Adaptation of the CONOCES-BLS scale to citizens: evidence of content validity and preliminary psychometric evaluation

Highlights

• •A BLS knowledge scale was adapted for use among adult citizens.
• •Expert consensus provided initial evidence supporting the content validity of the instrument.
• •All items were considered essential for assessing BLS knowledge.
• •Preliminary psychometric testing showed adequate internal consistency (α = 0.82).
• •Further psychometric evaluation is required to support its practical use.

Abstract

Introduction

Out-of-hospital cardiac arrest constitutes a high-priority health emergency. The evidence underlines the importance of early actions of bystanders through the activation of the emergency system, basic cardiopulmonary resuscitation, and defibrillation. Nevertheless, basic life support training in the general population is irregular, and limitations exist on the availability of validated instruments to assess knowledge in this area.

Objective

To adapt the CONOCES-BLS scale for use in the adult general population and to obtain initial evidence supporting its content validity through expert consensus, followed by a preliminary psychometric evaluation.

Method

Methodological study on the adaptation and content validity and preliminary test of the instrument in three phases. In the first, the items were adapted to the context of an adult citizen. Then, content validity was assessed through a Delphi panel, and lastly, a pilot test was conducted for the initial psychometric analysis.

Results

The content validity showed high levels of agreement, >0.90 in every item, and K* coefficients between 0.91 and 1, providing initial evidence supporting the content validity of the adapted instrument. All the items were considered essential, although item ES12 was less clear (0.71). In the psychometric analysis, the scale obtained a Cronbach’s alpha of 0.82, demonstrating an adequate internal consistency. Items with a lower contribution to the construct were identified, especially ES2 and ES5.

Conclusion

The adapted CONOCES-BLS-Citizens scale provides initial evidence supporting its content validity for assessing BLS knowledge in the adult general population. As the first instrument specifically adapted for this purpose, it may support the evaluation of community BLS knowledge. Further studies are needed to strengthen the psychometric evidence supporting its use.

Introduction

An out-of-hospital cardiac arrest (OHCA) is one of the main healthcare emergencies worldwide. Survival depends on early detection, effective alerting of emergency services, and immediate performance of basic cardiopulmonary resuscitation (CPR) maneuvers with defibrillation within the first minutes after cardiac arrest.¹ However, the rate of survival in Spain is lower than 10%.²

Many international organizations recommend community education in basic life support (BLS) as a high-priority strategy for health promotion, the prevention of avoidable deaths, and the strengthening of citizen resilience.³ This recommendation is supported by evidence showing that community-based interventions, particularly CPR and AED training, are associated with increased rates of bystander CPR and AED use and improved survival after OHCA.⁴ The “chain of survival”⁵ underlines the relevance of the immediate action of bystanders as a critical link for saving lives. However, knowledge alone may not be sufficient to initiate CPR, as fear of causing harm and lack of confidence to act are relevant barriers among bystanders. Training may help increase confidence and willingness to intervene in a real emergency.⁶ Thus, the international consensus, including the European Resuscitation Council (ERC) Guidelines, supports the need to systematically train the lay population in reanimation techniques, providing specific recommendations for lay rescuers; therefore, teaching CPR to the general population is currently considered a priority in emergency systems worldwide.1, 7, 8, 9 Despite the efforts made, the preparation of the population with respect to CPR is still unequal, and that the current programs have limitations in terms of accessibility, continuity, and assessment of results.3, 10, 11, 12

Additionally, there is currently a gap with respect to the availability of assessment instruments supported by adequate validity evidence that allow for the rigorous measurement of the impact of BLS on the public. Indeed, recent studies assessing CPR knowledge among adults have relied on self-designed questionnaires, highlighting the need for standardized instruments supported by appropriate validity evidence.¹³ The development of reliable measurement instruments supported by validity evidence is an essential methodological requirement for the design of efficient interventions in the area of community nursing and health promotion,¹⁴ where nurses can play a key role in BLS education and training within the community.¹⁵ Having such instruments allows us to not only establish precise baselines, but also to monitor changes attributed to training or community programs, to optimize the assignment of resources, and to favor the data-driven decision making.¹⁶ In this sense, the construction and validation of specific scales contributes to the strengthening of the methodological quality of the interventions and to increasing their transfer to society.¹⁷ Therefore, this study aimed to adapt the CONOCES-BLS scale for use in the adult general population and to obtain initial evidence supporting its content validity and preliminary psychometric properties, addressing the need for reliable tools to assess citizens’ BLS knowledge and support the development of community-based educational interventions.4, 12, 18

Methodology

Study design

A methodological study was conducted to adapt the CONOCES-BLS scale¹⁸ for the adult general population and to gather initial evidence supporting its content validity and preliminary psychometric performance. With the expressed permission from the author of the original scale, the process was developed following the ACCORD guideline¹⁹ and the methodology for the contextual adaptation of assessment instruments.

Procedure and phases of the study

Following the methodology described for the adaptation of scales to another context,¹⁷ the study was conducted in three phases: adaptation, content validity and preliminary test.

Phase 1

The original CONOCES-BLS scale, developed for schoolchildren, was adapted by the research team for use in the adult general population updated according to the latest ERC Guidelines.1, 20 The adaptation process was initially conducted independently by the researchers, followed by several consensus meetings to review and discuss the proposed modifications. Any discrepancies identified during the individual review were discussed and resolved collectively during this process. The adapted version was subsequently submitted to an expert panel for content validation through an e-Delphi process.

Phase 2

To obtain the validity of the content of the items21, 22 the consensus from experts was sought through the Delphi method, and lastly, a pilot test was conducted with the instrument created to measure its psychometric behavior. This group-based decision method allows discovering the opinion of a group of experts on a topic in a structured manner through a questionnaire, in which the elimination or inclusion of the items is performed through agreement.23, 24, 25, 26, 27 The objective of the panel was to analyze the dimensionality of the first version proposed, in order to find the useful, clear, and relevant elements.

• •Expert panel data collection procedure:

For the collection and extraction of data, an online questionnaire was constructed using an eDelphi. The questionnaire was composed of various blocks: the sociodemographic data of the participants; the second block contained the adapted items from the CONOCES-BLS scale.¹⁸ For each item, the experts were asked to score, using a Likert scale (from 1 to 5; in which 1 meant complete disagreement and 5 complete agreement), the degree of relevance, clarity, and usefulness of each item proposed. In addition, an empty space was provided after each item so the experts could write any necessary comments about each of the items analyzed, and an empty space at the end of the questionnaire for comments on the scale as a whole or any other matter. The comments were independently reviewed by the researchers and subsequently discussed collectively. After considering the suggested modifications and comments, no discrepancies were identified during the process.

• •Consensus criteria and content validity analysis:

In the first round, to clarify the suitability of the content,²⁸ the measurements centered on the identification of the degree of agreement among the experts were used as the criteria. Two criteria were applied. On the one hand, the percentage of one or some of the answer categories was utilized. This criterion was used for scalable questions, in which two contiguous categories could be considered²⁴; an item was accepted if it obtained an agreement (Likert score = 4) or complete agreement (Likert score = 5) score by 70% of the experts.²³ In addition, the content validity of the instrument was assessed through quantitative measurements based on expert judgements.²⁰ Validity is understood as the degree in which the instrument measures what it intends to assess.²⁵ For this, the relevance scores were dichotomized, according to Polit and Beck,²⁹ considering values ≥4 as relevant. The item content validity index (I-CVI) and modified kappa (k*) were calculated to correct for chance agreement,22, 27, 29 and the scale content validity index (S-CVI), and the context validity ratio (CVR) to quantify the degree of agreement with respect to the suitability of each element and the instrument as a whole.²¹ These indices include adjustments to mitigate the potential bias introduced by the evaluators. The I-CVI was calculated as the number of experts who provided a “very relevant” rating to each item, divided by the total number of experts. The values varied from 0 to 1: when the I-CVI > 0.79 the element was considered relevant, between 0.70 and 0.79, the items needed to be revised, and if the value was lower than 0.70, it was eliminated.³⁰ Likewise, the Scale-level Content Validity Index was calculated using the average method (S-CVI/Ave), obtained by averaging the I-CVI values across all items. Universal agreement (UA) was also assessed as the proportion of items for which all experts rated the item as highly relevant (S-CVI/UA).³¹

• •Population and sample in Phase 2:

A panel of 27 experts with experience in BLS, teaching, and/or health research methodology from different professional fields was selected using purposive sampling.23, 32 The inclusion criteria were active professionals, older than 18 years of age with experience in BLS and/or psychometry, who voluntarily accepted to participate in the study. The sample size was determined following intervals recommended in the literature, as an exact number needed for a Panel has not been defined.23, 24, 33 Experts were purposively selected by the research team according to predefined eligibility criteria to ensure multidisciplinary representation. The panel included experts in nursing, emergency medicine, and emergency medical services, as well as academics, healthcare managers, clinicians, psychometricians, a member of the original scale development team, and experts from different geographical regions.

Additionally, a group of non-healthcare adult individuals (n = 4) were included as representatives of the target population, within the framework of Patient and Public Involvement (PPI), following current recommendations.³⁴ These participants completed the adapted questionnaire using a cognitive interviewing approach to identify potential difficulties in reading, comprehension, or interpretation of the items, and to assess the clarity, wording, and overall comprehensibility of the instrument before pilot testing.35, 36, 37 After completing the questionnaire, participants were asked to comment on the clarity, wording, and understanding of each item, and to identify any ambiguous or difficult-to-understand terms.

Phase 3

Lastly, a pilot project with the instrument created was carried, with a total of 142 citizens enrolled in BLS training workshops. The questionnaire was administered before the BLS training began, in order to measure the initial psychometric behavior of the new questionnaire following the taxonomy recommended.³⁸

• •Population and sample in Phase 3:

A total of 142 citizens older than 18 years of age, residents in the Community of Valencia who met the inclusion and exclusion criteria and were enrolled in BLS training workshops. All study data used for the preliminary psychometric analysis were collected before the training session began. As the primary aim of this study was to obtain preliminary evidence of the adapted instrument's psychometric performance rather than to establish its definitive psychometric properties, a sample of 142 participants was considered appropriate. Although no universally accepted standard exists for determining sample size in psychometric research, methodological recommendations commonly suggest including 5–10 participants per item, with at least 10 participants per item representing a conservative criterion for initial psychometric evaluation.³⁹ The adapted instrument comprises 12 items; therefore, the sample of 142 participants provided 11.8 participants per item, exceeding this conservative criterion.

Exclusion criteria of citizens:

• •Healthcare staff with recent BLS training (last 12 months).
• •Severe disability that does not allow an adequate participation in the program.
• •Having some language barrier that is insurmountable with the available resources, that prevents adequate participation in the program.

The participant flow across the different phases of the study is presented in Supplementary Fig. S1.

Variables

The suitability and validity of the content of the items were measured, using the variables: essentiality, clarity, relevance and usefulness. The preliminary assessment of the instrument included internal consistency, together with descriptive indicators of feasibility, such as completion time and response completeness.

Measurement instruments

The original CONOCES-BLS scale¹⁸ is an instrument designed to assess the level of BLS knowledge in a general or non-expert population, with a special orientation to educational and community contexts. As for reliability, it has an acceptable internal consistency, with a Cronbach’s α and McDonald omega higher than 0.70. It is composed of 12 items, with 3 possible answers per item. One point is assigned to each correct answer. Scores between 0 and 5 indicate insufficient knowledge, 6–10 sufficient knowledge, and 11–12 excellent knowledge.

Data analysis

A descriptive analysis of the sociodemographic and clinical variables was performed, with the results expressed as frequencies, percentages, means, and standard deviations according to variable type. Parametric or non-parametric tests were used according to the normality of the data, which was previously evaluated with the Shapiro-Wilk test.⁴⁰ The statistical significance was set at p < 0.05, and all the analyses were performed with the JAMOVI graphical interface.⁴¹

Results

In Phase 1 of the instrument adaptation process, the items were revised by the researcher’s team to verify that they followed the international guidelines from the latest recommendations published by the ERC 2025.¹ The original structure and conceptual meaning of the instrument were preserved, with no items or domains added or removed. At this stage, some modifications were included in items ES3, ES5, ES8 and ES10. See Table 1.

Table 1.

Example of initial modifications to the CONOCEs scale.

Original	Initial modifications according to the researchers' criteria: adaptation
ES3-When I find a person lying on the ground and after asking for help, I must check if they respond to both verbal stimuli (shouting at them) and painful stimuli (shaking them)	ES3-When I find a person lying on the ground, I must check that the scene is safe, then ask for help and check if they respond to both verbal stimuli (shouting at them) and painful stimuli (shaking them)
ES10-I can perform resuscitation maneuvers omitting ventilations and applying continuous chest compressions at a rate of 100 per minute	ES10-I can perform resuscitation maneuvers omitting ventilations and applying continuous chest compressions at a rate of 100–120 per minute

Source: Created by the authors.

In the Phase 2, a total of 27 experts were invited in the first phase of the study, who were contacted given their relevance in the topic and their professional trajectory in December 2025 to February. Lastly, a total of 21 experts fully participated in all the study rounds. From the total, 38.2% (8/21) were men and 61.69% (13/21) women, with a mean age of 42.8 years (SD ± 1.26). Most of the experts were trained as nurses 47.6% (10/21), worked in the area of healthcare (33.3%, 7/21) and had experience in BLS training for citizens (71.4%, 15/21). A more detailed description of the participating Panel members is shown in Supplemental Material (Table S1).

Then, after contacting the experts, a document was sent with the 12 items of the CONOCES-BLS-Citizens v2 scale to assess the content validity according to the context in which the scale would be applied. The experts assessed each item considering the criteria of acceptability, usefulness, clarity, and relevance/essentiality. The values obtained after the first consultation round are shown in Supplemental Material (Table S2).

Experts participated anonymously in two Delphi rounds. Afterward, the data obtained were analyzed, by revising all the items considering the opinions of the experts and analyzing all the qualitative comments provided by the experts. This analysis led to the substantial modification of 2 items, the clarification or refinement of item 5, the fusion of two of them into one (11 and 12), the incorporation of an additional item* related with the concept of Side Position of safety, in agreement with the comments from the experts. The CONOCES-BLS-Citizens v3 scale was comprised of 12 items for round 2. A detailed description of all item modifications throughout the adaptation process is provided in Supplementary Table S3.

After compiling all the answers, given the substantial modifications introduced with respect to the previous version, the degree of essentiality of the items that comprised version 2 of the scale was first analyzed. For this, the CVR of all the items was calculated, with a mean of 0.95 (SD ± 0.06) obtained for the scale as a whole. On its part, the mean modified Kappa coefficient was high in the dimensions of usefulness (K* = 0.98 ± 0.03), relevance (K* = 0.98 ± 0.03), and clarity (K* = 0.89 ± 0.07). The percentage of items with an excellent agreement in all the dimensions measured was 91.6% (11/12). The agreement index values obtained after the second round of consultation are shown below in Table 2.

Table 2.

Results from the Panel of Experts Round 2, CONOCES-BLS-Citizens v3.

Item	I-CVI	Usefulness (K*)	Relevance (K*)	Clarity (K*)	Global interpretation	CVR	Interpretation
ES1	0.96	0.99	0.99	0.87	Excellent	0.91	Essential
ES2	1	1	1	1	Excellent	1	Essential
ES3	1	1	1	0.87	Excellent	1	Essential
ES4	1	1	1	0.96	Excellent	1	Essential
ES5	1	1	1	0.83	Excellent	1	Essential
ES6	0.96	0.96	0.91	0.87	Excellent	0.91	Essential
ES7	0.96	0.96	0.96	0.96	Excellent	0.91	Essential
ES8	1	1	1	0.91	Excellent	1	Essential
ES9	1	1	1	0.96	Excellent	1	Essential
ES10	0.91	0.91	0.96	0.87	Excellent	0.83	Essential
ES11	1	1	1	0.96	Excellent	1	Essential
ES12*	0.91	0.91	0.96	0.72	Excellent/good	0.82	Essential

Source: Created by the authors in JAMOVI.

A total of 2 rounds were needed to attain universal agreement among the experts, with version 2 of the scale obtaining an S-CVI/Ave of 0.98 and 7 of the 12 items achieved universal agreement (S-CVI/UA = 0.58), providing initial evidence supporting the content validity of the adapted instrument in the context of the citizens.

Next, cognitive interviews with members of the target population were conducted using retrospective verbal technique to assess the clarity, comprehensibility, and interpretation of the items, particularly those involving more complex concepts. Participants reported no difficulties in understanding or interpreting the questionnaire, and no modifications were suggested. As no new issues emerged, data collection was stopped after the fourth interview due to data saturation.⁴²

Then, in Phase 3 a pilot study was performed with a sample of 142 citizens, to analyze the initial psychometric properties of the instrument. Most of the citizens were women (68.3%; 97/142), with a mean age of 48.2 years (SD ± 1.76), with higher education training (57.7%; 82/142), with most living in urban settings (59.9%; 85/142). Most of the participants had not received previous BLS training (71.1%; 101/142) and did not have to perform resuscitation maneuvers (96.5%; 137/142).

The mean time needed to complete the questionnaire was 8.45 min (±4.35), with an increasing trend observed in individuals older than 50 years of age, who obtained a mean time of 16 min (±8 min). The descriptive analysis showed item means ranging from 0.463 and 0.97. The total score of the questionnaire was high, with a mean of 7.97 SD ± 2.685; range = 1–12, with a median of 8. See Supplemental Material Table S4.

All participants completed all items of the instrument, with no missing responses. As a preliminary assessment of internal consistency, the scale showed a Cronbach’s alpha of 0.828 for the scale as a whole. The item-total correlation were analyzed, showing that most of the items obtained moderate correlations with the total scale, although lower correlations were observed for ES2 (0.154) and ES5 (0.289). See Table 3.

Table 3.

Results from the reliability CONOCES-BLS-Citizens.

	Cronbach's alpha if an element is discarded	Correlation of the element with others
ES1	0.818	0.455
ES2	0.832	0.154
ES3	0.826	0.327
ES4	0.800	0.655
ES5	0.835	0.289
ES6	0.813	0.500
ES7	0.814	0.495
ES8	0.790	0.735
ES9	0.814	0.489
ES10	0.797	0.663
ES11	0.807	0.568
ES12	0.820	0.421

Source: created by the authors in JAMOVI.

Discussion

The present study provides initial evidence supporting the content validity and preliminary psychometric properties of the CONOCES-BLS-Citizens scale for assessing BLS knowledge in the adult general population. Unlike a simple linguistic adaptation, the instrument underwent a systematic contextual update while preserving the conceptual framework of the original scale. Beyond adapting the wording to a community setting, several items were revised according to the latest ERC Guidelines, including scene safety, compression-only CPR, and current compression rate recommendations. In addition, the expert panel identified the need to incorporate a new item addressing the recovery position, strengthening the comprehensiveness of the instrument. As resuscitation guidelines evolve over time, keeping knowledge assessment instruments up to date is both a challenge and a necessity to ensure they continue to reflect current scientific evidence and educational priorities.⁴ This systematic adaptation process contributes to the evidence supporting the content validity and enhances the practical applicability of the instrument for its intended target population consistent with the recent ILCOR consensus recommendations advocating the use of valid and reliable outcome measures for evaluating knowledge acquisition in laypersons participating in resuscitation education programs.¹³

The findings are particularly relevant given the growing recognition that the adaptation of measurement instruments requires rigorous methodological procedures. The adaptation or modification of scales remains a topic of concern in the scientific community,20, 43, 44 given that the use of non-validated instruments can compromise the validity of the results and lead to erroneous conclusions.¹⁶ In studies on CPR training programs, it is frequent to find in the literature that authors use assessment questionnaires designed ad hoc or inappropriate versions of scales. The science and practice of behavioral, social, and medical sciences greatly depend on the good measurement of the results, and a significant part of these measurements are based on tests used in languages, cultures, or contexts that are different from those for which they were initially designed.20, 44 This underlines the importance of any modification of the original scale that is supported on a rigorous evaluation of its psychometric properties.20, 43, 44, 45

In this context, following the initial steps included in the TARES declaration to transparently report about the relevant adaptations of the tests,⁴⁴ the content validity of the CONOCES-BLS-Citizens scale was assessed by Experts. All the items obtained I-CVI values higher than 0.90, indicating an excellent agreement with respect to usefulness, relevance, and clarity.21, 22, 31

Nevertheless, one of the challenges of the study was obtaining answers from the experts, given that despite their commitment with their participation in the project, and the sending of multiple reminders, six of them did not complete any evaluation round. These results were similar to those found by other authors when applying the Delphi method.17, 23, 33, 46, 47, 48, 49

As for the initial psychometric properties, Cronbach’s alpha was used as a preliminary assessment of internal consistency, based on the structure of the original scale. However, this analysis should not be interpreted as evidence of unidimensionality. Most of the items positively contributed to the internal consistency; however, ES2 and ES5 showed low item-total correlations (0.154 and 0.289, respectively), and item ES5 showed a higher effect on the Cronbach alpha value after it was eliminated (0.835), which suggests that its revision or re-formulation could improve the overall consistency of the scale.⁵⁰

The present study has some limitations. First, the process of content validation process was based on the opinion of an expert panel selected can introduce a selection bias and limit the diversity of perspectives. Likewise, the inclusion of only four representatives of the target population, although consistent with current recommendations for public involvement and sufficient to achieve data saturation during cognitive interviewing, may not fully reflect the heterogeneity of the target population. Furthermore, the pilot study was conducted using a convenience sample recruited from a single geographical region, which may limit the generalizability of the findings. In addition, as this is a self-administered instrument, the presence of social desirability bias cannot be ruled out. Although this study provides preliminary evidence supporting the content validity and reliability of the adapted CONOCES-BLS-Citizens scale to assess BLS knowledge in the general adult population, further research in larger and more diverse samples is needed to strengthen the evidence supporting its psychometric properties, particularly to examine the dimensional structure of the adapted instrument using appropriate factor analytic methods, its ability to discriminate between different levels of BLS knowledge and training and to assess potential ceiling effects. Further studies will also help determine the applicability of the instrument for evaluating community-based BLS education. Despite these limitations, the adapted CONOCES-BLS-Citizens retained the conceptual structure and objectives of the original scale while incorporating adaptations intended to improve its relevance and comprehensibility for the general adult population.

Conclusions

The CONOCES-BLS-Citizens scale provides initial evidence supporting its content validity to assess BLS knowledge in the adult general population. As the first instrument specifically adapted for this purpose, it may be useful for the assessment of community BLS knowledge and potentially, for evaluating educational interventions. Further studies are needed to strengthen the psychometric evidence supporting its use.

Fundings

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Ethical considerations

The study was approved by the corresponding Research Ethics Committee (UV-INV_ETICA-4220151) and was conducted in accordance with the ethical principles established in the Declaration of Helsinki.

CRediT authorship contribution statement

Patricia Marín-Maicas: Writing – original draft, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Ángela Soler-Sanchis: Writing – review & editing, Validation, Investigation, Data curation, Conceptualization.

Informed consent

All participants provided informed consent for the use of their anonymized clinical data for research purposes.

Use of artificial intelligence

The authors declare that OpenAI was used exclusively as a support tool for linguistic revision, improvement of manuscript wording, and editing. The study design, data analysis, interpretation of results, and final version of the manuscript were entirely conducted, reviewed, and approved by the authors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

The following are the Supplementary material to this article: