Heart failure self-care interventions to reduce clinical events and symptom burden

Figures & data

Figure 1 PRISMA flowchart for selection of evidence.

Abbreviation: PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Table 1 Self-care and clinical events

Reference and aim	Sample/method/interventions	Clinical events	Strengths and limitations
Davis et al^Citation30 Aim To test the effect of a targeted intervention on self-care, HF knowledge, and 30-day readmissions USA	n=125 UC n=62 IG n=63 Mean age 59 (SD 13) years 53% male 31% white Measures 1. SCHFI α=0.65–0.76 2. Dutch HF knowledge Scale α=0.39 Interventions UC 1. Standard discharge teaching (verbal review of a HF education booklet with symptom recognition, management, exercise, dietary and fluid restriction, medication adherence) IG 1. Standard discharge teaching plus targeted self-care teaching intervention while hospitalized tailored to personal routine   a. Workbook with pictograms   b. Personalized self-care schedule (medications, appointments, symptom documentation)   c.Problem-solving session audio-taped and given to patient   d. Weight scale   e. Pill box   f. Measuring cups   g. Salt substitute   h. Follow-up telephone call 24–72 hours after discharge to conduct teach back session	No difference in readmission rates between the groups: IG 22%, UC 19% at 30 days Self-care IG had greater but not statistically significant improvement in self-care for all subscales Dutch HF knowledge scores increased significantly in IG (P<0.001), but decreased in UC	Strengths Heterogeneous sample in terms of gender and ethnicity Limitations Short intervention time Age 21 years and older Depression scale not validated in young adults Small sample size Young mean age Single academic center Sample mainly black patients; not representative of diverse populations
DeWalt et al^Citation31 Aim To compare two different self-care methods; single-session versus multisession literacy-sensitive self-care intervention and effect on all-cause hospitalization and death USA	n=605 UC n=302 IG n=303 Mean age 60.7 (SD 13.1) years 52% male 39% white 16% Hispanic 38% black Measures 1. HF quality of life: improving chronic illness care evaluation HF symptom scale (self-care education and symptom burden)* 2. Clinical events (all-cause hospitalization or death) Intervention UC 1. 40-minute training session on daily self-assessment, action plan for exacerbation, salt avoidance, exercise, medication adherence 2. Educational manual 3. Digital weight scale IG 1. 40-minute training session on daily self-assessment, action plan for exacerbation, salt avoidance, exercise, medication adherence 2. Educational manual 3. Digital weight scale 4. Instruction on daily weights to guide diuretic self-adjustment 5. 5–8 follow-up 10-minute phone calls over 4 weeks to reinforce education and guide patient toward better self-care skills	IG (multisession intervention) did not decrease all-cause hospitalization and death at 1 year IRR 1.01 (95% CI 0.83–1.22) Self-care IG (multisession group) had significantly greater improvement in self-care at 1 and 6 months, but not at 12 months Effect of multisession differed by literacy group Low literacy had lower hospitalization and death with IRR 0.75; 95% CI (0.45–1.25) High literacy had higher IRR 1.22; 95%, CI 0.99–1.50 (P=0.0048) HF-related hospitalization Among low literacy, IG had a lower incidence rate 0.53; 95% CI 0.25–1.12 high literacy had a higher rate of 1.32 (95% CI 0.92–1.88; P=0.005) An intensive multisession intervention did not change clinical outcomes compared with a single-session intervention. People with low literacy appear to benefit more from multisession intervention than people with high literacy	Strengths Large sample Limitations Cannot attribute the effects to any single element of the intervention Information on dosages of medications not collected nor number of outpatient visits Randomization did not result in even distribution between groups on HF quality of life scores (predictor of hospitalization) Conducted at only four academic centers, which may limit generalizability
Jurgens et al^Citation7 Aim To test the efficacy of an HF training program on a patient’s ability to recognize and respond to changes in HF symptoms USA	n=99 UC n=51 IG n=48 Mean age 67.7 (SD 12.1) years 67.7% male 88.9% white Measures 1. SCHFI at baseline and 3 months, α=0.43–0.83 2. HF somatic perception scale (symptom burden), α=0.87 3. Clinical events (HF hospitalization, emergency department visit, death) Intervention UC 1. Booklet on HF self-care 2. Weight scale IG 1. Booklet on HF self-care 2. Weight scale 3. HF symptom training intervention: one time one-on-one training by advanced practice nurse using 6-minute walk Test to increase effort for detecting differences in dyspnea and fatigue 4. Daily symptom graph of weight, dyspnea, fatigue, and presence of edema 5. Meaning of symptoms (eg, when to call provider or take an extra diuretic) 6. Follow-up visit 1 week after training to review instruction Primary endpoint: time to first HF hospitalization, ED visit for HF, or HF-related cause and death	Early but not sustained benefit on clinical events No difference between UC and IG in 90-day event-free survival (χ^2=1.53, P=0.216) Self-care IG had statistically significant improvements in self-care maintenance, self-care management and self-care confidence from baseline to 90 days UC had statistically significant improvements in self-care maintenance and management from baseline to 90 days IG had greater but not statistically significant improvements in self-care maintenance and self-care confidence compared with UC UC had greater but not statistically significant improvement in self-care management scores	Strengths Evaluated components of intervention Limitations Small sample Primarily white, male sample Inability to blind the individual collecting telephone data
Kommuri et al^Citation32 Aim To assess changes in HF knowledge assessments between a control group and a nurse educator education session USA	n= 265 UC n=137 IG n=128 Mean age 67 years 61% male 78% white Measures 1. HF knowledge questionnaire at discharge, 30 days, 60, 90, and 180 days to assess HF disease management knowledge, self-care management behaviors and medications, diet recommendations, and salt content* Intervention UC 1. Standard discharge information IG 1. Standard discharge information 2. One hour face-to-face nurse educator delivered HF program   a. Principles of HF   b. Dietary sodium   c. Limiting fluids   d. Diuretic mechanisms   e. Rationale for medication   f. Daily weight monitoring   g. Self-care behaviors   h. Medication compliance   i. Smoking cessation   j. Avoid nonsteroidal medications   k. Response to worsening symptoms   l. Avoid alcohol   m. HF guidelines in lay terms Primary endpoint: clinical events of hospitalization or death within 180 days Change in HF knowledge scores from baseline to 30 days and 180-day follow up	IG had significantly higher knowledge scores at 3 months compared with UC. Clinical events associated with lower HF knowledge (10 versus 11, P=0.002) and HF management scores (4 versus 5, P=0.001) when compared with patients without clinical events HF nurse education at the time of hospital discharge results in improved patient knowledge and reduced risk of readmission Self-care Strong association between HF knowledge and adherence to HF self-care measures (median, IQR 1.0 to 4 versus 0, −2 to 2, P=0.007) Significant improvement in IG disease management knowledge scores (P<0.001) No improvement or difference between groups for dietary and sodium knowledge	Strengths Large sample Limitations HF knowledge questionnaire lacks reliability and validity data Unable to show improvements in dietary knowledge due to weakness in educational program
Lee et al^Citation34 Aim To test the effect of a comprehensive daily symptom diary intervention (education and counseling about HF symptoms, how to recognize symptoms and what to do with escalating symptoms) on event-free survival and health-related quality of life USA	n=44 UC n=21 IG n=23 Mean age 60 (SD 12) years 52.3% male 61.4% white Measures 1. Event-free survival assessed at 1 month and 3 months 2. SCHFI (two items assessing weight and edema monitoring)* 3. MLHF (physical symptom subscale)* 4. Symptom diary Intervention UC 1. Discharge education booklet describing HF, a low sodium diet, and medication instructions IG 1. One face-to-face education and counseling session in patient’s home   a. How HF symptoms are related to fluid retention and diet   b. How to monitor symptoms   c. What to do for worsening symptoms (eg, when to call provider or take an extra diuretic)   d. Review of medication regimen   e. Use of symptom diary using symptom clusters   f. How to weigh daily 2. Five booster sessions via biweekly phone calls for 3 months	IG participants experienced longer event-free survival (P=0.03) at 3 months UC had higher risk for a clinical event (HR 3.17, 95% CI 1.09–9.16) controlling for NYHA functional class Self-care No difference between groups at baseline Self-care at 1 and 3 months not reported	Strengths Heterogeneous sample in terms of gender Limitations Small sample size
Seto et al^Citation36 Aim To investigate the effects of a mobile phone-based telemonitoring system on HF management and outcomes Canada	n=100 UC n=50 IG n=50 Mean age 53.7 (SD 13.7) years 79% male 72% white Measures 1. SCHFI* 2. MLHF* 3. Clinical events Intervention UC 1. Standard care which included   a. Visits to clinic once every 2 weeks to once every 3–6 months, depending on severity of HF   b. Education during preliminary visits to clinic   c. Use of telephone clinic as necessary IG 1. Standard care 2. Telemonitoring system   a. Daily morning weight   b. BP readings   c. Weekly ECG 3. Answered daily morning symptom questions on mobile phone 4. Individual training session on how to use system, provided technical support	No differences between groups for HF hospitalizations, ED visits, or number of nights in hospital at 6 months IG group had higher number of HF clinic visits (P=0.04) related to the cardiologist asking them to come in based on telemonitoring alerts Self-care Both IG and UC had significant improvements in self-care maintenance scores. IG had greater improvement than UC on self-care maintenance (P=0.03) Both IG and UC had significant improvements in self-care management, but no significant differences between groups (P=0.7)	Limitations Study inadequately powered to detect effects of telemonitoring on mortality and hospitalization Sample included patients new to HF clinic and stable HF patients Homogeneous population in terms of gender and ethnicity Young mean age
Shao et al^Citation35 Aim To determine the effectiveness of a self-management program based on the self-efficacy construct in older people with HF Taiwan	n=93 UC n=46 IG n=47 Mean age 72 (SD 5.5) years 68% male Measures 1. Self-efficacy for salt and fluid control, α=0.94–0.96 2. HF self-management behavior scale (lower scores indicate better self-management), α=0.71 3. HF symptom distress scale, α=0.67 4. Health service use (include patient clinical, ED visits, and hospitalization) Intervention UC 1. Educated on improving general knowledge of HF, causes, signs and symptoms, complications, medications, and activity and dietary recommendations 2. Phone calls at weeks 3, 7, and 11 IG 1. Educated on improving general knowledge of HF, causes, signs and symptoms, complications, medications, and activity and dietary recommendations 2. Home visit within 3 days after enrollment 3. Phone calls at weeks 1, 3, 7, and 11 4. Diary of daily sodium and fluid intake and daily weight Primary endpoint: HF-related symptoms, hospitalization at 12 weeks	No significant differences between groups in health service use at 12 weeks Self-care Self-management behavior improved significantly in IG versus UC (mean 27.15 versus 30.13, P<0.001)	Limitations Short follow-up for measuring outcomes Lack of objective measures to support difference between and UC Homogenous population Two medical centers limits generalizability Unable to conclude which intervention was effective because the effectiveness of four information sources could not be determined
Shively et al^Citation33 Aim To determine the efficacy of a patient activation intervention compared with UC on patient activation, self-management, hospitalization, and ED visits in patients with HF USA	n=84 UC n=41 IG n=43 Mean age 66 (SD 11) years 99% male 77% white Measures 1. Patient activation measure 2. SCHFI, α=0.42–0.88 3. Medical outcomes study specific adherence scale 4. Hospitalization and E Intervention UC 1. Routine medical care in primary care or non-HF specialty clinic IG 1. Routine medical care in primary care or non-HF specialty clinic 2. Activation assessed and individualized health behavior goals set 3. Six sessions one-to-one education by telephone or in person by advanced practice nurse 4. Educational booklet 5. BP cuff 6. HF self-management D 7. Weight scale 8. Pedometer Measures taken at baseline, and at 3 and 6 months	IG had fewer hospitalizations in low/high activation group at 6 months (F=2.57, P=0.041) IG had more hospitalizations in medium activation group Self-care No significant differences in group-by-maintenance, management or confidence scores IG had significant increase in patient activation scores	Limitations Small sample size Homogeneous sample in terms of gender and ethnicity Small number of clinical events

Note:

* Cronbach’s alpha not reported.

Abbreviations: BP, blood pressure; CI, confidence interval; HF, heart failure; ECG, electrocardiogram; ED, emergency department; UC, usual care group; IG, intervention group; IQR, interquartile range; IRR, incidence rate ratio; SCHFI, Self-Care of HF Index; MLHF, Minnesota Living with HF Questionnaire; HR, hazard ratio; NYHA, New York Heart Association (functional class);, Cronbach’s alpha; SD, standard deviation.

Table 2 Self-care and symptom burden

Reference and aim	Sample/method/interventions	Symptom burden	Strengths and limitations
Baker et al^Citation39 Aim To examine the effect of two different levels of self-care learning on the adoption of key self-care behaviors and on HF-related quality of life USA	n=605 UC n=302 IG n=303 Mean age 60.7 (SD 13.1) years 52% male 39% white Measures S-TOFHLA ICICE* Heart Failure Symptoms Scale (used as measure of quality of life)* Intervention UC Brief Educational Intervention 40-minute educational session on daily self-assessment, action planning in case of exacerbations, salt avoidance, and medication adherence Educational manual Digital weight scale IG Teach to goal 40-minute educational session on daily self-assessment, action planning in case of exacerbations, salt avoidance, and medication adherence Educational manual Digital weight scale Specific instructions on daily weights to guide diuretic self adjustment 5–8 phone calls over 4 week period to reinforce education and guide patient towards better self-care skills Time frame of 4 weeks	IG had significant improvement in HF symptoms compared with UC (P<0.001) Symptoms were unchanged in UC group Self-care IG had a significant increase compared with UC 0.70 versus 0.30; P=0.008 Changes in self-efficacy Teach-to-goal group had a significant increase compared with brief educational intervention group (1.0 versus 0.4; P=0.006)	Strengths Large sample Heterogeneous sample in terms of gender and ethnicity Limitations High baseline knowledge of study participants Young mean age of sample
Lee et al^Citation34 Aim To test the effect of a comprehensive daily symptom diary intervention (education and counseling about HF symptoms, how to recognize symptoms, and what to do with escalating symptoms) on event-free survival and health-related quality of life USA	n=44 UC n=21 IG n=23 Mean age 60 (SD 12) years 52.3% male 61.4% white Measures 1. Event-free survival assessed at 1 and 3 months 2. MLHF (physical symptom subscale)* 3. Symptom diary Intervention UC 1. Discharge education booklet describing HF, a low sodium diet, and medication instructions IG 1. One face-to-face education and counseling session in patient’s home   a. How HF symptoms are related to fluid retention and diet   b. How to monitor symptoms   c. What to do for worsening symptoms   d. Review of medication regimen   e. Use of symptom diary using symptom clusters   f. How to weigh daily 2. Five booster sessions via biweekly phone calls for 3 months	MLHF physical subscale scores (proxy for symptom burden) were not significantly different between groups over 3 months Self-care No difference between groups at baseline Self-care at 1 and 3 months not reported	Strengths Heterogeneous sample in terms of gender Limitations Small sample size Young mean age of sample
Seto et al^Citation36 Aim To investigate the effects of a mobile-based telemonitoring system on HF self-care and clinical management Canada	n=100 UC n=50 IG n=50 Mean age 53.7 (SD 13.7) years 79% male 72% white Measures 1. SCHFI* 2. MLHF* Intervention UC 1. Standard care including   a. Visits to clinic once every 2 weeks to once every 3–6 months, depending on severity of HF   b. Education during preliminary visits to clinic   c. Use of telephone clinic as necessary IG 1. Standard care 2. Telemonitoring system   a. Daily morning weight   b. BP readings   c. Weekly ECG 3. Answered daily morning symptom questions on mobile phone 4. Given individual training session on how to use system, provided technical support during trial	No difference between groups on symptom burden (MLHF physical subscale) IG had greater improvement on MLHF emotional subscale (P=0.05) Self-care Both IG and UC had significant improvements in self-care maintenance scores. The IG had greater improvement than UC on self-care maintenance (P=0.03) Both IG and UC had significant improvements in self-care management, but no significant differences between groups (P=0.7)	Limitations Study inadequately powered to detect effects of telemonitoring on mortality and hospitalization Primarily white male sample Sample included patients new to HF clinic and stable HF patients Young mean age of sample
Shao et al^Citation35 Aim To determine the effectiveness of a self-management program based on the self-efficacy construct in older people with HF Taiwan	n=93 UC n=46 IG n=47 Mean age 72 (SD 5.5) years 68% male Measures 1. Self-efficacy for salt and fluid control, α=0.94–0.96 2. HF self-management behavior scale, α=0.71 3. HF symptom distress scale, 4. Health service use (include patient clinical, ED visits, and hospitalization) Intervention UC 1. Educated on improving general knowledge of HF, causes, signs and symptoms, complications, medications, activity, and dietary recommendations 2. Phone calls at weeks 3, 7, and 11 IG 1. Educated on improving general knowledge of HF, cause, signs and symptoms, complications, medications, and activity and dietary recommendations. 2. Home visit within 3 days after enrollment 3. Four phone calls at weeks 1, 3, 7, and 11 4. Diary of daily sodium and fluid intake and daily weight Primary endpoint: HF-related symptoms, hospitalization at 12 weeks	IG had lower symptom distress scores than the UC at time 2 and time 3, indicating improvement in HF symptom distress Self-care Self-management behavior improved significantly in IG versus UC (mean 27.15 versus 30.13, P<0.001), with lower scores indicating better self-management	Limitations Short follow-up for measuring outcomes Lack of objective measures to support difference between and UC Homogenous population Two medical centers limits generalizability Unable to conclude which intervention was effective because the effectiveness of four information sources could not be determined

Note:

* Cronbach’s alpha not reported.

Abbreviations: BP, blood pressure; HF, heart failure; ECG, electrocardiogram; ; ED, emergency department; UC, usual care group; ICICE, Improving Chronic Illness Care Evaluation; IG, intervention group; SCHFI, Self-Care of HF Index

Davis KK, Mintzer M, Dennison Himmelfarb CR, Hayat MJ, Rotman S, Allen J. Targeted intervention improves knowledge but not self-care or readmissions in heart failure patients with mild cognitive impairment. Eur J Heart Fail. 2012;14:1041–1049.

 Web of Science ®Google Scholar

DeWalt DA, Schillinger D, Ruo B, et al. Multisite randomized trial of a single-session versus multisession literacy-sensitive self-care intervention for patients with heart failure. Circulation. 2012;125:2854–2862.

 Web of Science ®Google Scholar

Jurgens CY, Lee CS, Reitano JM, Riegel B. Heart failure symptom monitoring and response training. Heart Lung. 2013;42:273–280.

 Web of Science ®Google Scholar

Kommuri NV, Johnson ML, Koelling TM. Relationship between improvements in heart failure patient disease specific knowledge and clinical events as part of a randomized controlled trial. Patient Educ Couns. 2012;86:233–238.

 PubMed Web of Science ®Google Scholar

Lee KS, Lennie TA, Warden S, Jacobs-Lawson JM, Moser DK. A comprehensive symptom diary intervention to improve outcomes in patients with HF: a pilot study. J Card Fail. 2013;19:647–654.

 Web of Science ®Google Scholar

Seto E, Leonard KJ, Cafazzo JA, Barnsley J, Masino C, Ross HJ. Mobile phone-based telemonitoring for heart failure management: a randomized controlled trial. J Med Internet Res. 2012;14:e31.

 PubMed Web of Science ®Google Scholar

Shao JH, Chang AM, Edwards H, Shyu YI, Chen SH. A randomized controlled trial of self-management programme improves health-related outcomes of older people with heart failure. J Adv Nurs. 2013;69:2458–2469.

 Web of Science ®Google Scholar

Shively MJ, Gardetto NJ, Kodiath MF, et al. Effect of patient activation on self-management in patients with heart failure. J Cardiovasc Nurs. 2013;28:20–34.

 PubMed Web of Science ®Google Scholar

Baker DW, Dewalt DA, Schillinger D, et al. The effect of progressive, reinforcing telephone education and counseling versus brief educational intervention on knowledge, self-care behaviors and heart failure symptoms. J Card Fail. 2011;17:789–796.

 Web of Science ®Google Scholar