Development and Testing of a Tool to Assess Self-Care Agency in Adults with Sickle Cell Disease

This article summarizes the psychometric evaluation of the Chronic Illness Assessment Interview for Sickle Cell Disease (CIAI-SCD), an instrument based on a model of self-care for adult patients with chronic medical conditions. The CIAI-SCD was administered to 104 adults with sickle cell disease. A factor analysis identified three factors that reflected the psychological constructs of Personal Satisfaction and Perceived Control (Factor 1), Feeling Concerned and Worried (Factor 2), and Feeling Supported (Factor 3). Preliminary evidence for the internal consistency, test-retest reliability, content validity, and construct validity of the CIAI-SCD was obtained. After further refinement and validation, the CIAI-SCD may be a useful tool for assessing factors related to self-care skills among adults with sickle cell disease. Article: Sickle cell disease (SCD) is a chronic illness characterized by a variety of complications associated with malformations of red blood cells (Reed & Vichinsky, 1998). The most common consequences of SCD are pain and organ damage arising from vaso-occlusive episodes (Reese & Smith, 1997). Whereas SCD was formerly characterized primarily as a childhood illness, advances in treatment have led to expanding survival rates, making adult adjustment to SCD an important health concern (Edwards, Telfair, Cecil, & Lenoci, in press). Interestingly, disease severity accounts for little of the variance in SCD adjustment (Telfair, 1994), suggesting that psychosocial and behavioral factors may play an influential role. In particular, adequate management of such chronic conditions as SCD may require improving self-care to promote optimal health and functioning (Reed & Vichinsky, 1998; Reese & Smith, 1997). Promoting self-care has become a central component of nursing interventions for patients with chronic illnesses (Carter, 1998; Coates & Boore, 1995; Craddock, Adams, Usui, & Mitchell, 1999; Cutler, 2001). Individuals diagnosed with a chronic disease such as SCD have to deal with a complex array of tasks in order to manage their illness and prevent SCD-related complications. Modifications of diet and activity are typically recommended as a means of reducing these symptoms. Actions such as these, which the individual undertakes to improve health and limit disease, are termed self-care and are an important factor in the prognosis for chronic illness (Horsburgh, 1999; Lee, 1999). Self-care refers to the process wherein patients take an active role in their health and well-being, a process that encompasses the skills, attitudes, and abilities needed to master the settings in which individuals with a chronic illness function. (Carter, 1998; Connelly, 1987; Orem, 1995; Von Korff, Gruman, Schaefer, Curry, & Wagner, 1997). Researchers and health professionals have become increasingly interested in self-care because of the growing evidence of its benefits for patients with a chronic disease. Improved self-care practices are related to enhanced quality of life (QOL) and improved health status in medical populations and the general public (Acton & Malathum, 2000; Testa & Simonson, 1996; Von Korff et al., 1997). For example, a retrospective case-controlled study of a self-management education program for asthmatics found significant benefits for those receiving self-care training (Boulet, Boutin, Cote, Leblanc, & Laviolette, 1995). One year after the intervention, the 42 asthma patients showed increased QOL and reductions in emergency department (ED) visits when compared with the control group. Moore, Von Korff, Cherkin, Saunders, and Lorig (2000) prospectively studied 226 patients with chronic low back pain assigned to either usual care or to a self-care promotion group consisting of two 2-hour sessions that included informational materials related to selfmanagement of low back pain. At the 3and 6-month follow-up periods, participants in the self-care promotion group had significantly less pain and physical disability. Finally, another recent study noted diminished costs and improved health status as a function of increased self-care practices among a national sample of older adults (Stearns et al., 2000). Although this study did not include an intervention, the researchers did find significant relationships between reported self-care practices and self-reported health status and health care utilization within a very large sample of elderly adults. In spite of these clear benefits associated with improving self-care, relatively little is known about the variables that might affect self-care practices. A number of factors are presumed to underlie the capacity and willingness to engage in effective self-care behavior (McDonald-Miszczak, Wister, & Gutman, 2001). Some time ago, Catherine Connelly (1987) proposed a model of the process of self-care in chronic illness. She described this model as a modification and extension of the health belief model (HBM) (Rosenstock, 1974), which posits that health behaviors (e.g., self-care) are determined within an expectancy-value framework by comparison of the costs of health behaviors with the value of these actions for reducing the likelihood or severity of an illness. Connelly’s model of self-care practices suggests that the process of self-care is affected by predisposing and enabling factors that she classified to represent a total of seven constructs (i.e., selfconcept, psychological status, health motivation, characteristics of the health regimen, social support, cues to action, and characteristics of the health care system). Self-concept refers to the general sense of one’s own functioning and well-being. Psychological status refers to the typical mood and habitual levels of anxiety and distress of an individual. Health motivation describes a person’s perceptions of the importance of health and an individual’s willingness to perform health care behaviors. Characteristics of the health regimen refers to an individual’s level of involvement in health care planning and the costs and benefits of performing health behaviors. Social support refers to the perceived degree and adequacy of assistance received from others. Cues to action represents perceived manifestations of illness that motivate an individual to act. Finally, characteristics of the health care system includes satisfaction with health care, personal contact with medical care personnel, and accessibility of health services. Each of these factors is presumed to influence the performance of self-care activities. Connelly incorporated these factors into a Chronic Illness Assessment Interview (CIAI), which she suggested could be used as a means of assessing constructs related to self-care practices in a variety of medical conditions. The CIAI was designed as a tool to be used in nursing for the evaluation of personal and environmental characteristics, as perceived by the individual with a chronic illness, that may play a role in shaping self-care practices. The original report by Connelly (1987) was a theoretical paper intended to guide research in the domain of self-care. To date, no psychometric evaluations of the CIAI have appeared in the literature. The purpose of this study was to examine the structure, reliability, and validity of this self-care instrument for adults with SCD. Only by using instruments with demonstrated validity and reliability can relationships between constructs be conclusively demonstrated (Janda, 1998). In summary, self-care training benefits patients and providers by reducing symptoms and costs among individuals with chronic illnesses. Self-care practices in individuals with SCD, however, have rarely been investigated, and we know relatively little about the factors that may play important roles in influencing selfcare. The CIAI (Connelly, 1987) is an instrument designed to measure constructs such as self-concept, health motivation, psychological status, social support, cues to action, and perceived characteristics of the self-care regimen and health care delivery system that are presumed to underlie self-care practices, providing a foundation on which the self-care process rests. The purpose of this study was to investigate the reliability and validity of the CIAI in a sample of adults with SCD. Because self-care is an important aspect of managing SCD symptoms (Reese & Smith, 1997), validation of an assessment instrument for individuals with this condition would be a potentially important step in SCD management. In fact, some recent research has indicated that self-care behaviors and the factors that influence them may vary across chronic illnesses (McDonald-Miszczak et al., 2001). In this study, disease symptoms and illness beliefs predicted self-care behavior among adults with arthritis, whereas perceptions of self-efficacy and general well-being predicted self-care behavior among adults with hypertension. Thus, it may be important to use disease-specific measures when assessing the self-care process in chronic illness. This study is a psychometric evaluation of a modified CIAI, adapted for use with individuals with SCD, henceforth referred to as the CIAI-SCD. A series of hypotheses guided the statistical analyses in this study. First, we hypothesized that the CIAI-SCD would be reliable as measured by estimates of internal consistency and test-retest reliability over the course of 6 months. Second, we hypothesized that the instrument would be content valid as determined by a panel of individuals with expertise in the field of SCD. Third, we hypothesized that factor analysis would show the CIAI-SCD to be composed of factors roughly corresponding to those identified by Connelly (1987). Fourth, regarding construct validity, we hypothesized that the factors of the CIAI-SCD would be related to previously validated measures of related constructs. We utilized measures of self-esteem, self-efficacy, sense of mastery, health-care utilization, pain severity, and satisfaction with medical care in order to assess construct validity. In addition, we hypothesized that CIAI-SCD scores would be related to measures of reported self-care behavior among individuals with SCD as predicted by Connelly’s model.