The number of cancer survivors continues to increase,1,2 and for many survivors, their fear of cancer recurrence (FCR) is among their top concerns.3 Fear of cancer recurrence is defined as the “fear, worry, or concern relating to the possibility that cancer will come back or progress”4(p3265) and is experienced by as many as 72% of survivors.5 Higher FCR is associated with an increased use of healthcare resources6,7 and a number of negative consequences for cancer survivors, including poorer quality of life,8 higher anxiety,9 and higher depression.10
Cancer survivors want help from healthcare providers (HCPs) to cope with FCR11,12; however, HCPs, many of whom are clinical nurses, desire further training about FCR.13 One way to facilitate this training is to enhance clinical oncology nurses’ ability to identify survivors at increased risk of heightened FCR, especially because FCR tends to be underrecognized by clinical HCPs.13 To identify survivors at increased risk of heightened FCR, an understanding about the predictors of FCR would be useful, and Leventhal’s Commonsense Model (CSM)14,15 is relevant in this regard. The CSM is helpful to understand how individuals make sense of their illness experiences16 and has been used to conceptually describe FCR.17,18 Leventhal and colleagues15 suggest that an individual’s characteristics of the self (eg, self-esteem, or response-specific self-efficacy) and the social-cultural context (eg, labeling and detection of symptoms, or decisions to seek care from health practitioners) influence their making sense of their illness experiences. In this regard, the CSM14,15 is useful to guide an examination about the predictors of FCR that can be categorized as sociodemographic characteristics, clinical characteristics, or characteristics of the self.
Leventhal et al14,15 posit that the social and cultural contexts within which one functions influence a person’s illness representations: Such contexts would include the sociodemographic variables captured in studies that have explored the predictors of FCR. Research findings about these predictors reveal contradictory results between FCR and age, marital status, ethnicity, education level, and employment-related variables. Systematic reviews19–21 summarizing these results were without conclusions about the relationships among most of these predictors with FCR because of the contradictory findings. Other sociodemographic variables, such as parental status or having dependent children, have only moderate support for their association with FCR,21 whereas immigration status and geographical location of residence (eg, urban or rural) have not been examined with FCR. Drawing from the CSM, immigration status, or number of years living in a country22 may inform how individuals make sense of illness experiences, such as FCR. In support of this, a systematic review revealed that rural cancer survivors have a higher psychological morbidity than their urban counterparts.23 Collectively, results suggest further examination between FCR and these sociodemographic variables is necessary.
The CSM14,15 indicates that social observations and comparisons,15 as well as interpretations of concrete, perceptual experiences,24 contribute to illness sense-making. This suggests that cancer survivors’ associations with cancer, such as knowing someone with a cancer recurrence, may contribute to FCR. Some research studies were developed to explore the relationship between FCR and patient perceptions, such as satisfaction with care25; however, few research studies have examined the association between FCR and patient perceptions.
As a result of illness experiences, each person formulates a repository of memories that inform their understanding about their future disease processes.15 Related to the clinical characteristics that may predict FCR, type of diagnosis, cancer stage, receipt of another cancer diagnosis (eg, metastatic disease or recurrence), time since diagnosis, and cancer treatment, have been studied; however, the relationships of these variables with FCR are varied as outlined in the summaries presented in systematic reviews.19–21 One of these reviews19 concluded that there was strong evidence relating FCR and number of physical symptoms or comorbidities, although the results of original research reports included in the review are heterogeneous. The CSM highlights the importance of somatic experiences upon how individuals make sense of their illness,15 suggesting that one’s perceptions about the cause of physical symptoms, commonly referred to as symptom attribution,26 are important to consider in relation to FCR; however, no known studies have examined this relationship.
Characteristics of the Self
According to the CSM, personal attributes contribute to a person’s sense of vulnerability to, or likely success in preventing, negative health effects.15 Self-esteem is identified as a paramount attribute of the self,15 but only one study27 has examined self-esteem with FCR. Another important attribute to consider is the generalized expectancies that one has toward an outcome,28 commonly dichotomized as optimism-pessimism.29 Although inverse associations between optimism and FCR are consistently reported in the literature,10,30,31 no known published results relating pessimism and FCR are available. Furthermore, research reports10,30,31 relating FCR and optimism appear to have measured optimism-pessimism in a unidimensional manner, whereas optimism and pessimism are most accurately measured as 2 distinct factors (eg, bidimensional32). For these reasons, self-esteem and generalized expectancies are important to further investigate in relation to FCR.
The CSM14,15 is a useful framework to guide empirical examinations about the predictors of FCR. However, as described in the previous sections, studies that have examined predictors of FCR have presented opposing results about variables that may be helpful for clinical oncology nurses to identify survivors at greatest risk of heightened FCR. Early identification of such high-risk survivors would allow for earlier education and/or referral to interventions that facilitate coping with FCR. Therefore, to provide empirical clarity about the relationships of these variables with FCR, the objective of this study was to examine sociodemographic characteristics, clinical characteristics, and characteristics of self as predictors of FCR within a large sample of cancer survivors. This objective is visually presented in Figure 1.
Sample and Setting
Study participants completed treatmenta for breast, testicular, melanoma, gastrointestinal, gynecological, thyroid, or lung cancer and were enrolled at a cancer survivorship clinic. The clinic provides cancer survivorship care (for approximately 5 years) to cancer patients being transitioned from tertiary to primary care. Participants were eligible for this study if they were (1) older than 18 years, (2) currently or had ever received care at the clinic, (3) currently free of cancer, (4) not cognitively impaired, (5) proficient in English, (6) able to provide informed consent, and (7) accessible by letter mail. Individuals were excluded if they had a diagnosis of childhood cancer.
Design and Procedures
Recruitment and data collection occurred between January and August 2015. Initially, the clinic’s medical director mailed an information letter describing the study and options for participating (either by hardcopy or an electronic questionnaire via the FluidSurveys33 platform) to all clinic patients. Patients who did not opt out from further contact were mailed a study package (consent form, questionnaire, and return-addressed postage-paid envelope) 10 days thereafter. Two weeks later, a scripted telephone call was made to those who had not returned a completed study package, and packages were remailed as requested. As completed packages were returned, the researcher conducted a medical chart review to confirm eligibility and extract study data. Hospital and university research ethics boards approved study procedures, and participants provided written informed consent.
FEAR OF CANCER RECURRENCE
Fear of cancer recurrence was assessed using the Fear of Cancer Recurrence Inventory (FCRI),8 a self-report measure that comprised 42 items on 7 subscales (Triggers, Severity, Psychological Distress, Functioning Impairments, Insight, Reassurance, and Coping Strategies). Responses are based on a 5-point Likert-like scale, where 0 indicates “not at all or never,” and 4 indicates “a great deal or all the time.”8 After responses to item 13 were reverse-scored, responses are summed to obtain subscale and overall FCRI scores. Higher scores indicate higher FCR.8 The validity and reliability of the FCRI have been demonstrated in samples of cancer survivors.8,34,35
Sociodemographic variables (age, gender, marital status, parental status, having dependent children, education level, employment status, ethnicity, immigration status, and urban/rural location) were collected via self-report. Additionally, data were collected for 3 associations with cancer variables: “In your personal life, is/was there someone close to you who had a diagnosis and treatment for cancer, and then cancer came back?” and “Has that person’s cancer returning affected your fear that your cancer may come back?” to which respondents answered “yes,” “no,” or “don’t know”; and “What is your relationship to the survivorship clinic?” to which respondents answered “I am currently being followed by the survivorship clinic” or “I have been discharged from the survivorship clinic.”
Clinical variables (diagnosis type and stage, receipt of another cancer diagnosis, time since diagnosis, cancer treatment received, and number of comorbidities36) were collected via self-report and medical chart review. Additionally, information about symptom attributionsb,26 was collected using the Revised Illness Perception Questionnaire Identity subscale.26 Participants indicated (yes/no) whether they have experienced any of the 14 symptomsc and whether they believed (yes/no) the symptom is related to cancer.26 Items to which participants reported “yes” on the second set of questions were summed; higher scores represented stronger beliefs about the symptom’s relationship to cancer.26 The validity of the Revised Illness Perception Questionnaire has been demonstrated26 and acceptable Cronbach’s α’s have been determined in samples of cancer samples.30,37
CHARACTERISTICS OF SELF
Participants completed the Rosenberg Self-esteem Scale (RSES),38 10 items on a 4-point Likert-type scale, ranging from 0 (strongly disagree) to 3 (strongly agree). Items were summed38 whereby an overall score was indicated for 2 self-esteem factors: a positively worded factor (items 1, 3, 4, 7, and 10) and a negatively worded factor where items 2, 5, 6, 8, and 9 were reverse scored prior to summation.39
The Revised Life Orientation Test (LOT-R)28 was used to assess generalized expectancies (dichotomized as optimism-pessimism). Participants indicated their level of agreement on ten 4-point Likert scale items, which, apart from the 4-filler items, were summed to obtain an overall score40 for 2 factors32: items 1, 4, and 10 were summed to reflect the optimism factor, and items 3, 7, and 9 were reverse scored and summed to reflect the pessimism factor.
SPSS version 24 (IBM Corporation, New York, NY) was used to conduct descriptive analyses. Cases missing 1 or more items on a measure without published guidelines for missing data were excluded listwise (n = 18–47).
To explore sociodemographic, clinical, and characteristics of the self as predictors (otherwise referred to as direct effects) of FCR, a structural equation modeling (SEM41) analysis was conducted using Mplus version 8.42 Structural equation modeling is unique from other statistical approaches, such as regression analysis and path analysis, and is described as a synthesis of path analysis and confirmatory factor analysis (CFA) in that it simultaneously tests 2 types of models: a theoretical/structural model and a measurement model.41 In this way, the relationships among the variables are explored (similar to that of multiple regression or path analysis), while measurement error associated with the measures used (similar to that of CFA) is corrected. For these reasons, SEM was an appropriate analysis to use in this study.
Initially, CFA was used to explore the fit of each measure to study data.41 After an acceptable fit was achieved per measure, the full measurement model—FCRI, RSES, and LOT-R together—was examined. Thereafter, relationships (or direct effects) were tested in a structural regression model,41 and the overall fit of the model and coefficients of relationships43 were reviewed.
Model fit was assessed using root mean square error of approximation (RMSEA)44 for which “good fit” ranges from 0.0645 to 0.07.46 For this study, a good fit of the models was indicated by an RMSEA of 0.07 or less.46 Other commonly referred to fit indices include the comparative fit index (CFI) and the Tucker-Lewis Fit Index (TLI). For this study, good fit of the models was indicated by a CFI of 0.94 or greater and TLI of 0.94 or greater.45 A χ2 test was reported for informational purposes only since it is often statistically significant in large samples.41 All analyses used the robust maximum likelihood method estimator (MLR)42 and .05 level of significance (2-tailed).
Of the 2143 survivors on the clinic roster, 129 were ineligible because of limited English (n = 54), cognitive impairment (n = 3), current cancer treatment (n = 6), unreachable by mail (n = 52), or deceased (n = 14). Among the remaining 2014 patients, 1001 consented to participate (49.70% participation rate), and 955 (95.40%) completed questionnaires in hardcopy (the remaining 46, or 4.6%, completed the questionnaires using the FluidSurveys33 platform). Respondent characteristics were not compared with nonrespondents because ethical clearance did not permit access to nonrespondent data.
The average age of the sample was 61.12 years (range, 23–98 years); the majority identified as female (85.01%) and white (77.12%). Participants were an average of 9.07 years postdiagnosis (range, 1–36 years), and most had been diagnosed with breast cancer (65.93%) and had received cancer treatment (89.11%). The mean FCRI score was 57.77 (SD, 28.64),d and the overall LOT-R and RSES scores were 16.01 (SD, 4.29) and 23.77 (SD, 5.23), respectively. Participants’ sociodemographic and clinical characteristics are detailed in Tables 1 and 2.
Table 1 -
Sociodemographic Characteristics of Participants (n = 1001)
| Married or common-law
| All other groups
| Not parent
|Have dependent children
| Up to some undergraduate
| Undergraduate graduate or higher and other
| Actively employed
| Not actively employed
| White, Caucasian, or European descent
| All other ethnicitiesb
| Not born in Canada
| Born in Canada
|Rural or urban location
|Survivorship clinic status
|Know someone with recurrence
| No/don’t know
|Knowing someone with recurrence affects FCR
| No/don’t know
Abbreviation: FCR, fear of cancer recurrence.
aMean (SD) used.
bChinese, Southeast Asian, Korean, Japanese (6.61%); Filipino (3.90%); South Asian (east Indian, Pakistani, Sri Lankan, etc) (3.21%); black or African American/African Canadian (2.71%); Hispanic, Latino, Mexican American, or Central American (1.30%); Arab or West Asian (1.30%); native Canadian (Inuit, Indigenous) (0.20%); mixed (parents are from 2 different groups) (1.30%); other (2.10%).
Table 2 -
Clinical Characteristics of Participants
|Time since diagnosis,a y
| Breast cancer
| All other cancersb
| Stages 0–1
| Stages 2–4
|Any cancer treatment receivedd
|Chemotherapy treatment received
|Radiation treatment received
|Other cancer treatment receivede
Abbreviations: AJCC, American Joint Committee on Cancer; FCR, fear of cancer recurrence.
aMean (SD) used.
bGastrointestinal (8.83%), testicular (5.23%), gynecological (7.95%), melanoma (7.72%), thyroid (3.61%), other (0.50%).
cn = 984.
dIncludes only chemotherapy, radiation therapy, or other cancer therapy (eg, aromatase inhibitors).
eAromatase inhibitors, and so on.
Confirmatory Factor Analysis
None of the sociodemographic characteristics, clinical characteristics, or characteristics of the self were highly correlated with each other: bivariate correlations were used for continuous independent variables, and linear regression was used for nominal independent variables. The CFA for each measure revealed good to acceptable fits (Table 3, Appendix), and the overall measurement model demonstrated an acceptable fit to the data (Table 3).
Table 3 -
Characteristics of Measures
||RMSEA Estimate (90% CI)
|Fear of Cancer Recurrence Inventory (FCRI)a
|Generalized expectancies (LOT-R) b
|Full measurement model
Abbreviations: α, α coefficient (internal consistency); CFI, comparative fit index; CI, confidence interval; df, degrees of freedom; RMSEA, root mean square error of approximation; TLI, Tucker-Lewis Fit Index.
aSecond-order confirmatory factor analysis.
bFirst-order confirmatory factor analysis.
cP ≤ .001.
dP ≤ .0001.
Structural Model Analysis
The RMSEA for the overall structural model indicated a good fit to the data, although the CFI and TLI indicated a slightly less than optimal fit: χ2 = 7411.32, P < .0001, df = 2884, RMSEA = 0.040 (0.039, 0.041), CFI = 0.873, TLI = 0.868. The coefficients, standard errors, and levels of significance for each predictor are indicated in Table 4; the standardized coefficients are visually presented in Figure 2.
Table 4 -
Effects of Sociodemographic, Clinical, and Characteristics of Self on FCR
|Have dependent children
|Level of education
|Know someone with a recurrence
|Believe knowing someone with recurrence affects FCR
|Survivorship clinic status
|Time since diagnosis
|Recur/Metastatic/Another Primary Diagnosis
|Other cancer treatmenta
|Any cancer treatmentb
|No. of comorbid conditions
Characteristics of Self
|Generalized expectancies (LOT-R25)
|Positively worded items factor
|Negatively worded items factor
Abbreviations: AJCC, American Joint Commission on Cancer; B, unstandardized coefficient; β standardized coefficient; FCR, fear of cancer recurrence; LOT-R, Revised Life Orientation Test; RSES, Rosenberg Self-esteem Scale; SE B, standard error.
aAromatase inhibitors, and so on.
bIncludes only chemotherapy, radiation therapy, or other cancer therapy (eg, aromatase inhibitors).
Results (Table 4) indicated that younger age was associated with higher FCR, whereas being male was associated with lower FCR. Survivors who knew someone with a cancer recurrence had lower FCR, whereas those who believed that knowing someone with a recurrence affected their own level of FCR had a higher level of FCR. In comparison to those discharged from the survivorship clinic to their primary care provider, survivors who continued to be followed at the clinic had higher FCR. None of the other sociodemographic characteristics were significantly associated with FCR. Similarly, none of the diagnostic variables (eg, cancer type or stage) or treatment variables influenced FCR; however, higher FCR was associated with increased time since diagnosis. Although number of comorbid conditions was not associated with FCR, survivors who reported more symptoms attributed to cancer had higher FCR. Only the self-esteem factor with negatively worded itemse was negatively associated with FCR. Finally, more pessimistic survivors had lower FCR, whereas optimism did not have a significant effect on FCR.
This study examined predictors of FCR using a theory useful to understand FCR (the CSM) and analyzed data using SEM techniques. In this large sample of cancer survivors, the strongest predictor of higher FCR was belief that knowing someone with a recurrence affects one’s own FCR, although knowing someone with a recurrence predicted lower FCR. Findings revealed that higher FCR was predicted by (in descending order of strength) higher symptom attribution, lower self-esteem, younger age, female gender, lower pessimism, longer time since diagnosis, and active follow-up at the cancer survivorship clinic.
Of the sociodemographic variables explored, only younger age and female gender were found to predict higher FCR. Although the finding between age and FCR is congruent with existing literature,8,9 this association has been revealed primarily within samples that comprised either only or predominantly breast cancer survivors, meaning that this relationship may not be fully understood in survivors of other cancers. In fact, age was not associated with FCR in samples of head and neck12,30 or colorectal48 cancer survivors, suggesting that FCR may be less of an issue for younger non–breast cancer survivors. However, this proposition requires further examination because these studies included survivors of various sex/genders, and sex/gender has been inconsistently associated with FCR.8,12,48
Varying results have been reported for the association between FCR and diagnostic or treatment-related variables for which the current study found only that longer time since diagnosis predicted higher FCR. This finding may suggest that FCR could increase over time; however, results from longitudinal studies25,49 oppose this proposition in that FCR remained stable over time. The current sample’s disease stage at diagnosis may provide insight for this finding, given that prostate cancer survivors with positive margins were found to have a level of FCR that increased over time.50 Indeed, more than 57% of the current sample was diagnosed with higher stage cancers, which may provide additional insight about FCR among survivors diagnosed with more advanced stages of cancer.
Most clinical variables assessed in this study did not assess the influence of clinical observations and comparisons15 or interpretation of clinical experiences,24 which the CSM includes as important considerations in formulating illness representations. However, when survivors’ perspectives were analyzed (eg, the symptom attribution and associations with cancer variables), they were found to predict FCR. Higher symptom attribution predicted higher FCR in the current and other studies,37,51 indicating that survivors experiencing more symptoms believed to be related to cancer have higher FCR. These symptoms, and continued follow-up at a cancer survivorship clinic, which also predicted higher FCR in the current study, may serve as reminders of the cancer experience. Such reminders have been associated with higher FCR,52 suggesting that cancer survivors’ interpretations or perceptions of their clinical experiences are powerful influences on FCR.
Study results provide support for the relationship between FCR and characteristics of the self. Higher self-esteem predicted lower FCR, which is consistent with the only other known report27 that has examined this relationship. Among the generalized expectancy variables, only pessimism was associated with FCR. This presents a new finding and is the first known analysis of FCR examining optimism and pessimism as distinct constructs.32 More pessimistic survivors may have lower FCR as a result of the coping tendencies that they tend to have. Pessimistic breast cancer patients were found to have a higher use of denial, positive reframing, self-distraction, and behavioral disengagement coping responses,53 and pessimism was found to completely mediate the relationship between denial and self-distraction coping responses on distress.53 These findings suggest that the relationship between pessimism and FCR is one among a complex series of relationships whereby the types of coping responses used by cancer survivors remove their attention from any FCR that they may be experiencing.
Implications for Nursing
Many of the cancer survivors in this sample continued to have heightened FCR, even years after completing cancer treatment. In fact, as many as 21% to 58% of the sample had a level of FCR that could be categorized as clinically significant,34,47 suggesting that approximately one-fourth to one-half of cancer survivors that oncology nurses care for have a level of FCR that may require professional attention. This finding and the finding that cancer survivors want help to cope with FCR,11,12 which is often underrecognized13 and a need that remains unmet,3 present important opportunities for oncology nurses to assess, care for, and provide intervention for an increasing number of cancer patients.1,2
Results of this study are consistent with previous work,54 indicating that sociodemographic characteristics and characteristics of the self have a much larger influence on FCR than clinical characteristics. This finding poses a unique opportunity for oncology nurses who are expected to provide holistic care55,56 that addresses the developmental, physical, psychosocial, and spiritual needs of patients.55,56 Such holistic assessments are within this study’s categorizations of sociodemographic characteristics and characteristics of the self and are therefore characteristics of patients that oncology nurses should assess and have knowledge about. The predictors of FCR identified in this study could be useful to educate clinicians, either by means of traditional education classes or by use of screening tools, pertaining to the characteristics of survivors likely to have highest levels of FCR. These resources could facilitate the identification of these survivors in oncology settings. For example, knowing that younger cancer survivors and women may have higher levels of FCR could prompt clinicians to initiate a conversation about FCR with these survivors. Subsequently, a more detailed assessment of FCR could take place, and appropriate resources could be enacted. In these ways, oncology nurses have unique insights about the characteristics of patients that are associated with heightened FCR, making oncology nurses pivotal to the identification of persons at increased risk of heightened FCR. Future work in FCR could build upon these findings to examine the implementation of these assessments into routine cancer care and the impact of these assessments upon FCR.
Strengths and Limitations
The first strength of this study is its large sample size. The second is the use of SEM, a powerful statistical method that simultaneously tests hypothesized relationships among a number of variables and errors in measurement.41 These features are particularly important when questionnaires are used to assess conceptual variables (eg, FCR) that were a major focus in this project.
Study limitations pertain to generalizability of findings. An underlying intention of this study was to extend knowledge about FCR beyond populations commonly reported in the literature (eg, breast cancer survivors). For this reason, a sample of cancer survivors with varied diagnoses living in a multiethnic center was surveyed; however, the sought-after variability was not obtained in the resulting data collected. Furthermore, the current sample was drawn from a clinic utilizing a novel model of care to transition survivors from tertiary to primary care, a model not routinely used across cancer systems. These limitations illuminate the need for future research to specifically seek out populations that are underrepresented in the FCR literature.
The results of this study may have clinical implications for oncology nurses. Most notably, the strongest predictors of FCR revealed in this study’s findings fall within the scope of oncology nursing assessments and the care expected of these nurses. In this way, oncology nurses can use these identified predictors of FCR to screen for, assess, provide education, and facilitate interventions for cancer survivors at increased risk of heightened FCR.
1. Canadian Cancer Society’s Advisory Committee on Cancer Statistics. Canadian Cancer Statistics 2016
. Toronto, Ontario, Canada: Canadian Cancer Society; 2016.
2. Bluethmann S, Mariotto A, Rowland J. Anticipating the “silver tsunami”: prevalence trajectories and comorbidity burden among older cancer survivors in the United States. Cancer Epidemiol Biomarkers Prev
3. Harrison JD, Young JM, Price MA, Butow PN, Solomon MJ. What are the unmet supportive care needs of people with cancer? A systematic review. Support Care Cancer
4. Lebel S, Ozakinci G, Humphris G, et al. From normal response to clinical problem: definition and clinical features of fear of cancer recurrence. Support Care Cancer
5. Costa DS, Dieng M, Cust AE, Butow PN, Kasparian NA. Psychometric properties of the Fear of Cancer Recurrence Inventory: an item response theory approach. Psychooncology
6. Thewes B, Butow P, Bell ML, et al. Fear of cancer recurrence in young women with a history of early-stage breast cancer: a cross-sectional study of prevalence and association with health behaviours. Support Care Cancer
7. Lebel S, Tomei C, Feldstain A, Beattie S, McCallum M. Does fear of cancer recurrence predict cancer survivors’ health care use?Support Care Cancer
8. Simard S, Savard J. Fear of Cancer Recurrence Inventory: development and initial validation of a multidimensional measure of fear of cancer recurrence. Support Care Cancer
9. Liu Y, Perez M, Schootman M, Aft RL, Gillanders WE, Jeffe DB. Correlates of fear of cancer recurrence in women with ductal carcinoma in situ and early invasive breast cancer. Breast Cancer Res Treat
10. Deimling GT, Bowman KF, Sterns S, Wagner LJ, Kahana B. Cancer-related health worries and psychological distress among older adult, long-term cancer survivors. Psychooncology
11. Hodgkinson K, Butow P, Hunt GE, Pendlebury S, Hobbs KM, Wain G. Breast cancer survivors’ supportive care needs 2–10 years after diagnosis. Support Care Cancer
12. Rogers SN, Scott B, Lowe D, Ozakinci G, Humphris GM. Fear of recurrence following head and neck cancer in the outpatient clinic. Eur Arch Otorhinolaryngol
13. Thewes B, Brebach R, Dzidowska M, Rhodes P, Sharpe L, Butow P. Current approaches to managing fear of cancer recurrence; a descriptive survey of psychosocial and clinical health professionals. Psychooncology
14. Leventhal H, Brissette I, Leventhal EA. The common-sense model of self-regulation of health and illness. In: Cameron L, Leventhal H, eds. The Self-regulation of Health and Illness Behaviour
. New York, NY: Routledge; 2003.
15. Leventhal H, Halm EA, Horowitz C, Leventhal EA, Ozakinci G. Living with chronic illness: a contextualized, self-regulation approach. In: Sutton S, Baum A, Johnston M, eds. The Sage Handbook of Health Psychology
. Thousand Oaks, CA: Sage Publications Inc; 2004:197–240.
16. Hagger MS, Orbell S. A meta-analytic review of the common-sense model of illness representations. Psychol Health
17. Lee-Jones C, Humphris G, Dixon R, Hatcher MB. Fear of cancer recurrence—a literature review and proposed cognitive formulation to explain exacerbation of recurrence fears. Psychooncology
18. Fardell JE, Thewes B, Turner J, et al. Fear of cancer recurrence: a theoretical review and novel cognitive processing formulation. J Cancer Surviv
19. Simard S, Thewes B, Humphris G, et al. Fear of cancer recurrence in adult cancer survivors: a systematic review of quantitative studies. J Cancer Surviv
20. Koch L, Jansen L, Brenner H, Arndt V. Fear of recurrence and disease progression in long-term (>/= 5 years) cancer survivors—a systematic review of quantitative studies. Psychooncology
21. Crist JV, Grunfeld EA. Factors reported to influence fear of recurrence in cancer patients: a systematic review. Psychooncology
22. Lara M, Gamboa C, Kahramanian MI, Morales LS, Bautista DE. Acculturation and Latino health in the United States: a review of the literature and its sociopolitical context. Annu Rev Public Health
23. Butow P, Phillips F, Schweder J, et al. Psychosocial well-being and supportive care needs of cancer patients living in urban and rural/regional areas: a systematic review. Support Care Cancer
24. Cameron LD, Leventhal H. Self-regulation, health, and illness. In: Cameron LD, Leventhal H, eds. The Self-regulation of Health and Illness Behaviour
. Routledge: New York, NY; 2003:1–13.
25. Ashing KT, Cho D, Lai L, et al. Exploring characteristics, predictors, and consequences of fear of cancer recurrence among Asian-American breast cancer survivors. Psychooncology
26. Moss-Morris R, Weinman J, Petrie K, Horne R, Cameron L, Buick D. The Revised Illness Perception Questionnaire (IPQ-R). Psychol Health
27. Skaali T, Fossa SD, Bremnes R, et al. Fear of recurrence in long-term testicular cancer survivors. Psychooncology
28. Scheier MF, Carver CS. Optimism, coping, and health: assessment and implications of generalized outcome expectancies. Health Psychol
29. Carver CS, Scheier MF, Segerstrom SC. Optimism. Clin Psychol Rev
30. Llewellyn CD, Weinman J, McGurk M, Humphris G. Can we predict which head and neck cancer survivors develop fears of recurrence?J Psychosom Res
31. Van den Beuken-van Everdingen MH, Peters ML, de Rijke JM, Schouten HC, van Kleef M, Patijn J. Concerns of former breast cancer patients about disease recurrence: a validation and prevalence study. Psychooncology
32. Herzberg PY, Glaesmer H, Hoyer J. Separating optimism and pessimism: a robust psychometric analysis of the revised Life Orientation Test (LOT-R). Psychol Assess
33. FluidSurveys. Canadian Survey Software. Cited May 16, 2017. https://fluidsurveys.com/popular-topics/canada/
34. Simard S, Savard J. Screening and comorbidity of clinical levels of fear of cancer recurrence. J Cancer Surviv
35. Lebel S, Simard S, Harris C, et al. Empirical validation of the English version of the Fear of Cancer Recurrence Inventory. Qual Life Res
36. Chaudhry S, Jin L, Meltzer D. Use of a self-report–generated Charlson Comorbidity Index for predicting mortality. Med Care
37. Freeman-Gibb LA, Janz NK, Katapodi MC, Zikmund-Fisher BJ, Northouse L. The relationship between illness representations, risk perception and fear of cancer recurrence in breast cancer survivors. Psychooncology
38. Rosenberg M. Society and the Adolescent Self-image
. Princeton, NJ: Princeton University Press; 1965.
39. Corwyn RF. The Factor Structure of Global Self-esteem among adolescents and adults. J Res Pers
40. Scheirer MF, Carver CS, Bridges MW. Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem). a reevaluation of the Life Orientation Test. J Pers Soc Psychol
41. Kline RB. Principles and Practice of Structural Equation Modeling
. 3rd ed. New York, NY: The Guilford Press; 2011.
42. Muthén L, Muthén BO. Mplus User’s Guide
. 8th ed.Muthén & Muthén: Los Angeles, CA; 2017.
43. Schreiber JB, Stage FK, King J, Nora A, Barlow EA. Reporting structural equation modeling and confirmatory factor analysis results: a review. J Educ Res
44. Hooper D, Coughlan J, Mullen M. Structural equation modelling: guidelines for determining model fit. Electron J Bus Res Methods
45. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Modeling
46. Steiger JH. Understanding the limitations of global fit assessment in structural equation modeling. Personal Individ Differ
47. Fardell JE, Jones G, Smith AB, et al. Exploring the screening capacity of the Fear of Cancer Recurrence Inventory–Short Form for clinical levels of fear of cancer recurrence. Psychooncology
48. Steele N, Haigh R, Knowles G, Mackean M. Carcinoembryonic antigen (CEA) testing in colorectal cancer follow up: what do patients think?Postgrad Med J
49. Ghazali N, Cadwallader E, Lowe D, Humphris G, Ozakinci G, Rogers SN. Fear of recurrence among head and neck cancer survivors: longitudinal trends. Psychooncology
50. Hong YM, Hu JC, Paciorek AT, Knight SJ, Carroll PR. Impact of radical prostatectomy positive surgical margins on fear of cancer recurrence: results from CaPSURE. Urol Oncol
51. Phillips KM, McGinty HL, Gonzalez BD, et al. Factors associated with breast cancer worry 3 years after completion of adjuvant treatment. Psychooncology
52. Ziner KW, Sledge G, Bell CJ, Johns SA, Miller KD, Champion VL. Predicting fear of breast cancer recurrence and self-efficacy in survivors by age at diagnosis. Oncol Nurs Forum
53. David D, Montgomery GH, Bovbjerg DH. Relations between coping responses and optimism-pessimism in predicting anticipatory psychological distress in surgical breast cancer patients. Pers Individ Differ
54. Koch L, Bertram H, Eberle A, et al. Fear of recurrence in long-term breast cancer survivors—still an issue. Results on prevalence, determinants, and the association with quality of life and depression from the cancer survivorship
—a multi-regional population-based study. Psychooncology
55. Oncology Nursing Society. Oncology Nurse Generalist Competencies
. Pittsburgh, PA; 2016.
56. Canadian Association of Nurses in Oncology/Association canadienne des infirmières en oncologie (CANO/ACIO). Practice standards and competencies for the specialized oncology nurse. 2006. Cited August 23, 2019. https://cdn.ymaws.com/www.cano-acio.ca/resource/resmgr/standards/CONEP_Standards2006September.pdf
. Accessed July 18, 2019.
As indicated in the main article, a structural equation modeling (SEM1) was conducted using Mplus version 8.2 Initially, confirmatory factor analyses (CFAs) were used to explore the fit (see paragraph below for details) of each measure to study data.1 After acceptable-to-good fits were achieved, the full measurement model—which included the Fear of Cancer Recurrence Inventory (FCRI), Rosenberg Self-esteem Scale (RSES), and Revised Life Orientation Test (LOT-R) together—was examined. Thereafter, relationships (or direct effects) were tested in a structural regression model,1 and the overall fit of this model and coefficients of relationships3 were reviewed.
Good model fits were indicated by a root mean square error of approximation (RMSEA) ≤0.07,4 a comparative fit index (CFI) ≥.94,5 and Tucker-Lewis Fit Index (TLI) ≥0.94.5 A χ2 test was reported for informational purposes only since it is often statistically significant in large samples.1 All analyses used the robust maximum likelihood method estimator (MLR2) and .05 level of significance (2-tailed).
Fear of Cancer Recurrence Inventory
The factor structure of the FCRI has been supported,6,7 and fear of cancer recurrence was represented as a single second-order factor in the current analysis.
Rosenberg Self-esteem Scale
The Rosenberg Self-esteem Scale8 was originally intended as a unidimensional measure of self-esteem and was therefore tested as a single-factor CFA. This model proved to be a poor fit to the data (χ2 = 327.495, P < .0001, df = 35, RMSEA = 0.091 [0.083, 0.101], CFI = 0.900, TLI = 0.871), and therefore, a 2-factor model whereby the negatively worded items and positively worded items were loaded onto separate factors9 was tested. The 2-factor model resulted in a better fit (see Table 3 in the main article) and was therefore retained in subsequent analyses.
Revised Life Orientation Test
Originally, only the 6 “nonfiller” items10 of LOT-R were explored in a first-order CFA, which resulted in a poor-fitting measurement model (χ2 = 280.945, P < .0001, df = 9, RMSEA = 0.174 [0.157, 0.192], CFI = 0.767, TLI = 0.612), as has been similarly reported in other studies.11,12 Following these previous CFA studies using the LOT-R,11,12 a 2-factor model whereby the “nonfiller” items were loaded separately onto optimism or pessimism factors was tested and greatly improved the fit of the measurement model (see Table 3 in the main article). For comprehensiveness, a 3-factor model whereby the “nonfiller” items were loaded onto a optimism or pessimism factor and the filler items were loaded onto a third factor was tested and proved to have a poorer fit than the previously tested model (χ2 = 197.553, P < .0001, df = 32, RMSEA = 0.072 [0.063, 0.082], CFI = 0.925, TLI = 0.894). Therefore, the 2-factor model using only the “nonfiller” items loaded onto their respective optimism and pessimism factors was retained for subsequent analyses.
1. Kline R. Principles and Practice of Structural Equation Modeling. 3rd ed. New York: NY: The Guilford Press; 2011.
2. Muthén L, Muthén BO. Mplus User’s Guide. 8th ed. Los Angeles, CA: Muthén & Muthén; 2017.
3. Schreiber J, Nora A, Stage FK, Barlow EA, King J. Reporting structural equation modeling and confirmatory factor analysis results: a review. J Educ Res. 2006;99(6):323–337.
4. Steiger JH. Understanding the limitations of global fit assessment in structural equation modeling. Pers Individ Differ. 2007;42(5):893–898.
5. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Model A Multidiscip J. 1999;6(1):1–55.
6. Lebel S, Simard S, Harris C, et al. Empirical validation of the English version of the Fear of Cancer Recurrence Inventory. Qual Life Res. 2016;25(2):311–321. d
7. Liu J, Mahendran R, Chua SM, et al. Validation of the English and Mandarin versions of the Fear of Cancer Recurrence Inventory in an Asian population. J Health Psychol. August 2017.
8. Rosenberg M. Society and the Adolescent Self-Image. Princeton, NJ: Princeton University Press; 1965.
9. Corwyn RF. The Factor Structure of Global Self-esteem among adolescents and adults. J Res Pers. 2000;34:357–379.
10. Scheier MF, Carver CS. Optimism, coping, and health: assessment and implications of generalized outcome expectancies. Health Psychol. 1985;4(3):219–247.
11. Glaesmer H, Rief W, Martin A, et al. Psychometric properties and population-based norms of the Life Orientation Test Revised (LOT-R). Br J Health Psychol. 2012;17(2):432–445.
12. Herzberg PY, Glaesmer H, Hoyer J. Separating optimism and pessimism: a robust psychometric analysis of the Revised Life Orientation Test (LOT-R). Psychol Assess. 2006;18(4):433–438.