Future studies should con-sider randomised controlled trials to estimate the impactof brief therapy; these studies should also involve reportsfrom both mothers and fathers, observation of children’sbehaviour and ratings from the mental health team. Thereare practical difficulties in arranging appropriate therapytrials in rural settings but the current dearth of evidencefor effectiveness highlights the need for a more rigorousapproach to evaluation.

In conclusion, parents whose children had brief ther-apy generally reported a high level of professionalismfrom the multidisciplinary team. Although parent satisfac-tion with the team was not related to improvement inchildren’s mental health, it may have been an importantfactor in their initial engagement: parents who were dis-satisfied may not have entered therapy and may be under-represented in the study. The outcomes of brief mentalhealth therapy were mostly acceptable to rural parentswho engaged in the program, especially when children didnot have severe problems. It was concerning that asubgroup of children continued to have clinically signifi-cant problems during the follow-up period. In future prac-tice with brief therapy, there may be benefits in routinefollow-ups with parents and referrers to assess whether theimprovement has been maintained, to give further adviceabout preventing relapse and in some instances to offerfurther therapy.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge project funding fromthe South Australian Department of Human Services andthe assistance of the CAMHS Country Team (in particularRobyn Coleman) in the implementation of the study.Robyn Duckworth (Southern Child and Adolescent MentalHealth Service) provided helpful comments on an earlierdraft of this paper.

REFERENCES

1 Reid M, Solomon S. Improving Australia’s Rural Health andAged Care Services (background paper no. 11). Melbourne:National Health Strategy, 1992.

2 Allison S, Roeger L, Dadds V, Wood A, Martin G. Muchbriefer assessments of children and adolescents. Aus-tralasian Psychiatry 1999; 7: 338–340.

3 Aveline MO. Brief therapy: Benefits and limitations includ-ing cognitive and behavioural therapies. Current Opinion inPsychiatry 1996; 9: 193–197.

4 Kirby RJ, Smyrnios KX. Child-oriented family therapy out-come research: Comparisons between brief family therapyand an alternative treatment. Australian and New ZealandJournal of Family Therapy 1990; 11: 75–84.

5 Weisz JR, Weiss B. Assessing the effects of clinic basedpsychotherapy with children and adolescents. Journal ofConsulting and Clinical Psychology 1989; 57: 741–746.

6 Smyrnios KX, Kirby RJ. Long-term comparison of brief ver-sus unlimited psychodynamic treatments with children andtheir parents. Journal of Consulting and Clinical Psychol-ogy 1993; 61: 1020–1027.

7 Weisz JR, Thurber CA, Sweeney L, Proffitt VD, LeGagnoux GL. Brief treatment of mild-to-moderate childdepression using primary and secondary control enhance-ment training. Journal of Consulting and Clinical Psychol-ogy 1997; 65: 703–707.

8 Rey JM, Plapp JM, Simpson PL. Parental satisfaction andoutcome: A four-year study in a child and adolescent men-tal health service. Australian and New Zealand Journal ofPsychiatry 1999; 33: 22–28.

9 Martin G, Allison S. Therapeutic alliance: A view con-structed by a family therapy team. Australian and NewZealand Journal of Family Therapy 1993; 14: 205–214.

10 Davis BA, Duffy E. Patient satisfaction with nursing care ina rural and an urban emergency department. AustralianJournal of Rural Health 1999; 7: 97–103.

11 Humphreys JS, Mathews-Cowey S, Weinand HC. Factors inaccessibility of general practice in rural Australia. MedicalJournal of Australia 1997; 166: 577–580.

12 Dillman DA. Mail and Telephone Surveys: The Total DesignMethod. London: John Wiley, 1978.

13 Achenbach TM. Manual for the Child Behavior Checklist.Burlington: University of Vermont Department of Psychi-atry, 1991.

166 AUSTRALIAN JOURNAL OF RURAL HEALTH

14 Alfons AM, Crijnen AM, Achenbach TM, Verhulst FC.Comparisons of problems reported by parents of children in12 cultures: Total problems, externalising and internalising.Journal of the American Academy of Child and AdolescentPsychiatry 1997; 36: 1269–1277.

15 Byles J, Byrne C, Boyle M, Offord DR. Ontario child healthstudy: Reliability and validity of the general functioningsubscale of the McMaster Family Assessment Device.Family Process 1988; 27: 97–104.

16 Sawyer MG, Savris A, Baghurst PA, Cross D, Kalucy RS.Family assessment device: Reports from mothers, fathersand adolescents in community and clinic families. Journalof Marital and Family Therapy 1988; 14: 287–296.

17 Kurtz LF. Measuring member satisfaction with a self-helpassociation. Evaluation and Program Planning 1982; 5:233–237.