Abstract
Background
Heart failure (HF) self-care is an important component of disease management and the focus of many interventions.
Aim
The aim of this study was to evaluate the validity and reliability of the 9-item European HF Self-Care Behavior Scale (EHFScB-9) in a sample of 200 adults from the United States with symptomatic HF.
Methods
Psychometric tests included item and confirmatory factor analyses, convergent and discriminant validity, and internal consistency.
Results
Item-total correlations ranged from 0.25–0.65. Many fit indices for the EHFScB-9 and the 4-item consulting behaviors reached thresholds of acceptability. As expected, the EHFScB-9 was associated with other measures of HF self-care but not with quality-of-life. Coefficient α was 0.80 for the EHFScB-9 and and 0.85 for the consulting behaviors subscale.
Conclusion
The EHFScB-9 was a valid and reliable measure of HF self-care among English-speaking U.S. adults with symptomatic HF.
Background
Self-care of heart failure (HF), a critical element of disease management, is associated with multiple health outcomes and a common target of therapeutic interventions.1–3 The reliable and valid measurement of HF self-care is central to advancing our understanding of how we can optimize self-care and help patients with HF best influence their own health.4 There are two instruments commonly used to measure self-care of HF; namely the Self-Care of HF Index (SCHFI v.6)5 and the 9-item European Heart Failure Self-care Behavior Scale (EHFScB-9).6 While the SCHFI was developed and evaluated in a United States (U.S.), there are no data published on the psychometric properties of the EHFScB-9 scale among U.S. adults. Accordingly, the aim of this study was to evaluate the validity and reliability of the EHFScB-9 in a sample of U.S. adults with symptomatic HF.
Methods
We completed an analysis of enrollment data in a prospective cohort study of symptoms among adults with symptomatic HF who were recruited through a single advanced HF clinic in the Pacific Northwest of the U.S. between October, 2010 and October, 2012. Eligible participants were 23 years old or greater, had current HF symptoms (New York Heart Association (NYHA) functional class II-IV) and were English speaking. Patients were excluded if they received heart transplantation or a ventricular assist device, had major cognitive impairment, or had major and uncorrected visual impairments. Written informed consent was obtained from all study participants; this study conforms to the principles of the Declaration of Helsinki, and was reviewed and approved by the institutional review board.
Measurement
Self-reported socio-demographics were assessed using a questionnaire. Clinical and treatment characteristics were collected by review of the electronic medical record. NYHA class was assessed by the HF cardiologist immediately prior to enrollment. Comorbidities were assessed during the medical record review using the Charlson Comorbidity Index.7
The European Heart Failure Self-Care Behavior Scale
The 12-item European Heart Failure Self-care Behavior Scale was developed a decade ago as a valid, reliable and practical measure of HF self-care behaviors,8 and has been translated and validated in several languages and in several countries.9–13 In 2009, the instrument was shortened to include nine items (EHFScB-9), each rated by five response options ranging from 1 (I completely agree) to 5 (I don’t agree at all).6 Scores on the EHFScB-9 range from 9–45; lower scores indicate better self-care.6 The EHFScB-9 also has a 4-item “consulting behavior” subscale that captures patients’ endorsement of contacting providers when symptoms occur;6 the consulting behavior subscale ranges from 4–20.
The Self-Care of Heart Failure Index
We used the Self-Care of HF Index (SCHFI v.6)5 for self-care convergent validity testing. In the SCHFI, 22 items are provided with four to five response options. Responses are standardized into three scores ranging from 0–100 that represent self-care maintenance (routine daily behaviors), self-care management (symptom recognition, evaluation, and treatment), and self-care confidence (confidence in self-care behaviors); higher values on the SCHFI indicate better self-care.5 In this sample, Cronbach’s α was 0.64, 0.60, and 0.84 on the SCHFI maintenance, management, and confidence scores respectively.
The Minnesota Living with Heart Failure Questionnaire
We used the Minnesota Living with HF Questionnaire (MLHFQ)14 for health-related quality-of-life (QOL) discriminant validity testing. The MLHFQ measures the influence of HF and its treatment on preventing participants from living as they wanted. 21 items are provided with six response options from 0 (No) to 5 (very much); higher scores indicate worse QOL. Summary scores include physical (range 0 to 40), and emotional (range 0 to 20) QOL indices. In this sample, Cronbach’s α was 0.93 and 0.90 on the MLHFQ physical and emotional scores, respectively.
Analysis
Item response means and SDs, and corrected item-total correlations were quantified. Item difficulty/endorsement was assessed by quantifying the proportion of participants who provided the best possible response (completely agree). Item difficulty of 0.3 indicates that few (30%) participants endorsed the item, and 0.7 indicates that many (70%) participants endorsed the item; 0.3 and 0.7 is the best range for item difficulty. Item discrimination was quantified by comparing item difficulty scores between participants with EHFScB-9 total scores in the top and bottom thirds of the distribution. Confirmatory factor analyses of the EHFScB-9 and the 4-item consulting behavior subscale were performed in Mplus v.6 (Los Angeles, California) using weighted least square parameter estimation using a diagonal weight matrix with standard errors, and mean- and variance-adjusted statistics that use a full weight matrix (i.e. WLSMV); all indicators were appropriately identified as ordered categorical data. Results are presented in parameter estimates and standard errors. To assess model fit, overall model χ2 tests, root mean square errors of approximation (RMSEA), weighted root mean square residuals (WRMR), comparative fit indices (CFI), Tucker-Lewis indices (TLI), normed fit index (NFI), and adjusted goodness-of-fit index (AGFI) were calculated using common thresholds of acceptability.15
Pearson’s correlations were used to quantify convergent and discriminant validity. Based on finding in the original EHFScB-9 psychometric paper,6 we assumed there would be strong relationships between the EHFScB-9 and the 4-item consulting behavior subscale and the SCHFI maintenance and management scores (convergence), and a weak an insignificant relationship between the EHFScB-9 and the 4-item consulting behavior subscale and indices of QOL (divergence). Cronbach’s alpha and 95% confidence intervals (CI) were calculated as an index of internal consistency.
Results
The sample was predominantly male and Caucasian and most participants had low comorbid burden (Table 1). The average age of the sample was 57 years and a majority of participants (60%) were classified as NYHA functional class III or IV.
Table 1.
Characteristics of the Sample (n=200)
| Patient Characteristics: | Mean±SD or n (%) |
|---|---|
| Age (years) | 57.0 ± 13.3 |
| Female | 100 (50%) |
| Caucasian | 170 (85%) |
| Body Mass Index (kg/m2) | 30.7 ± 7.4 |
| Charlson Comorbidity Category: | |
| Score of 1 or 2 (low) | 124 (62%) |
| Score of 3 or 4 (medium) | 64 (32%) |
| Score of 5 or more (high) | 12 (6%) |
| Heart Failure Characteristics: | |
| Left Ventricular Ejection Fraction (%) | 28.5 ± 12.3 |
| NYHA Functional Class: | |
| Class II | 80 (40%) |
| Class III | 113 (56.5%) |
| Class IV | 7 (3.5%) |
| Last Known Cardiac Index (L/min/m2) | 2.0 ± 0.5 |
| Last Known PCWP (mm/Hg) | 18.9 ± 8.8 |
| European Heart Failure Self-care Behavior Scale | |
| 9-item (range 9–45) | 18.1 ± 6.8 |
| Consulting Behaviors (range 4–20) | 8.4 ± 4.3 |
| Self-Care of Heart Failure Index | |
| Maintenance (range 0–100) | 70.2 ± 15.1 |
| Management (range 0–100) | 65.7 ± 20.1 |
| Confidence (range 0–100) | 63.3 ± 21.2 |
| Quality-of-Life: | |
| MLHFQ Physical Dimension (range 0–40) | 20.0 ± 11.1 |
| MLHFQ Emotional Dimension (range 0–20) | 10.0 ± 7.4 |
Abbreviations: MLHFQ = Minnesota Living with Heart Failure Questionnaire; NYHA = New York Heart Association; PCWP = pulmonary capillary wedge pressure; SCHFI = Self-Care of Heart Failure Index (v6); SD = standard deviation
Item-total correlations on the EHFScB-9 ranged from 0.25 (taking mediations as prescribed) to 0.65 (if I gain 5 pounds in one week) (Table 2). Item difficulty scores ranged from 0.22 (exercise regularly – the most difficult to endorse) to .91 (taking medications as prescribed – the easiest item to endorse). Most items were discriminatory regarding the top and bottom 33.3% of HF self-care performers. In contrast, taking medications as prescribed was not helpful in discriminating between participants who reported better or worse self-care.
Table 2.
Item Analysis for the EHFScB-9 in a U.S. Sample (n=200)
| Item | Mean ± SD | Item-total correlation‡ | α if deleted | Distribution of Item Responses
|
Item Discrimination | ||||
|---|---|---|---|---|---|---|---|---|---|
| completely agree | don’t agree at all | ||||||||
| 1 | 2 | 3 | 4 | 5 | |||||
| I weigh myself every day | 2.02±1.41 | 0.453 | 0.79 | 58.0% | 10.5% | 14.5% | 5.5% | 11.5% | 0.47 |
| If my shortness of breath increases † | 1.94±1.23 | 0.636 | 0.76 | 56.0% | 13.0% | 16.0% | 11.0% | 4.0% | 0.70 |
| If my feet/legs become swollen † | 2.06±1.30 | 0.638 | 0.76 | 50.8% | 15.1% | 19.1% | 7.5% | 7.5% | 0.71 |
| If I gain 5 pounds in one week † | 1.97±1.32 | 0.645 | 0.76 | 56.1% | 15.7% | 12.1% | 8.1% | 8.1% | 0.68 |
| I limit the amount of fluids I drink | 2.05±1.19 | 0.461 | 0.79 | 46.7% | 18.6% | 22.1% | 8.0% | 4.5% | 0.67 |
| If I experience increased fatigue † | 2.49±1.36 | 0.618 | 0.77 | 33.0% | 20.5% | 22.5% | 13.0% | 11.0% | 0.92 |
| I eat a low salt diet | 1.67±0.97 | 0.429 | 0.79 | 59.5% | 22.0% | 13.0% | 3.5% | 2.0% | 0.54 |
| I take my medications as prescribed | 1.13±0.44 | 0.246 | 0.81 | 90.5% | 6.5% | 2.5% | 0.5% | 0.0% | 0.12 |
| I exercise regularly | 2.88±1.40 | 0.341 | 0.80 | 22.3% | 19.3% | 23.4% | 17.8% | 17.3% | 0.68 |
I contact my doctor or nurse
Corrected item-total correlation
Abbreviations: EHFScB-9 = European Heart Failure Self-care Behaviors Scale; SD = standard deviation
The confirmatory factor analysis of the EHFScB-9 and consulting behaviors subscale are presented in Table 3. Half of the fit indices reached and others were close to reaching thresholds of acceptability; thus, the fit of the EHFScB-9 as a single scale and the fit of the 4-item consulting behaviors subscale could be improved in this population but are acceptable.
Table 3.
Confirmatory Factor Analyses for the EHFScB-9 and Consulting Behaviors Scale in a U.S. Sample (n=200)
| EHFScB-9 | Consulting Behaviors | |
|---|---|---|
| Standardized Parameter Estimates ± Standard Errors | ||
| I weigh myself every day | 0.54±0.06 | |
| If my shortness of breath increases† | 0.82±0.04 | 0.83±0.04 |
| If my feet/legs become more swollen† | 0.84±0.04 | 0.88±0.04 |
| If I gain 5 pounds in one week† | 0.83±0.03 | 0.81±0.04 |
| I limit the amount of fluids I drink | 0.58±0.05 | |
| If I experience increased fatigue† | 0.77±0.04 | 0.80±0.03 |
| I eat a low salt diet | 0.56±0.06 | |
| I take my medications as prescribed | 0.43±0.10 | |
| I exercise regularly | 0.40±0.07 | |
|
| ||
| Goodness of Fit | ||
| χ2 (df) | 103 (27) | 9.54 (2) |
| p-value | <0.001 | 0.008 |
| RMSEA‡ | 0.118 | 0.137 |
| WRMR | 0.997 | 0.397 |
| CFI | 0.942 | 0.992 |
| NFI | 0.923 | 0.990 |
| TLI | 0.922 | 0.976 |
| AGFI | 0.897 | 0.971 |
I contact my doctor or nurse
Abbreviations: AGFI = Adjusted Goodness-of-fit Index; CFI = Comparative Fit Index; df = degrees of freedom; EHFScB-9 = 9 Item European Heart Failure Self-care Behaviors Scale; NFI = Normed Fit Index; TLI = Tucker-Lewis Index; RMSEA = root mean square error of approximation; WRMR = weighted root mean square residuals.
90% confidence interval 0.093–0.142, p<0.001, for the EHFScB-9, and 0.03–0.20, p0.097, for the consulting behaviors subscale (calculated by necessity from models not considering the ordered categorical nature of these data).
Thresholds for Acceptable Fit:15
RMSEA = 0.05–0.08
WRMR <1.0
CFI and TLI ≥ 0.95
AGFI ≥ 0.85
NFI ≥ 0.90
Convergent validity testing of the EHFScB-9 with the SCHFI, and discriminant validity testing of the EHFScB-9 with the MLHFQ are presented in Table 4. There were moderate to strong correlations among the EHFScB-9 and consulting behaviors subscale and both the SCHFI maintenance and management scores. The EHFScB-9 and consulting behaviors subscale were not correlated with SCHFI confidence scores or with indices of physical and emotional QOL as measured by the MLHFQ.
Table 4.
Convergent and Discriminant Validity for the EHFScB-9 in a U.S. Sample (n=200)
| Linear correlations | SCHFI Maintenance | SCHFI Management | SCHFI Confidence | MLHFQ Physical | MLHFQ Emotional |
|---|---|---|---|---|---|
| EHFScB-9 | −0.594† | −0.424† | −0.038 | 0.132 | 0.164 |
| EHFScB-CB | −0.415† | −0.338† | −0.010 | 0.109 | 0.104 |
p<0.0001 for all correlations with Bonferroni correction for multiple measures
Abbreviations: CB = EHFScB Consulting Behaviors; EHFScB = European Heart Failure Self-care Behaviors Scale; MLHFQ = Minnesota Living with Heart Failure Questionnaire (physical and emotional domains); SCHFI = Self-Care of Heart Failure Index (v6).
Cronbach’s alpha of the EHFScB-9 was 0.80 (95% CI was 0.76–0.84). Single item deletion did not result in significant improvement of internal consistency. Cronbach’s alpha was 0.85 (95% CI 0.81–0.88) on the 4-item consulting behaviors subscale.
Discussion
In this sample of 200 U.S. adults with symptomatic HF, the EHFScB-9 was a valid and internally consistent measure of HF self-care behaviors. There are some minor limitations of the EHFScB-9 regarding the item on medication adherence, which had the lowest item-total correlation, poor discrimination and was the easiest item to endorse. It is also known that HF patients overestimate adherence to medications using self-report measures compared with objective indices.16, 17 Conceptually, however, removing an item on medication adherence from a measure of HF self-care would be difficult to justify from a validity perspective. Additionally, the EHFScB-9 has sufficient internal consistency that would not be significantly improved with the removal of the medication adherence item. Thus, the EHFScB-9 will be useful in models predicting HF self-care or using HF self-care to predict other outcomes in this population without further adjustment.
Consistent with a review of the psychometric properties of HF self-care measures,4 linear associations between scores in this study indicate that the EHFScB-9 is most closely associated with the routine self-care behaviors (SCHFI maintenance) and moderately associated with symptom response behaviors (SCHFI management) but not confidence in self-care behaviors (SCHFI confidence). In addition, both the EHFScB-9 and consulting behaviors subscale were not significantly associated with QOL metrics; these findings are consistent with results of other EHFScB-9 psychometric analyses,6 and indicate that self-care is a different construct than QOL.
The 4-item consulting behaviors subscale that captures behaviors in response to signs/symptoms of congestion also had acceptable fit and internal consistency. The consulting behaviors subscale was moderately associated with routine self-care behaviors but not to confidence in self-care or QOL. Although there are conceptual dissimilarities, the consulting behaviors subscale was moderately associated with the SCHFI management score. Thus, patients who are better at consulting with providers in response to signs/symptoms of congestion are also better at recognizing and engaging in self-initiated strategies to ameliorate HF symptoms.
The relatively young age, functional limitations, and English-speaking ability of the sample may limit the generalizability of these findings. The cross-sectional nature of this study also impairs our ability to comment on the utility of the EHFScB-9 over time. Finally, due to the relative small size of this sample, further psychometric testing is warranted in larger and more diverse patient populations.
Conclusion
Self-care of HF is an important element of disease management and a focus of a large and expanding body of international clinical research. The EHFScB-9 and consulting behaviors subscale are valid and reliable measures of self-care among English-speaking U.S. adults with moderate to advanced HF.
Acknowledgments
Funding Acknowledgment: This work was supported by an award from the American Heart Association (11BGIA7840062), and, in part, by award number K12 HD043488-08 from the U.S. Office of Research on Women’s Health and National Institute of Child Health and Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the American Heart Association, Office of Research on Women’s Health, or the National Institutes of Health.
Footnotes
Declaration of Conflicting Interests: None Declared
Contributor Information
Christopher S. Lee, Oregon Health & Science University School of Nursing, Portland, OR.
Karen S. Lyons, Oregon Health & Science University School of Nursing, Portland, OR.
Jill M. Gelow, Oregon Health & Science University School of Medicine, Portland, OR.
James O. Mudd, Oregon Health & Science University School of Medicine, Portland, OR.
Shirin O. Hiatt, Oregon Health & Science University School of Nursing, Portland, OR.
Thuan Nguyen, Oregon Health & Science University School of Medicine, Portland, OR.
Tiny Jaarsma, Linköping University, Sweden.
References
- 1.Riegel B, Moser DK, Anker SD, et al. State of the science: promoting self-care in persons with heart failure: a scientific statement from the American Heart Association. Circulation. 2009;120:1141–63. doi: 10.1161/CIRCULATIONAHA.109.192628. [DOI] [PubMed] [Google Scholar]
- 2.Riegel B, Lee CS, Dickson VV. Self care in patients with chronic heart failure. Nat Rev Cardiol. 2011;8:644–54. doi: 10.1038/nrcardio.2011.95. [DOI] [PubMed] [Google Scholar]
- 3.Moser DK, Dickson V, Jaarsma T, Lee C, Stromberg A, Riegel B. Role of self-care in the patient with heart failure. Curr Cardiol Rep. 2012;14:265–75. doi: 10.1007/s11886-012-0267-9. [DOI] [PubMed] [Google Scholar]
- 4.Cameron J, Worrall-Carter L, Driscoll A, Stewart S. Measuring self-care in chronic heart failure: a review of the psychometric properties of clinical instruments. J Cardiovasc Nurs. 2009;24:E10–22. doi: 10.1097/JCN.0b013e3181b5660f. [DOI] [PubMed] [Google Scholar]
- 5.Riegel B, Lee CS, Dickson VV, Carlson B. An update on the self-care of heart failure index. J Cardiovasc Nurs. 2009;24:485–97. doi: 10.1097/JCN.0b013e3181b4baa0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jaarsma T, Arestedt KF, Martensson J, Dracup K, Stromberg A. The European Heart Failure Self-care Behaviour scale revised into a nine-item scale (EHFScB-9): a reliable and valid international instrument. Eur J Heart Fail. 2009;11:99–105. doi: 10.1093/eurjhf/hfn007. [DOI] [PubMed] [Google Scholar]
- 7.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 8.Jaarsma T, Stromberg A, Martensson J, Dracup K. Development and testing of the European Heart Failure Self-Care Behaviour Scale. Eur J Heart Fail. 2003;5:363–70. doi: 10.1016/s1388-9842(02)00253-2. [DOI] [PubMed] [Google Scholar]
- 9.Lupon J, Gonzalez B, Mas D, et al. Patients’ self-care improvement with nurse education intervention in Spain assessed by the European Heart Failure Self-Care Behaviour Scale. Eur J Cardiovasc Nurs. 2008;7:16–20. doi: 10.1016/j.ejcnurse.2007.05.004. [DOI] [PubMed] [Google Scholar]
- 10.Kato N, Ito N, Kinugawa K, Kazuma K. Validity and reliability of the Japanese version of the European Heart Failure Self-Care Behavior Scale. Eur J Cardiovasc Nurs. 2008;7:284–9. doi: 10.1016/j.ejcnurse.2007.12.005. [DOI] [PubMed] [Google Scholar]
- 11.Pulignano G, Del Sindaco D, Minardi G, et al. Translation and validation of the Italian version of the European Heart Failure Self-care Behaviour Scale. J Cardiovasc Med (Hagerstown) 2010;11:493–8. doi: 10.2459/JCM.0b013e328335fbf5. [DOI] [PubMed] [Google Scholar]
- 12.Yu DS, Lee DT, Thompson DR, Jaarsma T, Woo J, Leung EM. Psychometric properties of the Chinese version of the European Heart Failure Self-care Behaviour Scale. Int J Nurs Stud. 2011;48:458–67. doi: 10.1016/j.ijnurstu.2010.08.011. [DOI] [PubMed] [Google Scholar]
- 13.Shuldham C, Theaker C, Jaarsma T, Cowie MR. Evaluation of the European Heart Failure Self-care Behaviour Scale in a United Kingdom population. J Adv Nurs. 2007;60:87–95. doi: 10.1111/j.1365-2648.2007.04371.x. [DOI] [PubMed] [Google Scholar]
- 14.Rector TS, Kubo SH, Cohn JN. Validity of the Minnesota Living with Heart Failure questionnaire as a measure of therapeutic response to enalapril or placebo. Am J Cardiol. 1993;71:1106–7. doi: 10.1016/0002-9149(93)90582-w. [DOI] [PubMed] [Google Scholar]
- 15.Schnermelleh-Engel K, Moosbrugger H, Muller H. Evaluating the fit of structural eqaution models: tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research Online. 2003;8:23–74. [Google Scholar]
- 16.Wu JR, Moser DK, Chung ML, Lennie TA. Objectively measured, but not self-reported, medication adherence independently predicts event-free survival in patients with heart failure. J Card Fail. 2008;14:203–10. doi: 10.1016/j.cardfail.2007.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Nieuwenhuis MM, Jaarsma T, van Veldhuisen DJ, van der Wal MH. Self-reported versus ‘true’ adherence in heart failure patients: a study using the Medication Event Monitoring System. Neth Heart J. 2012 doi: 10.1007/s12471-012-0283-9. [DOI] [PMC free article] [PubMed] [Google Scholar]