Abstract
Context
Experts in patient-reported outcome (PRO) measurement emphasize the importance of including patient input in the development of PRO measures. Although best methods for acquiring this input are not yet identified, patient input early in instrument development ensures that instrument content captures information most important and relevant to patients in understandable terms.
Objectives
The M. D. Anderson Symptom Inventory (MDASI) is a reliable, valid PRO instrument for assessing cancer symptom burden. We report a qualitative (open-ended, in-depth) interviewing method that can be used to incorporate patient input into PRO symptom measure development, with our experience in constructing a MDASI module for ovarian cancer (MDASI-OC) as a model.
Methods
Fourteen patients with ovarian cancer (OC) described symptoms experienced at the time of the study, at diagnosis, and during prior treatments. Researchers and clinicians used content analysis of interview transcripts to identify symptoms in patient language. Symptoms were ranked on the basis of the number of patients mentioning them and by clinician assessment of relevance.
Results
Forty-two symptoms were mentioned. Eight OC-specific items will be added to the 13 core symptom items and six interference items of the MDASI in a test version of the MDASI-OC based on the number of patients mentioning them and clinician assessment of importance. The test version is undergoing psychometric evaluation.
Conclusion
The qualitative interviewing process, used to develop the test MDASI-OC, systematically captures common symptoms important to patients with ovarian cancer. This methodology incorporates the patient experience recommended by experts in PRO instrument development.
Keywords: Signs and symptoms, patient-reported outcome assessment, qualitative research, ovarian cancer, M. D. Anderson Symptom Inventory, MDASI
Introduction
A patient-reported outcome (PRO) is “any report of the status of patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else.”1 International experts have issued recommendations for PRO instrument development to support regulatory claims,1,2 but methods for implementing the recommendations are not explicitly stated.2,3 In the past, selection of content and item generation for PROs often has been based on literature review and expert opinion.1 Patient input is sometimes included for psychometric testing purposes and possibly for limited cognitive debriefing, but typically only once instrument content has been defined. Experts recommend that content development and item generation include open-ended, qualitative individual or group interviews with members of the target patient population to ensure that the concept fully reflects the patient perspective and that items are relevant to its appraisal.1,2 Experts further recommend qualitative techniques to confirm the content validity of existing measures that were not developed with early patient input.2 Techniques for scientifically rigorous qualitative research may be unfamiliar to many instrument developers.
The effect of disease-related and treatment-related symptoms on the ability of patients to function as they did before diagnosis has been recognized as an additional burden to the patient and is termed “symptom burden.”4 Researchers at The University of Texas M. D. Anderson Cancer Center have developed a questionnaire (the M. D. Anderson Symptom Inventory [MDASI]) that measures the symptom burden of patients with cancer and includes symptoms that are common for patients with any type of cancer.5 Disease-specific versions of the MDASI have been developed;6–14 the ovarian cancer (OC) version is under development. Here, we describe qualitative research methods that were used to identify symptoms of importance to patients with a specific cancer, using the development of an item pool for the MDASI-OC as an example. Subsequent steps in the development of this instrument will include standard components of psychometric validation, cognitive debriefing, item reduction, and tests of sensitivity to changes in treatment and disease status.
Methods
The MDASI
The MDASI is a multisymptom PRO containing 13 core symptom items that have the highest frequency and/or severity in patients with a range of cancer types, and six items relating to the interference of symptoms with daily life.15 Disease-specific modules are available for use with patients with various cancers and receiving different treatments.6–14 Patients rate the severity of each symptom or the level of interference caused by all symptoms in the previous 24 hours on a numerical rating scale from 0 to 10, where “0” is “not present” or “no interference” and “10” is “as bad as you can imagine” or “complete interference.”15 The MDASI can be completed by paper-and-pencil self-report, interview, telephone-based automated interactive voice response system, or web-based or electronic tablet-based applications.15
Qualitative Interviews
To develop a MDASI-OC module, patients with OC who were receiving treatment in the outpatient Gynecologic Oncology Center at M. D. Anderson were invited to participate in qualitative interviews. Patients were selectively recruited and interviewed to gain information from a representative cross-section of the population of patients with OC. Prior to the start of the study, characteristics that might influence symptom burden in women with OC (e.g., age, disease stage, length of time since diagnosis, and types of treatment) were identified. No minimum sample size was defined; patients were recruited for interviewing until saturation of the data was reached (i.e., descriptions became increasingly redundant).16 The instrument development protocol was approved by the M. D. Anderson Institutional Review Board. Informed consent was obtained from all patients participating in the interviews.
An experienced qualitative researcher trained staff to conduct qualitative interviews. After initial didactic instruction in the fundamentals of qualitative interviewing, the interviewers observed and then conducted mock interviews. Once they were able to satisfactorily conduct a mock interview, they observed and then conducted actual research interviews. Throughout the study, the experienced researcher periodically monitored the performance of the interviewers and offered critiques to improve and maintain their interviewing skills.
In-depth interviews began with a question about present symptoms, such as “Tell me about symptoms you are experiencing now.” Subsequent questions referred to symptoms patients experienced when first diagnosed and during any treatment or surgeries. Additional probe questions were asked to clarify comments and to ensure that the interviewer fully understood the patient’s symptom description. Throughout the interview, patients were prompted to recall any symptoms they may have forgotten to mention (e.g., “Were there any other symptoms you were having at the time?”). All interviews were digitally recorded and transcribed verbatim for analysis.
After each interview, patients completed the 13 symptom items and the six interference items of the core MDASI. Patients were asked to describe any of the symptoms in the MDASI they had experienced if this had not been specifically mentioned during the interview.
Demographic, disease, and treatment data were collected from each patient’s medical record and analyzed using descriptive statistics to delineate the sample characteristics.
Subject Recruitment
Qualitative research uses purposive sampling to ensure a range of subjects that represent the target patient population; in this case, women with OC who were receiving or had received treatment. Recruitment was based on target percentages of patients to represent each characteristic identified as potentially influencing symptom burden in women with OC (e.g., age, disease stage, length of time since diagnosis, and types of treatment).
Analysis
Three researchers (L.A.W., S.A., A.K.S.) independently reviewed each interview transcript, and each researcher compiled a list of symptoms mentioned by the patients. Exact patient quotes representing each symptom also were extracted into separate electronic documents. The researchers met to compare the lists and compile a single list to accurately and comprehensively capture all symptoms described by the patients. No qualitative analysis software was used.
Wording consistent with the patients’ descriptions of the symptoms was used so that the final items in the instrument were in terms patients could understand. The number of patients who mentioned each symptom was counted to provide information about the frequency and typicality of the symptoms.17 The results of the qualitative interviews also were compared with the answers patients provided on the MDASI questionnaire. The interviews were continued until all three researchers felt that they had not heard any new or significant symptoms mentioned in the previous three interviews.
Initial Item Reduction
An expert panel composed of a gynecologic oncologist (D.C.B.), a gynecologist (S.A.), and an ovarian cancer researcher (C.C.S.) reduced the number of items in the symptom list. Item elimination was based on the following parameters: 1) determining items of least relevance on the basis of the number of patients mentioning the symptom and the OC experts’ judgment; 2) removing any items that were objectively measurable during clinical assessment (e.g., weight gain/loss); and 3) combining items that used different wording to describe the same symptom. Appropriate wording for items developed by combining symptoms was determined by referring to interview transcripts for patient terminology. Items were retained if they were MDASI core items or were consistently reported by at least 20% of patients. Some frequently reported items also were eliminated if they were known to be unrelated to OC and its treatment. The final list became the test version of the MDASI-OC. This version underwent psychometric testing and cognitive debriefing (reported elsewhere) to determine which items to retain for the final MDASI-OC module.
We confirm credibility of the results of the qualitative analysis during cognitive debriefing conducted along with the psychometric testing of the PRO instrument. A subset of the first patients completing the instrument (usually 20 to 40) are queried about the understandability of the questions and are asked if they believe that any important symptoms are missing from the instrument. If a symptom is identified as difficult to understand or missing by at least three patients, interviews and analysis are reviewed again to determine if the symptom was missed, misnamed, or misclassified in the initial analysis. A group of researchers and clinicians meet to determine if the symptom should be renamed or added for additional testing. If necessary, additional qualitative interviews may be conducted to further clarify the importance of the symptom to the patient experience.18
Results
For the MDASI-OC, interviews were conducted with 14 patients between March 31 and May 7, 2010 (Table 1). Only one participant had an active chronic comorbid condition (diabetes mellitus treated with oral medication). Mean (standard deviation) self-rated health-related quality of life for the previous week was 8.3 (2.1 [range, 2–10]), on a scale from 0 (bad) to 10 (good).
Table 1.
Demographic and Clinical Characteristics of Patients with Ovarian Cancer Who Participated in the Qualitative Interviews (N = 14)
| Characteristic | Mean (SD; range) |
|---|---|
| Age in years | 57.4 (9.0; 40–69) |
| Years since ovarian cancer diagnosis | 2.9 (2.8; 0–9) |
| n(%) | |
| Race/ethnicity | |
| White non-Hispanic | 13 (92.9) |
| Black non-Hispanic | 1 (7.1) |
| Marital status | |
| Married | 11 (78.6) |
| Not married | 3 (21.4) |
| Employment status | |
| Full- or part-time work outside the home | 5 (35.7) |
| Not employed outside the home | 9 (64.3) |
| Education | |
| ≤Grade 12 | 4 (28.6) |
| >Grade 12 | 10 (71.4) |
| Pathologic diagnosis | |
| High-grade papillary serous carcinoma | 14 (100) |
| Cancer stage at diagnosis | |
| Stage II | 1 (7.1) |
| Stage III | 11 (78.6) |
| Stage IV | 1 (7.1) |
| Unknown | 1 (7.1) |
| Previous treatments | |
| Chemotherapy regimens, 1–11a Cisplatin-paclitaxel (n = 11) Paclitaxel alone (n = 3) Carboplatin alone (n = 2) Topotecan alone (n = 2) Liposomal doxorubicin-aflibercept (n = 2) Other b (n = 10) Intraperitoneal cisplatin-paclitaxel (n = 2) |
13 (92.9) |
| Surgery more than 6 weeks ago | 13 (92.9) |
| Radiotherapy | 2 (14.3) |
| Current treatment | |
| Chemotherapy4, 3–10 cycles Carboplatin or cisplatin (n = 3) Gemcitabine (n = 3) Paclitaxel (n = 2) Otherb(n = 4) |
7 (50.0) |
| Surgery in last 6 weeks | 1 (7.1) |
| Radiotherapy | 0 (0.0) |
| Current disease status | |
| Complete response | 7 (50.0) |
| Partial response | 3 (21.4) |
| Stable disease | 2 (14.3) |
| Progression/recurrence | 2 (14.3) |
| Not applicable/not available | 8 (57.1) |
| Current ECOG performance status | |
| Grade 0 (fully active) | 9 (64.3) |
| Grade 1 (restricted but ambulatory) | 3 (21.4) |
| Grade 2 (ambulatory, unable to work) | 2 (14.3) |
SD = standard deviation; ECOG = Eastern Cooperative Oncology Group.
Some patients received more than one previous regimen.
Received by only one patient.
Some patients received more than one drug.
Qualitative Interview Results
Initial qualitative interviewing identified 42 separate symptoms that were mentioned by at least one participant; these included 12 of the 13 MDASI core items (excluding dry mouth). Eighteen symptoms were mentioned by at least 20% of patients (Table 2), with the most common being fatigue and pain. The participants also mentioned ways in which symptoms affected their day-to-day functioning (Table 3).
Table 2.
Symptoms Reported by ≥20% of Participants with Ovarian Cancer During Qualitative Interviews(N= 14)
| Symptom | n(%) |
|---|---|
| Fatigue | 13 (92.8) |
| Pain | 12 (85.7) |
| Abdominal pain | 9 (64.3) |
| Abdominal fullness/bloating | 9 (64.3) |
| Nausea | 8 (57.1) |
| Distress | 6 (42.9) |
| Neuropathy | 6 (42.9) |
| Disturbed sleep | 5 (35.7) |
| Lack of appetite | 5 (35.7) |
| Constipation | 5 (35.7) |
| Leg cramps/leg muscle pain | 4 (28.6) |
| Pain in the side | 4 (28.6) |
| Hair loss | 4 (28.6) |
| Feelings of sadness | 3 (21.4) |
| Vomiting | 3 (21.4) |
| Urinary urgency/frequency | 3 (21.4) |
| Back pain | 3 (21.4) |
| Eye problems | 3 (21.4) |
Table 3.
Quotes from Qualitative Interviews Illustrating Findings
| Symptom | Sample Quote |
|---|---|
| Fatigue | I was feeling very tired…. I didn’t have the energy that I was used to having. 66-year-old |
| Pain | It hurt to sit…. I was in a lot of pain at that point, just very uncomfortable. 59-year-old |
| It hurts. My feet are very sensitive. When I wear … shoes, it burns…. And that actually goes up into my legs. My legs, my muscles, are kind of … sore. 60-year-old | |
| Abdominal pain | I had pain in my abdomen on one side, kind of piercing pain. And it was intermittent. 59-year-old |
| Abdominal fullness/bloating |
I kept complaining about feeling bloated and then the fullness that you could see. 59-year-old |
| Interference | Sample Quote |
| General activities | It’s a big interruption in your life. 66-year-old |
| Relations with others | It (fatigue) prevents me from doing the things that I want to do with my grandchildren. 66-year-old |
| Work | Two months ago, I experienced cramping in my hand and that, because I’m an artist, did get in the way of holding tools for an extended period of time. 59-year-old |
| Symptom Occurrence | Sample Quote |
| At diagnosis | I was fatigued before this was diagnosed. I just got tired real easy, and I didn’t know why. 69-year-old |
| I had a period of abdominal pain…. I had my period, (and) about a week later, I’d have 3 or 4 days of discomfort right around my pubic bone, but it would go away. That had started in January. When it happened again in March, I’m like, “That’s not normal.” 45-year-old | |
| The pain was very severe for a couple of months, and I didn’t go to a doctor…. I thought maybe I had pulled a muscle…. So I didn’t really pay any attention to it…. And then it started moving down more towards my abdomen area, and then it kind of went to my right side…. I went to the doctor, and they gave me some muscle relaxers. And he said this should cure my pain. Well, it didn’t. 54-year-old | |
| Chemotherapy | With the chemo I’m getting now, … (I get) flu-like symptoms and a little nausea….I get it (chemo) on Thursday, on Friday and Saturdays and Sundays I am a little bit fatigued and feeling a little bit of nausea. It goes away by Monday or Tuesday…. Almost 6 days later I’m totally fine. 63-year-old |
| Radiotherapy | I had a very difficult time all summer with vomiting and diarrhea and really a very hard time getting over that radiation. 66-year-old |
| Surgery | There was pain because, of course, it was an abdominal surgery. 54-year-old |
General MDASI Responses
Symptoms reported to be most severe in the last 24 hours on the MDASI were disturbed sleep, fatigue, problems with remembering things, being distressed, and feeling sad (Table 4). Patients reported that symptoms interfered most with work, general activities, and walking but had little impact on their relations with others.
Table 4.
Results of the MDASI Questionnaire Completed by Patients With Ovarian Cancer Undergoing Qualitative Interviewing (N= 14)
| MDASI Items | Mean (SD) Score |
Patients Reporting Each Level of Severity, n | |||
|---|---|---|---|---|---|
| None | Mild (Scores 1–4) |
Moderate (Scores 5–6) |
Severe (Scores 7–10) |
||
| Symptom Items | |||||
| Disturbed sleep | 3.0 (2.9) | 4 | 6 | 1 | 3 |
| Fatigue | 2.9 (3.0) | 2 | 9 | 0 | 3 |
| Problems remembering | 2.3 (2.0) | 4 | 8 | 2 | 0 |
| Feeling sad | 2.2 (1.8) | 3 | 9 | 2 | 0 |
| Distress | 2.2 (2.6) | 5 | 6 | 2 | 1 |
| Feeling drowsy | 2.0 (3.0) | 7 | 5 | 0 | 2 |
| Lack of appetite | 2.0 (3.2) | 8 | 3 | 1 | 2 |
| Dry mouth | 1.7 (2.2) | 6 | 6 | 1 | 1 |
| Numbness | 1.6 (2.7) | 9 | 3 | 1 | 1 |
| Pain | 1.6 (2.2) | 7 | 4 | 3 | 0 |
| Nausea | 1.1 (2.8) | 11 | 2 | 0 | 1 |
| Shortness of breath | 1.0 (2.7) | 10 | 3 | 0 | 1 |
| Vomiting | 0.7 (2.7) | 13 | 0 | 0 | 1 |
| Interference Items | |||||
| Work | 2.4 (3.8) | 7 | 4 | 0 | 3 |
| General activities | 2.2 (2.9) | 6 | 5 | 2 | 1 |
| Walking | 2.1 (3.0) | 7 | 3 | 2 | 2 |
| Mood | 2.1 (3.0) | 7 | 4 | 1 | 2 |
| Enjoyment | 1.7 (2.7) | 7 | 4 | 2 | 1 |
| Relations with others | 0.7 (1.4) | 9 | 4 | 1 | 0 |
MDASI = M. D. Anderson Symptom Inventory module for ovarian cancer; SD = standard deviation.
Expert Panel Review and Test MDASI-OC Development
The expert panel review resulted in the inclusion of 21 symptom severity items—the 13 MDASI core symptom items and eight test items (Table 5)—into the psychometric testing version of the MDASI-OC, in addition to the six interference items from the core MDASI. These 21 symptom items include one MDASI core item (dry mouth) that was not mentioned in the initial interviews but that was identified by eight participants (54%) when they completed the MDASI after the interviews.
Table 5.
Symptoms to Be Included in the Test M. D. Anderson Symptom Inventory Module for Ovarian Cancer
| Symptom |
|---|
| M. D. Anderson Symptom Inventory core items |
| Pain |
| Fatigue |
| Nausea |
| Disturbed sleep |
| Distress |
| Shortness of breath |
| Trouble remembering |
| Lack of appetite |
| Feeling drowsy |
| Dry mouth |
| Feeling sad |
| Vomiting |
| Neuropathy |
| Ovarian cancer test items |
| Abdominal pain |
| Abdominal fullness/bloating |
| Constipation |
| Problem paying attention |
| Urinary urgency/frequency |
| Pain or burning with urination |
| Back pain |
| Leg cramps or leg muscle pain |
Three symptoms were mentioned by more than 20% of the participants in the interviews but were not included in the test instrument. Pain in the side was mentioned by four participants (29%) but coincided with pain in the abdomen in three of these cases, suggesting that pain in the side was not a separate item. Hair loss was mentioned by four patients as a side effect of treatment but was not included in the list of symptoms for testing because it is an observable sign rather than a symptom. Eye problems were mentioned by three participants (21%) but were not included in the test instrument because two of the three participants had eye conditions unrelated to their disease or treatment.
One test item (pain or burning with urination) was identified by only one (7.1%) participant but was included in the test instrument because clinicians believed it was an important symptom to include. One additional item (problem with paying attention) was added on the basis of clinician judgment.
Discussion
Although experts recommend qualitative patient input for the development of PRO measures and for content confirmation of existing measures, there are currently no recognized best practices for this process.1,2 Here, we describe a methodology we have used in the generation and confirmation of patient-reported symptom assessment measures.
The MDASI was initially developed without patient input for content description or item generation.5 Patient input on the relevance and acceptability of the MDASI core items was obtained in subsequent module development, such as the MDASI–lung cancer, the MDASI–gastrointestinal cancer, the MDASI–chronic myeloid leukemia, and the MDASI–chronic graft-versus-host-disease instruments.8,10–12 In confirming the core MDASI items and developing disease-specific and treatment-specific modules, we have developed a process of qualitative interviewing and analysis that is practical to use, comprehensively identifies issues of importance to patients, and generates items that are both understandable to patients and sensitive in measuring symptom burden.
The choice of interview techniques is guided by the goals of the interviews, the target population, and the feasibility of conducting interviews with the particular sample. Qualitative interviews can be conducted either in focus groups or individually. Focus groups require less researcher time, as four to six participants can be interviewed at the same time. These interviews usually take one to two hours to conduct and require an interviewer skilled in managing group dynamics. Focus groups are especially good for reaching consensus on questions. It can be difficult, however, to schedule a convenient time for all participants, and focus groups can sometimes miss issues that are significant to quieter participants or that are about sensitive topics. Individual interviews usually take 30 to 60 minutes, can explore individual patient issues and concerns in depth, and can include topics of a sensitive nature.
At our large cancer center, we have found that individual interviews are more feasible because they are more easily incorporated into the patient’s schedule. Although individual interviews can be conducted by phone, this method requires a more skilled interviewer to ensure that adequate rapport is developed and runs the risk of a participant getting distracted during the call.
Common ways of conducting and analyzing qualitative interviews are based on Phenomenology or Grounded Theory.3,19 However, we find it effective to have participants describe their disease and treatment experience as a story, starting with the present and then going back and describing what happened in the past. This method allows participants to think about and relate their experiences in a manageable way.21
Our experience has shown that a minimum of 10 interviews is required to ensure redundancy in participant responses, but the complexity of the condition and heterogeneity of the target population may necessitate more interviews. Prior to starting the interviews, we estimate how many participants we will need based on the condition and target population. We then interview the estimated number of participants. If redundancy is not found, we continue interviewing in increments of five participants. For the OC population, we initially estimated needing 15 participants, and we recruited this number. One declined to participate after initially consenting. Analysis of interviews from 14 participants identified redundancy.
Although a variety of analytical methods exist for qualitative research, we chose a modified descriptive exploratory analysis for symptom measure development.20 Each researcher independently identifies and names symptoms in patients’ terms by reviewing interview transcripts and selecting quotes that describe each symptom. Together the researchers compare lists of symptoms, resolve differences, make decisions on combining similar symptoms, and reach consensus on a final list. If symptom names are questioned, the researchers return to participant quotes to determine the symptom name.
The qualitative interview transcripts are a rich source of data about participant experiences. After meeting the primary goal of identifying symptoms, the researchers review the transcripts again to identify additional themes that may have implications for symptom assessment. The participants in our study frequently mentioned changes in symptoms over time based on disease status and treatment and the interference of symptoms with daily activities. Thus, our interviews confirm the importance of developing methods that are brief and easy to complete multiple times to monitor changes and assess symptoms associated with OC and treatment. We also identified the need to include symptom interference with daily activities along with symptom incidence and severity in an overall measure of symptom burden.
Qualitative interviewing elicits more items than is practical or necessary to include in an instrument. The quality, completeness, and reliability of data can be compromised by a PRO instrument that is too long or imposes undue physical, emotional, or cognitive strain on patients.1 Therefore, we used an expert panel to reduce the list of items generated from the interviews to a reasonable number to be included in the instrument for psychometric assessment. Expert panels, which may evaluate the relevance of items by formal rating scales or by discussion, may include physicians, nurses, other health care providers, patients with the disease, and family/caregivers of patients.
Psychometric testing of the MDASI-OC, to be reported separately, uses statistical techniques to evaluate the sensitivity and incidence of the OC-specific symptoms in a large number of patients with OC. Patients of all disease stages, ages, educational levels, races, ethnicities, performance status, and those receiving different therapies will be tested. Items that are not sensitive to changes in disease status, treatments, or performance status will be eliminated, as will items with a very low incidence of occurrence. The remaining items, along with the core MDASI items, will comprise the final validated version of the MDASI-OC.
Limitations
Participants selected for this study primarily had stage III disease to ensure that descriptions of therapy-related symptoms were obtained. Although these participants were questioned about symptoms they experienced early in the course of their disease, the interview responses may not have been completely representative of earlier-stage disease. The basic validity and reliability of a PRO instrument should be assessed as it is used with different patient groups. Any or all validation steps may need to be repeated and the instrument revised if it is discovered that performance is suboptimal or major changes have occurred in treatments or disease trajectories.
Conclusions
Qualitative research is an essential element of PRO development, as recommended by international experts.1,2 It is frequently neglected because expertise in qualitative research is unavailable and because such research is perceived as being subject to bias.22 It is important to establish clear practices and a methodology to standardize the data-gathering process, documentation of findings, and analytical methods, as is done in quantitative research.3,22 In this article, we describe a method of symptom-item generation for patients with a specific disease that seems to be consistent with expert recommendations1,2 and best practice in qualitative research,22 and that reflects the evolution of experience, expertise, and process improvement of the M. D. Anderson Cancer Center researchers involved in the development of symptom assessment measures. The methodology used in the initial steps of the MDASI-OC development will result in a valid and reliable PRO instrument that will add to the inventory of disease-specific PRO tools available to assess symptom burden in patients with cancer.
Acknowledgments
Disclosures and Acknowledgments
The project described was supported in part by the M. D. Anderson Cancer Center Support Grant CA016672 from the National Institutes of Health. Partial salary support for Sonika Agarwal was provided by grant funding from the Hawn Foundation, Dallas, Texas. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.
Support for third-party writing assistance was provided by Genentech, Inc. Diane C. Bodurka is a consultant for Genentech. Charles S. Cleeland is a consultant for Genentech, Amgen, and Exelixis.
The authors acknowledge the help and support of Elaine Yu, PharmD, Genentech, Inc., South San Francisco, CA, and Stephanie Wendlberger, CodonMedical, San Bruno, CA, in the drafting and editing of this manuscript; and Ibrahima Gning, DrPH, The University of Texas M. D. Anderson Cancer Center, Houston, TX, for the quantitative data analysis.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.U.S. Food and Drug Administration. Guidance for industry. [Accessed October 26, 2011];Patient-reported outcome measures: use in medical product development to support labeling claims. Available from http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM071975.pdf.
- 2.Rothman M, Burke L, Erickson P, et al. Use of existing patient-reported outcome (PRO) instruments and their modification: the ISPOR good research practices for evaluating and documenting content validity for the use of existing instruments and their modification PRO task force report. Value Health. 2009;12:1075–1083. doi: 10.1111/j.1524-4733.2009.00603.x. [DOI] [PubMed] [Google Scholar]
- 3.Lasch KE, Marquis P, Vigneux M, et al. PRO development: rigorous qualitative research as the crucial foundation. Qual Life Res. 2010;19:1087–1096. doi: 10.1007/s11136-010-9677-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cleeland CS. Symptom burden: multiple symptoms and their impact as patient-reported outcomes. J Natl Cancer Inst Monogr. 2007;37:16–21. doi: 10.1093/jncimonographs/lgm005. [DOI] [PubMed] [Google Scholar]
- 5.Cleeland CS, Mendoza TR, Wang XS, et al. Assessing symptom distress in cancer patients: the M. D. Anderson Symptom Inventory. Cancer. 2000;89:1634–1646. doi: 10.1002/1097-0142(20001001)89:7<1634::aid-cncr29>3.0.co;2-v. [DOI] [PubMed] [Google Scholar]
- 6.Fadol A, Mendoza T, Gning I, et al. Psychometric testing of the MDASI-HF: a symptom assessment instrument for patients with cancer and concurrent heart failure. J Card Fail. 2008;14:497–507. doi: 10.1016/j.cardfail.2008.01.012. [DOI] [PubMed] [Google Scholar]
- 7.Gning I, Trask PC, Mendoza TR, et al. Development and initial validation of the thyroid cancer module of the M. D. Anderson Symptom Inventory. Oncology. 2009;76:59–68. doi: 10.1159/000178809. [DOI] [PubMed] [Google Scholar]
- 8.Mendoza TR, Wang XS, Lu C, et al. Measuring the symptom burden of lung cancer: the validity and utility of the lung cancer module of the M. D. Anderson Symptom Inventory. Oncologist. 2011;16:217–227. doi: 10.1634/theoncologist.2010-0193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rosenthal DI, Mendoza TR, Chambers MS, et al. Measuring head and neck cancer symptom burden: the development and validation of the M. D. Anderson Symptom Inventory, Head and Neck Module. Head Neck. 2007;29:923–931. doi: 10.1002/hed.20602. [DOI] [PubMed] [Google Scholar]
- 10.Wang XS, Williams LA, Eng C, et al. Validation and application of a module of the M. D. Anderson Symptom Inventory for measuring multiple symptoms in patients with gastrointestinal cancer (the MDASI-GI) Cancer. 2010;116:2053–2063. doi: 10.1002/cncr.24920. [DOI] [PubMed] [Google Scholar]
- 11.Williams LA, Cortes JE, Ault PS, et al. Including patient input in patient-reported outcome instrument development: example from the MDASI-CML [abstract 2573]; Presented at the American Society of Hematology Annual Meeting; December 4–7, 2010; Orlando, FL. [Google Scholar]
- 12.Williams LA, Couriel DR, Mendoza TR, et al. A new measure of symptom burden in chronic graft-versus-host disease. Biol Blood Marrow Transplant. 2010;16(Suppl 2):S177. [Google Scholar]
- 13.Armstrong TS, Cohen MZ, Eriksen L, Cleeland C. Content validity of self-report measurement instruments: an illustration from the development of the Brain Tumor Module of the M. D. Anderson Symptom Inventory. Oncol Nurs Forum. 2005;32:669–676. doi: 10.1188/05.ONF.669-676. [DOI] [PubMed] [Google Scholar]
- 14.Armstrong TS, Gning I, Mendoza TR, et al. Reliability and validity of the M. D. Anderson Symptom Inventory-Spine Tumor Module. J Neurosurg Spine. 2010;12:421–430. doi: 10.3171/2009.10.SPINE0943. [DOI] [PubMed] [Google Scholar]
- 15.Cleeland CS. The M. D. Anderson Symptom Inventory user guide. Houston, TX: University of Texas M. D. Anderson Cancer Center; 2010. [Accessed July 30, 2012]. Available from http://www.mdanderson.org/education-and-research/departments-programs-and-labs/departments-and-divisions/symptom-research/symptom-assessment-tools/MDASI_userguide.pdf. [Google Scholar]
- 16.Sandelowski M. Sample size in qualitative research. Res Nurs Health. 1995;18:179–183. doi: 10.1002/nur.4770180211. [DOI] [PubMed] [Google Scholar]
- 17.Sandelowski M. Real qualitative researchers do not count: the use of numbers in qualitative research. Res Nurs Health. 2001;24:230–240. doi: 10.1002/nur.1025. [DOI] [PubMed] [Google Scholar]
- 18.Lincoln YS, Guba EG. Naturalistic inquiry. Newbury Park, CA: SAGE Publications; 1985. [Google Scholar]
- 19.Creswell JW. Qualitative inquiry and research design: choosing among five traditions. 1st ed. Thousand Oaks, CA: SAGE Publications; 1997. [Google Scholar]
- 20.Smith MJ, Liehr P. The theory of attentively embracing story. In: Smith MJ, Liehr P, editors. Middle range theory for nursing. 1st ed. New York: Springer; 2003. pp. 167–187. [PubMed] [Google Scholar]
- 21.Parse RR, Coyne AB, Smith MJ. Nursing research: qualitative methods. Bowie, MD: Brady Communications; 1985. [Google Scholar]
- 22.Brod M, Tesler LE, Christensen TL. Qualitative research and content validity: developing best practices based on science and experience. Qual Life Res. 2009;18:1263–1278. doi: 10.1007/s11136-009-9540-9. [DOI] [PubMed] [Google Scholar]