• My File Drawer 
• Journals 
• Featured Journal 
• Journal Supplements 
• Insiders 
• Recommended Reading 

• My CE Planner 
• Browse or Search CE 
• Recommended CE 
• CESaver Collection 
• CESaver Accounts 
• CEConnection Employer Accounts 
• Oncology Times CE Tests 
• Nursing Assistant CE Articles 

• Clinical Updates 
• Drug Updates 
• AHA Connection 
• Focus On: Nursing Ethics 
• Focus On: Diabetes 
• Focus On: Trauma 
• Focus On: Asthma 
• Patient Education
• White Papers

• Free eNewsletters 
• NursingCenter eNews 
• eNews Archives 
• Conferences 
• Career Article Archive 
• Your Guide to Certification 
• How to Get a License 
• Future of Nursing Initiative

• Surveys 
• Puzzlers 

Background: Rheumatoid arthritis (RA) is a chronic illness with a constellation of symptoms. The management of which is important for quality of life. Our review of the literature indicated that there is currently no standardized approach to measuring the symptom status of persons living with RA.

Objective: The purpose of this study is to report the development and initial validation of a structure to measure symptom status in persons living with RA.

Methods: For this secondary analysis, there were 901 female patients with complete symptom checklists available from the Arthritis, Rheumatism, and Aging Medical Information System. A tentative structure using exploratory factor analysis was developed, the structure was replicated in a separate sample using confirmatory factor analysis, and then hypothesized relationships with an external criterion (functional health) was tested using structural equation modeling. The symptom checklist contains 31 symptoms. The stem question is, "Have you had any of the following symptoms during the past 6 months; if yes mark all that apply."

Results: A two-factor structure for measuring symptom status was identified, RA Pain Symptoms and General Symptoms. Using the external criterion, we also demonstrated that the two factors were different and that the RA Pain Symptoms Factor had a stronger impact on functional health. This provides evidence of the discriminant as well as predictive validity of the RA Pain Symptoms Factor.

Discussion: Effective symptom management is an important outcome for nursing practice. Because the assessment of symptoms is the first step in symptom management, the identification of a measurement structure is an essential step.

Rheumatoid arthritis (RA) is a chronic illness with a constellation of symptoms including joint pain, joint swelling, and fatigue. The management of symptoms is important for the quality of life of the patient. Our review of the literature indicated that there is currently no standardized approach (measurement models) for measuring the symptom status of persons living with RA. The purpose of this study is to report the development and initial validation of a structure to measure symptom status in these individuals. We developed a tentative structure using exploratory factor analysis (EFA), replicated the structure in a separate sample using confirmatory factor analysis (CFA), and then tested a hypothesized relationship with an external criterion using structural equation modeling.

The identification and measurement of appropriate patient outcomes is a first step in pursuing quality patient care. For several decades, researchers have been exploring the factors that constitute health-related quality of life (HRQOL) and how they can be measured. Wilson and Cleary (1995) proposed a conceptual HRQOL model that suggests causal links among biological variables, symptom status, functional health, general health perceptions, and overall quality of life. Wilson and Cleary specifically hypothesized that symptom status predicted functional health; therefore, functional health was used as the external criterion in this analysis. The measurement of symptom status in patients living with RA has received little attention in the literature.

Numerous studies have documented the impact of symptom status on patients' HRQOL. Valente, Saunders, and Uman (1993) reported that the number of symptoms experienced was positively correlated with both depression and change toward unhealthy self-care behaviors. Mazonson, Lubeck, Bennett, Fifer, and Fries (1992) found a significant relationship between chronic diarrhea and a decrease in HRQOL. Lubeck and Fries (1993) also found that higher symptom scores were related to lower HRQOL scores. Sousa and Kwok (2004), using data from the AIDS Time-Oriented Health Outcome Study, tested the relationship between symptom status and reported HRQOL. In this study, symptom status was represented by a second-order factor structure, which consisted of one higher order general factor (symptom status) and six lower order factors (malaise/weakness/fatigue, confusion/distress, fever/chills, gastrointestinal discomfort, shortness of breath, and nausea/vomiting). The relationship between symptom status and HRQOL was significant ([beta] = -.61; z = -13.895). Using a parallel process growth modeling approach, Sousa and Kwok (2005) also showed that the rate of change in an individual's symptom status had a substantial negative impact on the rate of change in HRQOL across time. In fact the slope of symptom status accounted for 95% of the variance in the HRQOL slope. According to other researchers, an individual's symptom status plays an important role because it signals a change in functioning and represents a reason for a patient to seek health care (Dodd et al., 2001; O'Neill & Morrow, 2001; University of California, San Francisco, School of Nursing Symptom Management Faculty Group, 1994).

Symptoms refer to (a) sensations or expressions reflecting changes in a person's biopsychosocial functions and (b) a patient's perception of an abnormal physical, emotional, or cognitive state, or the perceived indicators of change in normal functioning as experienced by patients (Henry, Holzemer, Weaver, & Stotts, 1999; Holzemer et al., 1999; Rhodes, McDaniel, & Matthews, 1998; University of California, San Francisco, School of Nursing Symptom Management Faculty Group, 1994). Wilson and Cleary (1995) defined symptom status as a patient's perception of an abnormal physical, emotional, or cognitive state. Hegyvary (1993) suggested that symptoms were perceived indicators of change in normal functioning as experienced by patients.

Symptoms are subjective by nature and are experienced by the individual. Consequently, self-report is considered the gold standard of potential symptom measures. Patients with identical results from physical examination and laboratory tests can report different experiences in terms of such symptoms as pain and nausea. The patient's recognition and interpretation of symptoms are influenced by a host of personal, environmental, and health- or illness-related variables. Personal variables include demographic characteristics, psychological factors, sociological variables, and physiological factors. Environmental variables include sociocultural orientation, beliefs, values, and characteristics of the physical and social settings in which the person lives. Health- or illness-related variables involve perceived health status and presence of disease and risk factors (Dodd et al., 2001; University of California, San Francisco, School of Nursing Symptom Management Faculty Group, 1994).

The assessment of symptoms shifts the focus from specific cells and organs to the individual as a whole (Wilson & Cleary, 1995). Symptoms are an expression of subjective experiences that summarize and integrate data from an array of different sources. In chronic conditions such as RA, symptoms form the criteria for monitoring and evaluating the effectiveness of treatment. Inadequate symptom management can be incapacitating and can affect the individual's ability to function as usual, to perform activities of daily living, and to care for themselves. Uncontrolled symptoms, poor symptom management, may lead to comorbid conditions and increased utilization of health services. The untoward and debilitating consequences of uncontrolled symptoms underscore the importance of adequately managing symptoms.

Effective symptom management has been considered an important outcome for nursing practice and integral to evaluating the efficacy of nursing interventions (Lang & Marek, 1991; Mitchell, Ferketich, & Jennings, 1998). McMahon and Coyne (1989) stated that effective symptom management is important because symptoms and their related emotional distress may lead to delay or could reduce or terminate treatment, which could have a negative impact on optimal health.

The effective management of symptoms begins with a comprehensive and accurate assessment. Symptom management involves (a) identifying and recognizing symptoms, (b) appropriately interpreting the symptoms, (c) monitoring and evaluating the symptoms, (d) selecting strategies to relieve the symptoms that are experienced to enhance the patient's functioning and well-being, and (e) evaluating the effectiveness of these strategies (Sidani, 2003). During previous work developing measurement models to evaluate the Wilson and Cleary (1995) HRQOL model, it became apparent that there is currently no standardized approach for measuring symptom status of persons living with RA.

Measurement begins with the constructs. Constructs representing phenomena of interest cannot be adequately transformed into measurements until the nature of the attribute to be measured is fully specified (Polit & Hungler, 1999). A valuable construct has a theoretical basis, which is disclosed through clear operational definitions involving measured items. From a structural equation perspective, all concepts represented by instruments can be depicted by a measurement model, with the constructs represented by latent variables and the questions represented by measured items.

Nunnally and Bernstein (1994) stated that instruments serve three major functions: (a) establishment of a statistical relationship with a particular variable, (b) representation of a specified universe of content, and (c) measurement of the traits of interest. Corresponding to those three functions, they identified three types of validity: (a) content validity, (b) construct validity, and (c) predictive validity.

Validity is the extent to which an instrument measures the construct of interest (symptom status). In terms of structural equation modeling, Bollen (1989) suggested that validity assesses whether a latent variable relates to the measured items in a way that is consistent with theoretically derived predictions. Nunnally and Bernstein (1994) described three steps to obtain evidence for validity. First is specifying the domains of the measured items [content validity]. Second is determining the extent to which the measured items tend to assess the same thing [construct validity]. Third is performing studies of individual differences and/or controlled experiments to determine the extent to which supposed measured items produce results that are predictable from highly accepted theoretical hypotheses concerning the construct. Predictive validity is the extent to which a scale predicts scores on some external criterion, in this case functional health.

The purpose of this study is to report the development and initial validation of a structure to measure symptom status in patients living with RA. A tentative structure was developed using EFA, the structure was replicated in a separate sample using CFA, and hypothesized relationships were tested with an external criterion using structural equation modeling. This approach mirrors the steps identified by Nunnally and Bernstein (1994).

The Arthritis, Rheumatism, and Aging Medical Information System (ARAMIS) is a national arthritis data resource consisting of longitudinal, clinical data to assess long-term outcomes. The ARAMIS data bank describes the courses (over 25 years) of thousands of patients with rheumatic diseases. The data are collected from two general sources: medical records and patient questionnaires. ARAMIS data are collected with a prospective protocol using standard, defined data collection instruments, and it has been reported extensively (Fries, Spitz, Kraines, & Holman, 1980; Fries, Spitz, & Young, 1982; Ramey, Fries, & Singh, 1996).

For this secondary analysis, there were 901 female patients available. The patients have a diagnosis of RA and their average age is 61.57 years. The sample sizes of the initial sample (Sample 1) and the cross-validation sample (Sample 2) were n₁ = 444 and n₂ = 457, respectively. The descriptive statistics for demographic variables and symptom items are shown in Tables 1 and 2, respectively. The two samples are statistically similar to each other. There were 31 symptom variables in ARAMIS databank. Included in the analysis were only those items with a response rate greater than 90% in the sample, resulting in a total of 19 symptom variables.

TABLE 1. Frequencies, Means, and Standard Deviations of Demographic Variables

TABLE 2. Frequencies of Responses for the 19 Symptoms Used in the Analysis

A hypothesized measurement structure to represent symptom status that would be applicable for RA patients was not identified in the literature review. Consequently, an exploratory approach was used to develop a structure that was cross-validated in a second sample (Thompson, 2004). The first step of this strategy was to divide the data randomly into two samples of approximately equal size. A pseudo-random number generator based on a uniform distribution was used to select the two random samples. Exploratory factor analysis was used to analyze the first sample to determine the number of underlying factors and the relations between the factors and the observed symptom reported items. Then, the model was fitted and selected on the basis of the EFA results to the second sample using CFA [construct validity].

Using functional health as the external criterion, we also validated the final measurement model of symptom status [predictive validity]. To do this, a model comparison procedure originally suggested by Rindskopf (1984) was adapted. In this procedure, two models were compared: Unconstrained model and Constrained model. A chi-square difference test comparing the fit of the constrained and unconstrained models was evaluated (Bentler & Bonett, 1980). All the analyses were conducted by using Mplus 3 (Múthen & Múthen, 2004).

The symptom checklist in the ARAMIS data contains 31 symptoms in 7 areas: (a) general (fever, dizziness); (b) head, eyes, ears nose, mouth, and throat (blurred vision, ringing in ears, hearing difficulties, mouth sores, loss or change in taste, headache); (c) chest, lungs, and heart (chest pain, shortness of breath, wheezing); (d) musculoskeletal (joint pain, joint swelling, muscle pain, muscle weakness); (e) gastrointestinal tract (loss of appetite, nausea, heartburn, pain upper abdomen, pain in lower abdomen, diarrhea, constipation, vomiting); (f) neurologic and psychologic (depression, insomnia, nervousness, trouble thinking or remembering); and (g) skin (easy bruising, hives or welts, itching, rash). The stem question for the checklist is, "Have you had any of the following symptoms during the past 6 months; if yes mark all that apply." The data for each item were coded as 1 = present and 0 = absent.

The Health Assessment Questionnaire-Disability Index (HAQ-DI; Singh et al., 1991) is one of the most frequently used instruments for the evaluation of functional health, the external criterion. The HAQ-DI was developed in the late 1970s under the sponsorship of the Stanford Arthritis Center and has been used in two of their very large studies: ARAMIS, which is ongoing, and the AIDS Time-Oriented Health Outcome Study. The HAQ-DI was developed by assembling questions and components from a variety of instruments (Convery, Minteer, Amiel, & Connett, 1977). The HAQ-DI has evolved over numerous iterations through a series of subjective and objective assessments via statistical evaluation, physician appraisal, and patient feedback (Fries et al., 1980, 1982). The evaluations of the psychometric properties of the HAQ-DI have provided consistent and substantial evidence of both its reliability and validity across many applications and in different patient populations (Ramey et al., 1996). There is consensus that the HAQ-DI possesses face and content validity. Construct and convergent validity, predictive validity, and sensitivity to change have been established also in numerous studies (Ramey et al., 1996; Sousa & Kwok, 2004).

The HAQ-DI assesses the patient's level of functional ability and includes questions about fine movements of the upper extremities, locomotor activities of the lower extremities, and activities that involve both upper and lower extremities. The questionnaire was developed initially to assess functional health in patients with RA. It is self-administered and can be completed within 5 min. There are 20 questions in 8 categories: dressing, rising, eating, walking, hygiene, reach, grip, and usual activities. The stem for each question is: over the past week, "Are you able to…" perform a particular task. Scoring is patterned after the American Rheumatism Association functional classes: normal, adequate, limited, or unable (Steinbrocker, Trager, & Betterman, 1949). Each item has a 4-level difficulty scale that is scored from 0 to 3. Zero represents no difficulty, 1 is some difficulty, 2 is much difficulty, and 3 is unable to do. The highest score in each component determines the score for the category. In previous studies, the HAQ-DI has been shown to have a single common factor with eight functional health variables. Using CFA, Sousa and Kwok (2006) showed that this single-factor model fits the current ARAMIS data set (RMSEA = 0.083, SRMR = 0.028, and CFI = 0.97).

This section will report the results of the EFA, CFA, and validation with an external criterion as it applies to the development and initial validation of a structure to measure symptom status in persons living with RA.

One-factor, two-factor, three-factor, and four-factor models were estimated using Mplus 3 (Múthen & Múthen, 2004). The first five eigenvalues were greater than 1 (6.235, 1.846, 1.436, 1.284, and 1.169). Examination of the scree plot suggested a one-factor or two-factor solution. The two-factor solution produced a general symptoms factor and a second factor on which each of the specific arthritis pain-related symptoms (joint pain, joint swelling, muscle pain, and muscle weakness) had a high loading (>.45). Given that the two-factor solution had a straightforward interpretation, it was selected for further study.

The two-factor structure chosen from exploratory analysis was then cross-validated by conducting a CFA using Sample 2 (n₂ = 457). All of the general symptoms were constrained to load only on Factor 1, which was termed General Symptoms. The four pain-related measures were constrained to load on Factor 2, which was termed RA Pain Symptoms. The model was estimated using weighted least square estimation with robust standard errors and mean adjusted chi-square test statistic (WLSM).The WLSM provides appropriate robust estimates of models that include both continuous and categorical data. The results are summarized in Figure 1. The two-factor model found in Sample 1 was replicated in Sample 2 using CFA.

FIGURE 1. Confirmatory factor analysis of two-factor model (n2 = 457).

Functional health as measured by the HAQ-DI was used as an external criterion for validating the two-factor symptom status structure (General Symptoms and RA Pain Symptoms). As shown in Figure 2(a), the model supporting predictive validity did fit the data. The validity of the two-factor symptom status measurement model was supported by this analysis.

FIGURE 2. Validation of the symptom status two-factor model with functional health as the external criterion and chi-square difference test results comparing the fit of the constrained and unconstrained models.

As part of the validation of the measurement structure, the next step was to show that the two factors were distinct; RA Pain Symptom Factor predicted functional health, over and above the General Symptoms Factor (Crocker & Algina, 1986; Sechrest, 1963; West & Finch, 1997). This analysis would lend further support that the two factors are unique, representing two different aspects of symptom status.

Two models were compared: Unconstrained model-estimating two paths without any constraint (i.e., model [a] in Figure 2), and Constrained model-constraining the path from General Symptoms to functional health and the path from RA Pain Symptoms to functional health to be equal (i.e., model [b] in Figure 2). If the chi-square difference test comparing the fit of the constrained and unconstrained models was statistically significant, this would indicate that the two paths in the model differed in value (Bentler & Bonett, 1980); model (b) suggesting the paths were equal would not fit as well as model (a) with the pathways that were free to vary. The chi-square difference test was significant: [chi]²(1) = 9.921, p = .002, indicating the value of the path coefficient for the RA Pain Symptom Factor (b^ = .45, p < .05) was greater than the value of the path coefficient for the General Symptom Factor (b^ = .20, not significant). The results showed that the General Symptoms Factor and the RA Pain Symptoms Factor were distinct and that the RA Pain Symptoms Factor had a stronger impact on functional health.

These analyses reflected a process previously presented by Nunnally and Bernstein (1994). These analyses addressed the development and initial validation of a structure to measure symptom status in persons living with RA in which three major aspects of validation were examined (a) specifying the domains of the measured items, (b) determining the extent to which the measured items measure the same thing, and (c) determining the extent to which supposed measures of the construct produce results that are predictable from highly accepted theoretical hypotheses concerning the construct.

The EFA analysis suggested a two-factor solution; a general symptoms factor and a specific arthritis pain-related symptoms factor. The two-factor model was replicated using CFA in Sample 2. Using an external criterion, evidence was provided for the discriminant/predictive validity of the RA Pain Symptom Factor as well as the predictive validity of the two-factor structure. One of these factors, RA Pain Symptoms, is a cluster that seems to be specific to RA. This cluster was certainly distinct from the cluster of general symptoms. This specific symptom cluster (RA Pain Symptoms) could potentially be used to assess treatment effectiveness and quality care for patients living with RA. Using this specific cluster provides a focal point for nursing intervention, which would then improve the client's functional health. As a cluster, trends of these symptoms may be used in the assessment of patient care quality. After further evaluation, the RA Pain Symptom cluster may be used as a potential outcome for patients living with RA and its effect on their functional health.

Effective symptom management has been considered an important outcome for nursing practice and integral to evaluating the efficacy of nursing interventions (Lang & Marek, 1991; Mitchell et al., 1998). McMahon and Coyne (1989) stated that effective symptom management is important because symptoms and their related emotional distress may lead to delay or could reduce or terminate treatment, which could have a negative impact on optimal treatment regimens. Because the assessment of symptoms is the first step in the management of symptoms, the measurement structure identified is useful in the process of symptom management.

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606. [Context Link]

Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley. [Context Link]

Convery, F. R., Minteer, M. A., Amiel, D., & Connett, K. L. (1977). Polyarticular disability: a functional assessment. Archives of Physical Medicine and Rehabilitation, 58(11), 494-499. [Context Link]

Crocker, L. M., & Algina, J. (1986). Introduction to classical and modern test theory. New York: Holt, Rinehart, & Winston. [Context Link]

Dodd, M., Janson, S., Facione, N., Faucett, J., Froelicher, E. S., Humphreys, J., et al. (2001). Advancing the science of symptom management. Journal of Advanced Nursing, 33(5), 668-676. [Context Link]

Fries, J. F., Spitz, P. W., Kraines, R. G., & Holman, H. R. (1980). Measurement of patient outcomes in arthritis. Arthritis Rheumatology, 23, 137-145. [Context Link]

Fries, J. F., Spitz, P. W., & Young, D. Y. (1982). The dimensions of health outcomes: The Health Assessment Questionnaire, Disability and Pain Scales. Journal of Rheumatology, 9(5), 789-793. [Context Link]

Hegyvary, S. T. (1993). Issues in outcomes research. Journal of Nursing Quality Assurance, 5, 1-6. [Context Link]

Henry, S. B., Holzemer, W. L., Weaver, K., & Stotts, N. (1999). Quality of life and self-care management strategies of PLWAs with chronic diarrhea. Journal of the Association of Nurses in AIDS Care, 10(2), 46-54. [Context Link]

Holzemer, W. L., Henry, S. B., Nokes, K. M., Corless, I. B., Brown, M., Powell-Cope, G. M., et al. (1999). Validation of the Signs and Symptom Check-List for persons with HIV disease (SSC-HIV). Journal of Advanced Nursing, 30(5), 1041-1049. [Context Link]

Lang, N. M., & Marek, K. D. (1991). Outcomes that reflect clinical practice. Paper presented at the Patient outcomes research: Examining the effectiveness of nursing practice. Proceedings of the State of the Science Conference (pp. 27-38). Bethesda, MD. [Context Link]

Lubeck, D. P., & Fries, J. F. (1993). Health status among persons infected with human immunodeficiency virus-a community-based study. Medical Care, 31(3), 269-276. [Context Link]

Mazonson, P. D., Lubeck, D., Bennett, C., Fifer, S., & Fries, J. (1992). Quality-of-life among HIV patients with chronic diarrhea. Poster session presented at the International Conference on AIDS, Amsterdam. Abstracts 2, B178. [Context Link]

McMahon, K. M., & Coyne, N. (1989). Symptom management in patients with AIDS. Seminars in Oncology Nursing, 5(4), 289-301. [Context Link]

Mitchell, P. H., Ferketich, S., & Jennings, B. M. (1998). Quality health outcomes model. IMAGE: The Journal of Nursing Scholarship, 30, 43-46. [Context Link]

Múthen, L. K., & Múthen, B. O. (2004). Mplus user's guide (V.3.13). Los Angeles, CA: Múthen & Múthen. [Context Link]

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric Theory (3rd ed.). New York: McGraw-Hill. [Context Link]

O'Neill, E. S., & Morrow, L. L. (2001). The symptom experience of women with chronic illness. Journal of Advances in Nursing, 33(2), 257-268. [Context Link]

Polit, D., & Hungler, B. (1999). Nursing research: Principles and methods (6th ed.). Philadelphia: Lippincott. [Context Link]

Ramey, D. R., Fries, J. F., & Singh, G. (1996). The Health Assessment Questionnaire 1995-status and review. In B., Spiker (Ed.), Quality of life and pharmacoeconomics in clinical Trials (2nd ed., pp. 227-237). Philadelphia: Lippincott-Raven Publisher. [Context Link]

Rhodes, V. A., McDaniel, R. N., & Matthews, C. A. (1998). Hospice patients' and nurses' perceptions of self-care deficits based on symptoms experience. Cancer Nurse, 21(5), 312-319. [Context Link]

Rindskopf, D. (1984). Linear equality restrictions in regression and loglinear models. Psychological Bulletin, 96, 597-603. [Context Link]

Sechrest, L. B. (1963). Incremental validity: A recommendation. Educational and Psychological Measurement, 23, 153-158. [Context Link]

Sidani, S. (2003). Symptom Management. In D. M., Doran (Ed.), Nursing-sensitive outcomes: State of the science. Sudbury, MA: Jones & Bartlett. [Context Link]

Singh, G., Fries, J. F., Williams, C. A., Zatarain, E., Spitz, P., & Bolch, D. A. (1991). Toxicity profiles of disease modifying antirheumatic drugs in rheumatoid arthritis. The Journal of Rheumatology, 18(2), 188-194. [Context Link]

Sousa, K. H., & Kwok, O. (2004). Testing the Wilson and Cleary HRQOL conceptual model in persons living with AIDS using structural equation modeling [Abstract]. Paper presented at the International Society for Quality of Life Research Conference, Boston, MA. [Context Link]

Sousa, K. H., & Kwok, O. (2005). Examining the longitudinal impact of symptom status on QOL: A parallel process growth modeling [Abstract]. Quality of Life Research, 14(9), 2018. [Context Link]

Sousa, K. H., & Kwok, O. (2006). Putting Wilson and Cleary to the Test: Analysis of a HRQOL Conceptual Model using Structural Equation Modeling. Journal of Quality of Life Research, 15, 725-737. [Context Link]

Steinbrocker, O., Trager, C. H., & Betterman, R. C. (1949). Therapeutic criteria in rheumatoid arthritis. Journal of American Medical Association, 140, 659-662. [Context Link]

Thompson, B. (2004). Exploratory and confirmatory factor analysis: Understanding concepts and applications. Washington, DC: American Psychological Association. [Context Link]

University of California, San Francisco, School of Nursing Symptom Management Faculty Group. (1994). A model for symptom management. IMAGE: The Journal of Nursing Scholarship, 26, 272-276. [Context Link]

Valente, S., Saunders, J. M., & Uman, G. (1993). Self-care, psychological distress, and HIV disease. Journal of the Association of Nurses in AIDS Care, 4, 15-25. [Context Link]

West, S. G., & Finch, J. F. (1997). Personality measurement: Reliability and validity. In R., Hogan, J., Johnson, & S., Briggs (Eds.), Handbook of personality psychology (pp. 143-164). San Diego, CA: Academic Press. [Context Link]

Wilson, I. B., & Cleary, P. D. (1995). Linking clinical variables with health-related quality of life-a conceptual model of patient outcomes. Journal of American Medical Association, 273(1), 59-65. [Context Link]

Key Words: health-related quality of life; measurement validity; rheumatoid arthritis; symptoms