chronic fatigue syndrome: an examination of the phases

Chronic Fatigue Syndrome:An Examination of the Phases

Ä

Leonard A. Jason, Guy Fricano, Renee R. Taylor,Jane HalpertDePaul University

Ä

Patricia A. FennellAlbany Health Management Associates, Inc.

Ä

Susan KleinBowling Green State University

Ä

Susan LevineBeth Israel Hospital

The present study examined the Fennell Phase Inventory, an instrumentdesigned to measure the phases typically experienced by individuals withchronic fatigue syndrome (CFS). This inventory yields three factor scoresof Crisis, Stabilization, and Integration. These factor scores have beenemployed in a cluster analysis, yielding four clusters that matched the fourphases predicted by Fennell: Crisis, Stabilization, Resolution, and Integra-tion. The present study represents a partial replication study of a priorinvestigation of the Fennell Phase Inventory by Jason et al. (in press), butthat earlier study did not have an independent physician examination todiagnose patients with CFS. In the present study, 65 patients diagnosedwith chronic fatigue syndrome by a physician were recruited and admin-istered the Fennell Phase Inventory and other measures assessing CFS-related symptoms, disability, and coping. Each of the 65 patients wasclassified into one of four predefined clusters measuring a Crisis phase, aStabilization phase, a Resolution phase, and an Integration phase. Rela-tionships were explored between three of these cluster groupings and

Correspondence concerning this article should be addressed to: Leonard A. Jason, Department of Psychology,DePaul University, 2219 N. Kenmore Ave., Chicago, IL 60614; e-mail: [email protected].

JOURNAL OF CLINICAL PSYCHOLOGY, Vol. 56(12), 1497–1508 (2000)© 2000 John Wiley & Sons, Inc.

measures of symptoms, disability, and coping. Results confirmed Fennell’smodel, revealing significant differences between the three clusters in termsof levels of disability and modes of coping. Results suggest that the Fen-nell Phase Inventory accurately differentiates phases of adaptation to ill-ness experienced by individuals with CFS. © 2000 John Wiley & Sons,Inc. J Clin Psychol 56: 1497–1508, 2000.

Keywords: chronic fatigue syndrome; adaptation; phases

Chronic fatigue syndrome (CFS) appears to be a heterogeneous disease syndrome thatmay be triggered by a variety of factors (Friedberg & Jason, 1998; Fukuda et al., 1994).Quality-of-life investigations have demonstrated that CFS is one of the most debilitatingand functionally incapacitating medical conditions (Anderson & Ferrans, 1997). Phasetheories might provide investigators with new approaches for understanding the com-plexities of enigmatic chronic illnesses like CFS. Phase theories have been applied tounderstand medical service usage, to stop unhealthy behaviors, and to adopt preventivebehaviors (Weinstein, Rothman, & Sutton, 1998). In addition, they present a particularlyrobust lens for understanding the mediators of change and longitudinal processes involvedin coping with demands and challenges posed by serious chronic illness (Miller, 1992;Schweitzer, 1998).

Phase models might help researchers better understand the myriad of contradictoryresearch findings in the field of CFS (Jason, Richman, et al., 1997). If patients experiencequalitatively distinct phases of the CFS illness, their responses on standardized question-naires potentially could be dramatically different from one administration to the next,varying according to illness phase. If researchers collapse the responses of patients indifferent phases of the illness, the accuracy of findings might be obscured, as the patientsare experiencing fundamentally different processes. Patients’ experiences and responses,therefore, may need to be understood separately according to illness phase, rather thancombined into an unwieldy and heterogeneous category.

Fennell (1995) has proposed a four-phase model for understanding CFS. In Phase 1of the CFS illness, the individual moves into a crisis mode shortly after illness onset,wherein she or he experiences the traumatic aspects of a new illness. In Phase 2, theperson with CFS continues to experience chaos and dissembling, followed by the even-tual stabilization of symptoms. In Phase 3, the person with CFS moves into the resolutionmode as he or she works to accept the chronicity and ambiguity of this chronic illness andcreate meaning out of the illness experience. Finally, in Phase 4, the person with CFSachieves integration, wherein he or she is able to integrate pre- and post-illness self-concepts and respond to the illness in a more planful way.

Recently, Jason and associates (in press) examined the factor structure of the FennellPhase Inventory using a sample of 400 participants, who self-reported that a physiciandiagnosed them with CFS on a mail-in questionnaire. A three-factor solution emerged,yielding a Crisis score, a Stabilization score, and an Integration score for each individual.A cluster analysis then was conducted using the three mean factor scores for each indi-vidual, and four clusters emerged. These clusters matched the four phases predicted byFennell (1993, 1995a, 1995b).

The present study is a partial replication of the Jason et al. (in press) study, as well asa replication of Fennell’s (1993, 1995a, 1995b) model, with a sample of participantsdiagnosed with CFS by a physician according to the Fukuda et al. (1994) criteria in a

1498 Journal of Clinical Psychology, December 2000

tertiary-care medical setting. Individuals were clustered into the four phases that previ-ously were documented (Jason et al., in press), and relationships were explored betweeneach of these clusters and measures of symptoms, disability, and coping. It was hypoth-esized that individuals in the Crisis cluster would evidence more distress and shorterillness duration, and that those in the Integration cluster would evidence more effectivecoping skills.

Method

Sample Selection

In the fall of 1998, questionnaires were distributed to 100 patients who were diagnosedwith CFS based on the Fukuda et al. (1994) criteria. All of these patients were giventhorough medical evaluations to rule out other causes for their fatigue. Sixty-six of thepatients completed a consent form and returned the questionnaire to the physician’s office,for a completion rate of 66%. Because we were not given permission to include 34 of thepatients in the study, we did not seek out any additional information on these patients. Allparticipants were current patients of Dr. Susan Levine, a specialist in the diagnosis andtreatment of CFS. Prior to each participant’s enrollment in the study, Dr. Levine gaveeach patient a comprehensive evaluation, including laboratory testing recommended byFukuda et al. (1994), before diagnosing them with CFS.

Measures

Using an initial demographic questionnaire, participants were asked to provide the fol-lowing social-demographic information: age, gender, race, educational level, marital sta-tus, number of children, current occupation, and occupation prior to CFS diagnosis. Patientsalso were asked whether they currently were suffering from severe fatigue, extreme tired-ness, or exhaustion present for a period of six or more months. In addition, participantswere asked how long they had been experiencing CFS. Additional measures administeredare described below.

CFS Symptom Rating Form.Participants also were asked to complete the CFS Symp-tom Rating Form. Using this form, participants rated the severity of their fatigue and theeight CFS definitional symptoms (Fukuda et al., 1994) on a 100-point scale, with 05 nopain or problem and 1005 severe pain or problem. Social/recreational activities wererated on a 10-point scale, with 05 normal levels before the illness, 55 half of one’snormal level, and 105 activities completely curtailed. Work activities also were rated ona 10-point scale, with 05 full-time work without any difficulty, 55 half of one’s normallevel, and 105 not able to work at all. In a previous study (Jason, Ropacki, et al., 1997),a modified version of this form was demonstrated to have high test–retest reliability overa two-week period (test–retest agreement: 76%–92%).

Fennell Phase Inventory.The 20-item Fennell Phase Inventory was administered toeach participant, and each item was rated on a five-point scale (15 definitely do notagree, 55 very strongly agree). The Fennell Phase Inventory was factor analyzed in aprevious study (Jason et al., in press) using two samples. Three distinct factors emergedthat were consistent with the Crisis, Stabilization, and Integration phases of Fennell’s

Phases of the CFS Illness 1499

model (Fennell, 1993, 1995a, 1995b). Cronbach’sa was found to be in the moderaterange for the three factors (0.63–0.83). When these scores were employed in a clusteranalysis, four clusters emerged that were consistent with the four-phase model proposedby Fennell (1993, 1995a, 1995b).

CFS Severity Index.This is a one-item measure of disability in CFS. Respondentswere asked to check one of the following categories: (1) I am ill but not disabled at all; (2)I am minimally disabled; (3) I am mildly disabled; (4) I am moderately disabled; (5) I amseverely disabled; or (6) I am completely disabled. A higher score indicates a higher levelof disability.

Fatigue Severity Scale.The nine-item Fatigue Severity Scale (Krupp, La Rocca,Muir-Nash, & Steinberg, 1989) was used to measure fatigue severity. This scale is scoredon a seven-point scale, where one indicates strong disagreement and seven indicatesstrong agreement. Higher scores indicate higher levels of fatigue.

Medical Outcomes Study SF-36 Health Survey (MOS-SF-36).As a measure of dis-ability, we used the Medical Outcome Survey (MOS-SF-36) (Ware & Sherbourne, 1992).The MOS-SF-36, a 36-item broadly based self-report measure of functional status relatedto health, identifies nine health concepts as perceived by the individual. Higher scoresindicated better health or less impact on health or functioning.

The Illness Management Questionnaire (IMQ).The IMQ (Ray, Weir, Stewart, Miller,& Hyde, 1993) was developed specifically to assess coping in patients with CFS andhas been used extensively in studies of adults with CFS (Ray, Jeffries, & Weir, 1995).The IMQ has four factors: Maintaining Activity (i.e., attempting to ignore symptoms,disregarding possible adverse effects of activity), Accommodating to the Illness (i.e.,organizing and arranging one’s life to avoid exertion and manage stress), Focusing onSymptoms (preoccupation with symptoms, viewing one’s life as dominated by the ill-ness), and Information Seeking (searching for relevant information and an openness totry treatments). Higher scores indicate more agreement with the items measured in eachdomain.

Grouping Procedures

Using statistical procedures described below, each of the 60 participants was assigned toone of four groups reflecting Fennell’s four-phase model: a Crisis Group, a StabilizationGroup, a Resolution Group, or an Integration Group. Six of the 66 original participantswere excluded from this grouping process and all subsequent statistical analyses. Fivewere excluded because they did not complete all items on the Fennell Phase Inventory,and one was excluded because his response profile indicated that he did not fulfill criteriafor current CFS.

In an earlier investigation (Jason et al., in press), a factor analysis was conductedwith the Fennell Phase Inventory in a sample of 400 participants with CFS. The factoranalysis yielded a three-factor solution in which item groupings reflected a Crisisfactor, a Stabilization factor, and an Integration factor. Following the factor analysis,SPSS cluster analysis procedures were used to classify each of the 400 participants intoone of four clusters according to their mean scores on the Crisis, Stabilization, and


Integration factors. The four clusters were found to be consistent with the Crisis, Res-olution, Stabilization, and Integration phases of Fennell’s four-phase model (1993; 1995a;1995b).

Based upon results of this earlier investigation (Jason et al., in press), the followingprocedures were used for assigning each participant in the present study to one of fourgroups, each reflecting one of the phases of Fennell’s four-phase model (Crisis Phasegroup, Stabilization Phase group, Resolution Phase group, Integration Phase group). First,each individual’s mean score was calculated for the Crisis, Stabilization, and Integrationfactors of the Fennell Phase Inventory.1 In the present sample, the Crisis, Stabilization,and Integration factor mean scores ranged from 0 to 5, with higher scores indicating moreagreement with items measured by the corresponding factor.

Using Crisis, Stabilization, and Integration mean scores, each participant then wasassigned to one of the four groups according to the following algorithmic criteria de-rived from the cluster analysis in the Jason et al. (in press) study. Criteria for the Crisisgroup were a Crisis score of 3.00 or above and Stabilization and Integration scores of3.30 or below. Criteria for the Integration group were a Crisis score of 2.50 or below, aStabilization score of 2.80 or below, and an Integration score of 4.25 or above. Casesnot in either of these groups that had either a Crisis score of 3.10 or above, a Stabiliza-tion score 3.40 or above, or an Integration score of 3.75 or above were classified intoa Resolution group. Cases that did not meet any of the above criteria comprised theStabilization group. The sensitivity (correctly identifying cases) and specificity (correct-ly excluding non-cases) of these algorithmic criteria were measured using the 400-participant sample from the Jason et al. (in press) study. The algorithm used for thecrisis group yielded 94.3% sensitivity and 95.8% specificity. The algorithm used for theStabilization cluster yielded 61.6% sensitivity and 61.6% specificity. The algorithm forthe Resolution cluster yielded 95.8% sensitivity and 95.8% specificity. The algorithmfor the Integration cluster yielded 100% sensitivity and 100% specificity.

Using these algorithms, 11 CFS patients were classified into the Crisis Phase group,nine were classified into the Stabilization Phase group, 39 were classified into the Res-olution Phase group, and one CFS patient was classified in the Integration Phase group.Because there was only one CFS patient in the Integration Phase group, comparativestatistical analyses were limited to the three groups (Crisis, Stabilization, and Resolu-tion), with each of the three groups containing nine or more patients. Initial analyses wereconducted to assess for any significant differences between individuals in the Crisis,Stabilization, and Resolution groups in terms of the following social–demographic char-acteristics: gender, age, race, marital status, parental status, number of children, educa-tional status, pre-CFS occupational status, and current occupational status. The SPSSx2

procedure was used to analyze categorical dependent variables, and the SPSS one-wayANOVA procedure was used to analyze differences between the three groups on thecontinuous dependent variables. One significant difference was detected such that indi-viduals in the Crisis (M 5 6.36,SD5 1.96) group reported significantly lower occupa-tional status prior to CFS onset than individuals in the Stabilization (M 5 7.89,SD50.78) group (F (2, 56)5 3.42,p , 0.05). Initial analyses revealed no additional signifi-cant differences between individuals in the Crisis, Stabilization, and Resolution groups interms of the other demographic characteristics.

1The Crisis mean score was calculated by adding items 1, 5, 9, and 15 of the Fennell Phase Inventory anddividing by four. The Stabilization mean score was calculated by adding items 2, 3, 6, 7, 10, 11, 12, 13, 16, and17 of the Fennell Phase Inventory and dividing by ten. The Integration mean score was calculated by addingitems 4, 8, 14, and 18 of the Fennell Phase Inventory and dividing by four.


Results

Social–Demographic Characteristics

With respect to social–demographic characteristics, the sample was primarily female(84.8%) and white (90.9%). The mean age of participants was 45.4 years old. The major-ity of participants either was married (39.4%), never married (37.9%), or divorced (15.2%);only 4.5% were separated and none were widowed. Regarding educational level, 44.6%had a standard college degree or a graduate professional degree, 21.5% had partial col-lege, and 3.1% had a high-school degree or less. With respect to current occupation,72.7% scored at level 1 (unskilled worker or unemployed for a variety of reasons), whereasonly 3.0% respondents had a level-1 rating before becoming ill with CFS. Among par-ticipants, 9.1% were students, 21.2% were homemakers, 33.3% were unemployed, 12.1%were retired, and 75.8% were disabled.2 The average reported duration of CFS was 6.8years (range 1–60 years). We excluded the other participant who answered no to six ormore months of current chronic fatigue because that participant’s response profile indi-cated that he recently had recovered from CFS.

Correlational Analyses

Table 1 presents correlations between mean factor scores (Crisis, Stabilization, and Inte-gration), social–demographic characteristics, and measures of illness severity, disability,and coping. Crisis factor scores were correlated significantly with lower education levelsand a lower occupational status before CFS diagnosis. In addition, Crisis factor scoreswere associated significantly with lower scores on the Role-Emotional, Social Function-ing, Mental Health, and Vitality scales of the MOS. Crisis factor scores also were asso-ciated with a more severe self-rating of fatigue. Stabilization factor scores were correlatedsignificantly with higher Integration factor scores and with younger age. In addition,Stabilization factor scores were associated significantly with higher scores on the SocialFunctioning and Mental Health scales of the MOS. Integration factor scores also werecorrelated significantly with younger age. Integration factor scores also were associatedsignificantly with higher scores on the Accommodating to the Illness and InformationSeeking domains of the IMQ, and with higher scores on the Mental Health scale of theMOS. In addition, Integration factor scores were correlated significantly with higherseverity ratings on the symptom of post-exertional malaise.

Inferential Analyses

We then used the SPSS Analysis of Variance procedure (ANOVA) with Bonferroni post-hoc comparisons to test for differences between the three groups in terms of Crisis,Stabilization, and Integration factor scores, social–demographic variables, fatigue andillness severity ratings, CFS duration, coping styles, levels of disability, and activitylevels. Bonferroni post-hoc comparisons were conducted only when results from an over-all ANOVA were significant. Results are summarized below (see Table 2).

With respect to scores on Crisis, Stabilization, and Integration factors of the FennellPhase Inventory, there were significant differences between individuals in Crisis, Stabil-ization, and Resolution groups. First, individuals in the three groups differed signifi-cantly with respect to Crisis factor scores [F (2, 56) 5 9.11, p , 0.01]. Bonferroni

2Categories were not mutually exclusive, such that some participants endorsed more than one current occupation.


post-hoc comparisons revealed that individuals in the Crisis group evidenced signifi-cantly higher scores on the Crisis factor than individuals in the Stabilization and Reso-lution groups. There also were significant differences between individuals in Crisis,Stabilization, and Resolution groups in terms of Stabilization factor scores [F (2, 56)518.27,p , 0.01]. Bonferroni post-hoc comparisons revealed that individuals in the Crisisand Stabilization groups demonstrated significantly lower Stabilization factor scores thanindividuals in the Resolution group. Similarly, there were significant differences betweenindividuals in Crisis, Stabilization, and Resolution groups with respect to Integrationfactor scores [F (2, 56)5 26.34,p , 0.01]. Individuals in the Crisis group had signifi-

Table 1Means, Standard Deviations, and Intercorrelations with Fennell Item Mean Scores

Mean SD CRISIS STABIL INTEG

Crisis 2.94 .736Stabil 3.15 .613 2.21Integ 3.65 .748 2.24 .51**

Age 45.42 8.81 2.09 2.26* 2.28*Illness Severity 4.39 1.14 .018 2.03 .02Education 5.03 .810 2.39** .06 .14Fatigue Severity 55.95 10.79 .07 2.05 .20Children .730 1.07 .02 2.24 2.25*CFS Duration 6.75 8.28 2.05 2.09 .07Present Job 2.73 2.87 .05 2.08 2.18Pre-CFS Job 6.92 1.49 2.35** .02 2.03

IMQ:Accommodating to Illness 52.91 9.66 2.27 .17 .40**Focusing on Symptoms 30.05 8.07 .24 2.16 .03Information Seeking 25.15 5.47 2.03 .21 .44**Maintaining Activity 54.08 11.05 .12 .15 .08

MOS:General Health 10.18 3.13 2.23 2.02 2.03Physical Functioning 16.72 4.65 2.12 .18 .05Role—Physical 4.20 .640 2.18 .01 2.06Role—Emotional 4.36 1.32 2.29* .07 .03Social Functioning 4.45 1.80 2.30* .34** .17Bodily Pain 4.82 2.09 2.25 2.11 2.06Mental Health 19.20 5.09 2.52** .26* .30*Vitality 7.77 3.64 2.44* .14 .13

Social/Recreat Activities: 7.02 2.11 .11 2.10 .16Work Activities: 7.17 2.59 2.10 .10 .18Fatigue Rating: 74.40 20.38 .36** 2.08 .17New Type of Headaches: 55.92 30.12 2.03 2.14 .21Cognitive Impairment: 71.75 21.33 .20 2.23 .10Multiple Joint Pain: 64.95 30.80 .18 2.04 .02Muscle Pain: 66.73 30.17 .21 .03 .02Post Ex. Malaise: 75.86 22.83 .25 .03 .30*Sore Throat: 42.67 33.47 .00 .07 2.10Tender Neck/Lymph 42.73 31.26 2.01 .18 .13Unrefreshing Sleep: 74.92 25.52 .25 .07 2.05

* 5 p , .05; ** 5 p , .01.


cantly lower Integration factor scores than individuals in the Resolution group, and indi-viduals in the Stabilization group also evidenced significantly lower scores on theIntegration factor than individuals in the Resolution group.

With respect to coping styles, there were significant differences between the threegroups with respect to IMQ factors of Accommodating to the Illness [F (2,56)5 4.64,

Table 2Significant Cluster Differences (N = 60)

Crisis(n 5 11)

Stabilization(n 5 9)

Resolution(n 5 39)

Mean SD Mean SD Mean SD

Crisis Factor Score 3.70a** b** 0.37 2.56a** 0.21 2.87b** 0.77Stabilization Factor Score 2.65a** 0.45 2.70b** 0.46 3.45a** b** 0.47Integration Factor Score 2.84a** 0.56 3.02b** 0.42 4.02a** b** 0.57Age 49.27 7.10 46.11 7.83 43.74 9.41Illness Severity 4.55 0.93 4.33 1.12 4.49 1.14Education 4.45 1.04 5.00 0.87 5.11 .69Fatigue Severity 57.55 5.87 57.63 5.95 57.05 9.48Children 1.45 1.44 0.89 1.17 0.59 .94CFS Duration 8.00 9.43 6.22 4.60 6.74 9.48Present Job 2.09 2.47 3.33 3.54 2.74 2.84Pre-CFS Job 6.36 1.96 7.89 0.78 6.95 1.17IMQ:

Accommodating to the Illness 47.00* 5.69 52.22 9.30 55.82* 9.05Focusing on Symptoms 29.27 9.17 34.11 6.66 30.00 7.75Information Seeking 20.55** 5.96 24.89 4.31 26.79** 4.85Maintaining Activity 57.10 7.37 47.22 6.94 55.37 12.13

MOS:General Health 10.73 3.52 9.11 1.97 10.07 2.75Physical Functioning 16.00 4.45 15.63 4.03 16.82 4.56Role—Physical 4.09 0.30 4.33 0.71 4.13 0.41Role—Emotional 4.09 1.38 4.67 1.41 4.23 1.31Social Functioning 3.73 1.35 3.89 1.17 4.71 1.83Bodily Pain 3.95 1.30 5.85 2.85 4.61 1.98Mental Health 16.64 5.52 20.67 3.46 19.59 5.25Vitality 6.18 1.99 7.44 3.81 8.03 3.46

Social/Recreational Activities: 7.10 1.91 7.25 1.83 7.18 1.91Work Activities: 7.40 3.10 6.78 2.91 7.61 2.06Fatigue Rating: 80.20 16.21 70.56 21.86 75.64 18.22

Pre-CFS(%)

Current(%)

Pre-CFS(%)

Current(%)

Pre-CFS(%)

Current(%)

New Type of Headaches: 50.0 90.0 11.1 77.8 48.6 94.6Cognitive Impairment: 50.0 100.0* 0.0 77.8* 36.8 97.4*Multiple Joint Pain: 50.0 90.0 11.1 66.7 31.6 86.8Muscle Pain: 50.0 100.0** 11.1 55.6** 36.8 89.5**Post-Exertional Malaise: 40.0 100.0 0.0 100.0 38.5 92.3Sore Throat: 50.0* 60.0* 0.0* 77.8* 48.6* 94.6*Tender Neck/Lymph Nodes: 50.0* 60.0 0.0* 88.9 37.8* 83.8Unrefreshing Sleep: 30.0 100.0 11.1 88.9 47.4 100.0

Means with common superscripts across rows (a, a; b, b) are significantly different.* 5 p # .05; ** 5 p # .01.


p , 0.01] and Information Seeking [F (2,56)5 6.76, p , 0.01]. Bonferroni post-hoccomparisons revealed that individuals in the Crisis group were significantly less likely touse Accommodating to the Illness as a coping style and engaged in significantly fewerinformation-seeking practices than patients in the Resolution phase.

With respect to the percentage of patients in each group with a particular symptom,Pearsonx2 analyses revealed significant differences between the three groups for threecurrent CFS symptoms lasting six or more months (see Table 2). With respect to thesymptom of muscle pain for six or more months [X2 (2, N 5 59)5 8.92,p 5 0.01], allindividuals in the Crisis group reported muscle pain, a majority reported muscle pain inthe Resolution group, and a little over half reported muscle pain in the Stabilizationgroup. There also were significant differences for sore throat of six or more months [X2

(2, N5 59)5 8.25,p , 0.05]. Fewest individuals in the Crisis group reported sore throat,as compared with almost all individuals in the Resolution group and a majority of indi-viduals in the Stabilization group. In addition, there were significant differences withrespect to impairment in short-term memory or concentration of six or more months [X2

(2, N 5 59)5 6.28]. All participants in the Crisis group and almost all individuals in theResolution group reported memory or concentration impairment, as compared with 78%of individuals in the Stabilization group.

In addition, there were significant differences for two symptoms predating CFS onset,sore throat [X2 (2, N 5 59)5 7.48,p , 0.05] and lymph node pain [X2 (2, N 5 59)56.03,p5 0.05]. With respect to sore throat, half of the individuals in the Crisis group andalmost half of the individuals in the Resolution group reported sore throat before CFSonset, whereas no individuals in the Stabilization group reported sore throat before CFSonset. In terms of lymph-node pain, half of the individuals in the Crisis group reportedlymph node pain before CFS onset, whereas fewer individuals in the Resolution groupand no individuals in the Stabilization group reported lymph node pain.

Discussion

Findings from the present study were supportive of a recent study by Jason and associates(in press), which demonstrated that individuals with CFS can be sub-typed according tofour distinct illness phases: Crisis, Stabilization, Resolution, and Integration. The presentstudy demonstrates a partial replication of findings of the Jason et al. (in press) study,including only those patients who had been diagnosed with CFS by a physician. In addi-tion, the present study supplements results of the Jason et al. (in press) study in that itintroduces measures of psychosocial functioning and coping as related to the four CFSillness phases.

Results from the present study are consistent with results from a prior investigationof the four-phase model of CFS illness (Jason et al., in press) in a number of ways. In bothstudies, characteristics of individuals in the Crisis group were consistent with character-istics of individuals in the first phase of Fennell’s model (1993, 1995a, 1995b), the CrisisPhase. Findings from the present study suggest that the Crisis phase appears to be char-acterized by more profound illness severity, greater symptom severity, higher fatigueseverity, greater psychological distress, and higher levels of functional impairment ascompared with the other illness phases. With respect to coping styles, individuals in theCrisis group were significantly less likely to use Accommodating to the Illness as acoping style, and they engaged in significantly fewer information-seeking practices thanindividuals in the Resolution group. These findings are consistent with a certain level ofillness denial and less-active coping strategies that characterize the Crisis phase as describedby Fennell (1993, 1995a, 1995b). Demographic findings for a significant association


between Crisis factor scores and lower pre-CFS occupational status and lower educa-tional level suggest that the circumstance of having fewer socioeconomic resources maycomplicate adaptation to the CFS illness.

One unexpected, albeit insignificant, finding in the present study was that individ-uals in the Crisis group evidenced the longest CFS duration of all three groups. Theo-retically, this finding appears to stand in contrast to the results of a study by Buchwald,Pearlman, Kith, Katon, and Schmaling (1997). These researchers found an associationbetween shorter duration of illness and greater psychological distress. One explanationfor this discrepancy may involve the issue of temporality as it relates to Fennell’s four-phase model (1993, 1995a, 1995b). It is possible that more person-centered character-istics, such as the psychosocial resources one possessed prior to CFS, the psychologicalresources one possesses during CFS, the degree of support one has during the illness,and the level of illness severity are more likely to determine illness phase than externalfactors, such as the relative newness of CFS onset. It is possible that those individualswith greater illness severity also have more psychological distress, fewer resources, andgreater illness chronicity, and hence find themselves less able to progress out of theCrisis phase and through the other phases of adaptation to the illness. Without key re-sources to assist in recovery from a devastating and multifaceted loss such as CFS,patients will not be able to develop adequate coping strategies and compensatory re-sources over time (Hobfol, 1998). Further research is needed to better understand thesefindings.

Characteristics of individuals in the Stabilization group were consistent with char-acteristics of individuals in the second phase of Fennell’s model (1993, 1995a, 1995b).Findings suggest that the Stabilization phase appears to reflect some plateau or decreasein the perception and experience of CFS symptoms, as well as an increasing adaptation tothe illness. Individuals in the Stabilization group reported the lowest levels of functionalimpairment in the areas of physical and psychological functioning, and they demon-strated the lowest levels of fatigue, CFS illness severity, and CFS symptom severity withrespect to a majority of the minor symptoms (Fukuda et al., 1994). With respect to cop-ing, individuals in the Stabilization group were more likely to use strategies of Accom-modating to Illness and Information Seeking than individuals in the Crisis group, but lesslikely to use these strategies than individuals in the Resolution group. This pattern isconsistent with an intermediate movement away from denial-based coping and towardmore planful coping oriented toward self-care that is characteristic of the Stabilizationphase.

Characteristics of individuals in the Resolution group were consistent with charac-teristics of individuals in the third phase of Fennell’s model (1993, 1995a, 1995b). Ingeneral, findings supported the notion that individuals in the Resolution phase may expe-rience relapse or a renewed escalation in CFS symptomatology, but they psychologicallyare prepared more to cope with their symptoms than individuals in the Crisis phase. In thepresent study, individuals in the Resolution group reported intermediate levels of impair-ment in general health, physical, and psychological functioning that were higher thanindividuals in the Stabilization group but lower than individuals in the Crisis group.Similarly, individuals in the Resolution group demonstrated intermediate levels of CFSillness severity, and intermediate levels of CFS symptom severity for six of eight minorsymptoms (Fukuda et al., 1994). These findings support the notion that individuals in theResolution group appear to be moving toward a state of psychological transcendence, butcontinue to struggle with issues involving the integration of pre- and post-illness life-styles. Reflective of this greater sense of perspective, individuals in the Resolution groupwere significantly more likely than individuals in the Crisis group to use coping strat-


egies of Accommodation to the Illness and Information Seeking. Placed within a broadercontext of other theories relevant to coping with chronic illness, this particular finding is,in part, consistent with results from a study by Felton, Revenson, and Hinrichsen (1984).These researchers studied stress and coping in 170 middle-aged and elderly adults withchronic illnesses and found that cognitive coping strategies, including Information Seek-ing, were related to positive affect and better adjustment to the illness.

It is of interest that there was only one participant in the present study who wasclassified into the Integration phase. This finding may be attributable to the fact that oursample was drawn from a tertiary-care medical patient population. It is possible thatindividuals who are seeking care from a physician specializing in the treatment of peoplewith CFS tend to find themselves more within the earlier phases of the CFS illnessexperience. Accordingly, it is possible that those within the Integration Phase are morefamiliar with optimal forms of treatment for their symptoms, more accustomed to pre-dicting the course of their illness, and/or are more likely to seek care through primarycare or alternative medical practitioners.

Findings in the present study not only support findings from a previous study (Jasonet al., in press), but results from newly implemented measures of psychosocial function-ing lend further insight into ways in which individuals with CFS progress through dis-tinctive psychological, social, and physical phases of the illness experience. These resultssuggest that the Fennell Phase Inventory and the accompanying psychosocial measuresused in this study can be used by practitioners as tools to inform the assessment ofillness coping and adaptation in CFS patients. The Fennell Phase Inventory also mightbe used intermittently throughout the treatment process to assess patients’ progress inpsychotherapy.

Until the findings in the present study are replicated with additional samples, cautionneeds to be exercised in generalizing these findings to all individuals with CFS. Futurestudies of Phase theory (Fennell, 1995) should incorporate biological markers that areconsistent with patients’ reports of a physical illness course throughout the differentphases of CFS. McGregor and associates (1998) approximate this objective in that theyhave separated patients with CFS into subgroups based on host responses, infectiousresponses, sudden or gradual onset, and other differences in symptom and biochemicalconstellations. A second limitation of this study is that we chose to concentrate on explor-ing the relationship between problem-focused coping and the illness phases, and we didnot include measures of other types of coping. Given that previous research has foundthat emotion-focused coping is a common form of coping used by individuals facingvarious types of chronic illnesses, we would recommend that future studies of adaptationto CFS also incorporate measures of emotion-focused coping. Another limitation of thisstudy is that our theoretical model, in part, informed our choice of outcome measures. Asa result, our findings may have been more likely to support Fennell’s (1995) four-phasemodel, and we may not have included other important measures of functioning that maybe relevant to the topic of adaptation to CFS. Future studies should evaluate more rigor-ously this model by including a wider array of outcome measures to assess a broaderrange of functioning and adaptation to CFS. In addition to these issues, it is important torecognize that both physical and psychological changes in CFS can occur in a nonlinearfashion. People may regress from a more-integrated stage to a lower one, skip an entirestage, remain in a single stage without progression, or evidence characteristics of twostages simultaneously. Therefore, it is important not to view the Fennell (1995) four-phase model as a linear model that confines predictions to rigid movement through stagesbecause human beings are far more complex and less predictable than hypothesized inseveral of these theories.


References

Anderson, J.S., & Ferrans, C.E. (1997). The quality of life of persons with chronic fatigue syn-drome. The Journal of Nervous and Mental Disease, 185, 359–367.

Buchwald, D., Pearlman, T., Kith, P., Katon, W., & Schmaling, K. (1997). Screening for psychiatricdisorders in chronic fatigue and chronic fatigue syndrome. Journal of Psychosomatic Research,42, 87–94.

Felton, B.J., Revenson, T.A., & Hinrichsen, G.A. (1984). Stress and coping in the explanation ofpsychological adjustment among chronically ill adults. Social Science and Medicine, 18,889–898.

Fennell, P.A. (1993). A systemic, four-stage progressive model for mapping the CFIDJ experience.CFIDJ Chronicle, Summer, 40–46.

Fennell, P.A. (1995a). The four progressive stages of the CFS experience: A coping tool for patients.Journal of Chronic Fatigue Syndrome, 1, 69–79.

Fennell, P.A. (1995b). CFJ, sociocultural influences, and trauma: Clinical considerations. Journalof Chronic Fatigue Syndrome, 1, 159–173.

Friedberg, F.R., & Jason, L.A. (1998). Understanding chronic fatigue syndrome: An empiricalguide to assessment and treatment. Washington DC: American Psychological Association.

Fukuda K., Straus, S.E., Hickie, I., Sharpe, M.C., Dobbins, J.G., & Komaroff, A. (1994). Thechronic fatigue syndrome: A comprehensive approach to its definition and study. Annals ofInternal Medicine, 121, 953–959.

Hobfol, S.E. (1998). Stress, culture, and community. New York: Plenum.

Jason, L.A., Fennell, P., Klein, S., Fricano, G., Halpert, J., & Taylor, R.R. (in press). An investiga-tion of the different phases of the CFS illness. Journal of Chronic Fatigue Syndrome.

Jason, L.A., Richman, J.A., Friedberg, F., Wagner, L., Taylor, R.R., & Jordan, K.M. (1997). Poli-tics, science, and the emergence of new disease: The case of chronic fatigue syndrome. Amer-ican Psychologist, 52, 973–983.

Jason, L.A., Ropacki, M.T., Santoro, N.B., Richman, J.A., Heatherly, W., Taylor, R.R., Ferrari, J.R.,Haney-Davis, T.M., Rademaker, A., Dupuis, J., Golding, J., Plioplys, A.V., & Plioplys, S.(1997). A screening scale for Chronic Fatigue Syndrome: Reliability and validity. Journal ofChronic Fatigue Syndrome, 3, 39–59.

Krupp, L.B., La Rocca, N.G., Muir-Nash, J., & Steinberg, A.D. (1989). The fatigue severity scale:Application to patients with multiple sclerosis and systemic lupus erythematosis. Archives ofNeurology; 46, 1121–1123.

McGregor, N.R., Hoskin, L., Dunstan, R.H., Clifton Bligh, P., Butt, H.L., Fulcher, G., Roberts,T.K., Dunsmore, J., Zerbes, M., & Klineberg, I.J. (1998). Heterogeneity of symptom, onset andbiochemical profiles in “defined” CFS patients. Paper presented at the American Associationof Chronic Fatigue Syndrome annual meeting, Boston, MA.

Miller, W. (1992). The experience of nine women living with chronic fatigue syndrome, as dem-onstrated through mental imagery, drawings, and verbal descriptions. Unpublished doctoraldissertation. Cincinnati, OH: The Union Institute.

Ray, C., Weir, W., Stewart, D., Miller, P., & Hyde, G. (1993). Ways of coping with chronic fatiguesyndrome: Development of an illness management questionnaire. Social Science Medicine,37, 385–391.

Ray, C., Jefferies, S., & Weir, W.R.C. (1995). Life-events and the course of chronic fatigue syn-drome. British Journal of Medical Psychology, 68, 323–331.

Schweitzer, M.M. (1998, May 6). WECAN testimony to CFSCC [Online]. Available:http://www.cfids-me.org/wecan/[email protected].

Ware, J.E., & Sherbourne, C.D. (1992). The MOS 36-item Short-Form Health Survey (SF-36):Conceptual framework and item selection. Medical Care, June, 473–483.

Weinstein, N.D., Rothman, A.J., & Sutton, S.R. (1998). Stage theories of health behavior: Concep-tual and methodological issues. Health Psychology, 17, 290–299.


chronic fatigue syndrome: an examination of the phases

Documents