Self-Guided Online Cognitive Behavioral Strategies forChemotherapy-Induced Peripheral Neuropathy:A Multicenter, Pilot, Randomized, Wait-ListControlled Trial

Robert Knoerl,* Ellen M. L. Smith,† Debra L. Barton,† David A. Williams,‡

Janean E. Holden,† John C. Krauss,§ and Beth LaVasseur¶

*Phyllis F. Cantor Center for Research in Nursing and Patient Care Services, Dana-Farber Cancer Institute, Boston,Massachusetts.†School of Nursing, University of Michigan, Ann Arbor, Michigan.‡School of Medicine, University of Michigan, Ann Arbor, Michigan.§Comprehensive Cancer Center, University of Michigan, Ann Arbor, Michigan.¶St. Joseph Mercy Hospital, Ann Arbor, Michigan.

Abstract: The purpose of this pilot, parallel, randomized controlled trial was to examine the effi-cacy of a self-guided online cognitive and behaviorally-based pain management intervention (ProactiveSelf-Management Program for Effects of Cancer Treatment [PROSPECT]) to reduce “worst” pain forindividuals with chronic painful chemotherapy-induced peripheral neuropathy (CIPN). Secondary out-comes included “average” pain, nonpainful CIPN symptom severity, impression of change, and paininterference. Sixty patients with chronic painful CIPN were recruited from 5 outpatient academic andcommunity cancer centers. Patients were randomized in a 1:1 ratio to receive either 8 weeks of PROS-PECT or usual care. A 7-day electronic “worst” pain intensity diary and standardized measures of paininterference, nonpainful CIPN symptom severity, impression of change, and “average” pain were ad-ministered pre/post intervention. Postintervention mean scores were evaluated between groups usinganalysis of covariance adjusting for baseline. Individuals who received the PROSPECT intervention(n = 19) had significantly greater improvements in “worst pain” compared with individuals receivingusual care (n = 19; P = .046, d = .58). There were no significant differences in mean scores betweengroups for the secondary outcomes (n = 42). A larger, adequately powered study testing the PROS-PECT intervention is needed to determine if improvements in pain may be sustained, evaluate theeffect of the intervention on the secondary outcomes, and identify mediators of pain intensity-related improvement.Perspective: This study explores the efficacy of an 8-week online cognitive behavioral pain man-agement intervention for chronic painful CIPN. Intervention use resulted in greater improvements

Received August 8, 2017; Revised September 7, 2017; Accepted Novem-ber 27, 2017.This work was supported by the University of Michigan School of NursingNew Investigator Award and Rackham Graduate Student Research Grant.The funding source had no role in study design, data collection/analysis, or manuscript preparation.Creation of the Proactive Self-Management Program for Effects of CancerTreatment Web site was supported by the Damon Runyon Cancer Re-search Foundation (CI-53-10) to Norah Lynn Henry, MD, PhD. ProactiveSelf-Management Program for Effects of Cancer Treatment was adaptedfrom the Web site, Fibroguide.com, which was developed with supportfrom Grant numbers R01-AR050044 (National Institute of Arthritis andMusculoskeletal and Skin Diseases/National Institutes of Health), and DAMD17-00-2-0018 (Department of Defense).

Dr. Williams has served as a consultant to Community Health Focus, Inc,served as a grant reviewer for Pfizer independent grants for Learningand Change, and is currently the President of the American Pain Society.The other authors have no conflicts of interest to declare.Address reprint requests to Robert Knoerl, Phyllis F. Cantor Center forResearch in Nursing and Patient Care Services, Dana-FarberCancer Institute, 450 Brookline Avenue, Boston, MA 02215. E-mail:Robert_knoerl@DFCI.Harvard.edu1526-5900/$36.00© 2017 by the American Pain Societyhttps://doi.org/10.1016/j.jpain.2017.11.009

382

The Journal of Pain, Vol 19, No 4 (April), 2018: pp 382-394Available online at www.jpain.org and www.sciencedirect.com

in “worst” pain than usual care alone. The findings provide preliminary support for the efficacy of anonpharmacological intervention for chronic painful CIPN.

© 2017 by the American Pain SocietyKey words: Chronic pain, chemotherapy-induced peripheral neuropathy, cognitive-behavioral therapy,peripheral nervous system disease/chemically induced.

Chemotherapy-induced peripheral neuropathy (CIPN)is a common side effect of cancer treatment thatcan occur in up to 68% of individuals receiving neu-

rotoxic chemotherapy (eg, platinums, bortezomib, andtaxanes).6,31,51 The symptoms of CIPN include numbness,tingling, and pain in the hands and/or feet (symptomsgenerally present in a symmetrical, stocking-glovedistribution).53 In up to 40% of patients,31,62 CIPN maybecome chronically painful and persist for months to yearsafter the completion of chemotherapy.18,39,56 Patients withpainful CIPN often report decreases in quality of life andphysical function and may be required to stop poten-tially life-saving neurotoxic chemotherapy regimens.42,57

Despite the known negative effects that painful CIPNhas on physical function and quality of life, there are feweffective treatments for painful CIPN. Duloxetine 60 mg/dis currently the only medication recommended for thetreatment of painful CIPN.24,54 Because of their efficacyin other neuropathic pain populations, antidepressantsand anticonvulsants are often used to treat painful CIPN.24

However, adherence to these types of medications is poorbecause of side effects or lack of efficacy.20 Use of an ef-fective, nonpharmacologic intervention for painful CIPNmay decrease the need for drug therapy—potentially re-ducing the overall side effect burden for cancer survivorsand/or create a synergistic pain-reducing effect by theuse of multiple modalities. Thus, multimodal manage-ment approaches that incorporate nonpharmacologicapproaches for painful CIPN warrant further study.

One nonpharmacologic treatment used commonly forthe treatment of chronic pain (eg, back/neck, musculo-skeletal, and fibromyalgia) is therapist-administeredcognitive-behavioral pain management.7,40,46,61,64 This in-tervention is designed to help patients self-manage painand co-occurring symptoms such as anxiety, depression,and insomnia through cognitive and behavioral strate-gies such as relaxation, sleep hygiene, activity pacing, andcognitive restructuring.17,32 Cognitive-behavioral pain man-agement may reduce pain intensity by inducing structuralchanges in the prefrontal cortex (eg, increased gray mattervolume).27,50 This may provide individuals with increasedexecutive control function and subsequently, a greaterability to reappraise and gain a greater sense of controlover their pain. Structural changes in the prefrontal cortexthen may lead to the release of neurotransmitters (eg,norepinephrine and serotonin) that influence descend-ing pain inhibition mechanisms.27,50 Barriers related to thedelivery of therapist-administered cognitive behavioralpain management in practice include: 1) lack of accessto a reputable therapist, 2) cost associated with treat-ment, 3) negative stigma associated with psychologicaltherapies, and 4) transportation to the clinic.17,34 One wayto overcome these barriers is to offer this treatment ina self-guided online format. A self-guided cognitive-

behavioral pain management intervention providespatients with access to symptom management strate-gies that they can practice at their own pace without theneed to travel to meet with a therapist. There is strongevidence supporting the efficacy of self-guided cognitive-behavioral pain management for chronic pain.9,34,38,65

However, little is known about the efficacy of self-guided cognitive-behavioral pain management for chronicpainful CIPN.

PurposeThe purpose of this randomized, wait-list control pilot

study was to test the efficacy of a self-guided cognitive-behavioral pain management intervention called ProactiveSelf-Management Program for Effects of Cancer Treat-ment (PROSPECT) to reduce worst pain intensity forindividuals with chronic painful CIPN compared with in-dividuals receiving treatment as usual (ClinicalTrials.govIdentifier: NCT02760654). Secondarily, we explored theefficacy of the PROSPECT intervention to improve CIPNsymptom severity (eg, nonpainful numbness and tin-gling), pain interference, average pain severity, andpatients’ perceived global impression of change. Last, weexplored participant acceptability of and satisfaction withPROSPECT.

Methods

Design, Setting, and SampleThe study aims were examined via a parallel, 1:1 ran-

domized controlled trial design. Sixty patients wererecruited from 5 outpatient community and/or aca-demic oncology clinics in Southeast Michigan. Patientswere eligible if they: 1) were older than 25 years of age,2) self-reported ≥4 of 10 worst CIPN pain that persisted3 months or longer after the cessation of neurotoxic che-motherapy, 3) had at least National Cancer InstituteCommon Terminology Criteria for Adverse Events grade1 sensory CIPN,44 4) had a stable analgesic medicationregimen (≤10% change in dosage in the 2 weeks beforestudy enrollment), and 5) were able to access/use a com-puter. Participants were excluded if they had: 1) aprognosis of <3 months, 2) peripheral neuropathy fromother causes, 3) planned to receive neurotoxic chemo-therapy while enrolled in the study, or 4) participated incognitive-behavioral pain management in the past. Thisstudy was approved by the institutional review board as-sociated with each study site and written informedconsent was obtained from all enrolled participants.

Treatment GroupsParticipants were randomly assigned following simple

randomization procedures to either 8 weeks of PROS-

Knoerl et al The Journal of Pain 383

PECT or treatment as usual (control) in a 1:1 ratio usinga computer-generated random numbers table. Random-ization was stratified according to recruitment site tobalance out center effects. The principal investigator gen-erated the random allocation sequence, enrolled thepatients, and assigned participants to a study group. Thecomputer-generated random numbers table was storedon a spreadsheet. The principal investigator did not viewthe spreadsheet until informed consent was obtained andall baseline assessments were completed by the partici-pant. After informed consent and completion of allbaseline assessments, participants were informed of theirstudy group assignment. After the administration of thebaseline assessments by the principal investigator, trainedstudy staff administered all data collection proceduresat the subsequent time points. Health care providers werenot informed of their patients’ study group assignment.

PROSPECTThe password-protected PROSPECT Web site con-

tained cognitive-behavioral pain management strategiesand information designed to help individuals manage painand co-occurring symptoms after cancer treatment (eg,anxiety, depression, sleep, fatigue, and impairedcognition).16,43,58,60 Content (10 modules; Table 1) is pre-sented using written as well as video formats and patientscan download worksheets further describing the strat-egies. At baseline, participants were trained on how tonavigate the PROSPECT Web site and were encouragedto complete the “Steps for Me” link, which recom-mends modules on the basis of the patient’s responsesto questions about symptom severity and symptom man-agement practices. Participants were instructed to use themodules as much as they desired and did not receive anyadditional encouragement from the study staff after ob-taining access to the PROSPECT Web site.

Wait-List ControlParticipants in both groups continued to receive their

usual care from their primary provider (eg, routine on-

cology follow-up appointments and subspecialtyappointments). Participants randomized to the controlgroup received access to the PROSPECT Web site afterthe completion of all required surveys.

MeasuresAn 11-point numeric rating scale (NRS) of pain inten-

sity was used to measure worst and average CIPN painseverity (0, no pain; 10 pain as bad as you can imagine).8,12

The 11-point worst CIPN pain NRS was administered withina 7-day pain diary pre/post intervention, whereas theaverage CIPN pain NRS was administered via a single itemasking about average CIPN pain severity over the past 7days (worst pain was assessed in similar fashion at week4). Participants were coached to only report pain result-ing from CIPN (eg, painful numbness, tingling, burningin hands/feet), not pain from other sites or causes. The11-point NRS is recommended by the Initiative onMethods, Measurement, and Pain Assessment in Clini-cal Trials (IMMPACT)12 and several trials have supportedits reliability and validity.26,28,36 On the basis of evidencesuggesting that worst pain intensity is more highly cor-related with average functional interference (r = .65) thanaverage (r = .60) or current (r = .53) pain intensity,21,52 worstpain intensity was selected as the primary outcome.

Several other comorbid symptoms were assessed as rec-ommended by IMMPACT to increase our ability tocompare the results of this study with other trials testinginterventions for chronic pain. The Patient-Reported Out-comes Measurement Information System (PROMIS) PainInterference 4a (4 items; 1, not at all; 5 very much; trans-formed total score 41.6–75.6) subscale measures the effectof pain on the social, cognitive, and physical aspects ofone’s life over the past week.1 There is strong evidencesupporting the reliability and validity of the PROMIS paininterference item bank.1 In addition, to measure impres-sion of change after completion of the trial, we used thePatient Global Impression of Change (PGIC). The PGIC isa self-report item designed to assess patients’ overall im-pression of improvement over the course of a clinical

Table 1. Summary of PROSPECT ModulesMODULE CONTENT

About Late Effects Education about common cancer treatment-related side effects (ie, pain, fatigue, problems with memory,emotional distress, sleep problems)

Talk to Your Team Strategies to promote communication between the patient and their provider regarding cancertreatment-related symptoms

Get Your Body Moving Benefits of physical activity during cancer treatment and provides strategies to start and maintain regularphysical activity

Get a Better Night’s Sleep Sleep hygiene strategies to help individuals fall and stay asleep at nightSlow Your Body Down Step-by-step instructions for various relaxation techniques (ie, deep breathing, guided imagery,

progressive muscle relaxation)Improve Your Thinking Strategies to combat memory and thinking problemsSet Some Goals Strategies to set realistic goals and carry out planned goalsDon’t Over Do It Activity pacing strategiesTime for You Strategies to overcome barriers and challenges related to taking time out of the day to participate in

enjoyable activities to renew the mind and bodyAbout Peripheral Neuropathy Information about the symptoms of peripheral neuropathy, strategies to treat the symptoms of

neuropathy, and safety precautions to take because of the symptoms of neuropathy

This table describes the content of the 10 modules embedded within the PROSPECT Web site.

384 The Journal of Pain Self-Guided Cognitive Behavioral Strategies for Neuropathy

trial.63 The 7-point scale ranges from “very much worse”to “very much improved.”63

The European Organization of Research and Treat-ment of Cancer Quality of Life Questionnaire-Chemotherapy-Induced Peripheral Neuropathy Scale (QLQ-CIPN20) measures patient’s symptoms and functionallimitations related to CIPN in sensory, motor, and auto-nomic function domains (20 items; 1, not at all; 4, verymuch).48 The responses from each respective scale are thentransformed and scored from 0 to 100, with higher scoresrepresenting worsening symptoms.48 The internal con-sistency reliability α coefficient has been reported as .88,.88, and .78 for the sensory, motor, and autonomicsubscales, respectively.35 Further evidence shows that thesensory and motor subscales are moderately-highly re-sponsive to change (Cohen d = .82 and .48, respectively).35

Last, recent evidence supports the electronic adminis-tration of the QLQ-CIPN20 on the basis of high intraclasscorrelations (>.70) with paper/pencil administration.33

Last, we adapted questions from the AcceptabilityE-Scale to assess participant acceptability of and satis-faction with the PROSPECT Web site.3,59 The questionnairecontains 7 items that are scored on a 1 to 5 scale, withhigher scores representing greater acceptability and sat-isfaction. Survey questions ask participants how much theyenjoyed using PROSPECT and how helpful the moduleswere in improving their symptoms.

Participants completed a demographic survey (ie, sex,age, race, ethnicity, employment status, marital status,education, and previous computer use) before the comple-tion of the baseline assessments. Study staff abstractedparticipant cancer diagnosis (ie, disease and stage) andtreatment (eg, chemotherapy type), medication dosage,and comorbid condition-related information from the pa-tients’ electronic medical records.

Study Procedures

BaselineAfter informed consent (eg, participants were in-

formed of the nature of the PROSPECT and controlgroups), participants completed the baseline assess-ments (11-point worst and average CIPN pain NRS, PROMISPain Interference 4a, QLQ-CIPN20, demographic charac-teristics) via computer tablet. Participants were thennotified of their study group assignment and e-mailedthe 7-day worst pain diary. After completing the paindiary, the participants were e-mailed the PROSPECT Website link and password, or instructions about the controlgroup. Also at baseline, study staff abstracted cancerdiagnosis/treatment and medication dosage-related in-formation from the patients’ electronic medical records.

Week 4 and 8 Time PointsSurvey links to the same battery of assessments ad-

ministered at baseline were e-mailed to participants 4 and8 weeks after randomization. At these same time points,participants were contacted by telephone as a reminderto complete the electronic surveys and to answer struc-

tured interview questions about changes in medicationdosage/frequency. Unique to the week 8 time point, par-ticipants in both groups completed the PGIC andindividuals in the intervention group completed theAdapted Acceptability E-Scale and were contacted via tele-phone to discuss the most positive and negative aspectsof PROSPECT. Participants were also e-mailed a survey eachweek inquiring about the number of minutes spent usingPROSPECT or other symptom management resources. Atthe end of the study, control group participants weree-mailed the link to the PROSPECT Web site.

Statistical AnalysesAll data were analyzed using R version 3.4.0.49 Descrip-

tive statistics of the centrality and dispersion of all surveydata and demographic data were calculated. All analy-ses were calculated on the basis of the total number ofindividuals who completed the baseline and week 8outcome measures. A power analysis was not con-ducted because of the pilot nature of this work (eg,determine an effect size to conduct larger studies testingthis intervention) and because there were no previousstudies testing the PROSPECT intervention in individu-als with painful CIPN on which to base an effect sizeestimate.

Primary OutcomeWe averaged patient’s responses from the 7-day worst

CIPN pain diary for the baseline and week 8 time points.Participants must have completed 5 of the 7 daily worstpain ratings at the baseline and week 8 time points tobe included in the analysis.22 Week 8 mean scores in worstCIPN pain intensity (0–10 NRS diary) were comparedbetween groups using analysis of covariance adjustingfor baseline scores. The robustness of our findings wasexamined via a post hoc intent-to-treat analysis. Mul-tiple imputation (50 imputations; 50 iterations) (MICE RPackage)4 was used to handle missing data. The same re-gression analysis was then repeated using the pooled datafrom the multiply imputed data sets.

Secondary OutcomesWeek 8 mean scores in CIPN symptom severity (QLQ-

CIPN20 sensory and motor subscales), average pain(average CIPN pain NRS), and pain interference (PROMISPain Interference 4a) were compared between groupsusing analysis of covariance adjusting for baseline. Fisherexact test was used to analyze proportional differencesbetween those who experienced “improvement” (PGICscore ≥5) and “no improvement” (PGIC score ≤4) in the2 study arms.

Acceptability and SatisfactionDescriptive statistics for the items of the Adapted Ac-

ceptability E-Scale were calculated. We also summarizedresponses from the semistructured telephone inter-views to determine the most positive and negative aspectsof PROSPECT and the biggest barriers to accessing andusing the PROSPECT strategies.

Knoerl et al The Journal of Pain 385

Results

PatientsParticipant recruitment occurred from May 1, 2016 to

October 4, 2016. A review of the study sites’ electronicmedical records revealed 393 potentially eligible partici-pants (Fig 1). After telephone screening procedures, 311patients were deemed ineligible on the basis of inclusion/exclusion criteria. Of these 311 excluded participants, 18experienced CIPN-related numbness, tingling, and pain,but rated their worst pain severity as <4 of 10. Also, 22participants were eligible, but either declined to partici-pate or were not available to meet in person to provideinformed consent. Sixty participants were randomized tothe PROSPECT intervention (n = 30) or wait-list control(n = 30) groups. Patient follow-up occurred from May 23,2016 to December 19, 2016. After randomization, 13 par-ticipants terminated the study early because of personalreasons (eg, lack of time) or were lost to follow-up. Par-ticipants who terminated the study early did not completeany outcome measurements beyond the baseline timepoint. The attrition rate for the study was 22%. Overall,23 and 24 participants were eligible for analysis in thePROSPECT and control arms, respectively. There were no

adverse events due to study participation reported byparticipants.

Patient characteristics are described in Table 2. Indi-viduals in the PROSPECT group had higher levels of fatigueand sleep-related impairment compared with individu-als in the control group; otherwise, there were nodifferences. When comparing protocol completers versusnoncompleters, noncompleters had a greater percent-age of individuals with stage IV cancer (46%) thancompleters (19%), but baseline pain and co-occurringsymptom severity did not differ.

Primary OutcomeIndividuals with chronic painful CIPN who received the

8-week PROSPECT intervention had a mean change scoreof −.94 (SD = 1.36, range = −3.29 to 1.29, 95% confi-dence interval [CI] = −1.6 to −.28), whereas individuals inthe wait-list control group had a mean change score of0 in worst pain intensity (SD = 1.31, range = −3.43 to 2.86,95% CI = −.63 to .63; Table 3). The difference in worst CIPNpain intensity improvements between groups was sig-nificant (B = −.91, P = .046, 95% CI = −1.79 to −.02, d = .58,n = 38). Three participants receiving PROSPECT and 1

Figure 1. Consolidated Standards of Reporting Trials flow diagram. This shows the flow of participants through the duration ofthe study.

386 The Journal of Pain Self-Guided Cognitive Behavioral Strategies for Neuropathy

participant receiving treatment as usual experienced aclinically significant (>30%) reduction in worst CIPN painintensity immediately post-treatment (n = 38; Fig 2).19

Results of the post hoc intent-to-treat analysis (n = 60)indicated that the difference in worst pain intensity im-provement between groups was not statistically significant(B = −.64; P = .09; 95% CI = −1.38 to .10; d = .42).

Secondary OutcomesThere were no significant differences in mean change

scores for average pain (P = .18, d = .42), pain interfer-ence (P = .98, d = .01), or nonpainful CIPN sensory (P = .41,d = .23) or motor symptoms (P = .95, d = .02; n = 42;Table 3). Trends in average pain and pain interferenceacross time indicated that PROSPECT provided no clearbenefit over usual care. However, trends in nonpainfulCIPN symptoms suggested that individuals receiving PROS-PECT were experiencing considerable improvements asthe study progressed. There was also a greater numberof individuals in the PROSPECT group reporting im-proved impression of change after the completion of thetrial, but the difference between groups was not sig-nificant (P = .21, d = .48, n = 41; Table 3).

Acceptability and SatisfactionOverall, acceptability and satisfaction with the study

and PROSPECT Web site was moderate to high, with meanAdapted Acceptability E-Scale item scores ranging from3.26 to 4.58 of 5 (n = 19; Table 4). Participants reportedthat ease of use, the ability to print off work sheets onthe basis of the strategies they learned, and having accessto many relevant pain symptom management strategieswere all positive aspects of the PROSPECT intervention.Conversely, participants thought that there was not enoughinformation and/or strategies to help manage the symp-toms of nonpainful neuropathy (eg, numbness andtingling). Participants also cited lack of time and diffi-culty changing symptom management practices as barriersto implementing the strategies they learned. Last, par-ticipants thought that PROSPECT would have been morebeneficial if it contained additional cognitive strategiesrelated to pain management (ie, cognitive restructur-ing), features to interact with medical professionals toreview strategies and symptoms, or was provided in thebeginning stages of cancer treatment as they were be-ginning to experience painful CIPN symptoms.

PROSPECT UsageIn terms of intervention use, the mean number of

minutes individuals in the intervention group spent usingPROSPECT and other symptom management resources (eg,physical therapy, meditation, pool therapy, massage) washighest at the beginning of the study and declined asthe study progressed. The average amount of time par-ticipants in the control group spent using symptommanagement strategies (eg, ice therapy, massage, stretch-ing, yoga, distraction, exercise) varied across the 8-weekstudy period (Table 5). In terms of medication changes,

Table 2. Demographic, Cancer Treatment, andMedication Use Characteristics of theRecruited Sample

CHARACTERISTIC

PROSPECT(N = 30)*

CONTROL

(N = 30)*

SexFemale 23 (76.7) 22 (73.3)Male 7 (23.3) 8 (26.7)

Mean age (SD) 58.93 (9.33) 63.37 (8.36)Race

African American 2 (6.7) 1 (3.3)White 27 (90) 28 (93.3)Unknown 1 (3.3) 1 (3.3)

EthnicityHispanic 0 1 (3.3)Non-Hispanic 25 (83.3) 22 (73.4)Unknown 5 (16.7) 7 (23.3)

Education (n = 59)High school or less 4 (13.3) 6 (20.7)Some college 13 (43.3) 12 (41.4)College graduate 8 (26.7) 6 (20.7)Postgraduate degree 5 (16.7) 5 (17.2)

Employment statusEmployed 13 (43.3) 7 (23.3)Out of work 3 (10) 2 (6.7)Homemaker 0 1 (3.3)Retired 9 (30) 17 (56.7)Unable to work 5 (16.7) 3 (10)

Marital statusSingle 2 (6.7) 2 (6.7)Married 18 (60) 24 (80)Separated 1 (3.3) 1 (3.3)Divorced 7 (23.3) 2 (6.7)Widowed 2 (6.7) 1 (3.3)

Amount of computer useOnce a month 1 (3.3) 6 (20)Approximately once a week 1 (3.3) 1 (3.3)More than once a week 28 (93.3) 23 (76.7)

Cancer stageEarly I/II 13 (43.3) 9 (30)III 10 (33.3) 11 (36.7)Metastatic 6 (20) 9 (30)Unknown 1 (3.3) 1 (3.3)

Cancer TypeBreast 13 (43.3) 10 (33.3)Gastrointestinal 13 (43.3) 13 (43.3)Genitourinary 0 1 (3.3)Lung 1 (3.3) 3 (10)Multiple 1 (3.3) 2 (6.7)Lymphoma 2 (6.7) 1 (3.3)

Chemotherapy TypePlatinums 13 (43.3) 13 (43.3)Taxanes 12 (41.4) 8 (26.7)Bortezomib 1 (3.3) 0Vinca alkaloids 1 (3.3) 2 (6.7)Multiple 3 (10) 7 (23.3)

Pain-related symptoms (mean, SD)Anxiety 54.79 (9.42) 51.58 (7.83)Depression 52.92 (7.98) 49.51 (7.17)Fatigue 59.18 (8.26) 52.73 (8.25)Sleep-related impairment 58.65 (6.67) 55.07 (6.09)

Comorbid conditionsNone 12 (40) 7 (23.3)Chronic lung disease 1 (3.3) 1 (3.3)Coronary artery disease 0 2 (6.7)Diabetes 0 3 (10)Gastrointestinal disease 0 2 (6.7)Chronic kidney injury 1 (3.3) 0Obesity 1 (3.3) 2 (6.7)Hypertension 12 (40) 14 (46.7)Anxiety 3 (10) 4 (13.3)Depression 5 (16.7) 4 (13.3)Other 11 (36.7) 15 (50)

Baseline medicationsNeuropathic pain medications† 9 (30) 6 (20)Other analgesics‡ 6 (20) 8 (26.7)Neuropathic pain medications as well as other analgesics 9 (30) 8 (26.7)Antianxiety 8 (26.7) 8 (26.7)Sleep medications 3 (10) 2 (6.7)Antidepressants§ 4 (13.3) 5 (16.7)Fatigue medications 0 0

This table describes the demographic characteristics of the recruited sample at baseline.*Data are n (%) unless otherwise specified.†On the basis of the clinical practice guideline for painful CIPN treatment recommendations,23

neuropathic pain medications included: duloxetine, gabapentin, pregabalin, andantidepressants.‡Other analgesics included: oxycodone, morphine, ibuprofen, and acetaminophen.§Three participants (2 in control, 1 in PROSPECT) were receiving nortriptyline, a medicationthat has shown efficacy to treat pain in other neuropathic pain patient populations.

Knoerl et al The Journal of Pain 387

there were more individuals in the PROSPECT group thatincreased neuropathic pain medication frequency/doseand pain medication frequency/dose not indicated as afirst-line treatment for painful CIPN and/or neuropathicpain (“other analgesics”),11,24 but otherwise, there wereno considerable differences in the number of partici-pants changing medication frequency/dose betweengroups (Fig 3). To further explore how changes in pain

medication frequency/dose may have affected worst CIPNpain intensity improvement in individuals receiving PROS-PECT (within primary aim protocol completers; n = 38),we compared week 8 worst pain intensity mean scores(adjusting for baseline) between groups in 2 ways: 1) re-moving individuals in the PROSPECT arm who increasedtheir use of first-line neuropathic pain medications, and2) removing individuals in the PROSPECT arm who

Table 3. Mean Scores for Primary and Secondary Outcomes

OUTCOMES*INTERVENTION,MEAN (SD)

CONTROL,MEAN (SD) CONTRAST BETWEEN GROUPS†

Worst pain (n = 38)Baseline 5.04 (1.24) 4.78 (1.78) B = −.91, P = .046, 95% CI = −1.79 to .02, d = .58Week 4 (n = 39) 5.42 (2.32) 5.60 (2.5)Week 8 4.10 (1.81) 4.78 (1.93)

Average pain (n = 42)Baseline 4.37 (1.89) 3.91 (2.52) B = −.83, P = .18, 95% CI = −2.05 to .40, d = .42Week 4 (n = 39) 4.37 (2.11) 5.25 (2.36)Week 8 4.58 (1.87) 5.35 (1.99)

Pain interference (n = 42)Baseline 57.81 (8.16) 57.33 (7.07) B = .05, P = .98, 95% CI = −4.36 to 4.46, d = .01Week 4 (n = 39) 57.47 (8.49) 55.54 (5.90)Week 8 56.72 (7.96) 56.43 (7.74)

CIPN sensory (n = 42)Baseline 49.34 (16.68) 45.99 (20.1) B = −3.61, P = .41, 95% CI = −12.43 to 5.21, d = .23Week 4 (n = 39) 44.80 (23.10) 41.67 (18.70)Week 8 40.41 (18.66) 41.95 (17.37)

CIPN motor (n = 42)Baseline 34.21 (17.04) 25.83 (18.41) B = −.23, P = .95, 95% CI = −8.27 to 7.80, d = .02Week 4 (n = 39) 30.23 (23.59) 23.24 (13.57)Week 8 26.91 (17.71) 22.75 (13.3)

Impression of change (n = 41)8 Weeks (improved) 9/19 (47.3%) 6/22 (27.3%) OR = 2.35, P = .21, 95% CI = .55 to 10.84, d = .488 Weeks (no change or worse) 10/19 (52.7%) 16/22 (72.7%)

This table describes changes in primary and secondary outcome improvement over time between treatment groups.*Week 4 data are documented to provide information related to time to response.†Difference in week 8 mean scores adjusted for baseline scores. Impression of change scores were only compared at the week 8 time point.

Figure 2. Percent decrease in worst pain intensity because of use of PROSPECT or treatment as usual (n = 38). This figure showsthe percentage of participants reporting various reductions in worst pain intensity after completion of the study.

388 The Journal of Pain Self-Guided Cognitive Behavioral Strategies for Neuropathy

Table 4. Adapted Acceptability E-Scale Scores (n = 19)ACCEPTABILITY AND SATISFACTION; ADAPTED ACCEPTABILITY E-SCALE MEAN SD RANGE

1. How easy was it to access the Web site on your computer? 4.58 .84 2 to 52. How understandable was the content presented within the Web site? 4.58 .69 3 to 53. How much did you enjoy using the Web site? 3.26 1.05 1 to 54. How helpful was it to read and participate in the Web site activities to help manage your symptoms

related to CIPN (pain, physical functioning, anxiety, sleep disturbance, etc)?3.36 .96 1 to 5

5. Was the amount of time it took to complete the activities presented within the Web site acceptable? 4 1.11 1 to 56. Was the amount of time it took to complete the study questionnaires at the baseline, 4-week, and

8-week time points acceptable?4.42 1.02 1 to 5

7. Overall, how would you rate your satisfaction with the Web site? 3.95 .71 3 to 5

This table describes the descriptive statistics (mean, SD, range) of the Adapted Acceptability E-Scale Items at Week 8.

Table 5. PROSPECT or Other Symptom Management Resource Use According to Week

WEEK PROSPECT USE*SYMPTOM MANAGEMENT USE

(INTERVENTION)*,†SYMPTOM MANAGEMENT USE

(CONTROL)*,†

1 25.4 (0–90) (n = 14) 67.3 (0–390) (n = 15) 15.1 (0–120) (n = 10)2 12.1 (0–60) (n = 12) 63.8 (0–300) (n = 12) 2.5 (0–15) (n = 6)3 10 (0–20) (n = 6) 60 (0–210) (n = 6) 25 (0–160) (n = 8)4 17.1 (0–60) (n = 14) 19.2 (0–140) (n = 12) 27.9 (0–360) (n = 19)5 5.5 (0–30) (n = 10) 23 (0–180) (n = 10) 9.2 (0–60) (n = 10)6 16.4 (0–72) (n = 13) 49 (0–420) (n = 13) 2.7 (0–25) (n = 15)7 7.5 (0–35) (n = 13) 26.7 (0–120) (n = 13) 8.1 (0–45) (n = 12)8 8.6 (0–30) (n = 18) 6.7 (0–60) (n = 17) 32.1 (0–495) (n = 21)

This table describes the mean number of minutes individuals in the intervention or control arm spent using PROSPECT and/or other symptom management re-sources over time.*Mean number of minutes (range).†Symptom management use in the PROSPECT group refers to pain management strategies (eg, physical therapy, meditation, pool therapy, massage) used by par-ticipants in addition to the strategies embedded within the PROSPECT Web site, whereas symptom management use in the control group refers to strategies usedby participants to manage pain during the study (eg, ice therapy, massage, stretching, yoga, distraction, exercise). Type of symptom management strategies usedby participants in each treatment group varied according to individual.

Figure 3. PROSPECT versus control medication changes during study (n = 45). This figure shows the number of instances that in-dividuals (protocol completers) increased and/or decreased pain, anxiety, or sleep medication dosages. On the basis of the clinicalpractice guideline for CIPN treatment recommendations,23 neuropathic pain medications included: duloxetine, gabapentin, pregabalin,and antidepressants. Other analgesics included: oxycodone, morphine, ibuprofen, and acetaminophen.

Knoerl et al The Journal of Pain 389

increased their use of “other analgesics.” Results fromthese analyses indicated that the effect of PROSPECT onworst pain intensity was still statistically significant whenremoving individuals from the PROSPECT arm who in-creased their use of neuropathic pain medications(B = −1.07, 95% CI = −1.97 to −.15, P = .02; n = 36) andwhen removing individuals from the PROSPECT groupwho increased their use of “other analgesics” (B = −1.10,95% CI = −2.10 to −.10, P = .03; n = 33).

DiscussionThis 8-week, randomized controlled pilot trial showed

that a self-guided online cognitive behavioral pain man-agement program—PROSPECT—significantly improvedworst pain intensity in individuals with chronic painfulCIPN. Further, there were no significant differences inmean change scores between groups for the secondaryoutcomes of pain interference, nonpainful CIPN symp-toms, average pain, or global impression of change.

Individuals with chronic painful CIPN interacting withthe 8-week PROSPECT intervention reported a mean de-crease in worst pain intensity of .94. The mean decreasein pain intensity found in this study is comparable withthe effect of duloxetine, the only pharmacological agentcurrently recommended for the treatment of chronicpainful CIPN.24 The authors of a randomized, crossover,placebo-controlled study reported that use of duloxetine60 mg/d resulted in a mean decrease of 1.06 in averagepain intensity in individuals with chronic painful CIPN(P = .003, d = .513).54 On the contrary, PROSPECT had noeffect on the secondary outcome of average pain inten-sity. The comparison between average pain mean changescores between studies is challenging because the authorsof the duloxetine trial assessed average pain over the past24 hours, not 7 days.52,54 Nevertheless, the reported effectsizes and mean decreases for the 2 studies’ primary out-comes were similar for PROSPECT and duloxetine.

Although our findings supported the efficacy of PROS-PECT in improving worst CIPN pain intensity comparedwith individuals receiving treatment as usual, results frompost hoc sensitivity analyses suggest that these findingsshould be interpreted with caution. Few participants ex-perienced a clinically significant reduction in worst CIPNpain intensity after PROSPECT usage. Clinically signifi-cant improvements in pain intensity represent a 30%reduction in pain.19 The overall worst CIPN pain inten-sity mean change score for individuals receiving PROSPECTwas −.94 and only 3 individuals reported a clinically sig-nificant reduction in pain. Further, results from a post hocintent-to-treat analysis with the full sample (N = 60) re-vealed that the effect of PROSPECT on worst pain intensitywas not statistically significant (P = .09). Finally, becauseof the dearth of high-quality trials testing interven-tions for the treatment of painful CIPN, it is possible thatchanges in neuropathic pain medications and “other an-algesics” may produce an analgesic effect in individualswith chronic painful CIPN.23 Thus, changes in thefrequency/dosage of neuropathic pain medications or“other analgesics” may partially explain the efficacy ofPROSPECT on worst pain intensity. However, the results

from sensitivity analyses revealed that the effect of PROS-PECT on worst CIPN pain intensity remained statisticallysignificant after removing individuals in the PROSPECTarm who increased their use of neuropathic pain medi-cations and “other analgesics,” respectively.

There have been no published studies reporting theeffects of cognitive-behavioral pain management fornonpainful CIPN symptoms. Although nonpainful CIPNsymptoms such as numbness and tingling are a result ofperipheral nervous system damage (eg, dorsal root gangliaof primary sensory neurons),5,47 nonpainful CIPN symp-toms are still associated with changes in the centralnervous system. For example, the authors of a recent studyshowed that increased CIPN symptoms 1 month after neu-rotoxic chemotherapy are associated with changes inperfusion in brain areas associated with nociceptive pro-cessing (eg, anterior cingulate cortex and frontal gyrus).45

Thus, interventions that target centrally-mediated mecha-nisms such as cognitive behavioral pain management, mayalso be efficacious for the treatment of nonpainful CIPNsymptoms.27,50 The results of this current study supportthis conclusion because individuals receiving PROSPECThad greater improvements in nonpainful CIPN symp-toms over the 8-week trial period than individualsreceiving usual care, but, the difference was not signifi-cant. Further research is needed to determine ifinterventions targeting centrally-mediated mechanisms(eg, cognitive-behavioral pain management) can also in-fluence nonpainful CIPN symptoms.

There were also no significant differences in pain in-terference between groups. Although physical functionand 7-day recall of worst pain intensity are moderately-strongly correlated (r = .65),52 self-guided cognitive-behavioral pain management interventions are mosteffective when targeting a specific primary outcome.34

Therefore, perhaps if the PROSPECT intervention was tai-lored to include more physical activity training andeducational resources, it would have had a greater effecton pain interference.

On the basis of trends in PROSPECT usage and worstpain intensity improvement, the results suggest that par-ticipants interacted with the Web site frequently at firstto learn and practice the strategies, but then could in-corporate the strategies into their day-to-day activitiesto improve pain intensity independent of logging in toPROSPECT. However, little is known about the optimaldose of PROSPECT because we did not actively monitorpatient’s usage (eg, electronic tracking or use of strate-gies in day-to-day activities) and the response rate to theself-report measures was low. Significant predictors of self-management intervention usage include health careprofessional guidance/support, ample amount of time touse the intervention, and high satisfaction with inter-vention content.2 PROSPECT usage may be bolstered infuture studies if: 1) participants have an opportunity tointeract with a health care professional (eg, weekly videoor telephone call with nurses)25,29 to discuss pain-relatedsymptoms and strategy use, 2) participants have more timeto interact with the strategies of PROSPECT (eg, longerduration, scheduled time to interact with modules), and3) additional interactive features within PROSPECT are

390 The Journal of Pain Self-Guided Cognitive Behavioral Strategies for Neuropathy

designed (eg, achievement badges, message boards/support groups, visually appealing).2,37 Improving self-guided cognitive-behavioral pain managementintervention usage in future research is critical to estab-lish an optimal dose that can be prescribed in the clinicalsetting and to compare results across trials.

One common limitation of recent placebo-controlledrandomized controlled trials testing interventions for in-dividuals with chronic pain is high placebo response.13-15,30

Although this study did not contain a placebo control,psychoeducational interventions are prone to nonspe-cific effects (eg, study staff-participant interaction,participant motivation for participation, lack of blind-ing, credibility of treatment) that may have accountedfor the efficacy of the intervention.10 We attempted toimprove the assay sensitivity of the trial, or the abilityto distinguish an effective treatment from an ineffec-tive treatment14 by implementing strategies such as: 1)baseline participant worst CIPN pain intensity of 4 of 10or greater, participant pain duration of at least 3 monthssince completion of chemotherapy, 3) flexible interven-tion dosing, 4) standardized data collection proceduresfor both groups, and 5) <3 treatment arms.14,15 The ef-ficacy of these strategies in improving the assay sensitivityof this trial may be evidenced by the lack of control groupimprovement in worst pain intensity from the baselineto week 8 time point. However, an alternate explana-tion for this finding is that control group participants werenot blinded to the intervention they were receiving (eg,were informed that the alternative group received PROS-PECT) and thus, knew they were not receiving theintervention (nonspecific effect). Additional strategies wecould have implemented to minimize or control for non-specific effects include using a structurally equivalentcontrol group and administering manipulation checks todetermine if participants were using the treatment asintended.10 Nevertheless, future studies should con-tinue to implement strategies to improve assay sensitivityto facilitate the identification of effective treatments forindividuals with chronic pain.

There are several limitations to this study. First, thedropout rate was approximately 22%. This dropout rateis consistent with other self-guided cognitive behav-ioral pain management intervention studies38 and thedemographic characteristics of the completers andnoncompleters were similar between groups. However,because of the high number of individuals with stage IVcancers dropping out of the study, further research is

needed to examine the feasibility of administering PROS-PECT in individuals with advanced cancer. Second, we didnot conduct a power analysis because of the pilot natureof this study. Nevertheless, it is highly unlikely that thisstudy was adequately powered to detect differencesprimary or secondary outcomes between groups becauseof the small sample size and high dropout rate. Third,although worst pain intensity improved after PROS-PECT use, little is known as to what components led tothese improvements. The identification of mediators ofpain intensity improvement may allow for the develop-ment of self-guided cognitive behavioral painmanagement interventions containing the strategies hy-pothesized to improve the mediators. Fourth, greater useof other symptom management resources by individu-als in PROSPECT may have confounded the results of theprimary outcome. Fifth, we cannot generalize the resultsof this study to individuals with CIPN resulting from a spe-cific neurotoxic agent because we examined the PROSPECTintervention in individuals who received varying types ofneurotoxic drugs. Sixth, participants were not blinded towhat treatment (PROSPECT or wait-list control) they werereceiving and we did not assess for participants’ expec-tations of improvement on the basis of study groupassignment at baseline. Last, individuals receiving PROS-PECT had higher levels of baseline depression, anxiety,fatigue, and sleep-related impairment than individualsin the control group. Despite these differences in co-occurring symptom severity at baseline, individualsreceiving PROSPECT still experienced greater reduc-tions in worst CIPN pain.

Overall, currently there is 1 recommended pharmaco-logical agent and no nonpharmacological modalitiesrecommended for the treatment of chronic painful CIPN.This pilot study provides preliminary evidence support-ing the efficacy of a self-guided cognitive-behavioral painmanagement intervention for improving worst pain in-tensity in individuals with chronic painful CIPN. However,because of the small sample size and stated limitations,a larger study is needed to determine the true effect ofPROSPECT on pain intensity and the secondary out-comes. If shown to be efficacious in a larger study,PROSPECT should be tested alongside pharmacologicalagents for the treatment of chronic painful CIPN. Becausepain often clusters with other symptoms,41,55 multimodaltreatment approaches for chronic pain are beneficialbecause they target underlying pain as well as pain-related physiological mechanisms.

References

1. Amtmann D, Cook KF, Jensen MP, Chen WH, Choi S, RevickiD, Cella D, Rothrock N, Keefe F, Callahan L, Lai JS: Devel-opment of a PROMIS item bank to measure pain interference.Pain 150:173-182, 2010

2. Beatty L, Binnion C: A systematic review of predictors of,and reasons for, adherence to online psychological inter-ventions. Int J Behav Med 23:776-794, 2016

3. Berry DL, Blumenstein BA, Halpenny B, Wolpin S, FannJR, Austin-Seymour M, Bush N, Karras BT, Lober WB, McCorkle

R: Enhancing patient-provider communication with the elec-tronic self-report assessment for cancer: A randomized trial.J Clin Oncol 29:1029-1035, 2011

4. Buuren S, van Groothuis-Oudshoorn K: MICE: Multivari-ate Imputation by Chained Equations in R. J Stat Softw 45:1-67, 2011

5. Carozzi VA, Canta A, Chiorazzi A: Chemotherapy-inducedperipheral neuropathy: What do we know about mecha-nisms? Neurosci Lett 596:90-107, 2015

6. Cavaletti G, Cornblath DR, Merkies IS, Postma TJ, RossiE, Frigeni B, Alberti P, Bruna J, Velasco R, Argyriou AA,

Knoerl et al The Journal of Pain 391

Kalofonos HP, Psimaras D, Ricard D, Pace A, Galie E, BrianiC, Torre CD, Faber CG, Lalisang RI, Boogerd W, BrandsmaD, Koeppen S, Hense J, Storey D, Kerrigan S, Schenone A,Fabbri S, Valsecchi MG, CI-PeriNomS Group: Thechemotherapy-induced peripheral neuropathy outcome mea-sures standardization study: From consensus to the firstvalidity and reliability findings. Ann Oncol 24:454-462, 2013

7. Christiansen S, Oettingen G, Dahme B, Klinger R: A shortgoal-pursuit intervention to improve physical capacity: A ran-domized clinical trial in chronic back pain patients. Pain 149:444-452, 2010

8. Cleeland CS, Ryan KM: Pain assessment: Global use of theBrief Pain Inventory. Ann Acad Med Singapore 23:129-138,1994

9. Dear BF, Titov N, Perry KN, Johnston L, Wootton BM,Terides MD, Rapee RM, Hudson JL: The Pain Course: Arandomised controlled trial of a clinician-guided Internet-delivered cognitive behaviour therapy program for managingchronic pain and emotional well-being. Pain 154:942-950,2013

10. Donovan HS, Kwekkeboom KL, Rosenzweig MQ, WardSE: Nonspecific effects in psychoeducational intervention re-search. West J Nurs Res 31:983-998, 2009

11. Dworkin RH, O’Connor AB, Audette J, Baron R, GourlayGK, Haanpaa ML, Kent JL, Krane EJ, Lebel AA, Levy RM,Mackey SC, Mayer J, Miaskowski C, Raja SN, Rice AS, SchmaderKE, Stacey B, Stanos S, Treede RD, Turk DC, Walco GA, WellsCD: Recommendations for the pharmacological manage-ment of neuropathic pain: An overview and literature update.Mayo Clin Proc 85:S3-S14, 2010

12. Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA,Jensen MP, Katz NP, Kerns RD, Stucki G, Allen RR, BellamyN, Carr DB, Chandler J, Cowan P, Dionne R, Galer BS, HertzS, Jadad AR, Kramer LD, Manning DC, Martin S, McCor-mick CG, McDermott MP, McGrath P, Quessy S, RappaportBA, Robbins W, Robinson JP, Rothman M, Royal MA, SimonL, Stauffer JW, Stein W, Tollett J, Wernicke J, Witter J:IMMPACT: Core outcome measures for chronic pain clinicaltrials: IMMPACT recommendations. Pain 113:9-19, 2005

13. Dworkin RH, Turk DC, Katz NP, Rowbotham MC, Peirce-Sandner S, Cerny I, Clingman CS, Eloff BC, Farrar JT, KampC, McDermott MP, Rappaport BA, Sanhai WR: Evidence-based clinical trial design for chronic pain pharmacotherapy:A blueprint for ACTION. Pain 152:S107-S115, 2011

14. Dworkin RH, Turk DC, Peirce-Sandner S, Burke LB, FarrarJT, Gilron I, Jensen MP, Katz NP, Raja SN, Rappaport BA,Rowbotham MC, Backonja MM, Baron R, Bellamy N,Bhagwagar Z, Costello A, Cowan P, Fang WC, Hertz S, JayGW, Junor R, Kerns RD, Kerwin R, Kopecky EA, Lissin D,Malamut R, Markman JD, McDermott MP, Munera C, PorterL, Rauschkolb C, Rice AS, Sampaio C, Skljarevski V,Sommerville K, Stacey BR, Steigerwald I, Tobias J, TrentacostiAM, Wasan AD, Wells GA, Williams J, Witter J, Ziegler D: Con-siderations for improving assay sensitivity in chronic painclinical trials: IMMPACT recommendations. Pain 153:1148-1158, 2012

15. Dworkin RH, Turk DC, Peirce-Sandner S, He H, McDer-mott MP, Farrar JT, Katz NP, Lin AH, Rappaport BA,Rowbotham MC: Assay sensitivity and study features in neu-ropathic pain trials: An ACTTION meta-analysis. Neurology81:67-75, 2013

16. Eckhoff L, Knoop A, Jensen MB, Ewertz M: Persistenceof docetaxel-induced neuropathy and impact on quality of

life among breast cancer survivors. Eur J Cancer 51:292-300, 2015

17. Ehde DM, Dillworth TM, Turner JA: Cognitive-behavioraltherapy for individuals with chronic pain: Efficacy, innova-tions, and directions for research. Am Psychol 69:153-166,2014

18. Farquhar-Smith P: Chemotherapy-induced neuropathicpain. Curr Opin Support Palliat Care 5:1-7, 2011

19. Farrar JT, Young JP Jr, LaMoreaux L, Werth JL, Poole RM:Clinical importance of changes in chronic pain intensity mea-sured on an 11-point numerical pain rating scale. Pain 94:149-158, 2001

20. Gharibian D, Polzin JK, Rho JP: Compliance and persis-tence of antidepressants versus anticonvulsants in patientswith neuropathic pain during the first year of therapy. ClinJ Pain 29:377-381, 2013

21. Harris K, Li K, Flynn C, Chow E: Worst, average or currentpain in the Brief Pain Inventory: Which should be used tocalculate the response to palliative radiotherapy in pa-tients with bone metastases? Clin Oncol (R Coll Radiol) 19:523-527, 2007

22. Heapy A, Dziura J, Buta E, Goulet J, Kulas JF, Kerns RD:Using multiple daily pain ratings to improve reliability andassay sensitivity: How many is enough? J Pain 15:1360-1365, 2014

23. Hershman DL, Lacchetti C, Dworkin RH, Lavoie Smith EM,Bleeker J, Cavaletti G, Chauhan C, Gavin P, Lavino A, LustbergMB, Paice J, Schneider B, Smith ML, Smith T, Terstriep S,Wagner-Johnston N, Bak K, Loprinzi CL, American Societyof Clinical Oncology: Prevention and management ofchemotherapy-induced peripheral neuropathy in survivorsof adult cancers: American Society of Clinical Oncology clini-cal practice guideline. J Clin Oncol 32:1941-1967, 2014

24. Hershman DL, Lacchetti C, Dworkin RH, Lavoie Smith EM,Bleeker J, Cavaletti G, Chauhan C, Gavin P, Lavino A, LustbergMB, Paice J, Schneider B, Smith ML, Smith T, Terstriep S,Wagner-Johnston N, Bak K, Loprinzi CL, American Societyof Clinical Oncology: Prevention and management ofchemotherapy-induced peripheral neuropathy in survivorsof adult cancers: American Society of Clinical Oncology clini-cal practice guideline. J Clin Oncol 32:1941-1967, 2014

25. Heutink M, Post MW, Bongers-Janssen HM, Dijkstra CA,Snoek GJ, Spijkerman DC, Lindeman E: The CONECSI trial:Results of a randomized controlled trial of a multidisci-plinary cognitive behavioral program for coping with chronicneuropathic pain after spinal cord injury. Pain 153:120-128, 2012

26. Hjermstad MJ, Fayers PM, Haugen DF, Caraceni A, HanksGW, Loge JH, Fainsinger R, Aass N, Kaasa S: Studies com-paring numerical rating scales, verbal rating scales, and visualanalogue scales for assessment of pain intensity in adults:A systematic literature review. J Pain Symptom Manage 41:1073-1093, 2011

27. Jensen KB, Kosek E, Wicksell R, Kemani M, Olsson G,Merle JV, Kadetoff D, Ingvar M: Cognitive behavioral therapyincreases pain-evoked activation of the prefrontal cortex inpatients with fibromyalgia. Pain 153:1495-1503, 2012

28. Jensen MP, Karoly P, Braver S: The measurement of clini-cal pain intensity: A comparison of six methods. Pain 27:117-126, 1986

392 The Journal of Pain Self-Guided Cognitive Behavioral Strategies for Neuropathy

29. Jungquist CR, O’Brien C, Matteson-Rusby S, Smith MT,Pigeon WR, Xia Y, Lu N, Perlis ML: The efficacy of cognitive-behavioral therapy for insomnia in patients with chronic pain.Sleep Med 11:302-309, 2010

30. Katz N: Methodological issues in clinical trials of opioidsfor chronic pain. Neurology 65:S32-S49, 2005

31. Kautio AL, Haanpää M, Kautiainen H, Kalso E, SaartoT: Burden of chemotherapy-induced neuropathy—a cross-sectional study. Support Care Cancer 19:1991-1996, 2011

32. Kerns RD, Sellinger J, Goodin BR: Psychological treat-ment of chronic pain. Annu Rev Clin Psychol 7:411-434, 2011

33. Knoerl R, Gray E, Stricker C, Mitchell SA, Kippe K, SmithG, Dudley WN, Lavoie Smith EM: Electronic versus paper-pencil methods for assessing chemotherapy-inducedperipheral neuropathy, in Support Care Cancer. Berlin Heidel-berg, Springer, 2017, pp 1-10

34. Knoerl R, Lavoie Smith EM, Weisberg J: Chronic pain andcognitive behavioral therapy: An integrative review. WestJ Nurs Res 38:596-628, 2015

35. Lavoie Smith EM, Barton DL, Qin R, Steen PD, AaronsonNK, Loprinzi CL: Assessing patient-reported peripheralneuropathy: the reliability and validity of the European Or-ganization for Research and Treatment of Cancer QLQ-CIPN20 Questionnaire. Qual Life Res 22:2787-2799, 2013

36. Li L, Liu X, Herr K: Postoperative pain intensity assessment:a comparison of four scales in Chinese adults. Pain Med 8:223-234, 2007

37. Ludden GD, van Rompay TJ, Kelders SM, van Gemert-Pijnen JE: How to increase reach and adherence of web-based interventions: a design research viewpoint. J MedInternet Res 17:e172, 2015

38. Macea DD, Gajos K, Daglia Calil YA, Fregni F: The effi-cacy of Web-based cognitive behavioral interventions forchronic pain: a systematic review and meta-analysis. J Pain11:917-929, 2010

39. Marmiroli P, Scuteri A, Cornblath DR, Cavaletti G: Painin chemotherapy-induced peripheral neurotoxicity. J PeripherNerv Syst 22:156-161, 2017

40. McBeth J, Prescott G, Scotland G, Lovell K, Keeley P,Hannaford P, McNamee P, Symmons DP, Woby S, GkazinouC, Beasley M, Macfarlane GJ: Cognitive behavior therapy, ex-ercise, or both for treating chronic widespread pain. ArchIntern Med 172:48-57, 2012

41. Miaskowski C, Paul SM, Cooper B, West C, Levine JD,Elboim C, Hamolsky D, Abrams G, Luce J, Dhruva A, LangfordDJ, Merriman JD, Kober K, Baggott C, Leutwyler H, AouizeratBE: Identification of patient subgroups and risk factors forpersistent arm/shoulder pain following breast cancer surgery.Eur J Oncol Nurs 18:242-253, 2014

42. Mols F, Beijers T, Vreugdenhil G, van de Poll-Franse L:Chemotherapy-induced peripheral neuropathy and its as-sociation with quality of life: A systematic review. SupportCare Cancer 22:2261-2269, 2014

43. Moriarty O, McGuire BE, Finn DP: The effect of pain oncognitive function: A review of clinical and preclinical re-search. Prog Neurobiol 93:385-404, 2011

44. National Cancer Institute: Common Terminology Crite-ria for Adverse Events v4.0. Available at https://

ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/CTCAE_4.03.xlsx. Accessed January 9, 2018

45. Nudelman KN, McDonald BC, Wang Y, Smith DJ, WestJD, O’Neill DP, Zanville NR, Champion VL, Schneider BP, SaykinAJ: Cerebral perfusion and gray matter changes associatedwith chemotherapy-induced peripheral neuropathy. J ClinOncol 34:677-683, 2016

46. Otis JD, Sanderson K, Hardway C, Pincus M, Tun C,Soumekh S: A randomized controlled pilot study of acognitive-behavioral therapy approach for painful diabeticperipheral neuropathy. J Pain 14:475-482, 2013

47. Park SB, Goldstein D, Krishnan A V, Lin CS, FriedlanderML, Cassidy J, Koltzenburg M, Kiernan MC: Chemotherapy-induced peripheral neurotoxicity: A critical analysis. CA CancerJ Clin 63:419-437, 2013

48. Postma TJ, Aaronson NK, Heimans JJ, Muller MJ,Hildebrand JG, Delattre JY, Hoang-Xuan K, Lanteri-Minet M,Grant R, Huddart R, Moynihan C, Maher J, Lucey R, EORTCQuality of Life Group: The development of an EORTC qualityof life questionnaire to assess chemotherapy-induced pe-ripheral neuropathy: The QLQ-CIPN20. Eur J Cancer 41:1135-1139, 2005

49. R Development Core Team: R: A Language and Envi-ronment for Statistical Computing. Available at: https://www.r-project.org. Accessed January 9, 2017

50. Seminowicz DA, Shpaner M, Keaser ML, Krauthamer GM,Mantegna J, Dumas JA, Newhouse PA, Filippi CG, Keefe FJ,Naylor MR: Cognitive-behavioral therapy increases prefron-tal cortex gray matter in patients with chronic pain. J Pain14:1573-1584, 2013

51. Seretny M, Currie GL, Sena ES, Ramnarine S, Grant R,MacLeod MR, Colvin LA, Fallon M: Incidence, prevalence, andpredictors of chemotherapy-induced peripheral neuropathy:A systematic review and meta-analysis. Pain 155:2461-2470, 2014

52. Shi Q, Wang XS, Mendoza TR, Pandya KJ, Cleeland CS:Assessing persistent cancer pain: A comparison of currentpain ratings and pain recalled from the past week. J PainSymptom Manage 37:168-174, 2009

53. Smith EM, Bridges CM, Kanzawa G, Knoerl R, Kelly JP,Berezovsky A, Woo C: Cancer treatment-related neuro-pathic pain syndromes—Epidemiology and treatment: Anupdate. Curr Pain Headache Rep 18:459, 2014

54. Smith EM, Pang H, Cirrincione C, Fleishman S, PaskettED, Ahles T, Bressler LR, Fadul CE, Knox C, Le-LindqwisterN, Gilman PB, Shapiro CL, Alliance for Clinical Trials inOncology: Effect of duloxetine on pain, function, and qualityof life among patients with chemotherapy-induced painfulperipheral neuropathy: A randomized clinical trial. JAMA 309:1359-1367, 2013

55. Smith EM, Pang H, Ye C, Cirrincione C, Fleishman S,Paskett ED, Ahles T, Bressler LR, Le-Lindqwister N, Fadul CE,Loprinzi C, Shapiro CL, Alliance for Clinical Trials in Oncology:Predictors of duloxetine response in patients with oxaliplatin-induced painful chemotherapy-induced peripheralneuropathy (CIPN): A secondary analysis of randomised con-trolled trial—CALGB/alliance 170601. Eur J Cancer Care (Engl)26, 2015

56. Staff NP, Grisold A, Grisold W, Windebank AJ:Chemotherapy-induced peripheral neuropathy: A currentreview. Ann Neurol 81:772-781, 2017

Knoerl et al The Journal of Pain 393

57. Stubblefield MD, Burstein HJ, Burton AW, Custodio CM,Deng GE, Ho M, Junck L, Morris GS, Paice JA, Tummala S,Von Roenn JH: NCCN task force report: Management of neu-ropathy in cancer. J Natl Compr Canc Netw 7(Suppl 5):S1-S26, quiz S27-S28, 2009

58. Tang NK, Wright KJ, Salkovskis PM: Prevalence and cor-relates of clinical insomnia co-occurring with chronic backpain. J Sleep Res 16:85-95, 2007

59. Tariman JD, Berry DL, Halpenny B, Wolpin S, Schepp K:Validation and testing of the Acceptability E-scale for web-based patient-reported outcomes in cancer care. Appl NursRes 24:53-58, 2011

60. Taylor DJ, Mallory LJ, Lichstein KL, Durrence HH, RiedelBW, Bush AJ: Comorbidity of chronic insomnia with medicalproblems. Sleep 30:213-218, 2007

61. Thorn BE, Day MA, Burns J, Kuhajda MC, Gaskins SW,Sweeney K, McConley R, Ward LC, Cabbil C: Randomized trial

of group cognitive behavioral therapy compared with a paineducation control for low-literacy rural people with chronicpain. Pain 152:2710-2720, 2011

62. Ventzel L, Jensen AB, Jensen AR, Jensen TS, Finnerup NB:Chemotherapy-induced pain and neuropathy: A prospec-tive study in patients treated with adjuvant oxaliplatin ordocetaxel. Pain 157:560-568, 2016

63. William G: National Institute of Mental Health: ECDEUAssessment Manual for Psychopharmacology – Revised.Rockville, MD, US Department of Health, Education, andWelfare, 1976

64. Williams AC, Eccleston C, Morley S: Psychological thera-pies for the management of chronic pain (excludingheadache) in adults. Cochrane Database Syst Rev (11):CD007407, 2012

65. Williams DA, Kuper D, Segar M, Mohan N, Sheth M,Clauw DJ: Internet-enhanced management of fibromyalgia:A randomized controlled trial. Pain 151:694-702, 2010

394 The Journal of Pain Self-Guided Cognitive Behavioral Strategies for Neuropathy