America is facing an epidemic of chronic illness, with more than half of the population suffering from at least one chronic condition. As the population ages, this problem will be exacerbated by increasing multimorbidity. The failure of the American health care system to meet the challenge of preventing and managing chronic illness has been well documented by the Institute of Medicine.1 Wagner et al2 articulated the Chronic Care Model to provide a framework for redesigning practice to incorporate elements needed to improve chronic illness care. Widespread adoption of evolving practices will require that educational strategies for future physicians and other health professionals be developed and evaluated. Interprofessional team care has been demonstrated to improve care, but evidence in support of interprofessional education remains sparse.3
Medical and nursing students do not have consistently positive views of caring for patients with chronic illness, including disability and mental illness.4–6 Currently, medical and nursing student curricula include significant didactic and clinical exposure to chronic illnesses, but limited explicit instruction in core principles of chronic disease prevention and management, including team-based care or patient self-management. Because interprofessional team care is a critical element of a redesigned health care system,1 effective evaluation tools must address the impact of professional culture and educational interventions on physicians and other team members.
The purpose of this study was to develop an instrument and conduct a baseline assessment by measuring attitudes toward chronic illness among graduating and matriculating medical and nursing students before implementing changes in the curriculum, based on the Wagner Chronic Care Model.2
The university’s institutional review board determined that this study was exempt from full review.
A questionnaire was developed to measure students’ attitudes and beliefs about chronic illness. One of the authors (S.K.F.) prepared the first draft on the basis of a review of recent articles4,7 and discussions about chronic illness with faculty and residents in family medicine and internal medicine. She invited them to talk about their experiences with chronic illness and to express in their own words their personal thoughts about their patients. The draft was edited by a panel of faculty content experts in family medicine, internal medicine, geriatric medicine, nursing, pharmacology, and occupational therapy. The draft was pilot tested with a sample of 10 students and 24 residents, and it was revised using item-analysis procedures. The final version of the four-page questionnaire, the Chronic Illness Survey, included five short-answer and 35 Likert-scale items. It was administered to medical and nursing students who graduated in the spring of 2007, and it was subsequently administered to medical and nursing students matriculating in the fall of 2007.
Differences in frequency distributions were evaluated using χ2. Factor analysis was used to investigate relationships among the 35 Likert-scale items. Factor scores were calculated by summing the Likert responses (one to five) after inverting the responses for negatively worded items and standardizing scores to a mean of 50 and standard deviation of 10. t Tests were used to evaluate the effect of type of degree on factor scores and individual-item scores. The data were analyzed using Stata 10.0 (College Station, Texas).
One short-answer question asked the students to list three to five words that came to mind when they thought about caring for someone with a chronic illness. We performed content analysis using the data for the graduating medical students, who provided a total of 451 responses, including 158 unique words or short phrases. Two of the authors (J.J.V. and E.G.) independently reviewed the 158 terms with the goal of meaningful and reliable classification.
The first hypothesized classification structure was clinical responses (e.g., “goals of treatment,” “counseling,” “continuity”) versus emotional responses (“compassion,” “love,” “exhausting”). We tested another classification structure of physician-centered (e.g., “challenging,” “listening,” “support”) versus patient-centered (“anger,” “bravery,” “determination”). Neither approach led to reliable independent classifications of the responses.
Finally, a third classification structure described as “Constructive/Supportive” reactions to chronic illness versus “Discouraging/Pessimistic” was evaluated. Constructive/Supportive was defined as words representing professional clinical responsibilities or positive emotions. This included “aggressive treatment,” “patience,” “compassion,” “empathy,” “understanding,” “love,” “hopeful,” “caring,” and “perseverance.” Discouraging/Pessimistic, defined as practical problems and negative emotions, included “costly,” “thankless,” “frustrating,” “exhausting,” “sad,” and “tiring.”
Again, the two authors (J.J.V. and E.G.) independently reviewed and classified the 158 words as being either Constructive/Supportive, Discouraging/Pessimistic, or Mixed. Their initial classifications were identical for 136 (86%) words; they eventually reached consensus on all classifications. There were 85 responses classified as Constructive/Supportive, 65 as Discouraging/Pessimistic, and 8 that could not be classified (e.g., “chance,” “chronic,” “memories,” “selfness”).
All data for the 704 students were combined, yielding a total of 2016 responses representing more than 500 unique terms. The two authors reviewed the additional terms and classified them accordingly. The 10 most frequent responses and number of students that gave them included “patience” (185), “compassion” (147), “empathy” (69), “understanding” (68), “difficult” (58), “love” (43), “frustration” (37), “caring” (34), “support” (30), and “pain” (28).
Students were classified as Constructive/Supportive if at least 75% of their words were positive. They were Discouraging/Pessimistic if 75% were negative. Those remaining were categorized as Mixed.
Completed questionnaires were returned by 185 (83%) medical and 146 (87%) bachelor of science in nursing students who were about to graduate and by 258 (99%) medical and 115 (99%) nursing students who matriculated in the same year. The data analysis addressed the students’ experience with and overall understanding of chronic illness, their general attitudes toward chronic illness, and the differences between medical and nursing students.
Understanding of chronic illness.
When asked to list three of the most prevalent chronic diseases, the four responses of cancer, diabetes, heart disease, and hypertension accounted for more than 60% of responses. Other frequent responses included Alzheimer disease, arthritis, asthma, chronic obstructive pulmonary disease, depression, HIV/AIDS, and stroke.
A majority (78%) of students indicated they had a family member or friend with a chronic illness, and 54 (8%) indicated that they themselves had a chronic illness. The only difference between the professional programs was that 76% of medical students reported having a chronically ill family member compared with 66% of nursing students (P < .01). The medical students were more likely than the nursing students to mention a grandparent with chronic illness (67% versus 50%). The overall finding that 78% of students had personal experience with chronic illness is consistent with the findings of Turner, Pugh, and Budiani,4 who also researched medical students’ perspectives on chronic illness and reported that 79% of the students had a family member or another personal experience related to chronic illness.
The majority of all students underestimated the level of U.S. national health expenditures consumed by chronic illnesses, which is frequently reported to be around 75%. Their overall median estimate was 40%. The nursing students’ average of 30% was significantly lower (P < .01) than the medical students’ average of 45%. The median of 25% for matriculating nursing students was lowest, and the median of 50% for graduating medical students was highest.
General attitudes toward chronic illness.
Overall, 47% of the students were classified as Constructive/Supportive on the basis of their responses to the open-ended item asking for words that came to mind when they thought about chronic illness. However, 30% gave responses that were mostly Discouraging/Pessimistic. The remaining 23% used a mix of positive and negative words.
Factor analysis of the 35-item rating scale for all students yielded three factors with Eigen values greater than one, which accounted for 65% of variance. We also conducted two separate factor analyses of the medical and nursing students which accounted for 56% and 43% of variance, respectively. Each of the three factor analyses yielded three factors. In every case, the first and largest factor, which was named “Personal Interest in Patients,” involved the same six items. These included items such as “I enjoy working with patients’ families,” “I value the development of long-term relationships with my patients,” and “I try to understand how it must feel for someone to have to live with a chronic illness.” The alpha reliability estimate for this six-item scale was 0.68. Although there was some overlap in the other two factors identified in the three factor analyses, we found no consistent pattern.
We calculated mean scores on this Personal Interest in Patients factor for the students classified as Constructive/Supportive and those classified as Discouraging/Pessimistic. The mean of 51.0 for students classified as Constructive/Supportive was higher (P < .001) than the mean of 48.3 for those classified as Discouraging/Pessimistic.
Attitudes of medical and nursing students.
A majority (64%) of the entering nursing students were classified as Constructive/Supportive, which was significantly higher (P < .01) than the rate for entering medical students (45%) and significantly (P < .001) higher than the rates for the graduating medical (44%) and nursing students (43%). The pattern of mean scores on the Personal Interest in Patients factor was identical. The means ranged from highs of 54.1 for entering nurses and 51.1 for entering medical students, to lows of 49.1 for graduating nurses and 46.5 for graduating physicians.
Overall, the pattern of responses for the nursing and medical students was similar on the majority of the 35 rating scale statements. For example, the means on a scale of one to five for the item “No single health professional can meet all the needs of a patient with a chronic illness” were 4.1 for graduating medical students and 4.2 for graduating nursing students (P < .21). Similarly, for “I feel sorry for chronically-ill patients,” the means for both nursing and medical students were 4.0.
However, as shown in Table 1, there were 10 items in which the two groups differed significantly. Graduating medical students were more likely to agree with four items concerned with difficulties involving patient adherence and family support. They expressed less willingness to work with dying patients in hospice. The graduating nursing students were more likely to express interest in helping patients to manage multiple comorbidities and were more likely to agree that teaching patients about chronic illness is best done by nurses or health educators. The nursing graduates were more likely to agree that electronic medical records compromise patients’ privacy.
We designed this cross-sectional study to evaluate an instrument that could be used to measure differences in attitudes toward chronic illness among medical and nursing students, and to conduct a baseline assessment before implementing curricular changes. The survey yielded a large sample based on response rates over 80%.
Although the majority of students reported personal experience with chronic illness, we found measurable variation in their understanding of and attitudes toward chronic illness. Most students underestimated the economic impact of chronic illness. Although the absolute accuracy of the frequently cited value of 75% of U.S. health expenditures is debatable,2 the students’ average estimates of between 30% and 50% indicate that the vast majority of the estimates made by the 704 individual students in this study were low in relation to widely held beliefs about the relative cost of chronic illness.
We used qualitative analysis to develop a measure of overall attitudes toward chronic illness. Although we detected a generally positive feeling in about half the students, about one third expressed largely negative reactions to chronic illness. This pattern was consistent across groups, with the exception of entering nursing students, who were more positive. It seems that a reasonable goal of the educational program would be for graduating medical and nursing professionals to be realistic about chronic illnesses, but as positive as possible. Ideally, these professionals would be classified as either Constructive or at least Mixed/Indeterminate, using the methods developed in this study.
The goal of the factor analysis of the 35-item rating scale was to identify sets of related items. The common thread among the items was identified as Personal Interest in Patients. It is interesting that these six items stood independently and did not correlate with the other items related to chronic illness in the survey. Scores on these six items were related to students’ overall attitudes toward chronic illness by qualitative analysis. Although the alpha reliability for this scale was reasonable for a six-item scale and adequate for analyses of groups, it would not produce reliable scores for individuals. There might be an opportunity to develop this six-item scale further. The ideal professional would score high on the Personal Interest in Patients factor.
Although the sample included students at matriculation and graduation, we focused the analysis of items on differences between medical and nursing students. Although medical and nursing students responded similarly to most items, some differences were significant. The pattern of differences may reflect a tendency of medical students to perceive their role as healers, whereas nursing students may be reflecting their discipline’s role as patient advocates. These differences could be explored further through qualitative methods and may have important implications when designing interprofessional experiences for nursing and medical students.
We found that entering students had more positive attitudes toward chronic illness than graduating students. The cross-sectional design of this initial study, which was undertaken to evaluate the instrument and collect baseline data, limited the analysis of differences between students at matriculation and graduation. Although these differences could be due to the curriculum, they could also be related to the students’ aging, maturation, and other personal experiences during the educational program. It is possible that these attitudes reflect some students’ personal experiences with the chronic illnesses of family members, friends, or themselves over time.
It has been suggested that the development of unfavorable attitudes toward chronic illness among students and their impact on career choice may be the result of teacher role-modeling and narrow educational experiences in acute care settings involving patients with severe complications and advanced stages of chronic illnesses.8 Exposing students to highly functional individuals with chronic illnesses, modeling effective management, and demonstrating that chronic illness care is rewarding for clinicians may improve attitudes toward chronic illness. Nevertheless, the relative effects of all these variables can best be determined by a longitudinal follow-up of the same cohort of students from the beginning to the end of their training program using an instrument such as the one evaluated in this study.
This study contributes to the literature by demonstrating that the attitudes of medical and nursing students regarding chronic illness are similar. It has curricular implications for students of all health disciplines and for interprofessional education. Many of the entering students had chronic illnesses themselves or had family members with chronic illness and were still positive in their early attitudes. This study also provides initial data on an evaluation tool that can be used in longitudinal studies to assess the impact of changes in health professions and interprofessional education on students’ attitudes toward care for patients with chronic illnesses.
The authors sincerely appreciate the contributions of Shaun Franklin, MD, who contributed to the development and pretesting of the Chronic Illness Survey.
This study was supported in part by a grant from the Association of American Medical Colleges Enhancing Education for Chronic Illness Care Initiative, supported by the Josiah Macy, Jr. Foundation, 2006–2008.
1 Institute of Medicine. Crossing the quality chasm: A new health system for the 21st Century. Washington, DC: National Academy Press; 2001.
2 Wagner E, Austin BT, Davis C, Hindmarsh M, Schaefer J, Bonomi A. Improving chronic illness care: Translating evidence into action. Health Aff (Millwood). 2001;20:64–78.
3 Reeves S, Zwarenstein M, Goldman J, et al. Interprofessional education: Effects on professional practice and health care outcomes. Cochrane Database of Systematic Reviews, 2008.
4 Turner J, Pugh J, Budiani D. “It’s always continuing”: First-year medical students’ perspectives on chronic illness and the care of chronically ill patients. Acad Med. 2005;80:183–188.
5 Baxter H, Singh SP, Standen P, Duggan C. The attitudes of “tomorrow’s doctors” towards mental illness and psychiatry: Changes during the final undergraduate year. Med Educ. 2001;35:381–383.
6 Thompson T, Emrich K. The effect of curriculum on the attitudes of nursing students toward disability. Rehabil Nurs. 2003;28:27–30.
7 Lloyd-Williams M, Dogra N. Attitudes of preclinical medical students towards caring for chronically ill and dying patients: Does palliative care teaching make a difference? Postgrad Med. 2004;80:31–34.
8 Davis B, Nelson DB, Sahler OJZ, McCurdy FA, Goldberg R, Greenberg LW. Do clerkship experiences affect medical students’ attitudes toward chronically ill patients? Acad Med. 2001;76:815–820.