Psychological characteristics of cancer patients who use complementary therapies

PSYCHO-ONCOLOGY

Psycho-Oncology 14: 187–195 (2005)Published online 10 June 2004 in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/pon.834

PSYCHOLOGICAL CHARACTERISTICS OFCANCER PATIENTS WHO USECOMPLEMENTARY THERAPIES

ROBIN DAVIDSONa,*, LAURA GEOGHEGANb, LYNNE MCLAUGHLINc and ROGER WOODWARDb

aNorthern Ireland Cancer Centre, Belfast, IrelandbUniversity of Ulster, Belfast, IrelandcQueen’s University Belfast, Ireland

SUMMARY

There has been considerable research on the prevalence and demographic profile of cancer patients who opt tosupplement conventional therapies with the use of complementary therapy. There is rather less information on thepersonality and adjustment variables associated with the decision to use complementary therapy.The aim of the present study is to investigate the relationship between the use of complementary therapies by

cancer patients and their mental adjustment to cancer, recovery locus of control, life orientation andpsychopathology. Two groups were drawn from a regional centre which provides both conventional andcomplementary cancer treatments. Participants in Group 1 (n=61) opted for complementary therapies in addition toconventional treatments for cancer, while participants in Group 2 (n=56), chose conventional treatment only. Allparticipants completed the Mental Adjustment to Cancer Scale (MAC), the Recovery Locus of Control Scale(RLOC), the life orientation test (LOT), and the Hospital Anxiety and Depression Scale (HADS). Informationregarding demographic details and patients’ motivation for the use of complementary therapy was also collected.Those people who chose complementary therapy demonstrated a mental adjustment to cancer which is

characterised by significantly higher levels of fighting spirit and anxious preoccupation. This group had also a higherinternal recovery locus of control than those receiving conventional treatment alone. There were no significantdifferences between the groups on life orientation or psychopathology.The findings of this study do not support the argument that the use of complementary therapy is associated with

higher levels of psychopathology and distress. However, the data do indicate that for some patients the use ofcomplementary therapy fulfils an important psychological need. The finding that psychosocial variables like fightingspirit and locus of control may impact on an individual’s therapeutic choice can assist clinicians in tailoringinterventions to personality and adjustment characteristics. Copyright # 2004 John Wiley & Sons, Ltd.

INTRODUCTION

Increasingly, patients with cancer are opting tosupplement conventional treatments with a rangeof complementary and alternative interventions. Ingeneral, complementary therapies tend to empha-sise a patient orientated system of treatment, whilemedical interventions tend to be disease orien-tated. Dixon and Sweeney (2000) showed that

patient centred consultations are associated withpatients feeling understood and with greaterresolution of their concerns and symptoms.Studies investigating the prevalence of comple-mentary and alternative medicine (CAM) use incancer care have reported that between 7 and 64%of adult cancer patients use some form ofalternative or complementary medicine (Sollneret al., 2000), while Brigden (1995) reported the useof CAM as high as 80%. The discrepancies amongfigures being reported can be in part attributed tomethodological differences among studies, andalso to problems of definition and measurement.

The definition of CAM is notoriously difficult asthe term encompasses a whole range of different

Received 11 February 2004Copyright # 2004 John Wiley & Sons, Ltd. Accepted 12 May 2004

*Correspondence to: Gerard Lynch Centre, Northern IrelandCancer Centre, Belvoir Park Hospital, Hospital Road, BelfastBT8 8JR, Ireland. Tel.: 028 90699282; fax: 028 90641959.E-mail: [email protected]

therapies, remedies, nutrition and life style the-ories. More unusual and rare treatments have beensubject to controversy and have prompted debateson safety and effectiveness. One difficulty asso-ciated with adopting a broad definition is thatwidely accepted health practices such as acupunc-ture and aromatherapy are labelled together withmore highly controversial approaches. Peters et al.(2002) have defined complementary therapies toinclude such practices and ideas as self-defined bythe users for preventing or treating illness orpromoting health and well being. In the presentstudy, the therapies investigated were used along-side conventional medicine and categorised there-fore as complementary therapy by patients,medical staff and therapists alike. The comple-mentary therapies employed in this study arereflexology, aromatherapy, and acupuncture.

It has been argued, particularly in the medicalliterature, that people who use CAM are ‘mostlyrefugees from conventional medicine’ (Fulder, 1988).However, the majority of those cancer patients whoincorporate complementary therapies into theirtreatment do so as supplementary methods to helpcontrol symptoms, and only a small number adoptalternative cancer treatments in lieu of mainstreamtherapy (Cassileth and Chapman, 1995). Thisfinding, that only a small proportion of patientsuse CAM with curative intent, has been mirrored ina large number of studies, e.g. Vandecreek et al.(1999) and Verhoef et al. (1999). A persuasivemotive for using CAM is a desire to have morecontrol over disease management and these thera-pies tend to view the patient as an active participantin treatment when he or she is afforded more inputand control. Aulas (1996) has suggested that ifconventional treatments have been exhausted, un-conventional ones may increase the patient’s sense ofcontrol and well being, even if they do not result insurvival advantage. Ritvo et al. (1999) suggest thatthose who seek out complementary therapy do,however, tend to demonstrate more doubts aboutthe effectiveness of conventional treatment.

Faced with multiple motives, a number ofworkers have attempted to isolate and classifythe sources of patients’ attraction to complemen-tary therapy. Vincent and Furnham (1996) identi-fied four principal reasons for people’s choice ofalternative care:

1. belief in the positive value of alternative care,2. previous experience of orthodox medicine as

ineffective,

3. concerns about the adverse side effects ofmedical care,

4. poor communication with patients by orthodoxmedical practitioners.

Alternatively, Astin (1998) proposed threeexplanations:

1. dissatisfaction with conventional medicine,2. a need for personal control, and3. philosophical congruence, i.e. more compatible

with the patient’s values and world view.

Burstein et al. (1999) demonstrated that newusers of CAM could be differentiated from non-users in as much as there was more evidence ofdepression, lower levels of satisfaction with overallmedical care, greater fear of recurrence of disease,and more frequent and more severe somaticsymptoms. Kao and Devine (1999) argue that theuse of complementary therapies could be inter-preted as ‘a marker of distress’, or may suggest aneed to have more control over treatment choicesor the desire to explore all possible ways toinfluence outcome. It was reported that thosewho opted for CAM reported lower satisfactionregarding role functioning, cognitive functioning,social functioning, physical symptoms and globalquality of life. Frequently, CAM users are morehealth conscious and believe more strongly thatpeople can influence their health both throughlifestyle choices and maintaining a psychologicalequilibrium.

Generally, demographic studies have demon-strated that young, female, employed patients ofhigher socio-economic status appear more likely touse complementary therapy. Furthermore, diseaserelated variables such as type and stage of thetumour as well as patients with more advanceddisease and longer duration predict CAM use.Verhoef et al. (1999) however suggested that thefindings of greater pathology in users of CAMreported in some studies could be explained by thesimple fact that the surveys contained a higherproportion of more seriously ill patients thancontrol groups. Alternatively, it has been arguedthat patients with metastatic disease are morelikely to have felt disappointment with conven-tional therapy and this could explain the elevateduse among this group of patients. It is importantthat the design of any study takes account ofdifferences in disease severity.

R. DAVIDSON ET AL.188

Copyright # 2004 John Wiley & Sons, Ltd. Psycho-Oncology 14: 187–195 (2005)

There is an increasing literature on the associa-tion between physical and psychological outcomesand CAM use. For example, Redmond (1996)reported association between sense of control andaromatherapy. Meditation and tai chi, have beenshown to reduce depression and anxiety in patientswith cancer while acupuncture and aromatherapymassage have been associated with the reductionof pain and nausea (Cassileth, 1999).

The present study is an attempt to examine somepersonality, adjustment and psychopathologicalvariables associated with the decision to opt forcomplementary therapy. Greer et al. (1989)investigated the emotional adjustment of peoplewith a diagnosis of cancer. Fighting spirit wassignificantly associated with less psychologicaldistress and a similarly positive association wasfound for denial. An internal locus of control isassociated with a variety of positive healthpractices and improved physical and psychologicalfunctioning (Marks et al., 1986). A consistentfinding is that patients with an internal locus ofcontrol are more likely than those with externallocus of control to assume responsibility for theirmedical care. The general findings indicate thatcertain beliefs about locus of control can haveadaptive significance for people with seriousillnesses. It has been suggested that one of themotives for using CAM is a desire to have morecontrol over disease management (Warrick et al.,1999). These authors also suggest that the use ofCAM may represent an individual’s effort to assertresponsibility for their health. In the present study,the patients recovery locus of control is investi-gated and it is hypothesised that individuals usingCAM will perceive that they have greater personalcontrol over their recovery than those usingconventional cancer treatments alone.

Research has also implicated the personalitydimension of optimism and pessimism as animportant factor in a wide range of behaviouralchoices when people are confronted with adversity(Carver et al. 1993a; 1993b). With regard tocancer, optimistic patients have consistently shownbetter adjustment on a variety of variablesincluding mood change, coping and outcome.What is less clear, is how the effect of optimismis exerted. It has been suggested that optimists farebetter than pessimists and the relationship betweenoptimism and subjective well being is mediated bycoping style. Through the role of coping styles,optimism has been implicated in physical andpsychological cancer outcomes (Harrison and

Maguire, 1994). With regard to the interactionbetween optimism and emotional response tocancer, optimists are more likely to respond withfighting spirit, whereas pessimists are more likelyto demonstrate a helplessness–hopelessness re-sponse (Seligman, 1996). Wyatt et al. (1999)suggest that CAM users were significantly moreoptimistic than non-users, although this has notbeen replicated. Finally, with regard to psycho-pathology and CAM use, some studies have foundthat complementary therapy users were signifi-cantly more anxious, but there is some debate inthe literature as to whether or not depression is agood predictor of complementary therapy use(Burstein et al., 1999).

The present study then sets out to examinecancer patients’ who opt for CAM and a numberof hypotheses will be examined.

(a) That there will be a greater proportion ofindividuals with advanced disease opting forcomplementary therapy

(b) That adjustment style of those using comple-mentary therapy will be characterised byhigher levels of fighting spirit, anxious pre-occupation, but lower levels of fatalism thatthose receiving conventional treatments alone

(c) The individuals choosing complementarytherapy will tend to have an internal recoverylocus of control

(d) That individuals who opt for complementarytherapy will be more optimistic in their lifeorientation.

(e) Psychopathology, notably anxiety and depres-sion will be greater among those cancerpatients who opt for complementary therapy.

METHOD

Subjects

There were a total of 120 participants in thestudy, 28 males and 92 females, age range 20–89.Participants were both outpatients and inpatientsin a Regional Cancer Centre, which provides bothconventional and complementary therapy. Group1 (n=61) consisted of individuals who had optedfor complementary therapy in addition to conven-tional treatment for their cancer, while Group 2(n=59) were individuals who received conven-tional therapy alone. Given the exploratory nature

PSYCHOLOGICAL CHARACTERISTICS OF CANCER PATIENTS 189


of this study, patients suffered from a wide rangeof disease, and stages of illness. Sociodemographicvariables are summarised in Table 1 and Table 2presents a breakdown of diagnosis according togroup membership.

Design

The independent variable was the choice orotherwise to receive complementary therapy as asupplement to conventional treatment. Dependentvariables were mental adjustment to cancer,recovery locus of control, life orientation, anxietyand depression. The experimental group consistedof individuals who sought out complementarytherapy in the Regional Cancer Centre. Thecontrol group of consecutive admissions to theCancer Centre consisted of patients who did notactively opt for complementary therapy.

Instruments

The Mental Adjustment to Cancer Scale,Watson et al. (1989), assesses dimensions offighting spirit, helplessness/hopelessness, fatalism,avoidance, and anxious preoccupation. Whilethere is some ongoing debate about this scale bothpsychometrically and conceptually the MAC Scalehas been used widely in many studies, andOsborne et al. (1999) report on its psychometricproperties.

The Recovery Locus of Control Scale is a self-administered nine item questionnaire assessing ofan individual’s perception of control over theirrecovery. Five items reflect internal beliefs andfour items external beliefs. It is a 5 point Likert

response scale ranging from (1) strongly agree to(5) strongly disagree. The psychometric propertiesare reported by Partridge and Johnston (1989).

The life orientation scale (LOT) has beendeveloped to assess dispositional optimism, de-fined by Scheier and Carver (1987) as a habitualstyle of anticipating favourable outcomes. TheScale consists of four items keyed to the positivereaction, four keyed to the negative reaction, andfour filler items to disguise the purpose of the test.The LOT shows satisfactory internal consistency(Cronbachs Alpha=0.76), and test re-test relia-bility (0.79). Furthermore, the validity of this Scaleis indicated by positive correlations with internalcontrol beliefs, self-esteem and negative correla-tions with depression, hopelessness, and aliena-tion.

The Hospital Anxiety and Depression Scale(HADS) (Zigmond and Snaith, 1983) is a 14 itemscale to provide a brief measure of both anxiety(seven items) and depression (seven items). It wasdesigned to assess the severity of these variableswith minimum contamination of scores by reports

Table 1. Group membership and demographic variables

Complementary therapy Conventional treatment Values Sig.

Gender Male: 17.5% Male: 32.3% X2=3.751 p50.05

Female: 82.5% Female: 67.8%

Mean age 52.57 years 56.49 t=�1.862 N.S.

(Range: 20–90 years) (Range: 30–80 years)

Marital status Married: 60.3% Married: 83.1% X2=10.60 p50.05

Single: 23.8% Single: 5.1%

Education School only: 50.8% School only: 86.4% X2=17.80 p50.001

Higher education: 49.2% Higher education: 13.6%

Religiousness (Church attendance) Regularly: 79.4% Regularly: 83.1% X2=0.271 N.S.

Rarely: 20.6% Rarely: 16.9%

Table 2. Percentage of diagnoses according to group member-

ship

Diagnosis Complementary

therapy

Conventional

treatment

(%) (%)

Breast 60.3 44.1

Lung 4.8 6.8

Brain 3.2 3.4

Prostate 6.3 6.8

Testicular 3.2 6.8

Ovarian 4.8 5.1

Other 17.5 27.1



of physical symptomatology. Accordingly, it hasbeen widely used in the cancer literature. Psycho-metric properties for cancer samples have beenreported by Moorey et al. (1991).

Procedure

All potential participants were screened byrelevant healthcare professionals. Exclusion criter-ia were patients who had a learning disability, co-morbid psychotic illness, and those who wereacutely ill. People under 18 and over 90 years oldwere also excluded from the study. The research-ers, who were clinical psychology trainees, out-lined the nature and requirements of the study toeach participant and they were provided with awritten information sheet about the study.Through the use of a semi-structured question-naire each patient was initially interviewed. Thisprovided socio-demographic information, namelyage, marital status, level of education, and strengthof religious beliefs. There were also illness relatedquestions regarding date of diagnosis, satisfactionwith conventional treatments and if relevant, thelevel of satisfaction with complementary therapy.Furthermore, each participant who opted forcomplementary therapy was asked to describetheir reasons for this choice in as much detail aspossible.

All participants were assured that regardless ofwhether or not they participated, this would notaffect treatment availability. All of the question-naires were then completed with the researcherpresent. It was felt that this would enable theestablishment of good rapport, leading to aneasier, more open style of responding. Thequestionnaires used in the present study focusedon psychological variables and in an attempt toreduce intrusiveness, it was decided that medicalinformation would not be requested during thedata collection phase. Medical records wereconsulted in order to determine the exact clinicaldiagnosis, the conventional treatments received,illness duration, and type and stage of tumour.

RESULTS

Of the 120 individuals who participated in thisstudy, 61 had received complementary therapy inaddition to conventional cancer treatment. 63.5%

of this group sought complementary therapyindependently while 36.5% had requested itthrough their consultant oncologist. There wasno significant difference in the presence or absenceof metastatic spread between the two groups.There was also no significant difference betweenthe groups in terms of extent and duration ofconventional treatment.

Bi-variate analysis

All sub-scales of the MAC Scale were trans-formed into z scores. Between group comparisonswere then made using independent sample ‘t’ tests.The results showed that those in the complemen-tary therapy group had significantly higher scoresfor anxious preoccupation (t=2.458, DF=120,p50.05), and significantly lower scores for fightingspirit/helplessness (t=�3.58, DF=120, p50.001).The raw scores on this scale were universallytransformed into T-scores, i.e., a lower scoreindicates higher levels of fighting spirit. Therewas no significant difference between the groups infatalism. Comparison of the recovery locus ofcontrol revealed a higher score for the CAM group(�=35.1) than conventional group (�=32.6)(t=2.667, DF=120, p50.05). A relationshipbetween gender and recovery locus of controlwas also demonstrated, with female participantsdisplaying a higher recovery locus of control,�=34.4, males �=32.1 (t=�2.09, DF=120,p50.05).

Life orientation scores between the two groupswere not significantly different although interest-ingly there was significantly higher levels ofoptimism across groups among those who hadreceived tertiary education. There were no sig-nificant differences between groups on the anxietyand depression sub-scales of the HADS.

Correlational analyses

In order to explore the interrelationship betweenpsychological variables, all of the dependentmeasures were correlated using Pearson r and thematrix is presented in Table 3. There was asignificant negative correlation between trans-formed fighting spirit/helplessness and life orienta-tion (r=�0.47, p50.01). A significant negativecorrelation also exists between fighting spirit/helplessness and recovery locus of control



(r=�0.56, p50.01). Therefore, those individualswho show a higher level of fighting spirit are morelikely to present with an optimistic pattern ofresponding on the LOT, and a more internalrecovery locus of control. As would be expected,fighting spirit/helplessness is also positively corre-lated with both of the sub-scales from the HADS.In other words, high fighting spirit is associatedwith low psychopathology. Anxious preoccupa-tion is negatively correlated with scores on theLOT, which imply that individuals who scoremore highly for anxious preoccupation have lowerscores on life orientation (i.e. will have higherlevels of pessimism). Also of interest is thesignificant negative correlation between scores onthe LOT and HADS.

Discriminant function analysis

A discriminant function analysis was carried outin order to assess the power of the psychometricmeasures in discriminating between those whoavailed of CAM and those who did not. Themeasures used were: fighting spirit–helplessness;anxious preoccupation; fatalism; recovery locus ofcontrol; life orientation; HADS anxiety; HADSdepression. A single discriminant function success-fully discriminated between groups (Wilks’ Lamb-da=0.81; p50.001). Correlations of thediscriminant function with each of the measuresare given in Table 4.

As may be seen from Table 4, the strongestdiscriminating variable was fighting spirit/help-lessness, with a correlation coefficient of 0.67. Thenext strongest was recovery locus of control, but ina negative direction (�0.50). The discriminantfunction can thus be described as a measure of theclient’s determination to fight their cancer and alsoas a measure of the client’s assumption of owner-ship of their problem.

In order to appraise the face value effectivenessof the discriminant function analysis, participantswere classified into complementary or conven-tional groups solely on the basis of the discrimi-nant function.

The allocations derived were compared withactual group membership (Table 5) and a successrate of 68% was obtained. Predictions wereslightly more accurate for the complementarygroup (71.4%) than for the conventional group(64.4%).

Table 3. Correlation matrix of psychological variables (n=120)

FS/H AP F LOT RLOC HADS-A HADS-D

FSH 1.000 0.114 0.146 �0.471�� 0.565�� 0.328�� 0.320��

Pearson correlation sig (2-tailed) 0.212 0.108 0.000 0.000 0.000 0.000

AP 0.114 1.000 0.156 �0.250�� 0.029 0.507�� 0.289��


F 0.146 0.156 1.000 �0.175 �0.193� 0.051 0.063


LOT �0.471�� 0.250�� 0.175 1.000 0.209� �0.355�� 0.344��


RLOC �0.565�� 0.029 �0.193� 0.209� 1.000 �0.164 �0.255��


HADS-A 0.328�� 0.508�� 0.051 �0.355�� 0.164 1.000 0.615��


HADS-D 0.320�� 0.289�� 0.063 �0.344�� 0.255�� 0.615�� 1.000


��Correlation is significant at the 0.01 level (2-tailed). �Correlation is significant at the 0.05 level (2-tailed). FSH=fighting spirit/helplessness. F=fatalism. RLOC=recovery locus of control. HADS-D=HADS depression. AP=anxious preoccupation.LOT=life orientation test. HADS-A=HADS anxiety.

Table 4. Correlations between the first discriminant function

and psychometric variables

Fighting spirit–helplessness 0.672

Recovery locus of control �0.500

Anxious preoccupation �0.461

HADS anxiety �0.214

HADS depression 0.141

Fatalism 0.136

Life orientation �0.084



DISCUSSION

As it would be expected there were sociodemo-graphic differences between the groups, particu-larly age, gender and level of education. Younger,better educated, women tended to opt for com-plementary treatments, and this mirrors a consis-tent finding in the literature (Wyatt et al., 1999).There were no significant differences in theincidence of metastatic disease and patients beingtreated with palliative rather than curative intent.This would indicate that opting into complemen-tary therapy is not, as several researchers havesuggested, a frantic last minute attempt to find acure. Indeed, when questioned about the motiva-tion for using complementary therapy, only oneparticipant mentioned the idea of prolonging life.Conversely, most said that it was a pro-activechoice, irrespective of stage of disease to improvegeneral well being and relaxation. This suggeststhat disease severity is not an important predictorof CAM choice.

The results indicate that mental adjustment forthose who opt for complementary therapy, ischaracterised by significantly higher levels ofanxious preoccupation and fighting spirit. Whilethe latter is not unremarkable, the associationbetween anxious preoccupation and the adoptionof complementary therapies is less obvious. Greeret al. (1989) have described those displayinganxious preoccupation as reacting to their cancerwith marked and persistent anxiety, activelyseeking information about their illness, and tend-ing to interpret this pessimistically. They can seeka range of alternative treatments as was the case inthe present study. However, no evidence wasfound suggesting that those opting for comple-mentary therapy were different in life orientation.This is not consistent with the findings of Ritvoet al. (1999), who, using their risk adaptationmodel have suggested that positive expectancies,i.e. optimism, are important in estimating how an

illness can be modified and in the consequentselection of adaptive behaviours such as engagingin complementary therapy. It is possible that theheterogeneous nature of the CAM group in termsof stage of disease, i.e. early stage cancers,recurring cancers, and those receiving palliativecare, masked any relationship between the role ofoptimism and selection of CAM. It is also possiblethat those who select complementary therapyduring early stage cancer may be more optimistic.There was, however, an association between levelof optimism and anxiety and depression. In otherwords, those with a more pessimistic outlook willhave negative expectations of treatment success, ofdisease progression and the future. There was alsoan association between optimism and internalrecovery locus of control.

The present study found that individuals whoopt for complementary therapy possess a moreinternal recovery locus of control than those whoopt for conventional treatment alone. The decisionto use complementary therapy can be interpretedas a means of regaining control over the un-certainty associated with a diagnosis of cancer.Individuals with an internal recovery locus ofcontrol perceive their role in and responsibility fortheir health decisions as being important in theiroverall recovery. For those patients who opt forcomplementary therapy the use of these therapiesis a meaningful strategy to assume control overtheir disease treatment. This finding supports theTruant and Bottorff (1999) suggestion that use ofcomplementary therapy is part of regaining con-trol and living with cancer. The present study alsohighlighted the possible link between an externalrecovery locus of control and high levels ofdepression. However, the study did not findelevated levels of either depression or anxiety inthose participants who opted for complementarytherapy, which contradicts the view that comple-mentary therapy choice can be viewed as a ‘markerfor distress’ (Burstein et al., 1999).

The qualitative findings in the present studysupported the view that complementary therapywas almost universally viewed as a supportivetreatment for the purposes of relaxation and stressreduction (Table 6). Some participants, however,did express dissatisfaction with the increasinglytechnological approach to medicine, and fragmen-tation of care due to specialisation and clinicianobjectivity. Several of the participants describedtheir attraction to complementary therapy as beinga result of receiving ‘more compassion from

Table 5. Discriminant function classification table

Actual group

membership

Predicted group membership

Complementary Conventional

Complementary 44 (71.4%) 17 (28.6%) 61

Conventional 21 (35.6%) 38 (64.4%) 59

Total 65 55 120



complementary therapists’. A number of patientsincorporated complementary therapy into theirtreatment as a conscious choice to assist insymptom management and the side effects ofconventional treatments. This is consistent withthe view of Cassileth (1999), who demonstratedthat reflexology, aromatherapy and acupuncturehave been shown to alleviate some of the sideeffects of conventional treatment such as hotflushes associated with drug therapies like Tamox-ifen, and the side effects of nausea and fatigueassociated with chemotherapy. The key qualitativefinding is consistent with the quantitative locus ofcontrol findings that perception of ‘loss of control’over ones’ treatment and illness may lead indivi-duals to consider the use of complementarytherapy. One participant commented ‘before start-ing complementary therapy I felt that I had lostcontrol of my life and that I was just doing whatdoctors told me. Now I realise that I was startingto do positive things for me and looking aftermyself better’.

The present study does provide some indicationsthat personality and adjustment variables impacton the choice of complementary therapy. Therewere no significant differences between the twogroups in the key variables of age, cancer type andstage of disease. However, limitations of the studyinclude the non-matched, non-randomised natureof the sample. While a matched sample design isthe ideal, given the nature of this type of research itis rarely possible to match participants onvariables such as diagnosis and stage of illness.The absence of significant differences on somevariables particularly depression and life orienta-tion, may be partially due to the mixed sample.

For example, it is possible that those with endstage disease who opt for complementary therapymay demonstrate greater pessimism and psycho-pathology. Matching participants on illness vari-ables such as diagnosis, and extent of metastaticspread, may clarify further relationships betweensuch variables and use of complementary therapy.Only the three most common CAM therapies wereexamined in the study and it may be thatindividuals who opt for more esoteric therapiescould display different adjustment characteristics.The present study does, however, demonstrate thatthere are key predictors of complementary therapyuptake in the general, treatment seeking, cancerpopulation. The increase in use of complementarytherapy indicates that the psychological as well asthe physical needs of some patients are not beingmet by conventional cancer treatments.

ACKNOWLEDGEMENTS

The authors wish to thank Jan Daley, Lyn Lamont,Marita McMullan, Ashleigh Young and Angela Red-dick for their help with this paper.

REFERENCES

Astin JA. 1998. Why patients use alternative medicine:Results of a national study. J Am Med Assoc 279(19):1548–1553.

Aulas J. 1996. Alternative cancer treatment. Sci Am275(3): 162–163.

Brigden ML. 1995. Unproven (questionable) cancertherapies. West J Med 163: 463–469.

Burstein H, Gelber S, Guadagnoli E, Weeks J. 1999. Useof alternative medicine by women with early stagebreast cancer. N Engl J Med 340(22): 1733–1739.

Carver CS, Pozo C, Harris SD et al. 1993a. How copingmediates the effect of optimism on distress: A study ofwomen with early stage breast cancer. J Pers SocPsychol 65(2): 375–390.

Carver CS, Pozo-Kaderman C, Harris SD et al. 1993b.Optimism versus pessimism predicts the quality ofwomen’s adjustment to early stage breast cancer.Cancer 73(4): 1213–1220.

Cassileth BR. 1999. Complementary and alternativecancer medicine. J Clin Oncol 17(No 11 s): 44–52.

Cassileth BR, Chapman CC. 1995. Alternativeand complementary cancer therapies. Cancer 77(6):1026–1034.

Table 6. Reasons given for complementary therapy use

Reasons for using CAM %

Relaxation 81

General well being 32

Supportive relationship 32

Reduce stress 26

Regain control 26

Reduce anxiety 19

Insomnia 16

Counteract side effects 13

Learning experience 12.9

Pampering experience 9.7

Prolong life 6.5

Increase healing 3

Pain relief 3



Dixon M, Sweeney K (eds). 2000. The Human Effect inMedicine-theory, Research and Practice. RadcliffeMedical Press: Oxon.

Fulder S. 1988. The Handbook of ComplementaryMedicine. Oxford University Press: Oxford (cited inKelner and Wellman, 1997).

Greer S, Moorey S, Watson M. 1989. Patients’ adjust-ment to cancer: The mental adjustment to cancer(MAC) scale vs. clinical ratings. J Psychosom Res33(3): 373–377.

Harrison J, Maguire P. 1994. Predictors of psychiatricmorbidity in cancer patients. Br J Psychiatry 165:593–598.

Kao GD, Devine P. 1999. Use of complementary healthpractices by prostate carcinoma patients undergoingradiation therapy. Cancer 88(3): 615–619.

Marks G, Richardson JL, Graham JW, Levine A. 1986.Role of health locus of control belief as andexpectations of treatment efficacy in adjustment tocancer. J Pers Soc Psychol 51(2): 443–450.

Moorey S, Greer S, Watson M et al. 1991. The factorstructure and factor stability of the hospital anxietyand depression scale in patients with cancer. Br JPsychiatry 158 : 225–259.

Osborne RH, Elsworth GR, Kissane DW, Burke SA,Hopper JL. 1999. The mental adjustment to cancer(MAC) scale: Replication and refinement in 632breast cancer patients. Psychol Med 29: 1335–1345.

Partridge C, Johnston M. 1989. Perceived control ofrecovery from physical disability: Measurement andprediction. Br J Clin Psychol 28: 53–59.

Peters D, Chaitow L, Harris G, Morrison S. (eds). 2002.Integrating Complementary Therapies in PrimaryCare}A Practical Guide for Health Professionals.Churchill Livingstone: Edinburgh.

Redmond K. 1996. Advances in supportive care. Eur JCancer 5(2): 1–7.

Ritvo P, Irvine J, Katz J, Matthew A, Sacamano J,Shaw BF. 1999. The patient’s motivation in seekingcomplementary therapies. Patient Educ Counselling38: 161–165.

Seligman L. 1996. Promoting a Fighting Spirit: Psy-chotherapy for Cancer Patients, Survivors and TheirFamilies. Jossey-Bass: California.

Scheier MF, Carver CS. 1987. Dispositional optimismand physical well-being: The influence of generalizedoutcome expectancies on health. J Pers 55(2):169–210.

Sollner W, Maislinger S, DeVries A, Steixner E,Rumpold G, Lukas P. 2000. Use of complementarytherapy and alternative medicine by cancer patients isnot associated with perceived distress or poorcompliance with standard treatment but with activecoping behaviour. Cancer 89(4): 873–880.

Truant T, Bottorff JL. 1999. Decision making related tocomplementary therapies: A process of regainingcontrol. Patient Educ Counselling 38: 131–142.

VandeCreek L, Rogers E, Lester J. 1999. Use ofalternative therapies among breast cancer patientscompared with the general population. AlternativeTher Health Med 5(1): 71–83.

Verhoef MJ, Hilsden RJ, O’Beirne M. 1999. Comple-mentary medicine and cancer care: An overview.Patient Educ Counselling 38: 93–100.

Vincent C, Furnham A. 1996. Why do patients turn tocomplementary medicine? An empirical study. Br JClin Psychol 35(1): 37–48.

Warrick PD, Irish JC, Morningstar M, Gilbert R. 1999.Use of alternative medicine among patients with headand neck cancer. Arch Otolaryngol}Head NeckCancer 125(5): 573–560.

Watson M, Greer S, Bliss JM. 1989. Mental adjustmentto cancer (MAC) scale. User’s Manual. CRC Psycho-logical Medicine Research Group, Royal MarsdenHospital, Sutton, Surrey, UK.

Wyatt GK, Friedman LL, Given CW, Given BA,Beckrow KC. 1999. Complementary therapy useamong older cancer patients. Cancer Practice 7(3):136–144.

Zigmond AS, Snaith RP. 1983. The hospital andanxiety depression scale. Acta Psychiatr Scand 7:361–370.



Documents

Psychological characteristics of cancer patients who use complementary therapies