Jurgens, Corrine Y. PhD, RN, CS, ANP; Fain, James A. PhD, RN, BC-ADM, FAAN; Riegel, Barbara DNSc, RN, CS, FAAN
Heart failure is a significant public health problem that affects 5 million Americans.1 As a chronic and progressive clinical syndrome, heart failure negatively affects a patient's quality of life with unpleasant symptoms affecting well being, activities of daily living, and increases the risk of multiple hospitalizations for management of escalating symptoms. The ability of patients to monitor symptoms and maintain a complex regimen of multiple medications is essential for quality of life, avoidance of acute exacerbations as well as repeated hospital admissions. Nevertheless, self-management of heart failure is difficult because of the nonspecific symptom profile and the insidious nature of worsening symptoms. Thus, somatic awareness, defined as sensitivity to physical sensations and bodily activity secondary to physiological change, may improve patient ability to monitor symptoms. In this article, the development and testing of an instrument measuring the awareness or perception of heart failure symptoms is reported.
Any personal description of symptoms is subjective by nature and therefore open to question.2 Aside from daily monitoring of body weight, however, the self-management of heart failure relies on self-assessment of symptoms. Such self-assessments are often what bring patients to the attention of healthcare providers. Therefore, self-report of symptoms is extremely valuable for clinicians.
Existing instruments measuring somatic awareness are general in nature and not cardiac or heart failure specific.3,4 No measure of self-reported somatic awareness or perception of common heart failure symptoms is presently available. Those instruments that are heart failure specific are for the purpose of assessing the physical impact of symptoms on activities of daily living,5 limit symptom assessment to fatigue and dyspnea,6 exclude perceived symptom severity,7 or measure patient response to symptoms in relation to effects on quality of life.5,6
Available instruments that address facets of general somatic awareness are not directed toward heart failure symptoms per se. Some are measures of general bodily awareness,3 emotion-related bodily responses,8 beliefs about bodily attentiveness,4 and bodily awareness specific to the symptom of palpitations.9 Other instruments assess distress associated with common physical symptoms10 or separate the components of the symptom experience into occurrence and distress level for symptoms that are not specific to heart disease.11
Although none of these measures are heart failure specific, investigators have studied somatic or general bodily awareness in heart failure patients. Baas et al12 used the Body Awareness Questionnaire4 to explore the relationship between body awareness and somatization among patients with heart failure or after heart transplant. No differences in physiologic body cues or specific symptoms were found related to age, gender, or treatment. Nor were there differences in bodily awareness with regard to negative mood states such as anxiety or depression. Body awareness as related to physiological cardiac parameters was not reported in the study.
Despite emphasis on symptom monitoring in treatment guidelines for heart failure,13 available instruments remain limited to quality of life, number of symptoms, and symptom impact. None are adequate measures of perception or awareness of heart failure signs and symptoms. As noted earlier, instruments measuring general bodily awareness are not sufficiently focused on cardiac-related symptoms. As the constellation of symptoms typical of heart failure can be both nonspecific (eg, fatigue) and acute (eg, paroxysmal nocturnal dyspnea), both are needed for any measure that would be useful in monitoring symptom distress, patterns, and duration whether symptoms are subtle or overt. Therefore, the development of a symptom-specific instrument addressing both prodromal or early symptoms, as well as acute symptoms of a heart failure, was undertaken.
The Heart Failure Somatic Awareness Scale (HFSAS) is a 12-item Likert-type scale to measure awareness and perceived severity of signs and symptoms specific to heart failure. Heart failure, primarily an illness of older adults, often incurs fatigue, so ease of administration and participant burden are important considerations. Furthermore, previous studies report both age and gender as factors affecting symptom reporting.14-18 Consequently, evaluation for the presence of both age and gender bias was considered in the development and evaluation of this instrument.
The HFSAS was limited to 12 items to reflect the most common signs and symptoms of heart failure. A 4-point Likert-type scale was used to address these symptoms, and if present, to ascertain how much the patient was bothered by them at any point during the previous week. An even number of Likert responses was used as the instrument was designed to assess an event, not force a choice. Patients determined a symptom as present or not, and then scored the perceived severity on a 0 to 3 scale as follows: 0= Not at all; 1 = A little, slightly; 2 = A great deal, quite a bit; and 3 = Extremely, could not have been worse. Symptoms scored as 0 are symptoms not experienced by the patient. Scores range from 0 to 36, with higher scores reflecting higher perceived somatic awareness and symptom distress. Pedal edema, one common sign, is addressed in 2 separate items to account for patients who change footwear to accommodate swelling. The HFSAS is written in lay terminology and has a second-grade reading level.
The symptom list was derived from a review of the literature and previous studies citing symptom lists for heart failure.19-21 Two nursing experts in the area of heart failure reviewed the symptom list to establish content validity. No recommendations to edit, add, or delete items were suggested.
Testing of the HFSAS was divided into 2 phases which are presented in sequential order for the reader. In phase 1, feasibility factors, including ease of administration, comprehensibility for the patients, and subject burden, were evaluated. Phase 1 also included discriminant validity testing to address the relationship between the trait of anxiety and the HFSAS. The ability to discriminate between awareness of heart failure symptoms and symptoms secondary to anxiety was of concern for physiological and psychological reasons. Physiologically, sympathetic nervous system stimulation is a normal compensatory response to escalating heart failure symptoms resulting in an increased heart rate.22 Psychological anxiety can also produce sympathetic stimulation with similar symptoms, such as increased heart rate and sweating. Furthermore, patients in acute heart failure may experience difficulty breathing. Sympathetic symptoms of anxiety secondary to shortness of breath and emergent hospital admission are expected among patients with heart failure. Therefore, discriminate validity of the HFSAS was evaluated as the ability to discriminate between psychological anxiety and symptom distress specific to heart failure. We hypothesized that the HFSAS would measure awareness of symptom distress and not anxiety.
The Spielberger Trait Anxiety Inventory Form X-223 was used to assess the correlation between proneness toward anxiety and somatic awareness as measured by the HFSAS. The Spielberger Trait Anxiety Inventory is a 20-item, 4-point Likert-type scale measuring a person's general ongoing, typical level of anxiety. Scores range from 20 to 80 points, with higher scores representing greater proneness to anxiety. Internal consistency and reliability of the scale is high (alpha coefficients of >0.89 for Trait Scale). Test-retest reliability correlations are reported between r = 0.73 and r = 0.84.23
Phase 2 analysis addressed construct and criterion validity and reliability testing. To determine construct validity, an exploratory factor analysis was performed by varimax rotation of the principal components. Consistent with the definition of criterion-related validity, which is the utility of an instrument to predict some behavior external to the instrument itself,24,25 the HFSAS was evaluated for its usefulness as a predictor of symptom duration. In patients with heart failure, symptom severity is related to symptom duration,19 with symptom duration, severity, and novelty predicting care-seeking in community-dwelling elders.16 Further construct analysis compared the HFSAS to a general measure of somatic awareness, the Modified Somatic Perception Questionnaire, and tested for differences between general or symptom-specific bodily awareness in relation to symptom duration.
Phase 2 analysis also included an exploration of variability in HFSAS scores to assess for potential bias. The HFSAS scores were divided into quintiles to examine for relationships between the HFSAS scores and age, gender, and duration of heart failure symptoms.
Both phases testing the HFSAS were limited to community-dwelling adult patients emergently hospitalized with a diagnosis of acute heart failure. Patients enrolled in phase 1 were excluded from participating in phase 2. Patients were enrolled if they made their own healthcare decisions and understood or read English. Patients with obvious cognitive impairment or unstable medical status were excluded. Cognitive status was evaluated by the medical and nursing staff caring for the patient. Patients who met the eligibility criteria and the Framingham Diagnostic criteria for heart failure21 were enrolled from acute care facilities in the northeast region of the United States. Institutional Review Board approval was secured from all appropriate authorities and informed written consent was obtained prior to participation.
Phase 1 Procedure
Phase 1 testing was completed in the emergency department phase by patients admitted to a community hospital for symptom management. Participants completed the HFSAS and the Spielberger Trait Anxiety Inventory.
Phase 1 Sample
A convenience sample of 49 patients admitted with acute heart failure was enrolled. The primarily Caucasian sample was 63% female with a mean age of 73 years.
Phase 1 Results
All patients completed the surveys without difficulty. Patients reported that the HFSAS items were clear as to content and easy to answer with respect to heart failure symptoms. The HFSAS had a mean score of 18.96 (SD 2.9) and a range of 13 to 25 out of a possible score of 36. The Spielberger Trait Anxiety Inventory had a mean score of 47 (SD 7.5) out of 80 and the Cronbach alpha coefficient was 0.92. There was no significant correlation between trait anxiety and the HFSAS (r = 0.25, P = .08) supporting discriminant validity of the HFSAS and the hypothesis that the HFSAS measured awareness of symptom distress and not anxiety.
Phase 2 Procedure
Three sites were utilized in this phase, including 1 urban and 1 suburban tertiary care hospital and a community hospital. Patients were enrolled by the principal investigator or trained research assistants over a 15-month period, from October 2001 until January 2003. At their request, the questionnaires were read to the vast majority of the patients. Visual difficulties secondary to aging are a concern to elderly patients when completing surveys.26 As patients with heart failure are generally older, the option to have the instruments read to the patient was incorporated into the procedure. Interviews were done during the hospital stay and completed as soon as possible after admission to minimize the confounding influence of patient recall; 87% of the patients were enrolled within 3 days of admission. Demographic and clinical data were collected by interview and review of the medical record.
Phase 2 Sample
The convenience sample consisted of 201 participants. The mean age of the sample was 70 years (±12) and 39% were 75 years and older (Table 1). There were more males than females and the sample was predominately White. More than half of the patients reported previous admissions for acute heart failure. The majority of the patients had comorbid illnesses typical of heart failure, including coronary artery disease, hypertension, and diabetes. A minority (11%) participated in a heart failure clinic for management of their illness.
Twenty-eight potential subjects declined to participate, including 14 females and 14 males who, except for 1 Black male, were Caucasian. Reasons offered included time, the perception that the information was not important because they were elderly, or lack of interest. The demographic characteristics of the final sample are typical of the geographic area. The subjects who declined participation were demographically similar to the final sample.
Phase 2 Results
In the phase 2 sample, the mean score on the HFSAS was 13.17 (SD 5.52), with a range of 0 to 29 out of a possible maximum score of 36. There were significant differences in HFSAS scores by age, with younger patients having higher median and mean scores (Table 2). Differences in median HFSAS scores based on gender did not reach statistical significance (P = .08). The greatest variability in HFSAS scores occurred in relation to duration of symptoms of acute heart failure such as acute dyspnea (Fig 1). As indicated by the left to right increase in box-plot size in Figure 1, patients reporting the highest levels of distress secondary to the symptoms of heart failure had the most variability in terms of length of endurance of these symptoms prior to seeking care. Symptom duration did not differ significantly for patients who had been previously admitted for symptom management of heart failure and those admitted for the first time.
The reliability of the HFSAS was assessed utilizing theta reliability, an amplified Cronbach alpha useful for instruments with discrete items.24 Cronbach alpha, based on interitem correlations, is most useful when correlations range between 0.30 and 0.40.27 Only 8 of these 66 interitem correlations fell into or above this range (0.31-0.65), indicating that the items are largely discrete. Thus, theta reliability was appropriate. Theta reliability values are interpreted similar to Cronbach alpha.
Initial reliability of the HFSAS, calculated after the first 50 participants of phase 2 were enrolled, was 0.78. The theta was retested with the full final sample of 201 participants, including the first 50 participants. Theta reliability was 0.71 in the final sample. No item would have increased the reliability coefficient if deleted, so all 12 items were retained for the final HFSAS.
To assess the constructs associated with the HFSAS, an exploratory factor analysis was conducted using a principle component analysis with varimax rotation. Results of the analysis produced a 4-factor solution using the criterion of an eigenvalue greater than 1.028; 58% of the variance in scores was explained (Table 3). The 4 HFSAS factors corresponded to the pathophysiology associated with heart failure. Factor 1 included the more acute heart failure symptoms of orthopnea, dyspnea, and paroxysmal nocturnal dyspnea. Factor 2 comprised 2 items addressing symptoms related to peripheral edema. Factor 3, with the exception of the symptom of cough, assessed items related to symptoms typically associated with myocardial infarction. Cough had the lowest factor loading (0.381) of all 12 symptoms. Factor 4 represented the early symptoms of heart failure decompensation and included fatigue, weight gain, and dyspnea on exertion. Weight gain also loaded on Factor 2, but was retained in Factor 4 where it had the highest loading. Similarly, dyspnea loaded on Factor 4 and was retained in Factor 1 with the higher loading.
The HFSAS was analyzed for convergent and divergent construct validity in comparison with general somatic awareness as measured by the Modified Somatic Perception Questionnaire. The HFSAS had a low to moderate correlation with general somatic awareness (Spearman rho r = 0.48), indicating that the instruments measure something in common, but not the same construct.
The HFSAS was examined for its ability to predict duration of symptoms prior to care-seeking for acute heart failure. The HFSAS score was a significant predictor of symptom duration, with higher scores associated with increased time from symptom onset to arrival at the hospital. That is, in spite of increased somatic awareness, patients delayed longer. Further analysis revealed that the pattern of symptom onset was accountable for this counterintuitive result. Patients with a gradual symptom onset waited until their symptoms reached a severe level prior to seeking care. The HFSAS was also a significant predictor of duration of acute symptoms in the linear regression model after adjusting for demographic and clinical factors known to influence the symptom experience (Table 4).19,29-31 The duration of acute symptoms increased approximately 1 hour for every 1 point increase in the HFSAS score. Similarly, an increase in the HFSAS score was associated with an increase in duration of dyspnea, acute dyspnea, and dyspnea on exertion. Acute dyspnea was defined as a sudden onset or increase over baseline levels of chronic dyspnea. Dyspnea on exertion was shortness of breath noticed only with activity. A subscale of the HFSAS measuring the acute symptoms as a group (dyspnea, orthopnea, paroxysmal nocturnal dyspnea, and dyspnea on exertion) was predictive of an increase in acute dyspnea duration. The Modified Somatic Perception Questionnaire was not a significant predictor of symptom duration in the regression model, but HFSAS was predictive, after controlling for other factors such as age and pattern of symptom onset. Based on these analyses, the HFSAS was judged to have construct and criterion-related validity.
Patients with heart failure are asked to monitor their symptoms on a daily basis to facilitate early interventions that may avert hospitalization. Availability of a reliable and valid measure of somatic awareness may be useful in assisting patients to become more aware of symptoms. In particular, attention to and care of the early symptoms of heart failure decompensation is important in averting unnecessary hospitalizations. This first test of the HFSAS is promising. The HFSAS was shown to be reliable with content, discriminate, and construct validity. No gender bias was evident in scores.
The reliability of the HFSAS was adequate to support further testing, but it was lower than anticipated, which may reflect the limited number of items or the variable symptom profile among patients with heart failure. Borderline reliability could reflect the brevity of the instrument, which was intentional to minimize participant burden. In an effort to improve reliability, further development and testing of the HFSAS will address adding items that assess symptoms that are difficult to measure. Dyspnea on exertion is one such factor, as patients tend to decrease activity in order to accommodate symptoms.32
Somatic awareness, when measured with the symptom-specific HFSAS, was a significant predictor of duration of symptoms, thus supporting validity of the instrument. The Modified Somatic Perception Questionnaire, a general somatic tool, was less effective in predicting symptom duration. An instrument such as the HFSAS may enhance the ability of patients and healthcare providers to monitor symptom acuity based on patient status at baseline.
Interestingly, patients who had previously experienced an exacerbation of heart failure did not seek care any sooner than patients admitted for first time exacerbation of illness. One possible explanation is that previous experience with decompensation in heart failure does not provide sufficient knowledge for patients to avoid a hospital admission. These results contrast with those of Francque-Frontiero et al,33 who found that experience with heart failure predicted self-care ability. Another potential explanation for this difference in findings is the outcome variable. Francque-Frontiero et al found that patients with more experience had better self-care, but they did not assess the adequacy of self-care to prevent hospitalization, as we did. Our results suggest that it may take more than first-hand experience to teach patients how to avoid rehospitalization.
The accuracy or validity of patient self-reports is regularly questioned. Meek et al,2 studying a small sample of stable patients with chronic obstructive pulmonary disease, reported accurate recall of the average intensity of symptom experiences for a 2-week period. After regression analysis, however, the authors concluded that patient recall of symptom intensity for the 2 previous weeks was influenced by current symptom intensity. That is, when current symptoms are mild, the severity of relatively recent symptoms may be down-rated. Therefore, it may be advisable to have patients evaluate the number and intensity of symptoms currently being experienced, as well as those in the recent past. In a secondary analysis of heart failure patients in the SOLVD clinical trial,34 patient self-report of physical status predicted hospitalization as well as or better than physiologic or clinician assessments. In a study of acute and prodromal symptoms in women with myocardial infarction, McSweeney et al35 reported accurate symptom recall based on acceptable test-retest scores 7 to 14 days later.
Although not always feasible, asking patients to rate symptoms may require validation from others.36 Additionally, it may be difficult for patients with chronic, omnipresent symptoms to objectively rate their degree of distress especially when it involves incremental change, which must be assessed in the context of daily life. For example, a patient may feel a symptom is severe, but rate the discomfort as less bothersome if able to continue with daily activities. Qualitative studies of persons with heart failure illustrate that patients learn to live with symptoms over time.37,38
The HFSAS was tested in patients hospitalized for acute decompensated heart failure. The reliability and validity of the HFSAS among patients with chronic stable heart failure has not yet been explored and test-retest reliability has not been established. Furthermore, the effect of reading to the patient needs to be determined. A lack of ethnic diversity in the sample is also a limitation to generalizability.
Despite these limitations, the HFSAS may be useful in studies designed to improve symptom recognition and self-management. Using the HFSAS to establish baseline status may be clinically valuable in gauging the importance of the daily ebb and flow of symptoms. Fostering awareness of the early symptoms of decompensation may avert repeated hospital admission for symptom management. The ultimate goal is to facilitate effective self-monitoring of the symptoms of heart failure in this growing population.
1. American Heart Association. American Heart Association Heart Disease and Stroke Statistics: 2005 Update
. Dallas, Tex: American Heart Association; 2005.
2. Meek PM, Lareau SC, Anderson D. Memory for symptoms in COPD patients: how accurate are their reports? Eur Respir J
3. Main CJ. The modified somatic perception questionnaire (MSPQ). J Psychosom Res
4. Shields SA, Mallory ME, Simon A. The body awareness questionnaire: reliability and validity. J Pers Assess
5. Rector TS, Kubo SH, Cohn JN. Patients' self-assessment of their congestive heart failure, Part 2: Content, reliability and validity of a new measure, the Minnesota Living with Heart Failure Questionnaire. Heart Fail
6. Guyatt GH, Nogradi S, Halcrow S, et al. Development and testing of a new measure of health status for clinical trials in heart failure. J Gen Intern Med
7. Friedman MM, Griffin JA. Relationship of physical symptoms and physical functioning to depression in patients with heart failure. Heart Lung
8. Mandler G, Mandler JM, Uvillier ET. Autonomic feedback: the perception of autonomic activity. J Abnorm Soc Psychol
9. Barsky AJ, Cleary PD, Coeytaux RR, et al. The clinical course of palpitations in medical outpatients. Arch Intern Med
10. Cohen S, Hoberman HM. Positive events and social supports as buffers for life change stress. J Appl Soc Psychol
11. Rhodes VA, McDaniel RW, Homan SS, et al. An instrument to measure symptom experience: symptom occurrence and symptom distress. Cancer Nurs
12. Baas LS, Beery TA, Allen G, et al. An exploratory study of body awareness in persons with heart failure treated medically or with transplantation. J Cardiovasc Nurs
13. Grady KL, Dracup K, Kennedy G, et al. Team management of patients with heart failure: a statement for healthcare professionals from the Cardiovascular Nursing Council of the American Heart Association. Circulation
14. Cameron L, Leventhal EA, Leventhal H. Symptom representations and affect as determinants of care seeking in a community-dwelling, adult sample population. Health Psychol
15. Leventhal EA, Prohaska TR. Age, symptom interpretation, and health behavior. J Am Geriatr Soc
16. Mora P, Robitaille C, Leventhal H, et al. Trait negative affect relates to prior-week symptoms, but not to reports of illness episodes, illness symptoms, and care seeking among older persons. Psychosom Med
17. Rakowski W, Julius M, Hickey T, et al. Daily symptoms and behavioral responses. Results of a health diary with older adults. Med Care
18. Stoller EP, Forster LE, Pollow R, et al. Lay evaluation of symptoms by older people: an assessment of potential risk. Health Educ Q
19. Friedman MM. Older adults' symptoms and their duration before hospitalization for heart failure. Heart Lung
20. Friedman MM, King KB. Correlates of fatigue in older women with heart failure. Heart Lung
21. Ho KKL, Anderson KM, Kannel WB, et al. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation
22. Pepper GS, Lee RW. Sympathetic activation in heart failure and its treatment with beta-blockade. Arch Intern Med
23. Manual for the State-Trait Anxiety Inventory
[computer program]. Version. Palo Alto, Calif: Consulting Psychologists Press; 1983.
24. Carmines EG, Zeller RA. Reliability and Validity Assessment
. Beverly Hills, Calif: Sage Publications; 1979.
25. DeVellis RF. Scale development: theory and application
. Newbury Park, Calif: Sage Publications; 1991.
26. Warner CD. Somatic awareness and coronary artery disease in women with chest pain. Heart Lung
27. Gable RK, Wolf MB. Instrument development in the affective domain
. 2nd ed. Norwell, Mass: Kluwer Academic Publishers; 1993.
28. Munro BH. Statistical methods for health care research
. 3rd ed. Philadelphia, Pa: Lippincott; 1997.
29. Dracup K, Moser DK. Beyond sociodemographics: factors influencing the decision to seek treatment for symptoms of acute myocardial infarction. Heart Lung
30. Meischke H, Mitchell P, Zapka J, et al. The emergency department experience of chest pain patients and their intention to delay care seeking for acute myocardial infarction. Prog Cardiovasc Nurs
31. Sheifer SE, Rathore SS, Gersh BJ, et al. Time to presentation with acute myocardial infarction in the elderly: associations with race, sex, and socioeconomic characteristics. Circulation
32. Bennett SJ, Cordes DK, Westmoreland G, et al. Self-care strategies for symptom management in patients with chronic heart failure. Nurs Res
. May-June 2000;49(3):139-145.
33. Francque-Frontiero L, Riegel B, Bennett JA, et al. Self-care of persons with heart failure: does experience make a difference? Clin Excell Nurse Pract
34. Stull DE, Rempher K, Jairath N, et al. The utility of multidimensional assessments of physical health in patient care. AACN Clin Issues Adv Pract Acute Crit Care
35. McSweeney JC, O'Sullivan P, Cody M, et al. Development of the McSweeney Acute and Prodromal Myocardial Infarction Symptom Survey. J Cardiovasc Nurs
36. Riegel B, Carlson B, Glaser D. Development and testing of a clinical tool measuring self-management of heart failure. Heart Lung
37. Winters CA. Heart failure: living with uncertainty. Prog Cardiovasc Nurs
38. Zambroski CH. Qualitative analysis of living with heart failure. Heart Lung
Join us for an Exciting Heart Failure State of the Science Conference Sponsored by the AHA Council on Cardiovascular Nursing State of the Science: Promoting Patient Self-Care in Heart Failure May 7, 2006-8:30 to 11:30 AM
This state of the science conference is a preconference of the American Heart Association Quality of Care and Outcomes Research Conference, Washington, DC, Omni Shoreham Hotel, May 7-9, 2006 (http://www.americanheart.org/presenter.jhtml?identifier=3032938).
Attend both the State-of-the Science preconference and the Quality of Care and Outcomes Research Conference.
Overview: Despite recent advances made in heart failure science and care, the disorder remains a significant burden for patients and families. The complexity of heart failure necessitates a patient-focused, interdisciplinary approach to treatment and care. Ultimately, most care is done in the home by patients and their families or other caregivers; yet, the promotion of patient self-care has received relatively little systematic attention from researchers.
The foundation for self-care is comprehensive patient/family education and counseling, which includes skill building, behavioral strategies to increase adherence and alterations to the structure of healthcare delivery to accommodate patient self-care. There has not been a coordinated, systematic effort to review the existing state of the science in patient/family self-care to determine priorities for future research. Moreover, research efforts to date have been conducted in relative isolation, which has resulted in parallel lines of discovery that have failed to build upon each other. Thus, a systematic review is critical if we are to meet the challenges resulting from the rising numbers of heart failure patients who will be in need of care in the next decade.
The objectives of this conference are to:
1. synthesize the state of the science of heart failure patient self-care;
2. determine priorities for future research in this area; and
3. develop a synthesis document that includes specific strategies for moving priorities forward that will be published in a peer-reviewed journal.
© 2006 Lippincott Williams & Wilkins, Inc.