Injuries are a significant public health concern worldwide (1, 2) and it became the fourth leading cause of death among older adults (3). According to data from Ethiopia’s Health and Demographic Surveillance Sites, injuries constitute 6.4% of the 9,719 older adult deaths recorded in 3 years, with fall down injuries accounting for the majority (4). Now a days, fall down injuries have become one of the most prevalent and harmful injuries that affect older adults (5).

Age-related physical changes, pre-disposing comorbidities, and environmental factors like utilizing worn out shoes are usual causes of fall among the older adults (6). Regardless, a fall down injury might be fatal or non-fatal. Non-fatal fall injuries are highly correlated with a loss of independence (7) and it contributes to increased health care expenditures and a lower quality of life among older adults, and the expenses along with the health care services related to falls at older ages are rising dramatically on a global base (8).

Fall down injuries account for multiple complications, including fractures, among the older adult population (9). Comparably, it takes much longer for older adults to recover from fall-related injuries owing to diminishing functional capacity and physiological deterioration associated with age than adults (10, 11). Falls can lead to long-term disabilities and physical dependence (12). Falls cause not only physical injuries but also psychological repercussions such as a sense of unhappiness and discomfort (13).

Since falling exacerbates the chance of a recurrence of falling and is related to further detrimental health effects, including fear of falling (14, 15). Fear of falling, the level of certainty that someone feels in their ability to execute activities of daily living (ADL) without falling, by itself hinders the ability to perform routine activities for living, increasing the degree of dependence on assistance and reducing one’s autonomous (16). Moreover, falling increases the likelihood of falling after a certain period of time and corresponds to deconditioning, fragility, and gait disturbance (17). Commonly, surgical management, physiotherapy, and psychotherapy interventions are included in managing fall complications (18).

There are a number of fall risk assessment tools to measure the incidence of falls among community-dwelling older adults; however, fall risk level assessment among older adults following home health care services is untouched in the country. The John-Hopkins fall risk assessment tool (JH-FRAT) was first developed by Stephanie S. et al. to examine multi-factor fall risks among older adults in acute health care settings (19).

To apply JH-FRAT in the community for older adults following home health care services, the modified John-Hopkins fall risk assessment tool (mJH-FRAT) was cross-culturally adapted and validated (20) and it has demonstrated high specificity and sensitivity and is easy to use (20). The tool has seven items that help predict an older adult’s fall risk level. It has a total score of 35 and three categories, such as low risk (0–6), moderate risk (7–13), and high risk (14–34).

Taking into consideration the incidence of falls among older adults following home health care services and its negative impact on their ADL, psychological status, and quality of life, assessing their fall risk level with a valid and reliable measurement tool is a crucial step. Moreover, the measurement tool’s psychometric properties should be evaluated contextually to avoid its varying nature in relation to cultural context, literacy level, and age. However, there is no cross-culturally adapted, valid, and reliable fall risk measurement tool in Amharic to use among such a population in Ethiopia. Therefore, the aim of the present study is to cross-culturally adapt and examine the content validity and reliability of the English version of the modified John-Hopkins fall risk assessment tool on this population in Ethiopia.

This community-based cross-cultural adaptation, content validation and reliability study was conducted from November, 2023 to May, 2024. The current study followed a two-stage methodology to accomplish its objective. First, translation and cross-cultural adaptation of the English mJH-FRAT into Amharic, face and content validity were carried out. Next, the reliability of the Amharic version of mJH-FRAT was scrutinized. This study was carried out based on the Helsinki Declaration. The Institutional Review Board of the School of Medicine of the University of Gondar approved the study (ref no: SOM 575). The signed informed consent form was attained from every participant after providing a verbal and written account. Stage one:

The translation and cross-cultural adaptation of the English mJH-FRAT into Amharic language occurred based on the cross-cultural adaptation and translation of measurement tool guideline stated by Beaton’s (21). The detailed procedural steps are mentioned below.

Step 1: Two bilingual English-speaking forward translators who are fluent in both Amharic and English independently translated the original mJH-FRAT along with its specific instructions into Amharic (Am). As recommended, the first forward translator had a medical background, and he was from the department of physiotherapy at the school of medicine and health sciences at the University of Gondar, and he had information about the aim of the study. Whereas the second forward translator was an English language expert, and he is from the English language department, and he had no information about the aim of the study.

Step 2: The two forward translators and principal investigator combined the two forward translations (FT1 and FT2) into a common version (FT12) through consensus. This step resulted in a consensus translation of the Amharic version of mJH-FRAT (mJH-FRAT-Am I), and the principal investigator was there to mediate the synthesis of T12.

Step 3: Forward-translated Amharic version I of mJH-FRAT was translated back into English by two bilingual back translators independently. The two translators were neither aware nor informed about the concepts explored to avoid information bias and highlight unexpected meaning (22). Both the translators were from the department of English language, at the University of Gondar. They met and synthesized the back-translated English version of mJH-FRAT. This process was to make sure the content of the back-translated version was consistent with the content of the original version and spotting out possible imperfections.

Step 4: The expert panel was formed by five members: three physiotherapists (MSc), one from the language department (MSc), and one methodologist (PHD). The panel members, therefore, met and reviewed the translated version and discussed the clarity, understandability, and comprehensiveness of all the items in the questionnaire to reach a consensus on any possible discrepancy. Moreover, the panelists assessed the equivalence of the original and translated versions using four criteria: semantic equivalence, idiomatic equivalence, experiential equivalence, and conceptual equivalence (23). Finally, they proposed what would be considered the pre-final version of the questionnaire for further testing.

Step 5: The pre-testing of the pre-final Amharic version of mJH-FRAT was carried out using cognitive interviews with participants from the target population. The cognitive interview was carried out with a total of 30 older adults who follow home health care services in the study area. These participants were involved for the pre-testing purpose only. Hence, they did not participate in the reliability assessment. The clarity of instructions, language appropriateness, cultural suitability, and acceptability of the scale were evaluated as well. Lastly, the translated and cross-culturally adapted Amharic version of mJH-FRAT was prepared for face and content validation assessment.

Face validation is performed to assure that lay experts easily understand the items of the scale in terms of their feasibility, formatting, readability, and/or clarity in language (24). The adapted modified John-Hopkins fall risk assessment tool was checked for face validity by 15 randomly selected lay experts with yes or no responses to indicate favorable and unfavorable evaluation criteria, respectively. Lay experts were health practitioners (physiotherapists and clinical nurses) who were providing health care services at home. They were asked if the format and items were clear, understandable, and contextual language selection.

Content validity is also known as intrinsic validity, and it is a pre-request for further statistical validity. It is used to measure the comprehensiveness and representativeness of the content domain of items in a questionnaire (25). Moreover, it allows the researchers to obtain a clear picture of the limitations, dimensions, and components of the construct from the panel of experts (26). The content validity of the adapted mJH-FRAT was assessed by content validity determination and judgment quantification (27).

Two independent expert panels were formed for the content validity determination and quantification procedure. Each expert panel had 10 members from academia (physiotherapy, occupational therapy, and biostatistics departments). The criteria for panel member selection were expert’s knowledge of the subject matter, specific training, and work experience over 5 years. A consent form was sent to all the experts, and they were notified that they were taking part in the study voluntarily and that they had the right to withdraw at any time. The adapted modified John-Hopkins fall risk assessment questionnaire was then sent to the members of the expert panels through email after their consent to participate was secured.

The first expert panel was responsible for content validity determination. Additionally, they were also responsible for adding, removing, or modifying the items and evaluating the scale’s items for their representativeness, applicability, and feasibility in low-resource study settings. A scheduled face-to-face panel discussion was held to reach a consensus regarding their judgments to review and endorse the appropriate and feasible means of instruction.

The second expert panel was formed mainly to reduce over-or under-estimation of rating and judging. The panel members received the tool with a checklist to rate the preliminary scale’s items in terms of their relevancy and essentiality for the content domain of the scale. The rating process took no more than 15 min.

A robust procedure was followed to translate and cross-culturally adapt the original mJH-FRAT into Amharic. The forward and backward translations were carried out satisfactorily. Next, an in-person discussion was held by the expert committee to screen the establishment of equivalence between the translated and the original questionnaire based on the criteria mentioned earlier. Then, the committee approved that the translated questionnaire fulfilled all the equivalence criteria. Moreover, the instructional design was endorsed to be a face-to-face questionnaire administration method to make the tool feasible for the illiterate people. In conclusion, the translation and cross-cultural adaptation task was completed successfully.

The present study involves 15 lay experts for face validity assessment, and they reported that all the items were easy to understand. Additionally, they also asserted that the domain frame of the questionnaire was logical. Moreover, the experts support that the scale format was relevant to the measuring tool. Conversely, they revealed a jargon phrase that would be difficult to understand by the study population. The phrase in the item (ሲራምዱ ሚዛንዎን ለመጠበቅ ይቸገራሉ) was replaced with more clear words (ሲራምዱ የመንገዳገድ ችግር አለብዎት ወይ) as lay experts have recommended. Generally, the Amharic mJH-FRAT demonstrates excellent face validity.

The content validity assessment included a total of 20 panel experts in two different phases. All the experts approved the two identified domains and all the items in the first phase of the panel discussion. The panelists raised invaluable comments and added one item (“presence of fear of falling”) that is evidenced to be related to the content domain of the questionnaire and a cause for falling (46). Moreover, there was some modification in terms of item relocation to another domain and reordering of items with their respective domains. Then, the revised and adjusted version was resubmitted via email to experts for approval. Finally, the preliminary Amharic version of the modified John-Hopkins fall risk assessment tool, composed of eight items with two domains, was prepared for content validity quantification.

The content validity quantification phase shows that the item level CVI of all the items in both domains ranged from 0.8 to 1 (Table 1). The S-CVI based on average for the overall items was 0.96. The S-CVI based on the universal agreement value for the overall 8 items was 0.75. The kappa statistic coefficient value was ranged from 0.79 to 1 (Table 1). The CVR result was between 0.8 and 1 (Table 1).

The above findings support that all the items are relevant and essential for the content domain of the questionnaire, and the expert’s agreement was not affected by a chance agreement. This implies that the Am-JHFRAT is highly content valid to measure fall risk levels among older adults receiving home based health care services in Ethiopia.

The current study included a total of 150 older adult participants (75 for intra-rater reliability and 75 for inter-rater reliability) for the reliability assessment. The majority (43.4%) of the participant’s ages were between 70 and 75 years old. Above half (67.4%) of them were male. Only a few (12%) of participants achieved a college or higher education level. Additionally, the majority (62.7%) of them were married. Over half (68%) of them were Orthodox Christians. Furthermore, the majority (46%) of them had a moderate level of fall risk. The detailed socio-demographic data is illustrated in the Table 2.

The overall intra-rater reliability of the Amharic version of the modified John-Hopkins fall risk assessment scale was excellent (K = 0.94). In addition, the Cohen’s kappa coefficient value for the intra-rater reliability of each domain was 0.90 and 0.86, respectively. The result reveals that the overall tool has excellent intra-rater reliability and at the domain level as well.

Similarly, the overall inter-rater reliability of the Amharic version of the modified John-Hopkins fall risk assessment scale was excellent (K = 0.93). Moreover, the Cohen’s kappa coefficient values for the inter-rater reliability of each domain were 0.89 and 0.92, respectively, which supports the fact that the tool has excellent inter-rater reliability both at the scale and domain level.

The present study provides a contextual, content valid, and reliable fall risk assessment scale to apply during home-based health care services. The face validity was also evaluated, followed by a content validity examination. The study finding supports its use in the advancement of home based health care services and research work on this population to classify their fall risk level.

Despite this, the current study has a few limitations. To begin with, the study utilized a relatively small sample size with a convenient sampling technique, which may affect the representativeness of the result. Furthermore, due to the fact that there is no similar validated tool that measures the construction under consideration in an Ethiopian context, we were not able to scrutinize the tool’s construct and criterion validity, suggesting a need for further psychometric assessment of this measurement tool in future studies for better evidence with a large sample size and probability random sampling technique.