UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Verbal description of pain by patients with cancer Gagel, Mike Peter 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_gagel_mike.pdf [ 3.97MB ]
JSON: 831-1.0086104.json
JSON-LD: 831-1.0086104-ld.json
RDF/XML (Pretty): 831-1.0086104-rdf.xml
RDF/JSON: 831-1.0086104-rdf.json
Turtle: 831-1.0086104-turtle.txt
N-Triples: 831-1.0086104-rdf-ntriples.txt
Original Record: 831-1.0086104-source.json
Full Text

Full Text

VERBAL DESCRIPTION OF PAIN BY PATIENTS WITH CANCERbyMIKE PETER GAGELB.S.N., University of British Columbia, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCE IN NURSINGinTHE FACULTY OF GRADUATE STUDIESSCHOOL OF NURSINGWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1993© Mike Peter Gagel, 1993In presenting this thesis in partial fulfillment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the Head of myDepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.School of NursingUniversity of British ColumbiaVancouver, BCCanadaApril 30, 199311ABSTRACTResearchers have demonstrated an incongruence between patients' and nurses'perceptions of pain which may lead nurses to not appreciate or heed their patients'reports of pain. Part of this problem may lie with pain assessment. This descriptiveexploratory study was designed to determine what words, similes, and metaphorscancer patients use to describe their pain. The goal was to improve understanding ofthese patients' descriptions of pain as an essential basis for more accurate cancer painassessment. Also, the relationship of selected demographic variables (age, level ofeducation, mother tongue, and sex) and present pain intensity with the frequency andtypes of phrases containing pain descriptors were examined to determine if significantrelationships existed.A convenience sample of thirty-one patients at a large cancer treatment facilitywas asked to describe what pain feels like. The existing database of transcribedinterviews was analyzed using content analysis techniques. Coding of descriptorwords and phrases was based on dimensions of the Multidimensional Model of Pain(MMP) (Ahles, Blanchard, and Ruckdeschel, 1983; McGuire, 1987). Themes alsoemerged during data analysis and the following items were coded: similes, metaphors,and the number of times participants expressed difficulty in describing their pain.Data analysis revealed that participants used phrases from the Affective,Behavioural, Cognitive, Physiological, and Sensory Dimensions of the MMP. Themost frequently used descriptors matched those used by cancer patients in otherstudies. The participants spontaneously and frequently used 25 out of 73 descriptorscontained in the McGill Pain Questionnaire (MPQ); they also used 26 descriptors notfound in the MPQ. Participants primarily used affective and sensory words andphrases to describe their pain. Participants with Grade 12 or higher education used111more affective phrases than those with less than Grade 12. Participants who wereyounger than 65 and those with Grade 12 or higher used more sensory phrases thantheir counterparts. Participants with Grade 12 or higher also used more similes. Theremaining dimensions of the MMP and expressions of difficulty did not presentsignificant differences in the use of descriptor phrases by the study participants.Relationships based on participants' sex and mother tongue were not significant.Participants also used few metaphors to describe their pain.Based on the findings of this study, recommendations for nursing were offered.Namely, attention should be given to patients' descriptions of pain. Older or lesseducated patients may not use as many phrases to describe pain as younger or moreeducated patients (even though older patients likely experience the same intensity ofpain as others). Pain assessment tools may help overcome differences among patients'ability to describe pain, and nurses should implement these tools in their practice.However, for pain assessment tools to be practical, they must be easy to use and quickto administer. Revision of the descriptor portion of the McGill Pain Questionnairebased on the spontaneous descriptors used by cancer patients, is one area where nursescan contribute to the improved assessment and management of pain in patients withcancer.TABLE OF CONTENTSAbstract^ iiTable of Contents^ ivList of Tables viiList of Figures ^ viiiAcknowledgments ixChapter 1: Describing Pain^ 1Pain and Cancer 1Prevalence of Pain in Cancer Patients^ 1Causes of Pain in Cancer Patients 1Nurses' Awareness of the Problems of Pain Management in CancerPatients^ 2Nurses' Perception of Patients' Pain^ 3McGill Pain Questionnaire^ 4Background to this Study 6Research Question^ 6Assumptions 6Purposes of the Study 7Definition of Terms^ 7Significance of the Study 8Conceptual Model of Pain 8Summary^ 9Chapter 2: Theoretical Background and Literature review^ 11Multidimensional Model of Pain^ 11Pain Descriptor Studies^ 13Critique of the McGill Pain Questionnaire^ 13MPQ Pain Descriptor Studies and Cancer Patients ^ 16Classification of Disease by Pain Descriptor 16Selection of MPQ Pain Descriptors by Cancer Patients^ 17MPQ Descriptors and Cancer Patients — Meta-analysis 19Spontaneous Pain Descriptor Studies and Cancer Patients^ 21Summary and Conclusions^ 24Chapter 3: Methodology 26Instrument Study^ 26Data Collection 27Design^ 28Sample 29ivVSelection Criteria^ 29Selection Procedure 29Setting^ 30Ethics 31Data Analysis 32Content Analysis^ 32Steps Involved in Content Analysis^ 33Validity and Reliability^ 33Validity^ 33Reliability. 34Statistical Analysis.^ 35Data Analysis Tools 36Summary^ 38Chapter 4: Findings and Interpretations^ 39Sample Characteristics^ 39Spontaneous Descriptor Phrases and Pain Descriptors^ 40Descriptor Phrases 40Pain Descriptors^ 42Discussion of Descriptors^ 47Participant Variables and Descriptor Phrases^ 49Grouping of Variables 49Differences Among Variables^ 50Correlations Among Continuous and Grouped Variables, and CodedPhrases^ 51Discussion of Correlations^ 52Relationships Among Grouped Variables and Coded Phrases^ 52Affective Dimension 54Sensory Dimension^ 56Use of Similes 58Discussion of the Findings 60Summary and Conclusions^ 62Chapter 5: Conclusions, Implications, and Summary^ 65Conclusions^ 65Limitations 67Implications For Nurses^ 68Education^ 68Practice 70Future Research 72Summary^ 73viReferences^ 74Appendices 79A: McGill Pain Questionnaire^ 79B: Letter of Permission 80C: Interview Guide^ 81D: Data Collection Guide 82E: Clinical Investigations Committee Approval^ 83F: Ethics Approval^ 84G: Ethics Approval (Renewal)^ 85H: Letter to Participants 86K: Chi-Square Tables^ 92L: Correlation Matrix 93viiLIST OF TABLESTable 1: Affective Class of the MPQ^ 14Table 2: Selection Frequency of MPQ Pain Descriptors by Cancer Patients^ 20Table 3: Comparative Use of Spontaneous and MPQ Selected PainDescriptors by Patients with Cancer^ 24Table 4: Manifest Coding of Phrases — Multidimensional Model of Pain^40Table 5: Latent Coding of Phrases — Difficulty and Simile^ 41Table 6: MPQ Descriptors Used by Participants^ 43Table 7: Non-MPQ Descriptors Used by Participants 45Table 8: Descriptor-Modifiers Used by Participants^ 46Table 9: Spontaneously Used Pain Descriptors 48Table 10: Pearson Correlation Matrix^ 51Table 11: Probability Values for Coding Results ^ 53Table 12: Statistical Summary of Affective Phrases Grouped by Level ofEducation^ 55Table 13: Statistical Summary of Sensory Phrases Grouped by Age^ 56Table 14: Statistical Summary of Sensory Phrases Grouped by Level ofEducation^ 57Table 15: Statistical Summary of Similes Grouped by Level of Education^ 59Table 16: Level of Education by Age^ 61viiiLIST OF FIGURESFigure 1: Visual Analog Scale^ 28Figure 2: Number of Affective Phrases by Level of Education^ 55Figure 3: Number of Sensory Phrases by Age^ 57Figure 4: Number of Sensory Phrases by Level of Education^ 58Figure 5: Number of Similes by Level of Education^ 60Figure 6: Level of Education by Age^ 62ixACKNOWLEDGMENTSI would like to acknowledge the following individuals for their help incompleting this thesis:Consultants• Michêle Deschamps, RN, PhD, for introducing me to her ideas regarding painassessment and for permitting me to use her data in my thesis research.• Pat Porterfield, RN, MSN, for being a content expert.Committee Members • Anna-Marie Hughes, RN, EdD, for helping me develop my topic and seeing methrough to the end.• Jinny Hayes, RN, PhD, for her research expertise and our discussions regardinggender relations.• Carolyn Baker, RN, MScN, for her role as content expert and mentor.Family and Friends• My parents, family, and friends for their frequent yet gentle enquiries regarding"the paper" or "the 'T' Word."• Alexandra K. Dmitrasinovic, for her love, support, and unfailing encouragement.1CHAPTER 1: DESCRIBING PAINThe problem of adequate pain control and management has long beenrecognized by oncology nurses. Part of the problem with pain management lies withits assessment; researchers have demonstrated an incongruence between patients' andnurses' perceptions of pain which may lead nurses to not appreciate or heed theirpatients' reports of pain. Although the McGill Pain Questionnaire was an attempt toprovide an objective assessment of pain, there are reasons to argue for furtherinvestigation of the description of pain by patients with cancer.Pain and CancerPrevalence of Pain in Cancer Patients Pain in patients with cancer is, unfortunately, a widespread problem. Overall,40% of people with cancer experience pain (Health and Welfare Canada, 1984).Authors have noted that 50% to 87% of patients with highly advanced cancers havepain sometime during their hospitalization (Coyle & Foley, 1987; Donovan & Dillon,1987; Health and Welfare Canada, 1984; Pannuti, Rossi, Marraro, Strocchi, Piana, &Pollutri, 1980). In British Columbia, figures suggest that up to 26,000 people annuallymay experience pain due to cancer or its treatment (Deschamps, Coldman, Bradley, &Baker, 1991).Causes of Pain in Cancer Patients Patients with cancer may experience pain from a number of causes.Approximately 80% of pain in these patients is due to local or distant spread of cancer(Coyle & Foley, 1987). About 15% of pain in patients with cancer is related to cancertreatment (Coyle & Foley, 1987; Twycross & Fairfield, 1982). A small portion (about5%) of pain in patients with cancer is associated with causes other than cancer or itstreatment (e.g., migraine headache, arthritis, ulcer, deep vein thrombosis, bedsores,2etc.) (Twycross & Fairfield, 1982). This multicausal nature of cancer pain may makeit hard for patients to communicate about pain with their care providers. It also makespain assessment by care providers difficult and, therefore, may affect the provision ofadequate or optimal pain management.Nurses' Awareness of the Problems of Pain Management in Cancer Patients Nurses' awareness of problems with pain management for patients with cancerhas been well-documented. In Canada, the Canadian Association of Nurses inOncology (CANO) has published a position paper regarding pain management forcancer patients (1988). Degner (1987) reported a survey that specified painmanagement education for health professionals as the second highest priority topic(after stomatitis) for patient welfare and cancer nursing practice. Bramwell (1989)also reported a survey that ranked analgesia (pharmaceutical pain management) asthird among the top ten physical care problems (after nutrition, and nausea andvomiting) for patients with cancer.In the United States, the Oncology Nursing Society (ONS) has surveyed itsmembership several times during the past decade. In 1981, 1984, 1988, and 1991,pain control and management consistently ranked as one of the top four researchinterests or priorities for ONS members (Mooney, Ferrell, Nail, Benedict, &Haberman, 1991). The ONS also recently published a three-part position paper onpain in cancer patients (Spross, McGuire, & Schmitt, 1990a; 1990b; 1990c). Thispaper notes that pain control and management are research priorities for nurses;however, "despite 10 years of interest in nursing research on pain in people withcancer, only a limited body of research-based information exists" (Spross et al., 1990b,p. 758).3Nurses' Perception of Patients' PainDespite nurses' acknowledgment of the problem of pain management in cancerpatients, researchers in pain management have raised an important issue: patients' andnurses' perceptions of patients' pain may not be congruent causing patients' reports ofpain to go unheeded. For example, Teske, Daut, and Cleeland (1983) had two nurseobservers assess non-verbal pain behaviours in a sample of 34 patients with acute painand 37 patients with chronic pain. While they found no significant difference betweennurse observer ratings and self-reported ratings of acute pain patients, they did find asignificant difference between nurse observer ratings and self-reported ratings ofchronic pain patients (p<0.001). Rankin and Snider (1984) found that 89% of asample of 52 nurses felt their cancer patients were receiving adequate analgesia. Atthe same time, these nurses felt that 67% of these patients had moderate pain(implying the nurses believed moderate pain to be acceptable). Donovan and Dillon(1987) found in their sample that only 43% of 69 cancer patients with pain could recalla nurse having discussed pain with them. The patients' records tended to confirm thattheir reports of pain were disregarded as only 27% of the progress notes reported painas a problem within the previous 72 hours. Similarly, in another study, where patientinterviews were followed by an audit of nurses' notes, 30 cancer patients made verbalreports of pain to 22 nurses, but only 10 of the nurses were in general agreement withthe patients' statements about pain (Camp, 1988).Nurses' observations of pain may influence pain assessment. Dalton (1989)surveyed three groups of nurses — hospital nurses who had contact with cancerpatients, hospital nurses who did not have contact with cancer patients, and OncologyNursing Society members. Dalton found that approximately 75% of all the nurses inher sample stated that they first ask a direct question when assessing pain; however,4approximately 80% of nurses stated they also observe patient behaviour to determineif the patient is in pain. The nurses' behaviour is seemingly in contrast to these leadingexperts' assertion: "pain is whatever the experiencing person says it is" (McCaffery &Beebe, 1989, p. 7). Allan (N.D.), a neurologist specializing in pain, concurs bystating: "we must rely heavily on the subjective description of pain by the individualexperiencing it" (p. 7).Altogether, these studies suggest that nurses rely heavily on their ownperceptions or observations of patients' pain behaviours when assessing pain. Theseperceptions can exclude the importance of nurses looking closely at patients'subjective reports of pain. Because patients' verbal reports are not adequatelyconsidered by nurses, formal pain assessment techniques may be a more informativeway for nurses to assess cancer patients' pain. Recognition of the necessity to includepatients' subjective reports of pain has been demonstrated through the developmentand use of pain assessment tools.McGill Pain Questionnaire Pain assessment made a great gain with the development of the McGill PainQuestionnaire (MPQ) (Melzack, 1975). However, there have been few studies thathave specifically addressed pain assessment for cancer patients. Tearnan and Cleeland(1990) have noted, "most studies of pain quality have evaluated patients with chronicnonmalignant pain. Few have examined the pain of cancer patients" (p. 228). Theyalso note that the pain which cancer patients experience may be qualitatively differentthan that which patients with chronic pain experience.The MPQ has been tested to determine its suitability for use with patients witha variety of diseases and conditions including cancer. (Throughout this report, allMPQ citations refer to Melzack [1975].) Testing the idea that pain may be different5with different conditions, Dubuisson and Melzack (1976) administered the MPQ to 95patients with varying diagnoses (eight had cancer). They concluded that the MPQ wasuseful as a diagnostic tool because "77% of all the patients in the study [used paindescriptors that] resembled those of other patients in their own diagnostic categorymore than any other" (pp. 485-86). Later, McGuire administered the MPQ to 24cancer inpatients and concluded that the MPQ was a "reliable and valid measure ofcancer pain" (1984, p. 37).While the MPQ is a workable pain assessment tool for patients with cancer inits assessment of the psychological and physiological aspects of pain, it has twoshortcomings. First, administration of the MPQ can take twenty minutes or longer tocomplete (McGuire, 1984; Melzack, 1975; Zimmerman, Duncan, Pozehl, & Schmitz,1987). Such a lengthy administration time can potentially affect a tool's usefulness inclinical practice by contributing to patient fatigue or delaying initiation of pain reliefmeasures and making it impractical for nurses to use, especially in repeatedassessments of pain. Therefore, a tool with a quick administration time may be moreattractive for use in clinical settings.Second, words in the MPQ that a patient must select to describe his or her painwere originally gathered from medical and psychological literature and then tested onsubjects such as healthy university students (Melzack & Torgerson, 1971). As aresult, according to Deschamps, Band, and Coldman (1988), patients may notcomprehend or be able to use words contained in the MPQ. Pain descriptorscontained in the MPQ may not reflect terms cancer patients use. Accordingly, itbecomes important to identify the pain descriptors cancer patients spontaneously useto describe their pain. The spontaneous words provided by cancer patients may ormay not be present in the pain descriptors of the MPQ.6Background to this StudyThe nursing division of a prominent cancer treatment facility in BritishColumbia has identified pain as a research priority (Deschamps, M., personalcommunication, September, 14 & 16, 1992; Deschamps, 1991; Deschamps, Coldman,Bradley, & Baker, 1991). Accordingly, the division embarked on a two-phaseresearch programme: (a) To determine the prevalence and characteristics of painamong a sample of 8,564 adults with cancer in British Columbia; and (b) to develop aninstrument to evaluate cancer pain (hereafter referred to as the Instrument study).The Instrument study was the second phase of the research programme andwas designed to develop an instrument to evaluate cancer pain based on words usedby cancer patients to describe the sensory and affective components of their pain. Thenursing division of the cancer facility agreed to provide funding to initiate theInstrument study on the condition that external funding would be obtained to enablethe study to continue. Extra funding was not obtained, and only the first step of theInstrument study was carried out with a total of 39 interviews conducted.Research QuestionThese interviews, however, constitute a valuable (though small) source ofinterview data for an analysis of pain description. Hence, the question for this studywas developed: What pain descriptors do cancer patients spontaneously use todescribe their pain?Assumptions1. Pain is a problem for patients with cancer. It is multicausal and presentsdifficulties for effective assessment.2. A patient's description of pain is important for accurate pain assessment.73. A patient will honestly describe his or her pain, therefore, responses can betaken at face value.Purposes of the StudyThis descriptive exploratory study employs secondary analysis (using contentanalysis methods) of an existing database to address cancer patients' own verbaldescriptions of pain. The goal is to improve understanding of cancer patients' use ofpain descriptors and phrases as an essential basis for more accurate assessment of pain.The purposes of this study are to determine:1. The pain descriptors used by cancer patients with pain.2. The similes and metaphors used by cancer patients to describe their pain.3. The relationship of selected demographic variables (age, level of education,mother tongue, and sex) and present pain intensity with the frequency andtype of phrases containing pain descriptors.Definition of TermsFor the purposes of this study, the following terms are defined:Pain:^"an unpleasant sensory and emotional [affective] experienceassociated with actual or potential tissue damage, or described interms of such damage" (International Association for the Study ofPain 1979).Cancer pain:Cancer patient:Pain descriptor:pain due to cancer or its treatment.an adult (>17 years) who has a diagnosis of cancer and is receivingtreatment for cancer cure, control, or palliation.a word or words used to describe any of the various qualities ofpain as outlined by the Multidimensional Model of Pain (MMP)8Significance of the StudyNurses may not have a full appreciation of their patients' pain and a frequentlyused pain assessment tool (the MPQ) may be problematic. As a consequence, cancerpatients may not be obtaining as much relief from pain as possible. Identification ofwords meaningful to cancer patients with pain could prompt the development of amore useful pain assessment tool. A new assessment tool could provide a basis for amore accurate and practical assessment of pain.Conceptual Model of PainA conceptual model, the Multidimensional Model of Pain (MMP), originallydescribed by Ahles, Blanchard, and Ruckdeschel (1983) was used in this study. Ahlesand colleagues compared two groups of cancer patients, one group with pain and theother group without pain. As a result of their study, Ahles and colleagues (1983)conceptualized cancer pain as a multidimensional phenomenon consisting ofinterrelated components. In later work, McGuire (1987) added a socioculturalcomponent "to present a more complete picture of factors affecting the perception andinterpretation of pain" (p. 5). In the Multidimensional Model, as outlined by Ahles andcolleagues (1983) and modified by McGuire (1987), pain consists of the followingcomponents:1. Affective — emotional reactions to pain such as anxiety or depression;2. Behavioural — observable behaviours such as a patient's level of activity ordegree of medication intake;3. Cognitive — the way in which pain affects an individual's thoughts or theway an individual views himself or herself;4. Physiological — the organic cause of pain, physiological pain co-factors,and the pattern and duration of pain;95. Sensory — the location, intensity, and quality of pain; and6. Sociocultural — socioeconomic, cultural, demographic, ethnic and spiritualfactors.The Multidimensional Model of Pain was chosen because it addresses many causes ofpain and factors that can influence it. The interrelationship of the model's componentsis evident. For example, a depressed patient (Affective) may report a higher level ofpain intensity (Sensory) and may fear (Affective) that the pain indicates his or hercondition is worsening (Cognitive). In addition, because the model includes thecomponents of pain addressed in the Instrument study (Affective & Sensory), and themodel examines many more dimensions of pain, it provided a larger framework withwhich to explore the descriptors and phrases cancer patients use when they are askedto describe their pain.SummaryPain is an acknowledged problem for patients with cancer. While nurses areaware of the problem, nurses' perceptions of patients' pain and patients' subjectivereports of pain may not be congruent. This mismatch of nurses' perceptions andpatients' experiences may have a negative impact on pain assessment and management.The McGill Pain Questionnaire was developed to provide an accurate method toassess pain; however, its usefulness has been questioned due to its lengthyadministration and the origin of its pain descriptors. It is important, therefore, toincorporate patients' reports of pain during pain assessment to ensure that subjectiveexperiences of pain are heeded during decision-making for pain management. Byusing knowledge of patients' spontaneous use of pain descriptors, patients' subjectiveexperiences can be recognized and may aid in more accurate and practical painassessment conducted by nurses and other health-care providers. For this study, a10descriptive exploratory approach and the Multidimensional Model of Pain were usedto guide analysis of existing data from the Instrument study for occurrences ofspontaneous pain descriptors, similes, and metaphors used by cancer patientsexperiencing pain.This report is comprised of five chapters. Literature that explores thecomponents of the MMP, and studies that detail pain descriptors are examined inChapter 2. The study design, and data collection and analysis are detailed inChapter 3. Chapter 4 examines sample characteristics and details the words andphrases participants used to describe their pain. The study findings are summarized inChapter 5 which also presents recommendations for nursing education, practice, andresearch.11CHAPTER 2: THEORETICAL BACKGROUNDAND LITERATURE REVIEWThis chapter provides an overview of studies that support the components ofthe Multidimensional Model of Pain (MMP) and demonstrate the model'sconceptualization of the cancer patient's experience of pain. Further, the chaptercritiques the development and use of the McGill Pain Questionnaire (MPQ) andhighlights difficulties associated with its use with cancer patients. Studies of the MPQwhere cancer patients select pain descriptors to describe their pain and studies wherecancer patients use their own words to describe pain are presented to provide anassessment of research to date, and to place this study in context with related work inthe field.Multidimensional Model of PainA number of authors and researchers offer support for the individualdimensions of the Multidimensional Model of Pain. Briefly, the Affective Dimensionof pain in cancer patients is supported by Breitbart (1989), a psychiatrist, who statesthat pain has an impact on an individual emotionally, and psychological factors such asanxiety and depression can influence pain. McGuire and Scheidler (1990) state, "it isquite clear from the clinical perspective that there is an affective component to the painexperience and that it influences the person's response to pain" (p. 394). Ahles,Blanchard, and Ruckdeschel (1983) demonstrated increased scores on measures ofdepression in their sample of cancer patients with pain. In contrast, Cleeland (1984),found no significant difference in pain intensity measures between cancer patients whowere either labeled as depressed or not depressed. However, patients labeled asdepressed reported that pain interfered more with their mood than did patients whowere not labeled as depressed. Dorrepaal, Aaronson, and Van Dam (1989) found12two-thirds of the cancer patients in their sample reported that pain influenced theirmood (e.g., irritability, anger, sadness and depression, and apathy).The Behavioural Dimension of pain is evident when cancer patients spendless time walking or standing and more time lying down or guarding areas of theirbody to reduce pain (Allies et al., 1983; Donovan, 1985; Dorrepaal et al., 1989).Cancer patients rating their pain intensity as a 5 or higher on a 0 - 10 scale reportedthat pain interferes with their daily activities (Cleeland, 1984).In the Cognitive Dimension, Ahles and colleagues (1983) demonstrated thatcancer patients may believe pain to be a sign of worsening disease or the harbinger of apoor prognosis. Spiegel and Bloom (1983) found in their sample of patients withbreast cancer that belief in pain as an indication of worsening disease was significantlycorrelated with reports of more pain.It has been previously noted, in the Physiological Dimension, that pain isfrequently caused by cancer or its treatment. In addition, factors that enhance painhave been identified including fatigue, insomnia, and lack of sleep (McGuire, 1987).Melzack (1975) interprets the pattern and duration of pain as physiological. Thepattern may be described as brief, momentary or transient; rhythmic, periodic orintermittent; and continuous, steady or constant (Melzack, 1975). The duration ofpain is often categorized further as either acute or chronic. Acute pain subsides ashealing takes place and is usually of short duration (less than six months), whilechronic pain is usually prolonged (greater than six months) (McCaffery & Beebe,1989). Loeser and Cousins (1990) divide chronic pain into several categories, one ofwhich is chronic pain associated with cancer: "We assert that this is really acute painwhich is present on a long-term basis because of continued tissue damage" (p. 210).13In the Sensory Dimension, cancer patients may experience pain at two ormore sites. Pain intensity (e.g., none, mild, moderate, severe, intolerable, excruciating,bad, intense, etc.) is due in part to the location of pain and an individual's subjectivepain threshold (influenced by comfort, mood, medication, etc.) (McGuire, 1987;McGuire & Scheidler, 1990). Pain quality is described as what the pain feels like(pulsing, radiating, penetrating, burning, aching, etc.) (McGuire & Scheidler, 1990).McGuire (1987) added the Sociocultural Dimension to the MultidimensionalModel of Pain. She found that females and non-Caucasians often obtain a lower scoreon the McGill Pain Questionnaire than males and Caucasians (McGuire, 1984).Cleeland, Ladinsky, Serlin, and Nugyen (1988) found that despite similar cancers,Vietnamese patients reported more severe pain when compared to American patients(likely attributable to the lack of availability of potent opioids in Vietnam at the time ofthe study — highlighting a socioeconomic factor). Altogether, these findings endorseuse of the Multidimensional Model of Pain with cancer patients. (Throughout thisstudy, all MMP citations refer to Ahles, Blanchard, and Ruckdeschel [1983] andMcGuire [1987].)Pain Descriptor StudiesCritique of the McGill Pain Questionnaire As previously stated, the McGill Pain Questionnaire (Melzack, 1975) is a painassessment tool that was developed in the early 1970s and is used for the assessmentof pain. The MPQ was designed as a four-part tool. One part offers a diagram of thebody to indicate the location of pain. A second part contains a rating scale thatregisters the present intensity of pain. A third part has a grouping of wordsdescribing the pattern of pain. A fourth part has another grouping of words, divided14into 20 subclasses of pain descriptors, classified as sensory, affective, evaluative, andmiscellaneous (see Appendix A, p. 79).When the MPQ is used to assess pain, patients must select words that bestdescribe their pain, choosing only one word from each subclass within the four classes.Subclasses contain what some authors call qualitatively similar words (Deschampset al., 1988) arranged in rank order of pain intensity as determined by subjects inMelzack's original work with the MPQ (1975). To illustrate, Table 1 highlightssubclasses 11 - 15 that comprise the Affective Class of the MPQ.Table 1Affective Class of the MPQ AFFECTIVESUB-CLASSDESCRIPTOR11 TiringExhaustin12 SickeningSuffocatin13 FearfulFrightfulTerri^in14 PunishingGruellingCruelViciousKillin15 WretchedBlindinReference: Melzack (1975). (See also Appendix A, p. 79.)Over the years, the validity and reliability of the MPQ have been demonstratedin assessing pain in patients with cancer (Dubuisson & Melzack, 1976; Graham, Bond,15Gerkovich, & Cook, 1980; McGuire, 1984; Zimmerman et al., 1987); however, thetool has not escaped criticism. Two of these criticisms, the length of theadministration time and the nature of the pain descriptors used, will be discussedfurther. First, while researchers have demonstrated the reliability and validity of theMPQ, these researchers also state that administration of the questionnaire can take 20minutes or longer to complete (McGuire, 1984; Melzack, 1975; Zimmerman et al.,1987). Considering the need for repeated assessments, the administration time of theMPQ affects how useful the tool is for nurses due to time constraints (Zimmermanet al., 1987), possibly delaying medication administration. Also, it may be affected bythe attention span of patients (McGuire, 1984).The second criticism has been directed toward the pain descriptors used in theMPQ. Specifically, patients find some of the words difficult to understand(Deschamps et al., 1988; Graham et al., 1980; McGuire, 1984). This difficulty withthe descriptors in the MPQ may be due to the tool's development. Words in theMPQ, which a patient selects to describe his or her pain, were originally gathered frompsychological and medical literature and were then tested on subjects such as healthyuniversity students (Melzack & Torgerson, 1971). Examination of the original testpopulations reveals that subjects were predominantly young, well-educated males notexperiencing pain — potentially explaining why "the choice of some paindescriptors...are either difficult to comprehend and/or not spontaneously used bypatients with chronic pain" (Deschamps et al., 1988, p. 135). The only notation ofethnicity is that the students came from a variety of cultural and socioeconomicbackgrounds. A detailed look at studies reporting the evaluation of the MPQ's usewith cancer patients provides further information regarding the tool's usefulness.16MPQ Pain Descriptor Studies and Cancer PatientsA number of researchers have attempted to determine the descriptors cancerpatients most often use to illustrate their pain. The majority of these researchers havepresented their patients with the pain descriptors contained in the MPQ, noting thatcertain descriptors are selected approximately 30% - 40% of the time by cancerpatients (Dubuisson & Melzack, 1976; Graham et al., 1980; McGuire, 1984;Nicholson, McGuire, & Maurer, 1988; Zimmerman et al., 1987). However, most ofthe MPQ descriptors (approximately 60% - 70%) are either selected with a lowfrequency rate or are not selected by cancer patients. Other researchers using surveytechniques have attempted to determine what pain descriptors are used by cancerpatients by asking them to describe their pain using their own words.A review of studies that investigate cancer patients' use of pain descriptorsprovides a summary of the research already done in this area. Studies of cancerpatients and their selection of pain descriptors from the MPQ will be examined first.Then, reports of cancer patients' use of spontaneous pain descriptors will beexamined. Together these studies indicate what is already known about paindescriptors selected from MPQ lists or used spontaneously by cancer patients.Classification of Disease by Pain DescriptorMelzack, principle developer of the McGill Pain Questionnaire, along withfellow researcher Dubuisson, wanted "to determine whether patients with a given painsyndrome tend to choose a set of words which is distinctive enough to separate themfrom other groups of patients" (1976, p. 482). Patients were categorized into twogroups based on medical diagnosis. The patients then selected pain descriptors fromthe MPQ to describe their pain.17Dubuisson and Melzack's study consisted of 95 patients (8 were identified ashaving metastatic cancer). Patients had one of eight pain syndromes: menstrual,toothache, cancer, arthritis, labour, herpes, phantom, or disc disease. The results ofthe study indicated that patients with the same disease or pain syndrome tended tochoose the same pain descriptors (with 77% accuracy) to communicate their painwhen compared with patients with different types of pain. The researchers also foundthat there were seven MPQ pain descriptors (shooting, sharp, gnawing, burning,heavy, exhausting, and unbearable) the eight cancer patients chose with a 50%frequency. A sample of eight cancer patients is not sufficient, however, to determineprecisely what words cancer patients use to describe their pain.Selection of MPQ Pain Descriptors by Cancer PatientsGraham, Bond, Gerkovich, and Cook (1980) evaluated the MPQ as a measureof pain associated with cancer in a sample of 36 volunteers who were outpatients withcancer (24 female and 12 male, with a mean age of 52.7 years). Of the seven wordschosen by patients in the Dubuisson and Melzack (1976) study, five were chosen witha high frequency by at least one-third of the patients in this sample; the two exceptionswere the pain descriptors "unbearable" and "heavy." Besides the pain descriptorschosen by Dubuisson and Melzack's patients, the patients in Graham and colleagues'study selected additional MPQ pain descriptors (nagging, sickening, tender, stabbing,and throbbing) with a frequency range of 34.3% to 42.9%. The data from Grahamand colleagues' study point to a wider range of pain descriptors selected by cancerpatients in comparison to the data from the Dubuisson and Melzack study.McGuire (1984) used the MPQ as a measure of pain in cancer inpatients andcompared her findings with the studies of, among others, Dubuisson and Melzack(1976) and Graham and colleagues (1980). McGuire questioned whether cancer18patients in her study would select the same words to describe pain as cancer patients inother studies. Her convenience sample consisted of 24 patients (inpatients withcancer, 10 male and 14 female, 9 Caucasians and 15 non-Caucasians, with a mean ageof 51 years and a mean education of 10 years). McGuire's patients selected MPQ paindescriptors (unbearable, intense, tiring, tight, terrifying, tender, shooting, heavy,sickening, exhausting, stabbing, and torturing) with 29% to 54% frequency. AlthoughMcGuire reported the information appropriate to her sociocultural dimension, she didnot examine the effect of her patients' ethnicity, age, education, or sex on the selectionof pain descriptors. Although McGuire's patients selected a wide range of descriptors,many of the descriptors they selected match those selected by the patients in theprevious two studies.Zimmerman, Duncan, Pozehl, and Schmitz (1987) also wanted to ascertain theMPQ pain descriptors used by cancer patients to describe their pain. Theseresearchers' convenience sample consisted of 40 inpatients with a variety of cancers(24 female and 16 male; 39 Caucasian and 1 non-Caucasian; 32 were married; with amean age of 60 years, and a mean education of 12 years). The MPQ descriptors(nauseating, tight, troublesome, tugging, shooting, tender, miserable, sharp, sickening,exhausting, aching, throbbing, and tiring) were selected with 25% to 48% frequency.These researchers also did not examine the effect of their patients' ethnicity, age,education or sex on the selection of pain descriptors. As with McGuire's patients,these patients selected a wide range of MPQ descriptors including many of the samedescriptors selected by the patients in the previous studies.Another study was conducted by Nicholson, McGuire, and Maurer (1988) whowanted to describe the pain perception and response of 20 outpatients (13 male and 7female; 13 Black, 5 Caucasian and 2 Hispanic; with a mean age of 56.2 years). These19patients had a variety of head and neck cancers and had reported pain to their nurses.The most frequent MPQ pain descriptors (itchy, aching, throbbing, tender, sharp andmiserable) were selected with 40% to 60% frequency. These researchers again did notexamine the effect of their patients' ethnicity, age, or sex on the selection of paindescriptors (education was not reported). A new finding was the high frequency of40% for "itchy" which the researchers attributed to a radiotherapy side effect commonamong head and neck cancer patients. While this MPQ pain descriptor study had asmaller range of descriptors selected by the patients, with the exception of "itchy,"most of the descriptors are the same as those selected by patients in the other studies.The authors of the MPQ pain descriptor studies reviewed here interpreted theirresults in favour of the MPQ. All the authors recommended the MPQ as ameasurement tool for use with cancer pain. None of the authors, however, consideredthat pain descriptors may be added to or deleted from the MPQ to improve its usewith cancer patients. They also did not consider variables such as ethnicity, age,education, or sex and how these variables might affect the selection of paindescriptors.MPQ Descriptors and Cancer Patients — Meta -analysisWilkie, Savedra, Holzemer, Tesler, and Paul (1990) provide a summary of howcancer patients use MPQ pain descriptors in the form of a recent meta-analysis. Oneaspect about the studies they examined was the notation of pain descriptors selectedby patients experiencing pain. Table 2 (see p. 20) is based on information provided inWilkie and colleagues' report and summarizes the results of their meta-analysis ofMPQ pain descriptor-cancer patient studies.The collection of studies included nine with cancer patients for a total of 340cancer patients. (Of the MPQ pain descriptor-cancer patient studies reviewed earlier,20all except Nicholson et al. [1988] were included in the meta-analysis.) Table 2demonstrates a clustering of 18 descriptors in the 22% to 54% range when cancerpatients select MPQ pain descriptors to describe their pain. It is evident in the MPQpain descriptor-cancer patient studies that there is a cluster of MPQ descriptors thatthese patients select to describe pain. However, it is equally important to note thatcancer patients either failed to select or had low selection percentages (<22%) for themajority of pain descriptors in the MPQ (notable by their absence in Table 2).Table 2Selection Frequency of MPQ Pain Descriptors by Cancer Patients*MPQ PAINDESCRIP-TORMPQCLASS %MPQ PAINDESCRIP-TORMPQCLASS %Tiring A - 11 42.0 Throbbing S - 1 41.0Exhausting A - 11 51.0 Shooting S - 2 33.0Sickening A - 12 39.0 Stabbing S - 3 41.0Terrifying A - 13 38.0 Sharp S - 4 38.0Intense E - 16 29.0 Gnawing S - 5 33.0Unbearable E - 16 33.0 Burning S - 7 22.0Tight M - 18 30.0 Aching S - 9 40.0Nagging M - 20 34.0 Heavy S - 9 22.0Torturing M - 20 54.0 Tender S - 10 34.0Note. Under the column-heading, MPQ Class, S = Sensory, A = Affective, E = Evaluative andM = Miscellaneous. Numbers represent subclasses within each MPQ class. For example, theSensory class has ten subclasses listed as 1-10 in the MPQ (S - 1 to S - 10 in the table). TheAffective class has five subclasses listed as 11-15 in the MPQ (M - 11 to M - 15 in the table).(See also Table 1, p. 14.)* Reference: Wilkie, Savedra, Holzemer, Tesler, & Paul (1990).There are many more descriptors in the MPQ that cancer patients either do not selector select with a very low frequency. As it is important to know what descriptors21cancer patients spontaneously use to describe their pain, two articles that examine thisaspect of pain assessment were reviewed.Spontaneous Pain Descriptor Studies and Cancer PatientsA few researchers have attempted to ascertain what pain descriptors cancerpatients use and have looked beyond the descriptors found in the MPQ. By limitingpatients' selection of pain descriptors to those contained in the MPQ, informationabout additional descriptors cannot be obtained. Asking patients to use their ownwords to describe pain may lead to the discovery of new descriptors that could beincluded in a MPQ-type list designed for use with cancer patients. Those paindescriptors that could be eliminated from an MPQ-type list might also become evidentby noting those descriptors consistently omitted by cancer patients.Bressler, Hange, and McGuire (1986) attempted to characterize the painexperience of a group of cancer outpatients based on demographic and clinicalvariables and patients' subjective reports. One result of their study was a list of paindescriptors. Their purposive sample consisted of 58 outpatients with cancer (36 maleand 22 female; 30 Caucasian, 26 Black, and 2 Hispanic, with a mean age of 58 years,and a mean education of 10.8 years). Patients were asked to describe how their painfelt to them using their own words. The researchers reported that 12% - 40% of thepatients used the following pain descriptors: cramplike, pressure/numbness,movement of pain, burning/throbbing, intensity of pain, pattern of pain, sharp/knifelikeand aching/sore/tender.Certain descriptor percentages were reported by the researchers as MPQcategories rather than using the actual descriptors given by the patients (e.g., intensityof pain versus strong, hard, severe, etc.). While some of the pain descriptors are thesame as, or are similar to, pain descriptors found in the MPQ, some of the percentages22reported by these researchers are artificially inflated. The combination of separatecategories results with more impressive percentages (e.g., aching, sore, and tender fora cumulative 40%) versus the percentages that would have resulted had the categoriesbeen reported individually. As a result, information on the frequency of use ofindividual descriptors is not available. As with the MPQ descriptor studies, theresearchers in this study reported but did not examine the effect of their patients'ethnicity, age, education, or sex on the use of pain descriptors.Tearnan and Cleeland (1990) conducted a similar study where they surveyedcancer patients who had experienced pain within the previous month. Their sampleconsisted of all inpatients and outpatients (with breast, prostate, colon or rectum, orone of three gynecological tumours) in one hospital's oncology service. A total of 451eligible patients responded (with a mean age of 58 years) including 67% of eligibleinpatients (non-response was due to busy test schedules, fatigue, or declining toparticipate). Outpatients were reported to have a similar response rate. The purposeof the study was "to provide a descriptive analysis of words cancer patients use todescribe their pain" (p. 229). The patients were asked to describe their pain in theirown words. Tearnan and Cleeland (1990) found that the patients provided a total of129 distinct words to describe pain. Additionally, although a wide range of wordswere used by all the patients, most patients limited their pain description to a fewwords (each patient used a mean of 1.8 words). Sixty-seven percent of the 129 paindescriptors were used by 2.8% - 11.5% of the patients. The most frequently selecteddescriptors were: shooting, pressure, hurting, continual, dull, throbbing, burning,stabbing, sharp, dull-ache, and ache. The richness of Tearnan and Cleeland's (1990)data is diluted, however, because not all the 129 pain descriptors they identified are23reported in their article. Again, these researchers did not examine the effect of theirpatients' ethnicity, age, education, or sex on the use of pain descriptors.As with Bressler and colleagues' (1986) study, most of the words used byTearnan and Cleeland's (1990) patients are identical or similar to pain descriptorsfound in the MPQ. The number of spontaneous descriptors and their frequency of use,however, are low in comparison to the rates where patients are presented with theMPQ list of descriptors. While the studies by Bressler et al. (1986) and Tearnan andCleeland (1990), have produced lists of descriptors that are similar to those containedin the MPQ, the overall number of words and their frequency of selection is low inthese two studies, possibly because patients had to think of words themselves versushaving the opportunity to select words from a list.Just as the authors of the MPQ pain descriptor studies interpreted their resultsas supporting the MPQ, so did the authors of the spontaneous pain descriptor studies(due to the higher frequency of use of MPQ pain descriptors). The spontaneous paindescriptor authors also recommended the MPQ as a measurement tool for use withcancer pain. They, too, did not consider that pain descriptors may be added to ordeleted from the MPQ to improve it for use with cancer patients. They also did notconsider what effect their patients' ethnicity, age, education, or sex might have on theselection of pain descriptors. Table 3 (see p. 24) compares the pros and cons of theuse or selection of descriptors by patients.The findings noted in the above MPQ descriptor and spontaneous descriptorstudies suggest that MPQ pain descriptors help cancer patients describe their pain.However, the MPQ contains words whose percentage of use varies widely whenapplied to patients with cancer and it contains many words that may not be applicableto these patients. The question still arises whether the MPQ may lack certain words24cancer patients might use to describe their pain. Recall that Dubuisson and Melzack(1976) culled the MPQ descriptors from the literature rather than fromTable 3Comparative Use of Spontaneous and MPQ Selected Pain Descriptorsby Patients with CancerSPONTANEOUS MPQ SELECTEDPro • Pain descriptors arethe same as or similarto those found in the• Frequency of paindescriptors selected ishigh (22% - 54%)MPQCon • Frequency of paindescriptors used isvery low (2.8% -11.5%)• Many MPQ paindescriptors are choseninfrequently or not atall• Not known if allrelevant paindescriptors arecontained in the MPQor if irrelevant wordscan be deletedqualitative research with patients with pain. Since then, only two studies haveattempted to determine spontaneous cancer pain descriptors. Further study seemswarranted in this under-researched area.Summary and ConclusionsSupport for the Multidimensional Model of Pain (MMP) has beendemonstrated by noting the work of authors who have substantiated its variousdimensions. Because it is intuitively appealing and well documented, the MMPdimensions form the theoretical framework for this study and were used to guide theanalysis of pain descriptors gathered from the interview transcripts (as reported inChapter 4).25Patients respond favourably to lists containing pain descriptors (such as thatfound in the MPQ) as an aid to describing their pain (McGuire, 1984; Melzack, 1975).However, the review of the literature demonstrates that there are many words in theMPQ which cancer patients do not use (see Table 3, p. 24). This may be due to thesource of the MPQ descriptors, which originated in the literature and were first testedon healthy and well-educated subjects (Melzack & Torgerson, 1971), rather thanoriginating with cancer patients' own words when they are experiencing pain.Searching through words that do not match a patient's experience or words withwhich a patient is not familiar may contribute to lengthy administration times for theMPQ. In turn, lengthy administration may make it impractical for nurses to use theMPQ and could contribute to patient fatigue or delay the initiation of pain reliefmeasures.It is noteworthy that many MPQ pain descriptors were not used by cancerpatients in the studies reviewed. A new pain assessment tool that revises the paindescriptor section of the MPQ could be more precise and have a shorter administrationtime. Therefore, this study conducted further research into the spontaneous paindescriptors used by a group of cancer patients. This allows for assessment of newwords that may be included in a MPQ-type list of pain descriptors and addsinformation regarding nonessential descriptors that may be deleted. Ultimately, thisresearch may result in the development of an improved pain assessment tool thatcontains only the pain descriptors that have meaning for patients with cancer.Data collection, study design, ethical considerations, and data analysis aredetailed in Chapter 3.26CHAPTER 3: METHODOLOGYThis chapter discusses the methods used to examine the words and phrasescancer patients use to describe their pain. The Instrument study is discussed first toprovide context, including its sample selection and data collection. Then, the design,procedure, ethical considerations, and data analysis specific to this study will beoutlined. Finally, validity and reliability will be addressed.Instrument StudyData for this study were originally collected for a larger study entitled, Study ofpain in the British Columbia cancer population, phase II: Development of aninstrument evaluating cancer pain (Deschamps, Coldman, Bradley, & Baker, 1991),referred to here as the Instrument study. The interviews were transcribed but the datawere not analyzed because the study ended when further funding was not obtained.Because I was the person who conducted the interviews, I have received permission touse data for this study (see Appendix B, p. 80).As noted in Chapter 1 (see p. 6), nursing research in pain at the local provincialcancer treatment facility was focused in two areas: (a) determination of the incidenceand prevalence of pain in the BC cancer patient population and (b) the development ofa pain assessment tool. This study is concerned with the second area of the nursingresearch programme (i.e., the Instrument study). The first step of the Instrument studywas designed to identify pain descriptors spontaneously used by patients experiencingpain related to cancer or its treatment. The pain descriptors were to be classified intogroups based on the categories of the McGill Pain Questionnaire (i.e., sensory,affective, and evaluative). Its second step was designed to determine the intensitylevel of each pain descriptor and validate the classification scheme obtained in the firststep. Patients were to be instructed to assign intensity values to each of the pain27descriptors and to indicate which descriptors corresponded to the pain they werecurrently experiencing. Patients would then be instructed to state whether they agreeif pain descriptors in individual categories reflected a similar type of pain. Finally,patients were to be instructed to rank the pain descriptors in ascending order ofintensity.Data CollectionI conducted 39 interviews for the Instrument study and 31 have beentranscribed. It was not possible to transcribe all interviews due to poor sound quality(background noise, e.g., fire-alarm testing) and occasional technical difficulties withthe audio equipment. Along with the 31 transcribed interviews, other interviews wereconducted by a number of undergraduate nursing students participating in a researchcourse exercise. Unfortunately, the quality of those interviews is variable. Thestudents used a slightly different interview guide, and due to the large number ofinterviewers, inter-rater reliability is a confounding variable. The student interviewshave not been transcribed and were not used in this analysis.Participants were asked to use their own words to describe what their painfeels like. Most interviews lasted approximately one-half hour with none exceedingone hour. The interviews were conducted over a ten-week period during February,March, and April, 1991. The pain descriptors were collected in audiotaped interviewsby asking cancer patients to describe their pain with the aid of an interview guide (seeAppendix C, p. 81). Face and content validity of the interview guide were determinedby a Nurse-Research Consultant and an Oncology Clinical Nurse Specialist. Patientdemographics were recorded at the time of the interview on a data collection guide(see Appendix D, p. 82).Investigator Side0^1^2^3^4^5^6^7^8^9^10I ••••I••••I••••I••••I••••I••••I••••I••••I••••I••••I•INo painV4->IWorst pain everParticipant Side28In addition to the interviews, participants were asked to score the intensity oftheir pain using a Visual Analog Scale. The Visual Analog Scale used in this studyconsisted of an instrument with a sliding indicator. On one side of the instrument, thewords "no pain" and "worst pain ever" were printed on opposite ends. The other sideof the instrument was marked in centimetres (cm) with 0 cm corresponding to "nopain" and 10 cm corresponding to "worst pain ever." Participants were asked to movethe sliding indicator to a point between "no pain" and "worst pain ever" that indicatedtheir pain intensity. The instrument was then turned over to reveal placement of thesliding indicator on the centimetre scale for recording of the pain intensity. VisualAnalog Scales are usually well understood by patients, are sensitive to variations inpain intensity, and represent a ratio measurement scale enabling quantitativeexpression of pain intensity (Deschamps et al, 1988).Figure 1Visual Analoq ScaleDesignA descriptive exploratory design was used in this study to conduct a secondaryanalysis of data from the first step of the Instrument study. According to Woods andMitchell (1988), exploratory studies emphasize identification of factors related to aphenomenon of interest; "these studies frequently address health-related phenomena29such as pain, anxiety, or well-being" (p. 150). Accordingly, the focus of this study isthe identification of pain descriptors and phrases used by patients in the Instrumentstudy. Further, this study explored the relationship of the pain descriptors and phrasesto selected demographic variables (age, level of education, mother tongue, and sex)and present pain intensity. Analyzing the relationship of pain descriptor frequencieswith such variables could provide useful information between them and the paindescriptors used by patients with cancer. Therefore, data were analyzed using contentanalysis techniques and statistical tests.SampleParticipants in the Instrument study were recruited from a cancer treatmentfacility providing active treatment for people with cancer in British Columbia and theYukon Territory. The facility has four inpatient hospital units (one of which includesan outpatient chemotherapy programme), large ambulatory clinics (one of whichincludes a pain clinic), and a radiotherapy unit. Participants were recruited from thehospital units and the pain clinic.Selection CriteriaParticipants were selected using the following criteria:1. Currently experiencing pain related to cancer or its treatment.2. Older than 17 years of age.3. Able to speak English fluently.4. Had the willingness, energy, and time to meet with me.Selection Procedure 1. Patients were enrolled from among the four hospital units and the painclinic.302. Suitability of patients was determined by the Nursing Unit Manager of eachunit (or delegate, i.e., the In-Charge nurse) and me, using Selection Criteria1-4 (noted above).3. Patients were approached by the Nursing Unit Manager (or delegate) todetermine if the patient was willing to participate in the study.4. I approached patients after I was informed of their interest. At this firstinteraction, I explained the purpose of the study (as per the Instrumentstudy), answered questions, and then obtained written consent later thatsame day if the patient agreed to participate.5. Interviews were conducted at a time convenient for the patients.Sixteen women and fifteen men were interviewed. There average age was 57.2years and they had an average of 11.26 years of educational preparation. Theparticipants had a wide variety of primary cancers and they had a number of differentmother tongues, though all spoke English. These participants also reported that theywere currently experiencing pain with an intensity of 3.95 as rated on a 0 - 10 VisualAnalog Scale. The sample is more fully described in Chapter 4.SettingParticipants on the hospital units were interviewed in their rooms (all of whichwere one or two bed units), and participants in the pain clinic were interviewed inexamination rooms. All interviews were conducted in private. Hospitalized patientswere interviewed while sitting or lying in their beds and pain clinic patients wereinterviewed while sitting in chairs in the examining room. Occasionally, familymembers were present during the interview, however, they did not participate in thediscussion. Interviews of hospital patients were occasionally interrupted due tomedication rounds or porters arriving to escort patients for tests. Also, there were31frequent intrusions due to fire alarm testing that was performed over several months,coinciding with the period that the interviews were conducted.EthicsThe Instrument study was granted approval by the Clinical InvestigationsCommittee at the provincial cancer treatment facility (see Appendix E, p. 83) and bythe Behavioural Sciences Screening Committee for Research and Other StudiesInvolving Human Subjects at the University of British Columbia (see Appendices F& G, pp. 84 & 85).• A letter explaining the Instrument study was given to the participant in themorning (see Appendix H, p. 86), and if he or she agreed to participate, I would returnin the afternoon for the participant to sign the consent form (see Appendix I, p. 87).Prior to consenting, participants were informed that: (a) declining to participate orwithdrawing participation in the Instrument study would not affect their present orfuture treatment; (b) participation in the study would not result in any expected ill-effects either to themselves or anyone else; and (c) the study would not benefit themdirectly but it was hoped that information collected could be used in the future to helpwith the assessment of pain. Only one patient declined participation.To protect the confidentiality of the participants, unique study numbers wereused to code the data for each participant and all personal identifying information wasremoved from the data. No information which could identify individuals is discussedin this report. Only the content experts consulted, members of the thesis committee,and I had access to the data used for this study. Upon completion of this analysis alldata were returned to the principal investigator of the Instrument study for safekeeping according to the ethical stipulations outlined in the Instrument study.32Data AnalysisThe interviews were analyzed using content analysis methods to identifyspecific pain descriptors and contextual phrases describing pain. The relationshipbetween the frequency of pain descriptor phrases and patient variables noted earlierwas analyzed. In addition, the pain descriptors were categorized according to type(dimensions from the Multidimensional Model of Pain). Contextual phrases (similesand metaphors) describing pain were also identified and categorized according to type.Content Analysis Content analysis is "essentially a coding operation," and may be applied tovirtually any form of communication (Babbie, 1986, p. 271). It is described as theapplication of quantitative research methods to qualitative data (Catanzaro, 1988). Itis particularly useful in instances where the participant's own language is crucial to aninvestigation (Holsti, 1969; Polit & Hungler, 1991). Content analysis is usuallydescribed as consisting of two forms: latent or manifest content analysis. Latentcontent analysis codes the investigator's impression or interpretation of the underlyingmeaning of a communication, whereas manifest content analysis codes the visible orsurface content (Babbie, 1986). Concepts emerge from the data when latent contentanalysis is used. Conversely, data are coded or classified according to a conceptualframework with manifest content analysis (Babbie, 1986; Holsti, 1969). This studyapplied both methods of content analysis in coding the pain descriptors and phrasesused by cancer patients according to dimensions of the Multidimensional Model ofPain (MMP) (surface content) and similes, metaphors, and expressions of difficultywhen describing pain (items which emerged from the data).Objectivity was achieved by conducting the analysis on the basis of explicitlyformulated rules (Polit & Hungler, 1991). The MMP provided structure and helped33focus the categorization of pain language during the data analysis. For example,words and phrases that address the location, intensity, or quality of pain were coded asSensory (after the Sensory Dimension of the 1VIMP). Use of manifest content analysismethods enables researchers to make "replicable and valid inferences" about datawhen empirical and statistical methods are applied to data (Catanzaro, 1988, p. 437).Steps Involved in Content Analysis Wilson's (1985) three basic steps involved in a content analysis were used:1. The unit of analysis was identified as either a whole or a partial response(broken down into words, phrases, or sentences). Words and phrasesaddressing the description of pain were the units analyzed in this study.2. The development of categories occurred next. Classifications were set upin advance. Analysis involved reading through the data and coding unitsinto existing categories. Some categories for analysis were based on thedimensions of the Multidimensional Model of Pain. Other categoriesemerged from the data (similes, metaphors, and expressions of difficulty).Results were reported as frequencies of units among the various categories.3. Development of the coding rationale. The process of coding required thecoder to decide in which category a coding unit belongs. Therefore,categories were clearly and sufficiently defined. To accomplish this, thefollowing mechanisms were used.Validity and ReliabilityValidity.In conducting a valid content analysis, the researcher's ability to address thefollowing points is important (Wilson, 1985):341. Develop a rationale for and define the categories. "Rules [that] embodythe criteria for processing the content...must be as specific and complete aspossible" (Catanzaro, 1988, p. 438).2. Show how the categories are appropriate to the data and how the data canbe coded into the categories.3. Demonstrate the relevance of the categories to the research question.The Multidimensional Model of Pain provided the categories for the data: thetheoretical dimensions of the model. Pain descriptors were, therefore, coded into theMMP categories of Affective, Behavioural, Cognitive, and so forth. For example,when a patient discussed his or her emotional reactions to pain (such as anxiety ordepression), the data were coded as Affective. Finally, coding data according to thetheoretical dimensions of the model enabled the realization of two of this study'spurposes which were to determine what pain descriptors were used and the frequencyand type of descriptors used (according to dimensions if the MMP).Reliabilitv.Besides validating the categories, rater reliability is a concern when contentanalysis research methods are used (Holsti, 1969). Problems with category reliabilityoccur when the coder is unable to clearly formulate categories in the data (Holsti,1969). As the dimensions of the model were defined and formed the categories for thedata, coding was simplified. The coder determines which dimension is represented byeach descriptor or contextual phrase. For example, the phrase "the stabbing painmakes it difficult for me to sit" is coded as: "stabbing" — Sensory descriptor; "difficultfor me to sit" — Behavioural phrase.Inter-rater reliability occurs when two or more coders code data in aconsistently similar manner. As a test of my ability to identify and classify occurrences35of pain descriptors, content experts in pain management (an Oncology Clinical NurseSpecialist [CNS] and a Palliative Care Clinical Nurse Specialist) coded a randomsample of five unmarked transcripts. We independently highlighted sentences thatcontained a description of pain or a reference to pain description. There was anagreement of 0.75 between the Oncology CNS and me and 0.81 between the PalliativeCNS and me. The highlighted sentences were then each printed onto index cards.The same content experts and I independently sorted the index cards (containing paindescriptors and phrases) according to the dimensions of the MMP. The was anagreement of 0.70 and 0.73 between me and each of the CNSs. The consistentagreement between the CNSs and me indicates that as an analyst, I accuratelyidentified and classified descriptions of pain in the selected transcripts.Statistical Analysis.Once the content analysis was completed, data concerning the frequency ofdescriptors and phrases in each of the MMP dimensions was available. Selecteddemographic variables were then split into subgroups of two (male-female, mothertongue of English-Other, etc.), and comparisons were made between the number ofphrases used by each subgroup. These comparisons helped with the determination ofsignificant relationships and differences between the variables and descriptor phrases.The usual statistical tests to utilize in this situation are t-tests (or the ANOVAfor three or more groups), correlation coefficients, and Chi-Square tests (Shavelson,1988). However, the data in this study violated the assumptions normally associatedwith a t-test (for example, a normally distributed population) because they were basedon a convenience sample. The assumption of a normally distributed population is nottenable and the use of a parametric test such as the t-test, therefore, is not applicable.Instead of the t-test, a non-parametric test was used.36The Mann-Whitney U test is a non-parametric equivalent of the t-test(Shavelson, 1988) and was used here to compare relationships between groupedvariables and pain descriptors (e.g., the difference between older and youngerparticipants and the mean number of affective phrases used, etc.). Unlike the t-test,the Mann-Whitney U test is not based on the actual scores of participants. Instead, itis based on the rank order of raw scores from lowest to highest. The rank scores fortwo groups are used to compute a U value which is compared with values referencedin any statistics text to determine whether or not the U value is significant. In the caseof SYSTAT/SYGRAPH® (described below), the probability value for the U statistic isautomatically generated, obviating the need for comparison with a reference table.Mann-Whitney U tests were used to test for relationships among grouped variablesand the frequency of phrases used to describe pain.Correlations test for relationships among continuous variables. For example,correlations were computed among phrases coded according to the dimensions of theMMP and variables such as level of education. The Chi-Square was used to test fordifferences between nominal variables based on the number of responses that fell intocertain categories. Thus, for example, the Chi-Square test was used to test fordifferences between women and men who were younger or older.Data Analysis Tools.Three personal-computer programs were used to assist with the data analysis.The first program, The Ethnographe, manages computer files associated withqualitative data. The program assists with the mechanical tasks of managing data(such as interview transcripts) (Seidel, Kjolseth, & Seymour, 1988). It generatesprint-outs with line numbers alongside the data. In this case, I hand coded directly onthe print-outs, highlighting words and phrases and placing codes alongside the37highlights (e.g., S for Sensory data, A for Affective data, etc.). Once this coding wascompleted, I entered the codes into The Ethnograph© for each transcript according tothe line numbers associated with each code. The program was then used to generatesummaries of codes (e.g., the occurrence of affective phrases across all interviewtranscripts). All affective phrases were then saved into a separate file, labeled"affect.txt," for example.The second program, RightWriter®, is intended to assess grammar, style,usage and punctuation in word processor files (RightSoft, 1988). However, anadditional feature of the program can generate word frequency lists from wordprocessor files (a feature not available in The Ethnograph©), which allows detectionof every occurrence of a descriptor.After print-outs of the transcripts were coded, The Ethnograph© was used tocompile files containing coded phrases (affect.txt, behay.txt, etc.). Once the contentanalysis identified phrases used by participants to describe pain, word frequency listsbased on those phrases made it possible to accurately determine the frequency ofindividual pain descriptors. Word frequencies were tabulated from these files usingRightWriter® which provided the frequencies for all words in the files. As a result, Imanually culled extraneous words (e.g., "a," "an," "my," "the," etc.). The remainingdescriptors were then scrutinized within transcripts using the search feature of a word-processor. These words were examined in their original context to determine if theywere used as descriptors.The third program, SYSTAT/SYGRAPHO, is a compilation of statistical andgraphing programs designed for use with parametric or non-parametric data.SYSTAT/SYGRAPHO was used to compute Chi-Square tests, Pearson r Correlations,and Mann-Whitney U tests, and to generate illustrative tables and graphs.38SummaryThis study employed a descriptive exploratory design to identify descriptorsand phrases used by selected cancer patients to describe their pain. The data werecollected as part of a larger study to develop an instrument to evaluate cancer pain,using a convenience sample of patients. Thirty-one existing transcribed interviewswere analyzed for occurrences of words and phrases that describe pain. Contentanalysis methods were used to categorize pain descriptors and phrases used by thecancer patients, and the relationships between the descriptors and participant variablesare reported in the next chapter.The coding process borrowed categories (dimensions) from theMultidimensional Model of Pain which provided the theoretical framework for thisstudy. Coding involved sorting pain descriptors and phrases used by participantsaccording to the theoretical dimensions of the model. Reliability was established bythe use of two content experts who independently coded the same randomly selectedtranscripts and data. Although the use of a convenience sample limits thegeneralizability of the results of this study, these patients with cancer were articulate indescribing their pain.Chapter 4 presents characteristics of the sample, lists the pain descriptors usedby the participants, and examines the relationship of participant variables and use ofdescriptor phrases.39CHAPTER 4: FINDINGS AND INTERPRETATIONSThis chapter presents the findings of this study which were analyzed inresponse to the research question: What pain descriptors do cancer patientsspontaneously use to describe their pain? Presentation of the findings andinterpretations are presented in the following sections:1. Characteristics of the study participants.2. Spontaneous descriptors phrases and pain descriptors used by participants.3. Relationship of participant variables with descriptor phrases, expressions ofdifficulty, and similes.Throughout this chapter, these findings are compared with the results of studiesreviewed in Chapter 2.Sample CharacteristicsThe data for this study were gleaned from 31 transcripts and data collectionguides obtained from the Instrument study as discussed in Chapter 3. Data collectedincluded information on sex, type of cancer, age, mother tongue, education, and painintensity. The participants were comprised of 16 women and 15 men. They had awide variety of primary cancers: Breast=3, Cervix=2, Colon/Rectum=4,Esophagus=2, Intra Hepatic Bile Duct=1, Leukemia=1, Lung=5, Lymphoma=1,Melanoma=1, Ovarian=1, Prostate=3, Pyriform Sinus=1, Renal=1, Sarcoma=1,Stomach=1, Thorax=1, Tongue=1, and Unknown Primary=1. Their mean age (allages were calculated from a single reference) was 57.2 years (standard deviation=15.7years) with a range from 21 to 85 years.When asked about their mother tongue, the participants listed a variety oflanguages: Danish=1, Dutch=1, English=23, Finnish=1, French=1, German=2, andTagalog=2; however, all participants spoke fluent English during the interviews. The40participants also had a range of educational preparation, from Grade 5 tobaccalaureate degree, with a mean of 11.26 years in school (standard deviation=2.5years). Immigrants to Canada often provided a Canadian grade equivalent whenprompted for their educational level. All responses were taken at face value andrecorded as stated by the participants.Participants were asked to rate their present pain intensity and the highestintensity of their current pain using a Visual Analog Scale. Pain intensity was scoredfrom 0 - 10. The participants reported an average present pain intensity of 3.95(standard deviation=3.0) which ranged from 0 - 10. They also reported how intensetheir present pain became when it was at its worst. Their average highest painintensity was 8.60 (standard deviation=2.2) and ranged from 2 - 10.Spontaneous Descriptor Phrases and Pain DescriptorsDescriptor Phrases Responding to the question of what their pain feels like, participants used anassortment of descriptive phrases. Phrases were coded on print-outs of the interviewtranscripts using manifest content analysis according to dimensions of theTable 4Manifest Codind of Phrases — Multidimensional Model of Pain (n=31) AFFECT-WEBEHAV-IOURALCOG-NITIVEPHYSIO-LOGICALSENS-ORYMINIMUM 0 0 0 0 2MAXIMUM 13 11 5 12 46MEAN 4.419 2.226 1.419 3.452 15.677SD 3.990 2.617 1.649 3.213 11.125Multidimensional Model of Pain (MMP). Table 4 (see p. 40) summarizes thetabulated results of the manifest coding of phrases according to MMP dimensions.41The Sociocultural Dimension was the only dimension of the MMP not used byany of the participants when they described their pain. As is evident in Table 4, thenumber of responses varied across the dimensions with the lowest mean of 1.42occurring in the Cognitive Dimension (range 0 - 5) and the highest mean of 15.68occurring in the Sensory Dimension (range 2 - 46). The Cognitive and SensoryDimensions represent the opposite extremes of responses with respective ranges of 0 -13 and 2 - 46 responses..In addition to MMP dimensions, similes and metaphors were coded.Expressions of how difficult it is to describe pain ("expressions of difficulty") werenoted throughout the transcripts and were also coded so they could be included in thedata analyses. A summary of the tabulated results appears in Table 5. ParticipantsTable 5Latent Coding of Phrases — Difficulty and Simile (n=31) DIFFI-CULTY SIMILEMINIMUM 0 0Lo • .^14i iLl 7 12MEAN 1.645 2.419SD 1.836 2.861would mention how difficult it was to describe their pain, expressing this an average of1.64 times per interview. Some participants did not report any difficulty, while othersdid so several times. As participants tried to enhance my understanding, they used anaverage of 2.42 similes (phrases that contained like or as) per interview. Amongparticipants, the use of similes ranged from zero to a dozen per interview.Because metaphors were represented in the transcripts less than 10 times, theywere not included in the statistical analyses.42Pain Descriptors In total, there are 51 distinct descriptors that were used by 2 to 13participants (or 6% to 42% of the study sample). Table 6 (see p. 43) presents theindividual pain descriptors according to the dimensions of the MultidimensionalModel of Pain. Where an MMP dimension has subcategories, the subcategoriesare presented (e.g., Sensory Intensity and Sensory Quality). Descriptors with acommon root are grouped on the right with a bracket and are treated as a singledescriptor (e.g., ache, aches, and aching are grouped together). The numberimmediately to the right of a descriptor is its frequency (df in the table) and thenumber further to the right represents the number of participants using thatdescriptor (np). To keep the number of descriptors manageable, only thosedescriptors with a minimum frequency of five and which were used by at least twoparticipants are presented in Tables 6 and 7. (The complete list of descriptors ispresented in Appendix J, p. 88.)As illustrated in Table 6 (see p. 43), the participants spontaneously used anumber of MPQ pain descriptors. For example, the affective descriptors"annoyed" and "annoying" were used a total of five times by three of the studyparticipants. The sensory descriptors "ache," "aches," and "aching" were usedtwenty-five times by nine participants.Affectiveannoyedannoyingfearfearfultireddf np 1 }4^332}^35^3Behaviouraltiredtirednessdf np411^2Physiological• Duration}constant^8constantly^1df np 7Sensory• Intensitydf npjumping^41jumps 3j^2numb^21 4numbness^3143Table 6MPQ Descriptors Used by ParticipantsSensoryache^21aches 2aching^2burn 1burning^10cold 10dull^13hot 6hurtinghurts^8210}hurthurthurtinghurtsintenseintensity1 177pressingpressure^14pulling^41^10^pulls 1^4.1sharp^29 11shooting^6^2sore^14soreness^11^8df np^stabbing^5^212J1212181}sting^2stinging^5^39stings 1throbbed^13^throbbing^4^53^throbs^2,6^tight 52^tighten^}1^2tightening^113 41 2tiredtiredness^3J• QualityNote: df=descriptor frequency and np=number of participants.Most of the descriptors used by participants are in the Affective and SensoryDimensions, matching the structure of the McGill Pain Questionnaire. This finding,along with the International Association for the Study of Pain (IASP, 1979) definitionthat pain is an unpleasant affective and sensory experience, demonstrates that theparticipants' use of descriptors is consistent with the IASP definition of pain and theassumptions regarding pain underlying the MPQ. Examples from interview transcripts44for each dimension listed in Table 6 are provided as context to illustrateparticipants' use of MPQ descriptors. Due to the number of codes embeddedwithin some quotations, only a few of the descriptors from Table 6 are highlightedin italics.Affective:Just annoying really. I don't know what to describe? Well, it's very low,it's not very painful at all. It's just very dull, a very minor ache...Behavioural:Sometimes it gets so bad you don't give it much thought you know. Youjust get so tired all the time.Physiological (Duration):That's a constant ache I guess.Sensory (Intensity):This time it was the most excruciating, it's probably the worst I've had it.Just excruciating, sharp, sharp, intense pain.Sensory (Quality):It's so sharp a pain like, explain it like a tube and you've got a nerve sittingdown there, jumping, jumping, trying to get out.Table 7 (see p. 45), illustrates that the participants used 26 descriptors notfound in the MPQ. Again, words with common roots are bracketed on the rightand are rated as a single descriptor. The majority of non-MPQ descriptors were inthe Affective and Sensory Dimensions as were the MPQ descriptors. Examplesare provided to set the context for the use of these descriptors.45Table 7Non-MPQ Descriptors Used by ParticipantsAffectiveangrycrankinesscrankydepresseddepressingfrustratedfrustratingfrustrationgoodscaredscarytensetensionBehaviouralbreathbreathingwindspositionsitsittingwalkwalking461281Physiological• Pattern df np13^911^89^69^6comesdowngoesupSensory• IntensityexcruciatingpainfulscreamscreamingSensoryhardjerkjerksknife^51knives 1fmuscle^91muscular^21knitting needle 1}needle 1^5needles^3painful^131painfully^3 jrip^1ripped^2^3}ripping^3through^5^4df np5^4321^2411^584^625^482^5411^5df np72^318^35df np10^411^641 }^4• Quality4^df np6^41 }4^2379Note: df---descriptor frequency and np=number of participants.Affective:I guess pretty shitty. And if it carries on too long, I spoke of frustrating andone finally saying well it's not worth living, if you had to live with a pain likethat on a constant time, it wouldn't be worth it.Behavioural:It trickles out, it, it will wear out in an hour or two. And if I'm careful I won'thurt it again. Like there is just one position I can get that hip and it will hurt...46Physiolosical (Pattern):It's a burning kind of on and off again sort of a thing that comes and goes. Iassume from the treatment.Sensory (Intensity):Oh it's really terrible. I can't walk. I can walk once I get up on my feet. I can walkwith a walker a bit but it's verypainfid and my legs just seem to collapse under me.Sensory (Quality):Yeah, sometimes when I get up I get pins and needles and I just have to sit acouple minutes and that goes away.In addition to non-MPQ descriptors, the study participants used a number ofmodifiers listed in Table 8 with the descriptors. Modifiers were used by the participants toenhance or change the meaning of a pain descriptor. Because of their association withdescriptors, the majority of modifiers are found in the Affective and Sensory Dimensions.Table 8Descriptor-Modifiers Used by Participantsever 10 7Affective Sensorydf np little 6 5 • Qualitylotlotsmore4664df npbaddoesn'tdon't721 21613 bad 13 11not 13 13 not 17 14really 15 10severe 11 5Sensoryso 24 13• Intensity very 18 8df np worsebad 35 15 worst1417 11Note: df=descriptor frequency and np=number of participants.47Brief examples demonstrating the use of modifiers follow.Affective (modifier):You sure don't feel good anyway.Sensory Intensity (modifier):It's not very bad now, but it was really bad before.Discussion of DescriptorsWhen compared with studies reviewed in the literature section detailingspontaneous pain descriptors, commonalties appear. The average age ofparticipants in this study is close to the average age of participants in the tworeview studies (this study – 57.2 years; Bressler and colleagues' [1986] study – 58years; Tearnan and Cleeland's [1990] study – 58 years). The average educationallevel of participants in this study is also close to the average for participants inBressler and colleagues' (1986) study — 11.26 and 10.8 years respectively(Tearnan and Cleeland [1990] did not report educational level).The thirteen most frequently used descriptors listed in Bressler andcolleagues' (1986) study were used by 12% - 40% of their participants (see p. 21of this study). Nine of these descriptors were used by 6% - 43% of theparticipants in this study. Similarly, the eleven descriptors listed in Tearnan andCleeland's (1990) study were used by 2.8% - 11.5% of their participants (seep. 22 of this study). Again, nine of these descriptors were used by 6% - 43% ofthe participants in this study. Table 9 (see p. 48), details the most frequently usedspontaneous descriptors common to this study and the two studies mentionedabove.48Table 9Spontaneously Used Pain DescriptorsDESCRIPTOR STUDYaching* Bressler, Tearnan, & Gagelburning* Bressler, Tearnan, & Gagelpressing Bressler, Tearnan, & Gagelsharp* Bressler, Tearnan, & Gagelthrobbing* Bressler, Tearnan, & Gagelintense* Bressler & Gagel**knife/knifelike Bressler & Gagelnumb Bressler & Gagelsore Bressler & Gageldull Tearnan & Gagel**hurting Tearnan & Gagelshooting* Tearnan & Gagelstabbing* Tearnan & GagelNote:^Descriptors matching those found in Wilkie and colleagues' (1990) meta-analysisstudy of the MPQ.** There were no descriptors common to Bressler and colleagues' (1986) and Tearnanand Cleeland's (1990) studies that were not also used by participants in this study.Table 9 demonstrates a considerable overlap of spontaneously selected paindescriptors among subjects in Bressler and colleagues' (1986) study, Tearnan andCleeland's (1990) study, and the present study. In addition, single asterisks in thetable highlight the spontaneous pain descriptors that match seven of the mostcommonly selected MPQ pain descriptors as noted in Wilkie and colleagues' (1990)meta-analysis study (see Table 2, p. 20).The descriptor-modifiers used by participants in this study (see Table 7, p. 45),were reported in neither Bressler and colleagues' (1986) nor Tearnan and Cleeland's(1990) studies and are missing from the MPQ descriptor studies. This may be partlydue to the previous investigators' interest in descriptors only; however, the use of49modifiers was important for the participants in this study when they described theirpain. It would be likely that they are also important to other cancer patientsattempting to describe their pain. The use of modifiers in the description of painshould be further investigated to determine if this information may be helpful to theassessment of pain. Certainly, the high frequency of some of the modifiers warrantsfurther investigation (e.g., bad, little, so, very, etc. See Table 8, p. 46).Further comparison of the descriptors in this study with MPQ descriptorsdemonstrates that of the 51 descriptors spontaneously used by participants, 25 matchthe 73 descriptors found in the MPQ. However, there were just as many descriptorsused by participants in this study that are not found in the MPQ (n=26). Thesefindings suggest limitations of the McGill Pain Questionnaire when used with cancerpatients. The MPQ lacks a number of descriptors that are important to cancer patientsand may contain descriptors not useful for cancer patients attempting to describe theirpain.Participant Variables and Descriptor PhrasesThree types of statistical tests were employed to test for relationships anddifferences among the data: Chi-Square statistics were used to test for differencesamong nominal variables (sex and mother tongue); correlation coefficients were usedto test relationships among continuous and grouped variables (age, level of education,and present pain intensity) and participants' use of descriptor phrases; and Mann-Whitney U statistics were used to test for differences between the means of groupedvariables and participants' use of descriptor phrases.Grouping of VariablesIt was necessary to transform continuous variables into discrete groups inorder to perform the Mann-Whitney U tests. Continuous variables (interval or ratio50data) need to be placed into ordinal groups for U testing because scores are orderedaccording to their rank within a group during calculation of a U statistic (Shavelson,1988). Conversely, correlations can be calculated from either grouped or continuousvariables, and nominal variables such as sex and mother tongue are naturallydichotomously grouped.Participants were divided into subgroups based on a number that would permitfuture investigators to compare their results with the results of this study. For example,age was split into subgroups of age 64 or lower (20 participants) and age 65 or higher (11participants) since age 65 is generally used as an age for retirement in Canada. Educationwas grouped into Grade 11 or lower (14 participants) and Grade 12 or higher (17participants) since Grade 12 is considered a milestone in the school system. Participants'present pain was grouped into intensity less than or equal to 5 (21 participants) andintensity greater than 5 (10 participants); five is simply the midpoint of - 10. Groupsbased on mother tongue and sex divided into the following: English=23 participants andOther=8 participants, and Female=16 participants and Male=15 participants.Differences Amonq VariablesThe data were analyzed using Chi-Square to determine if there was asignificant difference between women and men and the grouped variables of age, levelof education, mother tongue, and present pain intensity. Mother tongue was testedseparately against age, level of education, and present pain intensity. None of theresulting Chi-Square statistics were significant at p50.05 (see Appendix K, p. 92).This is probably a result of the small sample size which does not allow for a minimumof 10 items per cell of a Chi-Square table. Therefore, there was no evidence ofstatistically significant differences in age, level of education, mother tongue, or presentpain intensity between women and men in this sample. There were also no significant51differences in age, education, or present pain intensity between study participants whohad English or another language as a mother tongue.Correlations Among Continuous and Grouped Variables, and Coded Phrases After coding of the transcripts was completed, the number of coded phraseswas manually tabulated and correlations and measures of central tendencies werecomputed using SY STAT/SYGRAPH® . Table 10 presents Pearson r correlationcoefficients that were based on participant variables and the number of phrases theyused to describe pain in three dimensions of the MMP. The rows in Table 10 containcontinuous variables (age and education [Age and Educ in the table]) and groupedvariables (Age-Grp and Educ-Grp in the table). Significant correlations arehighlighted, and only columns that contain significant results are presented in the table(see Appendix L, p. 93, for the complete correlation matrix).Table 10Pearson Correlation MatrixAFFECTIVE PHYSIOLOG-ICAL SENSORYAGE -0.333 -0.310 -0.289AGE-GRP -0.346 -0.285 -0.381*EDUC 0.428** 0383* 0.405*EDUC-GRP 0.362* 0.265 0.345Note: Only columns that contain significant results are presented.Shaded rows and all columns contain continuous variables.* p<0.05.** p<0.01.The correlations indicate positive relationships between education and each ofthe three dimensions listed in the table. Participants who had a higher education used52more affective phrases (both for education as a continuous variable, and as a groupedvariable — e.g., Grade 11 or lower and Grade 12 or higher). These participants alsoused more physiological phrases (education as a continuous variable), and sensoryphrases (education as a continuous variable). There was an inverse relationshipbetween age and the Sensory Dimension of the MMP. Participants who were older(age as a grouped variable — 64 years and below and 65 years and above) usedsignificantly fewer sensory phrases than their younger counterparts. It seems intuitivethat participants with a higher educational level would use more phrases to describetheir pain, presumably because they have developed a larger vocabulary due to theirschooling. One might also speculate that older people are more reticent aboutdescribing their pain based on the lower number of sensory phrases used by this group.However, such a statement may be attributable to "ageism" and an alternative theory isaddressed later.Discussion of CorrelationsTable 10 presents significant correlations among participants' age andeducation and their use of phrases in the Affective, Physiological, and SensoryDimensions of the MMP. Expressions of difficulty, similes, and the remaining MMPdimensions were not significantly correlated with age and education. The non-randomsampling and the small sample used for this study may be responsible for the relativelyfew significant correlations noted. As it is not possible to make inferences based oncorrelations, the Mann-Whitney U tests is the statistical test of preference for this data.Relationships Amonq Grouped Variables and Coded Phrases Mann-Whitney U tests were used to compare phrases based on dimensions ofthe MMP, expressions of difficulty, and similes with the grouped participant variablesof age, education, and present pain intensity. Table 11 lists all probability valuesLATENTMANIFEST (MMP)AFFECT-IVEBEHAV-IOURALCOGNI-TIVEPHYSIO-LOGIC-ALSENS-ORYDIFFI-CULTYAGE 0.141 0.983 0.117 0.209 0.013* 0.864 0.095EDUC 0.004* 0.597 0.180 0.198 0.039* 0.473 0.035*PAIN 0.621 0.665 0.841 0.507 0.687 0.631 0.363SIMILE53resulting from computation of Mann-Whitney U tests (results that are statisticallysignificant, i.e., p50.05, are highlighted).Table 11Probability Values for Coding ResultsBecause subjects in the Instrument study were asked to describe what their painfeels like, their responses may have been biased towards the Affective and SensoryDimensions of the Multidimensional Model of Pain. This may account for the lack of asignificant difference in the number of phrases used based on age and education and the useof behavioural, cognitive, and physiological phrases; and for the total lack of findings forsociocultural phrases. Examples from the Behavioural, Cognitive, and PhysiologicalDimensions are presented to illustrate participants' use of phrases in these dimensions.Behavioural:Then the pain will get so excruciating I just have to sit up right away,immediately.Cognitive:Well, you have to accept it. I have no choice. I sure would like it to go away butit's not about to.Physiological (Duration):And I was in constant pain, horrendous.54During the coding process, I noted that many of the participants commented onhow difficult it was for them to describe their pain. There were no significant differencesamong participants with respect to expressions of difficulty. Nevertheless, examples areprovided below to illustrate participants' use of expressions of difficulty.Difficulty:No, it is just that it gets very tense. It's really hard to describe.I don't know how you would describe pain.Tables 12 - 15, provided below, display results that support the assertion ofsignificant findings highlighted in Table 11 (see p. 53). In the tables, continuous variablesare grouped the way they were for the statistical computations. Along with the tables,scatter plots (Figures 2 - 5) graphically display relationships between descriptor phrasesand continuous variables upon which the discrete variables were based. Scatter plots haveregression lines that were generated by SYSTAT/SYGRAPH® to indicate the slope anddirection of relationships. In general, results of the Mann-Whitney U tests are in agreementwith the correlation findings. Note that only the statistically significant findings highlightedin Table 11 are addressed.Affective Dimension Education. There were significant differences between the participants' level ofeducation and their use of affective phrases. Participants with Grade 12 or highereducation used significantly more affective phrases than those with Grade 11 or lower.The mean number of affective phrases used by participants who had a Grade 12 or highereducation was 5.82. Those who had a Grade 11 or lower education used a mean of 2.71phrases. As with the correlations, it may be plausible to assume that with more years ofeducation, individuals have a greater vocabulary from which to draw descriptions of pain.55Table 12Statistical Summary of Affective Phrases Grouped by Level of Education (n=31)GRADE^11 ^12N. of PARTICIPANTS 14 17MINIMUM 0 2MAXIMUM 13 13MEAN 2.714 5.824STND DEVIATION 3.148 4.142RANK SUM 153.5 342.5MANN-WHITNEY U 189.5PROBABILITY 0.004Figure 2Number of Affective Phrases by Level of Education 1510502^7^12^17^22Years of Education56Sensory Dimension Age. In the Sensory Dimension, significant differences were demonstratedwith the variables age and education. Older participants used significantly fewerphrases in the Sensory Dimension than did younger participants. Participants whowere age 64 or lower used an average of 19.15 phrases in this dimension. Participantswho were 65 or older used half as many phrases; their average was 9.36. Again, thisfinding is in agreement with the correlation between increased age and decreased useof affective phrases. Once more, one might assume that older people are reticentwhen it comes to describing their pain.Table 13Statistical Summary of Sensory Phrases Grouped by Ade (n=31)AGE^64 ^65N. of PARTICIPANTS 20 11MINIMUM 5 2MAXIMUM 46 16MEAN 19.150 9.364STND DEVIATION 12.145 4.802RANK SUM 380 116MANN-WHITNEY U 170PROBABILITY 0.01357Figure 3Number of Sensory Phrases by Ace50Cl)a) 40u)cos.,4a.s..,' 300Cl)Za)Ci)c...o, 20s.a)Ez= 10 30^40^50^60^70^80^90Age in YearsEducation. As with the Affective Dimension, participants who had a higherlevel of education also had a greater number of responses in the Sensory Dimension.Table 14Statistical Summary of Sensory Phrases Grouped by Level of Education (n=31)GRADE_^ 11 .^ 12N. of PARTICIPANTS 14 17MINIMUM 2 5MAXIMUM 22 46MEAN 10.643 19.824STND DEVIATION 5.583 12.895RANK SUM 172 324MANN-WHITNEY U 171PROBABILITY 0.03958Participants with a Grade 12 or higher education used a mean of 19.82 sensoryresponses, while those with Grade 11 or below used a mean of 10.64 responses. Itappears that in this sample of participants, the higher one's educational level, the moreone is able to describe pain (based on the number of sensory and affective phrases).Figure 4Number of Sensory Phrases by Level of Education50cn 40cr)coG.30cuoci) 20zs..io0 2^7^12^17^22Years of EducationUse of Similes Education. For similes, the only significant difference demonstrated was witheducation. There was greater use of similes by participants with a higher level ofeducation. Participants who had a Grade 12 or higher education used an average of3.30 similes while those with Grade 11 or lower used an average of 1.36 similes. Thefollowing quotations give examples of participants' use of similes:59It starts like a sudden pain and sometimes it feels like being pricked by manyneedles and then it starts down there. It gets worse....it's a real sharp pain like somebody just shoves a knife in you and just keepstwisting and then starts pulling your insides out through the small hole and, itwould just take my breath away so I couldn't even breath and I would fall righton to the floor...Table 15Statistical Summary of Similes Grouped by Level of Education (n=31)GRADE_^ 1 1 _^ 1 2N. of PARTICIPANTS 14 17MINIMUM 0 0MAXIMUM 8 12MEAN 1.357 3.294STND DEVIATION 2.061 3.177RANK SUM 169.5 326.5MANN-WHITNEY U 173.5PROBABILITY 0.02760Figure 5Number of Similes by Level of Education 1510502^7^12^17^22Years of EducationDiscussion of the Findings The use of expressions of difficulty was not significantly different betweengroups of participants. The finding of no difference is understandable, however,because it is difficult to describe pain. Present pain intensity was also found not to besignificantly different between participants. This finding could be related to the smallsize of the study sample and the selection of suitable study participants by NursingUnit Managers or their delegates. The Unit Managers may have spared patients whowere experiencing high intensity pain from the rigours of study participation.Further caution in interpreting the findings in this section is suggested. Thefinding of significant differences in the use of phrases between younger and olderparticipants (sensory) may actually be due to education.61Age or Education? Older participants had less education than youngerparticipants. Participants aged 64 or lower had, on average, nearly two more years ofTable 16Level of Education by Age (n=31) AGE^64 ^65N. of PARTICIPANTS 20 11MINIMUM 5 6MAXIMUM 16 14MEAN 11.950 10.000STND DEVIATION 2.328 2.449RANK SUM 370 126MANN-WHITNEY U 160PROBABILITY 0.035education than their counterparts aged 65 or older (mean of 11.95 years of educationversus 10.00 respectively).In the Sensory Dimension, both the younger participants and participants withmore education used significantly more phrases than their older and their less educatedcounterparts. While it can only be stated that these variables were significant factorsin the Sensory Dimension, questions can be posed: does the higher level of educationof younger participants account for the higher number of phrases these participantsused in the Sensory Dimension, or does socialization in a different era influence olderindividuals to use fewer sensory phrases? While the Kruskal-Wallis test (a non-parametric equivalent to the ANOVA test) could help answer this question, it cannotbe used here because the sample size in this study is too small to accommodate enoughparticipants per group.62Figure 6Level of Education by Age221712220^30^40^50^60^70^80^90Age in YearsThe other significant finding, that of a difference in the average use of similesbetween groups based on education, could be due to the fact that all similes used werecontained within sensory phrases. Participants who used sensory phrases often usedsimiles within those phrases. Therefore, the use of similes parallels the use of sensoryphrases which were used more often by participants with Grade 12 or highereducation.Summary and ConclusionsExamination of the study sample revealed that there were 16 women and 15men whose average age was 57.2 years. Among them, the participants had a widevariety of cancers and reported an average intensity of pain at 3.95 on a 0 - 10 scale.Despite having a range of mother tongues, the participants all spoke fluent English. In63addition, their education level ranged from Grade 5 to baccalaureate degree with anaverage of about eleven years in school.Transcripts from the 31 interviews were analyzed to determine which wordsand phrases were used to describe pain. Demographic variables were also assessed todetermine if they were associated with the frequency use of descriptor phrases. Atotal of 51 cancer pain descriptors were identified. Twenty-five of these spontaneousdescriptors matched those found in the McGill Pain Questionnaire. There wasconsiderable overlap with the pain descriptors identified in this study with the mostfrequently occurring descriptors in Bressler and colleagues' (1986) study, Tearnan andCleeland's (1990) study, and Wilkie and colleagues' (1990) meta-analysis study.There were also 26 non-MPQ descriptors used by participants in this study that werenot reported in the literature. These non-MPQ descriptors played an important role inthe description of pain by the study participants. In addition, descriptor-modifierswere used which altered the meaning of descriptors. Further study into thesedescriptors and descriptor-modifiers would be warranted.When descriptor phrases were examined based on demographic variables, anumber of findings were noted. A number of significant correlations were highlightedand some of the correlations remained significant when the more rigorous Mann-Whitney U computations were performed. Significant findings included the increaseduse of affective phrases by more educated participants, the increased use of sensoryphrases by younger participants and participants with more education, and the greateruse of similes by more educated participants. However, as mentioned earlier, the useof similes was tied to the use of sensory phrases, and the effect of greater educationmay have accounted for the increased use of sensory phrases by younger participants.64The Behavioural, Cognitive, and Physiological dimensions did not havesignificant differences in the use of phrases based on any of the participant variables.There was no significant difference among the participants in their use of expressionsof difficulty. Present pain intensity was also not significantly associated with thenumber of phrases used.Altogether, the findings suggest that the descriptors noted in this studyaccurately describe pain as reflected by the overlap with descriptors in other studiesand the MPQ. When looking at the MPQ, the findings suggest that it contains asubset of descriptors that cancer patients use to describe their pain. Moreover, theMPQ is lacking a number of descriptors and modifiers important to cancer patientswith pain.Chapter 5 summarizes the study findings and offers implications for nurseswith respect to pain assessment in cancer patients.65CHAPTER 5: CONCLUSIONS, IMPLICATIONS, AND SUMMARYThis chapter presents conclusions based on the findings of this study andpresents implications for nursing education, practice, and research.ConclusionsCancer causes pain in many people because of its invasive nature and thetreatment required to cure, control, or palliate it. For this reason, as many as 26,000British Columbians with cancer may experience pain annually (Deschamps, Coldman,Bradley, & Baker, 1991). While nurses are aware of this problem and have expressedthat research in this area should be a priority, assessment and management of cancerpatients' pain fall short of ideal. For example, authors have noted that nurses' andpatients' perceptions of pain may not be congruent and that nurses may not heedpatients' reports of pain. Pain assessment tools are intended to solve such problems byproviding objective measurements of pain.The McGill Pain Questionnaire came into use in the mid 1970s and remains aninfluential tool today as evidenced by the number of MPQ-based studies cited in thisreport. However, there are problems with the MPQ, notably with its list of paindescriptors which originated in the literature instead of from patients describing theirpain. The Instrument study was an attempt to develop a new tool similar to the MPQbut it was intended to be based on words used by cancer patients.Although data collection for the Instrument study started, funds did notmaterialize and the study was not completed. As I was the research assistant for thatstudy, I have used the data I collected using secondary analysis to determine whatdescriptors and phrases the sample of cancer patients used to spontaneously describetheir pain. In addition, the use of descriptor-modifiers, similes, and expressions ofdifficulty in describing pain were examined. Relationships among selected66demographic variables and reports of pain intensity were also explored to determine ifthere were differences between both continuous and grouped data (e.g., youngerversus older participants) in the frequency of phrases they used to describe pain. Thefollowing conclusions have been drawn:1. With respect to the research question, this study found that participantsspontaneously used 51 words to describe their pain (words which wereused by two or more participants, five or more times). The use ofdescriptors by participants in this study was similar to other studies whereparticipants were asked to describe pain using their own words or to selectMPQ descriptors (average age and education was also similar among theparticipants). Of the descriptors used in this study, 25 match the 73 paindescriptors found in the McGill Pain Questionnaire. However, participantsused an additional 26 descriptors not found in the MPQ. The descriptors,found in Table 7 (see p. 45), include strong words such as "angry,""depressed," "frustrated," "scared," "screaming," and "painful." None ofthese words are included in the descriptor section of the MPQ and suggestthat it may not be entirely suitable for use with cancer patients.2. Variables which proved to be significant included age and level ofeducation. Higher educational preparation (Grade 12 or above versusGrade 11 or below) was associated with the higher use of both affectiveand sensory descriptors. Participants who were age 64 or younger alsoused more sensory phrases than did participants who were 65 or older. Apotential confounding factor, however, was the significantly highereducational preparation of participants who were younger than 65 versusthose who were older.673. Examination of the descriptor phrases revealed that there were nosignificant differences between groups based on the variables of sex, age,level of education, mother tongue, or present pain intensity in theBehavioural, Cognitive, and Physiological Dimensions of theMultidimensional Model of Pain. No sociocultural phrases were noted inany of the transcripts. There were no significant differences among theparticipants in the number of times they expressed difficulty in describingtheir pain. The use of metaphors by participants in this study was foundnot to be extensive (less than ten occurrences were found in the 31transcripts), but there were 72 occurrences of similes with the participantsusing them as an aid to describe the intensity or quality of pain. Whilethere was a significant difference in the use of similes between participantswith Grade 12 or higher education and those with less, all similes wereused in phrases coded as sensory. That is, this difference is unquestionablylinked to the use of sensory phrases.LimitationsWoods (1988) outlines several disadvantages that may be present when datafrom another study are analyzed. In particular, the quality of the database cannot beassumed — the database must be accepted as it exists. As the Instrument study wasintended to collect affective and sensory data (due to the structure of the MPQ), thismay have influenced the data collection procedure resulting in a disproportionatelyhigh number of affective and sensory words and phrases. Also, a researcheremploying a database for a secondary analysis is not often part of the original researchteam and the original researcher may not be accessible to provide information to thesecondary analyst. However, in this case I was the research assistant in the original68study and conducted all the interviews that comprise the data used in this study. Inaddition, while it was assumed that the quality of the Instrument study data wasacceptable, the original researcher was available for consultation when I had questionsconcerning the data.An additional limitation is due to the use of a convenience sample of patientsfrom one institution in the Instrument study. Patients seen at the cancer treatmentfacility may not be representative of the general population of cancer patients. Forexample, cancer patients in the terminal stage of the disease are treated at communityhospitals or palliative care clinics in their home communities rather than at theprovincial facility, therefore, there may be a disproportionately low number ofpalliative patients in the sample. In addition, it was not possible to access outpatientsin the Radiotherapy Department during the Instrument study's data collection period(these patients may also experience pain as noted by the Nicholson et al. [1988] study,see p. 18 of this study). Biased selection of patients may also have occurred becausethe Nursing Unit Manager or delegate was able to screen which patients would besuitable for the study. This may have resulted in a disproportionately high number ofparticipants in the sample whose pain was more controlled the "average" cancerpatient. Although the study participants may not be representative of all people whohave cancer, they were cancer patients experiencing pain. Though limitinggeneralizability, the results of this study should have relevance for nurses working withcancer patients experiencing pain.Implications For NursesEducation Education in thorough pain assessment can help address problems in cancerpain management. Students should be taught to be aware that there may be69differences in how patients communicate about their pain regardless of pain intensity.As this study demonstrated, there were significant differences in the use of affectiveand sensory phrases based on level of education and sensory phrases based on age.The less educated participants offered fewer expressions of pain in the Affective andSensory Dimensions of the MMP and older participants used fewer phrases in theSensory Dimension than their more educated and younger counterparts. Yet, thegroups of participants did not have significantly different pain intensities.Pain assessment tools may help overcome differences among patients' ability tocommunicate about pain, if nurses know how to use these tools. Although basicnursing education may provide beginning nurses with a grounding in pain and painmanagement, many students may not actually use pain assessment tools until theyenter nursing practice. Students should be introduced to assessment tools when theyare supervised in the clinical setting. They could be provided with the opportunity tocare for patients experiencing pain and to use tools to assess and monitor painintensity, and to respond with interventions that help lower or prevent pain.The interviews conducted for this study also touched on a variety of topicsrelated to pain. The opportunity to discuss pain can be valuable for both the patientand student nurse. By providing patients with the opportunity to discuss their pain,and through the use of a conceptual model such as the MMP, students can assess andintervene in areas related to pain. For example, in the Behavioural Dimension, someof the study participants noted how pain affected their ability to do everyday activities(e.g., driving, eating, washing dishes, etc.). In the Physiological Dimension,participants noted how cancer therapies or analgesics contributed to pain (e.g., layingon the radiotherapy table, constipation, etc.). Introduction to a model such as theMultidimensional Model of Pain would allow student nurses to view aspects of pain70other than just its intensity. With the knowledge of peripheral aspects of pain,students would develop a greater knowledge of pain and could develop strategies tointervene in those areas. Students with this type of introduction to pain assessmentwould be well-prepared to care for patients with pain due to cancer or other illnesses.PracticePracticing nurses may not have had the opportunity to learn current painassessment techniques in their basic nursing programs. The points noted in the sectionon implications for education have relevance for practicing nurses as well. On oneoccasion, I observed a nurse who objected to using a Visual Analog Scale because shestated she found it embarrassing to ask patients to score pain on a 0 - 10 scale.However, she probably did not appreciate the usefulness of this tool. (Recall thatDonovan and Dillon [1987] found in their sample that only 43% of 69 cancer patientswith pain could recall a nurse having discussed pain with them.) To effectively helppatients manage pain, nurses must have the knowledge, ability, and willingness to usetools that measure pain. Only when pain is measured repeatedly can nurses andpatients determine if pain is being effectively managed.This study also demonstrated the difficulty patients can have in describing theirpain. Participants often exclaimed with a nervous chuckle that I was asking them todo an impossible task. "How do you describe pain?" was the rhetorical question Ireceived when I asked participants to tell me what pain feels like. It was difficult todraw information from the participant but there will be occasions when specific wordsare important to a patient for his or her description of pain. These words should besolicited from the patient and communicated with other nurses through the patient'scare plan to ensure consistent assessment and management of pain.71The exercise of interviewing patients clearly demonstrated to me that use of apain assessment tool could make pain assessment an easier and more efficient task.However, nurses need tools they can seamlessly integrate into their practice. That is,they need tools which are quick to administer and easy to use. This study found thatparticipants spontaneously used only 25 out of 73 descriptors contained in the MPQ.Participants also used 26 descriptors not found in that tool. This finding suggests thatthe MPQ's usefulness for cancer patients may be limited in part because patients haveto search through words they would not use and cannot find some of those words theylikely would use to describe their pain. This undoubtedly takes time, and because ittakes 20 minutes or longer to administer the MPQ, nurses are not likely to integratethis tool into their everyday practice.Practicing nurses also need the opportunity to learn and gain skill with painassessment techniques. While it might seem obvious that patients with more educationuse more phrases to describe their pain, how often is this information taken intoconsideration? An informal search through patient charts will undoubtedly turn upseveral occurrences of the phrase, "No c/o pain." Older participants in this study didnot use as many sensory phrases as younger participants despite having similar painintensities. No complaint of pain should not be taken to imply "no pain." Nurses needto be aware that certain patients may not be as verbal about pain as others. Agencybased inservice education and community based distance education programs caneasily be designed that will update and inform practicing nurses on the current state ofknowledge for pain assessment and management. This may address the problem of themany articles in the review section of this report that detail the shortcomings ofnursing pain assessment.72Future Research The findings in this study point to potential answers to recognized problemswith pain assessment. Replication of this study, through the use of a randomized andlarger sample to enable generalizable findings, could refute or support these findings.Replication of this research would also need to address the potential bias of this studyregarding interview questions that may have solicited more affective and sensoryresponses. Notwithstanding the non-generalizability of the findings in this study, thefollowing points are offered as suggestions for future research.A pain assessment tool could be developed that is similar to the McGill PainQuestionnaire, building on its established validity and reliability. The new painassessment tool would incorporate spontaneous descriptors that are used by cancerpatients but which are not currently included in the MPQ. Also, the new tool wouldnot contain descriptors that this study and several others have demonstrated are notused by cancer patients to describe pain. As the MPQ has been found to be lengthy interms of its administration time, adjustment of the descriptor portion of the MPQ mayresult in quicker administration times if cancer patients are able to find descriptors thatmake sense to them. Attention should also be paid to the total length of a tool. Aspain assessment has to be repeated to monitor the progress of pain management, fasteradministration of a tool would be particularly beneficial. Finally, when patients areasked to spontaneously describe pain, differences may emerge based on variables suchas age or education. By providing a list of descriptors that more accurately reflects thepain cancer patients experience, a new pain assessment tool could allow for suchdifferences. Further research would then be needed to test and revise this new painassessment tool.73SummaryThis study was useful in answering the question of what words and phrasescancer patients spontaneously used to describe their pain. Support for modification ofthe McGill Pain Questionnaire was evidenced by the finding of 25 matching and 26non-matching words. New words should be added to an MPQ-like list and extraneouswords (such as lancinating) should probably be removed. Revisions based on wordsactually used by cancer patients would likely result in a more efficient and clinicallyappealing tool. This study also revealed how difficult it is for patients to describe theirpain and strongly pointed to differences among the study participants that might leadindividuals to use a greater or fewer number of phrases to describe their pain based ontheir age or education. Clearly, a new pain assessment tool for cancer patients needsto be developed that will address issues of speedy and practical administration and willalso account for differences among patient responses based on variables such aseducation and age. Nurses can play a key role in the development, testing, andrefinement of such a tool.74REFERENCESAhles, T.A., Blanchard, E.B., & Ruckdeschel, J.C. (1983). The multidimensionalnature of cancer-related pain. Pain, 17(3), 277-288.Allan, B.M. (N.D.). Assessment of chronic pain in cancer. Unpublished paper.Babbie, E.R. (1986). The practice of social research (4th ed.). Belmont, CA:Wadsworth.Bramwell, L. (1989). Cancer nursing — a problem-finding survey. Cancer Nursing,12(6), 320-328.Breitbart, W. (1989). Psychiatric management of cancer pain. Cancer, 63, 2336-2342.Bressler, L.B., Hange, P.A., & McGuire, D.B. (1986). Characterization of the painexperience in a sample of cancer outpatients. Oncology Nursing Forum, 13(6),51-55.Camp, L.D. (1988). A comparison of nurses' recorded assessments of pain withperceptions of pain as described by cancer patients. Cancer Nursing, 11(4), 237-243Canadian Association of Nurses in Oncology. (1988). Position statement regardingpain management. Toronto, ON: Author.Catanzaro, M. (1988). Using qualitative analytical techniques. In N.F. Woods &M. Catanzaro (Eds.), Nursing research: Theory and practice (pp. 437-456).St. Louis, MO: C.V. Mosby.Cleeland, C.S. (1984). The impact of pain on the patient with cancer. Cancer, 54,2635-2641.Cleeland, C.S., Ladinsky, J.L., Serlin, R.C., & Nugyen, C.T. (1988).Multidimensional measurement of cancer pain: Comparisons of US andVietnamese patients. Journal of Pain and Symptom Management, 3(1), 23-27.Coyle, N. & Foley, K. (1987). Prevalence and profile of pain syndromes in cancerpatients. In D.B. McGuire & C.H. Yarbro (Eds.), Cancer pain management(pp. 21-46). Orlando, FL: Grune & Stratton.75Dalton, J. (1989). Nurses' perceptions of their pain assessment skills, painmanagement practices, and attitudes towards pain. Oncology Nursing Forum,16(2), 225-231.Degner, L. (1987). Priorities for cancer nursing research: A Canadian replication.Cancer Nursing, 10(6), 319-326.Deschamps, M. (1991). Research programme in nursing at the British ColumbiaCancer Agency. Cancer Research News, 8(1), 3. (Available from B.C. CancerFoundation, 601 West 10th Avenue, Vancouver, BC, V5Z 1L3)Deschamps, M., Band, P.R., & Coldman, A.J. (1988). Assessment of adult cancerpain: Shortcomings of current methods. Pain, 32(2), 133-139.Deschamps, M., Coldman, A., Bradley, L., & Baker, C.S. (1991). Study of pain inthe British Columbia cancer population, Phase II: Development of aninstrument evaluating cancer pain. Unpublished manuscript.Donovan, M.I. (1985). Nursing assessment of cancer pain. Seminars in OncologyNursing, 1(2), 109-115.Donovan, M.I., & Dillon, P. (1987). Incidence and characteristics of pain in a sampleof hospitalized cancer patients. Cancer Nursing, 10(2), 85-92.Dorrepaal, K., L., Aaronson, N.K., & Van Dam, F.S.A.M. (1989). Pain experienceand pain management among hospitalized cancer patients: A clinical study.Cancer, 63, 593-598.Dubuisson, D., & Melzack, R. (1976). Classification of clinical pain descriptions bymultiple group discriminant analysis. Experimental Neurology, 51, 480-487.Graham, C., Bond, S., Gerkovich, M., & Cook, M. (1980). Use of the McGill PainQuestionnaire in the assessment of cancer pain: Replicability and Consistency.Pain, 8, 377-387.Health and Welfare Canada. (1984). Cancer pain. A monograph on the managementof cancer pain (Cat. No. H42-2/5-1984E). Ottawa, ON: Minister of Supply andServices Canada.Holsti, O.R. (1969). Content analysis for the social sciences and humanities.Reading, MA: Addison-Wesley.76International Association for the Study of Pain. (1979). Subcommittee on taxonomyof pain terms: A list with definitions and notes on usage. Pain, 6, 249-252.Loeser, J.D., & Cousins, M.J. (1990). Contemporary pain management. TheMedical Journal of Australia, 153, 208-216.McCaffery, M. & Beebe, A. (1989). Pain: Clinical manual for nursing practice.St. Louis, MO: C.V. Mosby.McGuire, D.B. (1984). Assessment of pain in cancer inpatients using the McGill PainQuestionnaire. Oncology Nursing Forum, 11(6), 32-37.McGuire, D.B. (1987). The multidimensional phenomenon of cancer pain. InD.B. McGuire & C.H. Yarbro (Eds.), Cancer pain management (pp.1-20).Orlando, FL: Grune & Stratton.McGuire, D.B., & Sheidler, V.R. (1990). Pain. In S.L. Groenwald, M.H. Frogge,M. Goodman, & C.H. Yarbro (Eds.), Cancer nursing: Principles and practice(2nd ed.) (pp.385-441). Boston, MA: Jones & Bartlett.Melzack, R. (1975). The McGill Pain Questionnaire: Major properties and scoringmethods. Pain, 1, 277-299.Melzack, R., & Torgerson, W.S. (1971). On the language of pain. Anesthesiology,34(1), 50-59.Mooney, K.H., Ferrell, B.R., Nail, L.M., Benedict, S.C., & Haberman, M.R. (1991).Oncology Nursing Society research priorities survey. Oncology Nursing Forum,18(8), 1381-1388.Nicholson, B.A., McGuire, D.B., & Maurer, V.E. (1988). Assessment of pain in headand neck cancer patients using the McGill Pain Questionnaire. The Journal of theSociety of Otorhinolarnyngology Head and Neck Nurses, 6(3), 8-12.Pannuti, F., Rossi, A.P., Marraro, E., Strocchi, A., Piana, E., & Pollutri, E. (1980).The natural history of cancer pain. In R.G. Twycross & V. Ventafridda (Eds.),The continuing care of terminal cancer patients (pp. 75-89). New York, NY:Pergamon.Polit, D.F., & Hungler, B.P. (1991). Nursing research: Principles and methods (4thed.). Philadelphia, PA: J.B. Lippincott.77Rankin, M.A., & Snider, B. (1984). Nurses' perceptions of cancer patients' pain.Cancer Nursing, 7, 149-155.RightSoft. (1988). RightWriter: User 's manual (Ver. 3.1). Sarasota, FL: Author.Seidel, J.V., Kjolseth, R., & Seymour, E. (1988). The Ethnograph. User's Manual(Ver. 3.0). Lettleton, CO: Qualis Research.Shavelson, R.J. (1988). Statistical reasoning for the behavioral sciences (2nd ed.).Boston, MA: Allyn & Bacon.Spiegel, D., & Bloom, J., R. (1983). Pain in metastatic breast cancer. Cancer, 52,341-345.Spross, J.A., McGuire, D.B., & Schmitt, R.M. (1990a). Oncology Nursing Societyposition paper on cancer pain: Part I. Oncology Nursing Forum, 17(4), 595-614.Spross, J.A., McGuire, D.B., & Schmitt, R.M. (1990b). Oncology Nursing Societyposition paper on cancer pain: Part II. Oncology Nursing Forum, 17(5), 751-760.Spross, J.A., McGuire, D.B., & Schmitt, R.M. (1990c). Oncology Nursing Societyposition paper on cancer pain: Part III. Oncology Nursing Forum, 17(6), 943-955.Tearnan, B.H., & Cleeland, C.S. (1990). Unaided use of pain descriptors by patientswith cancer pain. Journal of Pain and Symptom Management, 5, 228-232.Teske, K., Daut, R.L., & Cleeland, C.S. (1983). Relationships between nurses'observations and patients' self-reports of pain. Pain, 16(3), 289-296.Twycross, R.G., & Fairfield, S. (1982). Pain in far advanced cancer. Pain, 14(3),303-310.Wilkie, D.J., Savedra, M.C., Holzemer, W.L., Tesler, M., D., & Paul, S.M. (1990).Use of the McGill Pain Questionnaire to measure pain: A meta-analysis. NursingResearch, 39(1), 36-41.Wilson, H.S. (1985). Research in nursing. Reading, MA: Addison-Wesley.78Woods, N.F. (1988). Using existing data sources: Primary and secondary analysis.In N.F. Woods & M. Catanzaro (Eds.), Nursing research: Theory and practice(pp. 334-347). St. Louis, MO: C.V. Mosby.Woods, N.F., & Mitchell, P.H. (1988). Designing studies to explore association anddifference. In N.F. Woods & M. Catanzaro (Eds.), Nursing research: Theoryand practice (pp. 150-165). St. Louis, MO: C.V. Mosby.Zimmerman, L., Duncan, K., Pozehl, B., & Schmitz, R. (1987). Pain descriptors usedby patients with cancer. Oncology Nursing Forum, 14(4), 67-71.PRI: 9 A ^ E^ M.—.--.--PRI(T)(1 - 10)^(11- 16)^(16)^(17-20)PPI( 1-20)APPENDIX A: MCGILL PAIN QUESTIONNAIREPatient's NameMcGill Pain QuestionnaireDate Tim•^•rn/pmI^FLICKERINGQUIVERINGPULSING11^TIRINGE 2 HAuSTING12^SICKENINGTHROBBINGSUFKOCA TiNGSEATINGPOUNDING 13 FEARKulFRiGHTFuL2 JUMPING —I TERRIF yiNGFLASHINGSHOOTING _ 14 PuNISHfNG3 PRICKING GRUELLINGBOFIINGCRuELDRILLING — VICIOUSKILLINGSTABBINGLAKICINA TING — 15 *PIET C.IED• SHARPCIL to■OtNGcuT TING _. 16 ANNOYING^.._,LACERATING _. TRouSLESOmES PiNCHINGMISER•BLE—PRESSINGINTENSEGNAWINGuNBE AR•BLECFIAMPING 12 SPRE aDiNGCRUSHING ___, N•DIA TINGPENE TRATiNGI TUGGING•ULLINGP,E ACINGWRENCHING —, 113 TtG.T^_.NUmEl ._,oa•wING^_7 HOTBURNiNGSCALDINGSOuE2 ZINGTE •RINGSE ARiNGi9 coot. —8 TINGLINGiTcmyGoLo^—FREEZINGSMARTINGSTINGING 20 N•GGING^—9 OuLLNAuSEA TiNG^_AGONIZING^-,SOREORE ADFUL^—HUR TINGTORTURING^—ACHINGKEA vY PPI0^NO PAIN^—10 TENDER I^MILD —TAUTRASPiNG2^DISCOMFORT iNG —SPLITTING3^DIS TRESSING^_...• NORSIBLES^E ICRUCIATINGCOMMENTS 1 E :: EXTERNALI 1 :: INTERNAL#79BRIEF^__MOMENTARY ...TRANSIENT ...RH Y THMI C^_P E R IC) 0 i c —IN TERMIT TENTCONTINUOUSSTEADYCONSTANTAPPENDIX B: LETTER OF PERMISSIONBRITISH COLUMBIA CANCER AGENCYSeptember 14, 1992Mr Mike GagelPO Box 4043Vancouver, B.C.V6B 3Z4Dear Mike:It is with pleasure that I give my permission to use part ofthe data collected for the "Study of Pain the British ColumbiaCancer Population, Phase II: Development of an instrumentevaluating cancer pain" for the purpose of your Master of Sciencein Nursing at the University of British Columbia.I wish you the best in your endeavour and look forward the seethe results of your analysis.Yours sincerely,c_DA^(111Michele Deschamps, RN, Ph.D.Head, Research in Nursing SectionBritish Columbia Cancer Agencycc: Anna-Marie Hughes, RN, Ph.D.UBC School of NursingJenny Hayes, RN, Ph.D.UBC School of NursingCarolyn Baker, RN, MScNVancouver General Hospital80600 West 10th Avenue, Vancouver, B.C., Canada, V5Z 4E6 Tel. (604) 877-6000 Fax (604) 872-4596APPENDIX C: INTERVIEW GUIDEINTERVIEW GUIDE1. Explanation of the Study and ConsentReview "Patient Inrormation and Consent ror Participation" withsubject and obtain two written consents - one for the subject and onefor the study. Explain to the patient that he or she will be part ofa larger study involving 8,000 people. Ask the patient if he or shehas already received a questionnaire in the mail.2. Interviewi). Sensery- Where is your pain located?(If more that one pain, locate all painful areas.)- If more than one pain -> Let's start with the worst pain.- If more than one pain or the only pain -> How would you describethis pain?- If you had to describe this pain to a friend, what words would youuse to make your friend understand what you are feeling?- What words describe the intensity of the pain you have right now?- What is the intensity of the pain right now? (Use the VAS.)- If more than one pain -> Are your other painful areas similar tothe area with the worst pain?- If different from worst pain -> How would you describe this pain?- If you had to describe this pain to a friend, what words would youuse to make your friend understand what you are feeling?- What words describe the intensity of the pain you have right now?- What is the intensity of the pain right now? (Use the VAS.)- What do you think is causing your pain?- What are you taking for the pain? How satisfied are you with thepain relief you are receiving?ii). Affective- What feelings do you have about your pain?- If you had to describe this pain to a friend, what words would you• use to make your friend understand how you are feeling?- What is one word that describes how you feel about your pain?iii). Socio-Demographic Information - See the Data Collection Guide.iv). Thank you81c:\word\doc\intrview.doc 910228APPENDIX D: DATA COLLECTION GUIDEDATA COLLECTION GUIDEDate:^ Time:82Chart Number: ^Full Name:^Date of Birth:Ethnic Origin:Primary Site: ID Number:Sex:Educational Level:Mother Tongue:Secondary Site(s):Metastases:Pain Etiology:Analgesics:Comments:c:\word\doc\dataguid.doc 910218Date: ^D- 044^11 JMC/srDr. Alseph M. ConnorsChairmanClinical Investigations CommitteeGtA4-4\--83APPENDIX E: CLINICAL INVESTIGATIONS COMMITTEE APPROVALCCA BCCancer Control Agencyof British Columbia600 West 10th AvenueVancouver, B.C., CanadaV5Z 4E6(604) 877-6000Telex 04-507648Fax (604) 872-4596CLINICAL INVESTIGATIONS COMMITTEE (CIC)TO:^DR. MICHELE DESCHAMPSResearch ConsultantDivs. of Nursing and EpidemiologyRE:^C.I.C. DECISION CONCERNING PROPOSED TRIAL:CIC 89-09 STUDY OF PAIN IN THE BRITISH COLUMBIA POPULATIONPrincipal Investigator: Dr. M. DeschampsThis protocol has been granted FULL APPROVAL by the ClinicalInvestigations Committee of the CCABC.When the study has received approval of the UBC Screening Committee, itmay be activated. A copy of the written approval from the UBCScreening Committee should be sent to the Chairman and/or Secretary ofthe CIC when it becomes available.Please submit any future revisions or amendments to this protocol tothe CIC. This will facilitate maintenance of up-to-date records ofeach study. Changes affecting patient eligibility, treatment toxicityor informed consent must be reported to the CIC.FULL APPROVAL FORMUBC approval pendingRev. 12/8684APPENDIX F: ETHICS APPROVAL^ v& 171.161bU %..vIUMDla^ B89-064Office of Research ServicesBEHAVIOURAL SCIENCES SCREENING COMMITTEE FOR RESEARCHAND OTHER STUDIES INVOLVING HUMAN SUBJECTSCERTIFICATE^of APPROVALINVESTIGATOR: Deschamps, M.UBC DEPT:^NursingTITLE:^Study of pain in the British Columbia cancerpopulationNUMBER:^B89-064CO-INVEST:^Bradley, L.^Coldman, A.Hislop, G.APPROVED:^APR 2 6 1989The protocol describing the above-named project has been reviewedby the Committee and the experimental procedures were found to beacceptable on ethical grounds for research involving human subjects.Dr. R.G.C. Johnston, ChairmanBehavioural SciencesScreening CommitteeDr. R.D. pratleyDirector, Research ServicesTHIS CERTIFICATE OF APPROVAL IS VALID FOR THREE YEARSFROM THE ABOVE APPROVAL DATE PROVIDED THERE IS NOCHANGE IN THE EXPERIMENTAL PROCEDURESAPPENDIX G: ETHICS APPROVAL (RENEWAL)The University of British Columbia^B89-064ROffice of Research ServicesBEHAVIOURAL SCIENCES SCREENING COMMITTEE FOR RESEARCHAND OTHER STUDIES INVOLVING HUMAN SUBJECTSCERTIFICATE^of APPROVALINVESTIGATOR: Deschamps, M.UBC DEPT:^NursingTITLE: Study of pain in the British Columbiacancer populationNUMBER:^B89-064RCO-INVEST:^Bradley, L.^Coldman, A.Hislop, G.APPROVED:^89-04-26RENEWED:^92-07-13The protocol describing the above-named project has beenreviewed by the Committee and the experimental procedures werefound to be acceptable on ethical grounds for researchinvolving human subjects.85A .Dr. R.D Sprat eyDirector, Research Servicesand Acting ChairmanTHIS CERTIFICATE OF APPROVAL IS VALID FOR ONE YEARFROM THE ABOVE APPROVAL DATE PROVIDED THERE IS NOCHANGE IN THE EXPERIMENTAL PROCEDURESAPPENDIX H: LETTER TO PARTICIPANTSBRITISH COLUMBIA CANCER AGENCYPATIENT INFORMATION AND CONSENT FOR PARTICIPATIONThe Division of Nursing at the British Columbia Cancer Agency is conductinga study to identify words describing pain which will be used to develop aquestionnaire regarding cancer pain.Your physician at the Agency has been informed that we could ask for yourhelp with this study. Your participation is completely voluntary. Whether ornot you agree to help is entirely up to you and will in no way affect your carenow or later. For this study we would like you to be interviewed forapproximately 30 minutes. During the interview you will be asked to describepain you are experiencing. Although this study will not benefit you directly,we are asking you for a little of your time to help in this project. It is hopedthat this information can be used in the future to help with the assessment ofpain.No ill effects are expected for you or anyone else. All information willbe coded and your name will not be used. The interviews will be tape recordedto facilitate analysis of the data but these recordings will be destroyedimmediately following. All information is confidential and will be seen andlistened to only by members of the project team. If a report of results ispublished, only information about the total group will appear.Should you have any questions relating to this study, please contact Dr.Michele Deschamps at 877-6000, Local 5105.Thank you for your collaboration with this project.86600 West 10th Avenue, Vancouver, B.C.. Canada, V5Z 4E6 Tel. (604) 877-6000 Fax (604) 872-4596APPENDIX I: CONSENT FORMBRITISH COLUMBIA CANCER AGENCYCONSENT FORM*NAME: (Please print)I have read and/or had explained to me theunderstar.d, and agree to participate in this study.to refuse to answer any of the questions and that Iany time, without this affecting my care now or inThis form is being signed in duplicate and Iinformation.content of this form.^II understand that I am freecan end my participation atthe future.am retaining a copy for mySigned:Witness:Date: 600 West 10th Avenue, Vancouver, B.C., Canada, V5Z 4E6 Tel. (604) 877-6000 Fax (604) 872-45968788APPENDIX J: LIST OF DESCRIPTORSAffectiveafraidagitatedangryannoyedannoying^411anxious^2awful 1bad^7better 1bothers^2}botheringbottled^1calm 2care^1carry 1cheerful^1crankiness^2cranky^3crawl 1cross^1cruel 211crycrying^1damn 1depressed^4}depressing^1devastating 2disappointed 11don't^22doesn'tdown 3drain^1drained^2drives 1extremely^1fear^3fearful^2fed^1frustrated^8}frustrating^4frustration^2getting^4good 6great^1hardest^1hell 4helpless^1here 1hopeless^1hurt^2hyper 1irritated^2jittery 3kicking^1like 19living^2lousy 3mean^2mental 1miserable^2mood^1morale^1nervous^1nice^2not 13nuisance^1optimistic^1out^4quiet 1quivery^1rut^1relief 3scaredscary^28}scream^1shitty 1somewheresorrystressstressedsufferedsufferingsuppressingtalktearfultempertensetensionterriblethinkthinkinthrowing^1tired^5tough 1unbearable^1upsets^1upside 1want^4worried^1worry 1worrying^1worst^1worth 1Behaviouralappetite^3backing^1balance^1bend 3breath7breathing^2}careful^1choke 51choked^1dishes^32151321 11 1121141 1491}89disturb^1drink 2earlier^1eat 6 1eating^1favouring^1functioning^1gag^3getting^5jumping^3lay^11laying 1 jlearned^1lift^1lying 1maintain^1move 31moved^1nauseated^1position^8prepared^1quiet^3relax 2rolling^1scream^2sit^61sitting^4sleep 1stand^3still 2stop^21stops 1 jswallow^1talk^2tears 1tensing^1}tension^1tired 41tiredness^1up^18walkwalking^281weight^2winds 1Physiological• Durationall^5always^2consistent^1constant8constantly^1}continuous^1fleeting^1going 7instantaneous 2long^2quick 1sets^1short 1steady^2streak 1temporary^1trickles^1wear 1while^2Physiological• Patternarising^1building^1changes^2comes 13constant^1creeping^1crept 1down^11goes 9jerk^1}jerks3jump^1}jumps 3magnifies^1migrates^1night^1off 3on^4pulsating^2radiated^1}radiates^4ripples^11rippling^11rolling 1running^1slowly 1sneaks^1spasm 41spasms^1spread^1steady 1sudden^2travel 1up^9varies^4Sensory• Intensityanything^7attack^2bad 35big^1blew 1bother^1}bothered^1bothering^1bothers^1calm 1deal 1 so 30 consuming 1eases 1 soreness 12 }contracting 1everexcruciating1010sorersores 1crampcramps311extreme spots 1 crushing 2extremely 31} stand 4 difficultflares 2 stars 1 difficulty111hell 2 strength 1 discomfort 2high 1 tears 5 driveshit 3 terrible 4 driving1}3horrendous 1 tolerable 1 dull 13hurt 8 unbearable 2 electric 2hurting 1 varies 2 expanding 2hurts 12 vengeance 1 fire 1incredible 1 very 18 fleeting 1intense 121 winds 1 fractured 1intensity 2^j worse 141 frostbite 1jolt 1 worst 7 frozen 1little 7 gnawing 2lot 41 Sensory good 3lots 21 •^Quality gripping 1low 1 gout 1mad 1 ache hangover 1magnified 1 aches 221}hard 9medium 2 aching 2 headache 31mild 3 arthritis 4 headaches 11minor 2 asserting 1 heartburn 4more 8 aware 4 hot 6most 2 awful 2 hurt 10much 9 bad 13 hurting 2not 20 band 1 hurts 8painful 11 bothered 1 ice 4quite 3 break 21 indigestion 1really 15 broke 11 jabbing 1scream bruise 2 jam 1screaming4}2 burn1 0}jerksettled 1 burning jerks1}4severe 11 cold 10 jumping 41sharp 3 comfortable 1 jumps 3shooting 2 constricting 1 kicked 29091knife^51^pulling^4knives 1 j pulls11knitting needle 1^pulsating^2knot^21 punched^1knots 2 j^push^3liquid^1 radiated^11monotonous 2^radiates^1 jmuch^6 raw^1muscle^91^razor 2muscular^2 right^20nagging^1^rip 1nails 1 ripped^2needle^1}^ripping^3needles^3 ripples^21nerve 5^rippling^2,1numb^21 scalding^2numbness^3,1^sciatic^21ouch^1^sensitive^1overdone^1 sharp 29overextended 1^shooting^6painful^131 sickness^1painfully^3 j^skinned^1pee^2 slugged^1pinch 21^sore 141pinching^1 j soreness^1 .1pressing^1 1^spark 1pressure^14 stabbing^5prick 11^stiff^1}pricked^11 stiffness^1sting 2stinging 5}stings 1streak 1surges 1tender 3throbbed 1‘throbbing 4throbs 2,through 5tight 5'tighten 1tightening 1,tired 41tiredness 3jtoothache 4ulcers 1unpleasant 2watery 1weight 2worn 2wound 4wretched 1wrong 4off 8 sciatica^11^uncomfortable 3SEXLANGUAGETOTAL ^SEX15^MALE^649^656MALEFEMALE 14 2 16^FEMALETOTAL 23^8 31^TOTALPRES-PAIN_^ 11 I ^12^TOTAL12162193311571012 23517831TOTAL 14AGEENGUSH 16 7 23OTHER 4 4 8TOTAL 20^11 31^_11^^12 TOTAL.^64^^65 TOTAL IOTHER 3LANG11EDUCATIONLANGENGUSH92APPENDIX K: CHI-SQUARE TABLESSEXAGETOTAL ^SEXEDUCATIONTOTAL^64^^65 .^11^I^_^ 12MALE 7 8 15^MALE 9 6 15FEMALE 13 3 16^FEMALE 5 11 16TOTAL 20^11 31^TOTAL 14^17 31Note: x 2 =2.675, p>0.05 Note: x 2 =1.553, p>0.05Note: x 2 =1.790, p>0.05 Note: x 2 =1 059, p>0.05Note: x 2 =0.322, p>0.05 Note: x 2 =0 009, p>0.05PRES-PAINLANG ^11^^12 TOTALENGLISH 15 8 23OTHER 6 2 8TOTAL 21^10 31Note: x 2 =0.005, p>0.0593APPENDIX L: CORRELATION MATRIXAFFECT BEHAV COGNI PHYSIO SEN-SORYAGE -0.333 0.117 -0.170 -0.310 -0.289AGE-GRP -0.346 -0.095 -0.307 -0.285 -0.381EDUC 0.428 0.114 0.283 0.383 0.405EDUC-GRP 0.362 0.142 0.238 0.265 0.345PRS -PAIN 0.044 0.288 0.077 -0.056 0.032PP-GRP -0.088 0.106 0.052 -0.005 0.020DIFF SIMILE AGE AGE-GRP EDUCAGE 0.197 -0.208 1.000AGE-GRP 0.251 -0.249 0.767* 1.000EDUC -0.293 0.282 -0.227 -0.284 1.000EDUC -GRP -0.134 0.329 0.002 -0.158 0.791*PRS -PAIN 0.176 -0.000 0.019 -0.089 -0.104PP-GRP^_ 0.055 -0.121 0.215 0.159 0.008EDUC-GRPPRS-PAINEDUC-GRP 1.000PRS -PAIN 0.037 1.000PP-GRP 0.123 0.825** Not an exact correlation because the variable is continuous on one axis but is grouped on theother.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items