The LVAD Self-Care Behaviour Scale

A picture of a leaf in backlight

The Left Ventricular Assist Device Self-care Behaviour Scale.

Heart failure is a common and progressive condition affecting millions of people worldwide. For patients with advanced heart failure who are no longer responsive to optimal medical therapy, the implantation of a Left Ventricular Assist Device (LVAD) offers a life-prolonging treatment option. While LVAD therapy can significantly improve survival, it also requires patients and their caregivers to take on extensive self-care responsibilities in the home setting.

Adequate self-care behaviour is essential for patients with an LVAD in order to prevent complications, prolong life, and enhance quality of life. However, until recently, no comprehensive and validated instrument was available to assess self-care behaviours specific to this patient group. Therefore, following a literature review,¹ we developed scale items using the Delphi method,2 and assessed the scale's psychometric properties.3

The 20-item LVAD self-care behaviour scale is a valid and reliable instrument to measure self-care behaviour for LVAD supported patients in research and clinical practice.3 The scale is divided into three subscales: Monitoring, Heart failure self-care and LVAD self-care. The total score and subscale scores will be standardized, ranging from 0 to 100, with higher scores indicating better self-care. More information on scoring can be found lower on the webpage.

License

You are welcome to use the LVAD self-care behaviour scale. No license is needed, and it is free to use.

References

1. Kato N, Jaarsma T, Ben Gal T. Learning Self-care After Left Ventricular Assist Device Implantation. Curr Heart Fail Rep 2014;11:290-298. doi: 10.1007/s11897-014-0201-0
2. Kato NP, Jaarsma T, Casida JM, et al. Development of an Instrument for Measuring Self-Care Behaviors After Left Ventricular Assist Device Implantation. Prog Transplant 2019;29:335-343. doi: 10.1177/1526924819874358
3. Kato NP, Melnikov S, Denfeld QE, et al. Validity and reliability of the left ventricular assist device self-care behaviour scale. PLoS One 2023;18:e0275465. doi: 10.1371/journal.pone.0275465

Research group

Development of 33-item LVAD Self-Care Behavior Scale

Summary of the development process of 33-item LVAD Self-Care Behavior Scale

The development of the LVAD Self-Care Behavior Scale began with a narrative literature review that identified 56 self-care behaviors related to living with a Left Ventricular Assist Device (LVAD). These behaviors were drawn from both qualitative and quantitative studies as well as practical guidelines. Five qualitative studies contributed insights into the self-care experiences of patients and caregivers after LVAD implantation, highlighting issues such as dressing and showering, physical and psychological limitations, sterile dressing techniques, dealing with device-related problems, and the role of social support.

A pciture for scoring within self care

The Middle-Range Theory of Self-Care of Chronic Illness,1 was chosen as a fundamental theory of this scale. The theory includes three key components: self-care maintenance, self-care monitoring, and self-care management. A research team consisting of four nurses and one cardiologist reviewed the 56 items for overlap and relevance, reducing the list to 37 items that measures LVAD self-care behaviors.

Next, a two-round Delphi survey was conducted with 17 international experts in LVAD and heart failure care, including LVAD coordinators, nurse practitioners, cardiologists, and one surgeon. In the first round, 32 of the 37 items were rated as important by over 80% of panelists, and 33 were rated as relevant. Four items were removed due to being vague, overly subjective, or overlapping with others. For example, an item about avoiding infection was removed due to varied interpretations, but a more specific item about monitoring infection signs was retained. Several items were revised based on panel feedback, such as specifying alcohol limits (e.g., 1 unit/day for females, 2 units/day for males). Additionally, four new items were suggested by panelists: attending scheduled clinical visits, monitoring stroke symptoms, changing backup controller in case of emergency, and contacting an LVAD team when a patient is seen in another physician’s office or an emergency department. Two of these were included on the final scale.

In the second round of the Delphi survey, over 85% of panelists agreed to retain 30 high-importance items from the first round, but one item (item 35) was excluded due to overlap with item 7, resulting in 29 retained items. Of the 7 moderately important items, 5 (items 34, 36–39) were removed, while items 22 and 31 were reinstated after discussion despite partial agreement. Among 4 new proposed items, only 2 (items 18 and 24) were added based on majority support. As a result, the final version of the new LVAD Self-Care Behavior Scale included 33 items: 19 related to self-care maintenance, 10 to self-care monitoring, and 4 to self-care management.

Two response formats were considered: frequency of performing behaviors (“never” to “always”) and level of agreement (“totally disagree” to “totally agree”). The research team chose the frequency-based 5-point Likert scale, as it was easier for patients to use and aligned with standard self-care assessments in heart failure, hypertension,2 and coronary heart disease.3

Finally, face validity was tested with 25 LVAD patients in Israel and Japan. The patients found the items understandable and relevant to their self-care practices, and without any difficulty in answering or interpreting the questions. The questionnaire took approximately 5–7 minutes to complete.


A detailed description of the scale's development can be found in:

Cover of publication ''
Naoko Perkiö Kato, Tiny Jaarsma, Jesus M Casida, Christopher S. Lee, Anna Strömberg, Tuvia Ben Gal (2019)

Progress in transplantation , Vol.29 , s.335-343 Continue to DOI

The refinement of the 33-item scale into a 20-item version

The refinement of the original 33-item LVAD Self-Care Behaviour Scale into a psychometrically robust 20-item version involved a structured, four-step process grounded in both statistical analysis and clinical assessment. This process aimed to improve the scale’s clarity, relevance, and discriminative ability, ensuring it was suitable for use across diverse clinical and research settings.

Step I – Descriptive Item Analysis (Initial Item Reduction)

The first step involved examining item-level statistics including means, medians, skewness, kurtosis, floor/ceiling effects, and item-total correlations. Six items were removed due to their limited ability to differentiate among patients or their lack of clinical specificity.

Three items (Item 8: "follow the steps when changing the dressing", Item 23: "take my medicine as prescribed", and Item 33: "attend scheduled clinical visits") were excluded because they showed extremely high mean scores (≥4.9) and minimal variance, suggesting a ceiling effect. These items were universally endorsed by nearly all patients, thus offering little value in distinguishing levels of self-care behavior. Item 31 ("measure my blood pressure daily") was removed because not all patients had access to the necessary equipment or the ability to measure their own blood pressure, leading to potential inconsistencies and missing data. Two items reflecting general health behaviors (Item 27: "limit my alcohol intake" and Item 28: "avoid cigarettes and tobacco smoke") were also excluded. These behaviors, while important, are not unique to LVAD self-care and may reflect broader lifestyle habits rather than LVAD-specific adherence. Additionally, the alcohol item had a relatively high percentage of missing data (3.2%), likely due to variations in cultural norms and understanding of alcohol units.


Step II – Exploratory factor analysis (on 27 items)

After excluding 6 items, an exploratory factor analysis (EFA) using promax rotation was conducted on the remaining 27 items to explore underlying factor structure. Two items were removed at this stage: Item 1 ("I clean the controller, batteries and battery connection") and Item 4 ("I keep both the back-up battery and controller with me". These items did not contribute meaningfully to any factor and were thus deemed psychometrically weak.

Step III – Second EFA (on 25 items) and clinical review

A second EFA was conducted, leading to the removal of an additional five items based on low or negative factor loadings, clinical irrelevance, or redundancy. Item 9 ("check and record LVAD speed, flow, power and PI") showed negative loading and was considered clinically outdated due to changes in device design, Item 6 ("wear a stabilisation device") was deemed not universally relevant for home care across patient populations, Item 30 ("contact the LVAD/heart failure team in case of symptoms") overlapped conceptually with an existing item on responding to alarms, Item 21 ("talk with someone when feeling sad or worried") was considered redundant within the scope of the scale and Item 10 ("inspect all cable connectors and the driveline for dirt or damage") overlapped with other items focusing on cleanliness and care of the driveline, risking over-representation of that behavior.

These decisions were supported by both factor loadings and discussions among the multidisciplinary research team, ensuring clinical relevance was preserved.

Step IV – Final EFA and model confirmation (20-item version)

A final EFA was performed on the remaining 20 items. The Kaiser-Meyer-Olkin (KMO) measure verified sampling adequacy (KMO = 0.79), and Bartlett’s test of sphericity was significant (χ² = 834.8, p < 0.001), confirming the data’s suitability for factor analysis.

Three interpretable factors emerged, explaining 46.7% of the total variance:

1. Factor 1 – Monitoring: Capturing daily self-assessments of symptoms and physiological status (e.g., fever, leg swelling, shortness of breath).
2. Factor 2 – Heart Failure Self-Care: Reflecting maintenance behaviors such as diet, fluid intake, physical activity, and sleep.
3. Factor 3 – LVAD Self-Care: Focused on behaviors specific to LVAD device management and communication with the care team.

Each item demonstrated adequate loading on its respective factor, and the internal consistency for the total scale (Cronbach’s α = 0.80) and subscales (α = 0.81 for Monitoring, α = 0.67 for HF Self-Care, α = 0.63 for LVAD Self-Care) met or approached recommended thresholds.


A detailed description of the refinement of the 33-item LVAD Self-Care Behaviour Scale into the 20-item LVAD Self-Care Behaviour can be found in:

Cover of publication ''
Naoko Perkiö Kato, Semyon Melnikov, Quin E. Denfeld, Jesus Casida, Anna Strömberg, Tuvia Ben-Gal, Christopher S. Lee, Tiny Jaarsma (2023)

PLOS ONE , Vol.18 Continue to DOI

References

1. Riegel B, Jaarsma T, Stromberg A. A middle-range theory of self-care of chronic illness. ANS Adv Nurs Sci. 2012;35(3):194-204. doi:10.1097/ANS.0b013e318261b1ba.
2. Dickson VV, Lee C, Yehle KS, Abel WM, Riegel B. Psychometric testing of the Self-Care of Hypertension Inventory. J Cardiovasc Nurs. 2017;32(5):431-438. doi:10.1097/JCN.0000000000000364.
3. Vaughan Dickson V, Lee CS, Yehle KS, Mola A, Faulkner KM, Riegel B. Psychometric testing of the Self-Care of Coronary Heart Disease Inventory (SC-CHDI). Res Nurs Health. 2017;40(1):15-22. doi:10.1002/nur.21755.

Translated versions

The 20-item LVAD self-care behavior scale, published in 2023, has been translated into several languages. Please find the translated versions here. If there is no version in your desired language, please contact us. A manual of the translation procedure is available on this webside.

Scoring

Scoring (pdf)

Instructions for translation

5 steps for translation and back-translation of the scale

1. Use the English version as a basis.

2. Recruit 2 translators for the forward translation: English to ‘new language’. It is generally recommended that the forward translators should have the ‘new language’ as their mother tongue. It is recommended that one translator has expertise on the construct to be measured (e.g. nurse, health care professional), and the second one being a language expert, but naive on the topic. The translators should work independently from each other and be instructed to stay close to the English version.

So in step 2 you make two ‘new Language versions (1 per translator);

3. In this step you combine the 2 ‘new language’ versions into one. Discuss possible differences and use the words that are closest to meaning in the English version

4. Recruit 2 other translators for the backward translation: ‘New language’ (from step 3) to English.

The back translators should have the original language as their mother tongue (= English). They should be blind for the original version of the questionnaire. It is recommended that the back translators are both language experts and naïve to the constructs to be measured.

The translators should work independently from each other.

So in step 4 you make two English versions (1 per translator).

5. Combine the 2 English versions from step 4 into one. Discuss possible differences and make a final ‘back-translated English’ version. Send that version to Tiny Jaarsma for a final check.