Abstract
Introduction
The present study aimed to identify signs of frequent fall-related body dysfunction (depression/cognition) as exhibited in daily activities among older adults. The role of fall risk in mediating body dysfunction and daily activities was also explored.
Method
Participants included 123 non-institutionalised older adults. Depression and cognitive status were measured by the Geriatric Depression Scale (GDS-15) and the Montreal Cognitive Assessment (MoCA). Fall risk was determined by a questionnaire, supported by the Time Up and Go test (TUG). Executive functions (EF) were assessed by the Behavior Rating Inventory of Executive Function-Adult Version (BRIEF-A) and the Alternate Executive Function Performance Test medication management performance-based assessment. Daily life measures included the Barthel and Instrumental scale of activities of daily living, and World Health Organization Quality of Life questionnaire.
Results
Based on a falls risk score, 39 out of 123 participants (32%) were high-risk fallers. High-risk fallers showed greater body dysfunction, as recognised in daily activities. Structural equation modelling (SEM) revealed that fall risk mediated the associations among depression, executive dysfunction and daily activities.
Conclusion
Emotional and cognitive dysfunctions that affect people with high fall risk may manifest while older people perform daily activities. Community fall prevention programmes should screen for such fall-related dysfunction and provide strategies to minimise falls and enhance daily function.
Introduction
Falling is a major problem in older adults. Approximately 30%–40% of non-institutionalised individuals over the age of 65 fall at least once a year (Swift and Iliffe, 2014) and half of this group fall again in the following year. In order to minimise this risk, this group should have long-term care during their life time (Pellicer-García et al., 2017). The high health costs and severe personal consequences – including injuries, immobilisation, reduced independence in daily activities and even death – make the incidence of falls a major public health concern (Barban et al., 2017; De Oliveira Cruz et al., 2020).
Thus, understanding the factors that underlie resiliency or vulnerability among fallers should be identified. Early screening of body dysfunction and related restrictions in daily activities that predict falls may facilitate an increased awareness of fall risk and improve prevention programmes (Saftari and Kwon, 2018).
Literature review
Frail older adults are at greater risk of falling. Other geriatric syndromes, such as depression and executive dysfunction, are also more prevalent among people aged 65 and above with a history of falling (Pellicer-García et al., 2017). Depressive symptoms have been identified as risk factors for falls, independent of antidepressant use (Kvelde et al., 2015).
The relationships among falls, depression and executive dysfunction can be explained by changes that occur in biochemical mechanisms and in structures of the aged brain. These changes include grey matter loss in the somatosensory and motor areas, the prefrontal and inferior parietal cortices (McGinnis et al., 2011), as well as the loss of complexity of neuronal connections (Acker, 2004) that affect motor, emotional and cognitive abilities (Kvelde et al., 2015). The high cognitive abilities referred to as executive functions (EFs) are responsible for an individual’s controlled, goal-oriented behaviour (Luiten et al., 2013). Thus, EFs are essential for motor function, emotional regulation, daily activity performance and the maintenance of an active lifestyle (Caetano et al., 2018; Hahn et al., 2014). Attention problems and uncontrolled goal-oriented behaviour may significantly impair postural control and an individual’s ability to manoeuvre around obstacles in their environment (Bernard and Lacour, 2017). Brain imaging studies have shown increased brain connectivity within the executive network, mediated between the caudate grey matter volume and fall risk, via changes in attention among people with Parkinson’s disease (PD) (Rosenberg-Katz et al., 2015). Other studies have shown that EFs deteriorate prior to, or at the same time as physical performance (Caetano et al., 2018), suggesting that EFs are also associated with fall risk in older adults without PD.
The frequent combination of reduced balance, impaired emotional status and executive dysfunction may have far reaching effects on daily life for older adults. They may exhibit decreased independence in basic/instrumental activities of daily living (BADL/IADL) (Fitzgerald et al., 2016) and a significantly reduced quality of life (QoL) (Barban et al., 2017; Laurence and Michel, 2017). Yet, information about the relationship between these factors in non-institutionalised older adults is lacking (Pellicer-García et al., 2017).
Several important points may be drawn from the literature. First, it is critical to elucidate key factors that contribute to a loss of function as well as resiliency. Second, as key factors of fall risk, the causative relationships among depression, EF, falls and daily function should be further explored. Yet, many studies that have explored these relationships have used laboratory or neuro-psychological tools. For example, EFs are frequently evaluated by measures, such as the trail making test (TMT) (Reitan, 1955) and phonological fluency test (PF) (Borkowisky et al., 1967). These measures refer to specific EF components and do not provide a thorough understanding of impairments in daily life.
Information gathered from performance-based assessments with ecological validity, in which the task is the central focus is lacking. The use of such tools may enhance our understanding of an individual’s performance in real life (Josman et al., 2009) and reveal the precise manner in which balance-emotional-cognitive difficulties affect daily function. Moreover, by recognising the deterioration in emotional-cognitive aspects during the performance of daily routine activities, early intervention could be provided, which could thereby minimise fall risk.
This approach is supported by the International Classification of Functioning Disability and Health (ICF) of the World Health Organization (World Health Organization, 2001). According to this classification, disability is no longer determined by body dysfunction but by the interaction with the person’s ability to perform daily activities. Moreover, the performance of daily activities, as well as the QoL, is the main outcome of intervention efficiency (Engel-Yeger, 2019).
As falls, body dysfunctions that refer to depression and cognitive decline are usually recognised and detected in primary care, adequate and timely treatment is not always available. Thus, it is essential to increase awareness among older people, families and health care givers of the emotional-cognitive early signs of falls and recognise those signs in older people performing daily life activities.
Based on the above, the aims of the present study were to (1) examine the prevalence of fall risk among non-institutionalised older adults. (2) Identify signs of fall-related geriatric body dysfunction (depression and cognition/EF), as evident in daily activities, which differ between high- and low-risk fallers, using self-reports and a performance-based assessment. These assessments aim to reflect the implications of the body dysfunction on daily life. (3) Explore the role of fall risk in mediating between body dysfunction and daily life.
Method
Participants
Comparison of socio-demographic parameters and frail status between groups.
Qualitative variables: # chi-squared test; quantitative variables: ## t-test.
NS: not significant.
% within group risk.
Instruments
A demographic and health status questionnaire which also included the participants’ current medication uptake.
Depression and cognitive status were measured by the following:
The GDS (Yesavage et al., 1983) – a self-report questionnaire with 15 dichotomous items (yes/no questions) designed to screen for depression in older adults. The total score ranges from 0 to 15 points. A higher score indicated the presence of more symptoms of depression. A score of ≥6 indicated a need for a thorough medical/psychiatric assessment. A score of ≥11 was considered the cutoff for an indication of depression, and a higher score indicated higher depression severity.
The MoCA (Nasreddine et al., 2005) – this short screening tool profiled the cognitive status. The areas of examination included visuo-spatial abilities, EF, attention, language, short-term memory and orientation. The total score was derived by summing the correct answers, to give a maximum of 30 points. The test is highly sensitive in identifying people with mild cognitive impairment (MCI) (83%–90%) (Nasreddine et al., 2005). Based on a sample size of 8411, Lu et al. (2011) noted that the appropriate cutoff score for people without MCI or dementia and without formal education is 13/14. As our study included both participants with and without formal education, the MoCA cutoff score in the current study was set at ≥ 14.
The assessment of fall risk was based on a questionnaire, supported by clinical examination.
Based on the Israeli Ministry of Health (2017), fall risk was determined according to two parameters: (a) A questionnaire – that gathered information about the number of falls during the previous year; information about fractures or other significant injuries caused by falls and information about walking or stability difficulties. (b) Time Up and Go test (TUG) (Podsiadlo and Richardson, 1991) – this clinical examination measured mobility and lower extremity functions and was used as a screening tool for fall risk. The subject was requested to get up from a standard chair, without using upper extremity support, and walk 3 m ahead at a regular pace. The subject was then required to turn around and walk back to the chair and sit down. The subject could have used a walking aid if needed. No physical assistance was given throughout the test; however, the examiner followed the participant to prevent any incidence of falling. The test was conducted twice. However, only the second round was scored. The score reflected the performance time in seconds and was measured by a stopwatch. A shorter performance time indicated better performance. A performance time longer than 13.5 s indicated a greater fall risk (Herman et al., 2011).
The high-risk group was defined if one of the following three scenarios occurred: 1 – the participant fell twice or more during the previous year; 2 – the participant had one fall during the previous year with a significant injury and 3 – the participant had one fall during the previous year and a TUG score greater than 13.5 s.
In the present study, fall risk was calculated as a continuous parameter that included the TUG score (‘0’ = no risk or ‘1’ = fall risk) plus the summary of the fall questionnaire scores. The range of the fall risk score was 0–10 (mean 2.33 ± 2.53). Since this variable was positively skewed, a log transformation was performed for the SEM analysis.
Assessment of EF by a self-report questionnaire and performance-based assessment: (a) The Behavior Rating Inventory of Executive Function - Adult Version (BRIEF-A) (Ciszewski et al., 2014) was used to screen for possible executive dysfunction and indicated the subject’s awareness of their own self-regulatory functioning. The BRIEF-A assessed everyday behaviours associated with specific domains of EF in adults, which were summarised in two index scales, the Behavioral Regulation Index (BRI) and the Metacognition Index (MI), as well as another scale reflecting overall functioning (Global Executive Composite [GEC]). The BRI comprised four scales: Inhibit, Shift, Emotional Control and Self-Monitor. The MI comprised five scales: Initiate, Working Memory, Plan/Organise, Task Monitor and Organisation of Materials. Behaviour frequency was rated on a Likert scale ranging from ‘rare’ to ‘often’. Raw scores were transformed to t-scores (mean = 50, SD = 10). The t-scores of 65 or above reflected executive dysfunction. (b) The medication management subtest of the Alternate Executive Function Performance Test (aEFPT) (Hahn et al., 2014) – this was an additional part of the valid performance-based EFPT assessment, which measured EF while the subject was carrying out a daily task. The subject’s functional independence level and the amount of help required during the tasks were also recorded. The four original EFPT tasks were cooking oatmeal, telephone use, taking medication and paying a bill. The aEFPT had four additional tasks, which were similar to the four original tasks but with a novel component to prevent a learning effect from the original form.
In the medication management subtest, subjects are asked to sort medications into a 7-day pill sorter, instead of taking a medication as requested on the original form (Hahn et al., 2014). The subject is instructed to find and sort medicines in a weekly pill sorter. For successful performance, the subject has to ignore distractors (other bottles) and use prospective memory to follow the specific sorting instructions. This medication management task was selected as it is a daily common function among older adults, which does not require special facilities (like a stove top for cooking, which was one of the subtests) and can be completed within a relatively short time.
The scoring referred to five EFs: initiation, organising, sequencing, safety and judgement and completion. Each component was scored on a scale of 0–5 points, according to the level of assistance required to complete each the task. The points were allotted as follows: 5 – doing the task for the participant; 4 – physical assistance; 3 – verbal direct instruction; 2 – gestural guidance; 1 – verbal guidance and 0 – independent performance of the task. The score for the complete task ranged from 0 to 25, and a higher score indicated that more assistance was required (Hahn et al., 2014).
Daily life measures:
(a) The Barthel Index of ADL (Mahoney and Barthel, 1965), which measured the performance of BADL in 10 functional domains: eating, bathing, dressing, bowel and bladder control, personal hygiene, transfers, walking on a straight surface and stair climbing. Items were scored in 5-point intervals (0–15), depending on the level of assistance required by the participant. The total score ranged from 0 to 100. Higher scores indicated better functional ability. (b) Instrumental Activities of Daily Living Scale (IADL) (Lawton and Brody, 1969). This interview-based functional assessment consisted of eight components: telephone use, shopping, food preparation, housekeeping, laundry, transportation use, responsibility for personal medication and money. Item scores ranged from 0 to 3 or 0 to 4, according to the level of assistance required by the participant. The total score ranged from 0 to 20. Higher scores indicated better functional ability. (c) The World Health Organization Quality of Life Brief questionnaire (WHOQoL-BREF) (The WHOQOL Group, 1998). This questionnaire was an abbreviated 26-item version of the WHOQoL-100, which is the gold standard for measuring QoL in four domains: physical, psychological, social relationships and environment. Scores in each domain ranged from 0 to 100. Higher scores represented a higher QoL.
Procedure
The study was authorised by the Ethics Committee, approval number 007/18, year of approval 12/2017. Advertisements calling for study participants were published in various neighbourhoods and day-care centres in central Israel. Those who agreed to participate in the study contacted the study investigator by telephone. During this call, potential participants were required to answer questions that were designed to verify the eligibility criteria. For those who were eligible, a meeting was scheduled at their home or at the day-care centre (after obtaining approval from the management of the day-care centres). Participants signed an informed consent form and completed the evaluation battery of tests, first the demographic and health status questionnaire, then the MoCA, followed by the fall questionnaire, TUG test and the GDS, BRIEF-A, aEFPT - medication management task, ADL questionnaires and the WHOQoL-BREF. For most participants, all study variables were collected in a single meeting. However, in cases where participant grew tired, a second meeting was arranged, to avoid exhaustion. In this second meeting, participants completed the evaluations that were not performed in the first meeting. Data were collected by a qualified occupational therapist and by third year students who were trained to perform the study with all relevant procedures.
Statistical analysis
All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) for Windows 25.0. Chi-squared test is used to compare qualitative variables between high- and low-risk groups. Multivariate analysis of variance (MANOVA) examined whether significant differences existed between groups among all subscales. The t-test and ANOVA examined whether significant differences existed between groups in the total scores of the dependent measures.
Structural equation modelling (SEM) was used to examine the relationships among age, education, depression, EF, performance of activities of daily living, health-related QoL and fall risk. The model examined the role of fall risk in mediating between all other parameters and daily life, in terms of ADL performance and health-related quality of life (HRQoL). Fall risk was included as a continuous parameter and the evaluation was performed on the general sample. The following fit indices were evaluated: the goodness-of-fit statistic (GFI), root mean square error of approximation (RMSEA), standardised root mean square residual (SRMR), standardised RMR and comparative fit index (CFI). Chi-squared tests were used for nested model comparison. The level of significance was set at 0.05.
Results/findings
In the present sample, a high fall risk was prevalent among 32% of the sample.
Comparison of depression, cognitive status, EF, ADL and HRQoL between high-risk and low-risk fallers.
# Quantitative variables: t-test; ANOVA; MANOVA = multivariate analysis of variance.
MoCA: The Montreal Cognitive Assessment; EFs: executive functions; ADL: activities of daily living; HRQoL: health-related quality of life; MoCA: Montreal Cognitive Assessment; GDS: Geriatric Depression Scale; aEFPT: alternate executive function performance test; BRIEF-A: Behavior Rating Inventory of Executive Function - Adult Version; BRI: Behavioral Regulation Index; MI: Metacognition Index; GEC: global executive composite; BADL: basic activities of daily living; IADL: instrumental activities of daily living. BRIEF-A-MI components are marked in italics.
The correlations between the continuous fall risk score and the other variables were examined. A higher fall risk score was significantly correlated with increased age (r = 0.34, p < 0.0001), fewer years of education (r = −0.35, p < 0.0001), higher levels of depression (r = 0.34, p < 0.0001), lower cognitive level (r = −0.31, p < 0.0001) and lower total HRQoL (r = −0.41, p < 0.0001). Based on these significant correlations, the variables were entered into the SEM model. The SEM model revealed the following goodness of fit indices: χ2 (25) = 38.275, p = 0.04; CFI = 0.97; NFI = 0.93, RMSEA = 0.06. Standardised indirect effects were provided in the relevant analyses.
When referring to the beta coefficients, a significant correlation was noted between age and the number of years of education (r = −0.32, p < 0.0001), as well as between age and depression (GDS) (r = 0.34, p < 0.0001). Age was directly related to the aEFPT medication management score (β = 0.18, p = 0.03), as well as lower performance in IADL (β = −0.25, p < 0.0001). Older adults tended to have lower levels of education, higher levels of depression, reduced EFs and restricted IADL.
Higher levels of depression (GDS score) were correlated with fewer years of education (r = −0.39, p < 0.0001) and were directly related to lower metacognition and behavioural regulation (based on the BRIEF-A) (β = 0.55, p < 0.0001; β = 0.42, p < 0.0001, respectively), as well as lower performance in IADL (β = −0.23, p < 0.0001). Individuals with higher levels of depression were less educated, had reduced EF and lower performance in IADL. Furthermore, lower metacognition (based on the BRIEF-A) was directly related to a higher risk fall and lower HRQoL (β = 0.34, p = 0.005; β = −0.31, p < 0.0001, respectively).
Worse performance, as reflected by the aEFPT medication management score, was directly related to fewer years of education (β = −0.37, p < 0.0001) and higher fall risk (β = 0.32, p < 0.0001). Fall risk was directly related to lower performance in BADL and IADL (β = −0.41, p < 0.0001; β = −0.34, p < 0.0001) and lower HRQoL (β = −0.13, p = 0.05). Fall risk did not mediate any associations between BRIEF-A-BRI and BADL/IADL; or BRIEF-A-BRI and HRQoL.
However, aEFPT medication management and metacognition in the BRIEF-A and ADL and HRQoL were mediated by fall risk as follows: aEFPT and IADL (standardised indirect effect = −0.10) (95% CI: −0.23 to −0.04, p < 0.001); aEFPT and BADL (standardised indirect effect = −0.13) (95% CI: −0.28 to −0.04, p < 0.001); aEFPT and QoL (standardised indirect effect = −0.04) (95% CI: −0.11 to −0.003, p = 0.03); BRIEF-A-MI and IADL (standardised indirect effect = −0.11) (95% CI: −0.24 to −0.03, p = 0.003); BRIEF-A-MI and BADL (standardised indirect effect = −0.14) (95% CI: −0.29 to −0.04, p = 0.004) and BRIEF-A-MI and QoL (standardised indirect effect = −0.04) (95% CI: −0.13 to −0.002, p = 0.04); (see Figure 1). Structural equation modelling – the risk of falls mediates the association between depression, executive dysfunctions and daily life.*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001. GDS = Geriatric Depression Scale; BRIEF-A = Behavior Rating Inventory of Executive Function - Adult Version; aEFPT= Alternative Executive Function Performance Test; MI = Metacognition Index; BRI = Behavioral Regulation Index; BADL = basic activities of daily living; IADL = instrumental activities of daily living; HRQOL = health-related quality of life.
The model explains 27% of the fall risk, 17% of the performance of BADL, 34% of the performance of IADL and 55% of total HRQoL.
To summarise, in older adults, fall risk played a significant role in mediating the associations among depression, executive dysfunction and daily life, as reflected by reduced performance of BADL/IADL and lower HRQoL: higher depression was related to lower MI, that were associated with higher fall risk, which was correlated with lower IADL and BADL and lower HRQOL. As for age and years of education – older adults with lower years of education had a lower performance in aEFPT, and had higher risk fall, which was associated with lower performance on IADL and BADL and with lower HRQOL. Higher depression correlated with lower BRI, which was related to lower HRQOL.
Discussion and implications
Fall prevalence in non-institutionalised older adults was evaluated and its relationship with other known geriatric body dysfunction and activities of daily life. In line with the ICF model, this study used standard self-reports, clinical observation and performance-based assessments with ecological validity that mimic daily situations. Although this study was focused on older adults who are still relatively functional, about one-third of the study participants had a high fall risk and significantly worse body dysfunction than those with a low fall risk. This was evidenced by a tendency towards greater levels of depression, lower cognitive status, lower EF, restricted ADL and lower HRQoL.
Correlations between fall risk (continues score), socio-demographic parameters, body dysfunction parameters and HRQoL.
As mentioned above, and similar to previous reports (Laurence and Michel, 2017; Montero-Odasso and Speechley, 2018), the present study showed that depression is associated with reduced EF, as presented by reduced MI and BRI, while these reduced EFs are related to high fall risk. Most studies of EFs and falls have used traditional assessments and referred to specific domains of EF (e.g. inhibition and working memory) (Caetano et al., 2018). However, the present study used measures that reflected the expression of EF in daily life activities, including the BRIEF-A self-report questionnaire and the performance-based aEFPT medication management test.
The results showed that the greatest difference in EF between individuals with high fall risk and those with low fall risk was observed in the BRIEF-A-working memory domain. Earlier reports have shown that attention and memory are more affected by age and have a greater impact on posture control (Laurence and Michel, 2017). Studies have also reported that in older adults, EF mediates the association between motor performance and fall risk (Caetano et al., 2018). Other reports have shown that among individuals with a high fall risk, EF mediates the association between changes in memory and reduced performance in IADL (Royall et al., 2004).
According to the SEM model in the present study, older adults had lower EF and more restricted IADL. Restricted performance of the ADL was directly associated with fall risk. However, regarding EF, the model presented a new perspective that places fall risk as the mediator between executive functions and ADL. It is noteworthy that fall risk mediated between BRIEF-A-metacognition (MI) and ADL but not between BRIEF-A-Behavioural Regulation Index (BRI) and ADL.
A possible explanation for this finding is associated with the fact that MI components, such as initiation, working memory, planning, monitoring and organisation, are more essential for motor planning, motor control and navigating obstacles in the environment.
Another aspect included in metacognition is awareness (Toglia and Kirk, 2000). Awareness is critical for motor performance in challenging environments, especially among older adults. Awareness is a key factor in cognitive rehabilitation and is related to the use of improved strategies for enhanced performance of daily activities (Engel-Yeger et al., 2011; Toglia et al., 2010). These findings emphasise the fact that prevention and intervention programmes to minimise falls in older adults, should screen for EF problems, and give special attention to the individual’s metacognition and awareness of self-performance and environmental context and cues (Burgess et al., 2006).
In line with the ICF model, clinicians should gather information not only from isolated tasks in a lab setting but also from tasks that mimic daily life scenarios, with reference to the environmental context, where the regulation of balance takes place (Bernard and Lacour, 2017). With this information, clinicians may improve their selection of the best-adapted behavioural strategy (Hahn et al., 2014; Josman et al., 2009). For example, they may be able to determine how the activity should be approached within a certain environment, how to compensate for a slow reaction time or physical difficulties (Barban et al., 2017) or what environmental adaptations should be established to enhance the adaptive response to environmental demands at home or in the community (Mirelman et al., 2019).
The present study has a few limitations. The study was focused on a specific part of the population, non-institutionalised older adults. The distribution of participants was not equal across the two groups (high fall risk and low fall risk). Future studies should be focused on older adults who live in other settings, such as nursing homes, examine the differences between men and women and use larger sample sizes to examine group differences. Cohort studies are also recommended to better understand the effects of ageing on falls, as well as the associated body dysfunction and its expression in the daily lives of individuals.
Conclusion
To summarise, fall risk should be routinely screened among non-institutionalised older individuals and their engagement in physical activity should be encouraged (Ribeiro et al., 2017). Moreover, health services should note that although the ability of older individuals to function in daily activities is related to the effects of their emotional/cognitive/EF status, good balance and a low tendency to fall also play a major role in this relationship. Hence, fall prevention and intervention programmes should be informed by the impact of body dysfunction on daily life (Cohen and Kimball, 2003). Such prevention and intervention programmes should also determine impaired emotional-cognitive cues, as expressed during the performance of daily activities among older people, in their natural environment. Those cues may serve as warning signals that could predict the next fall and thus lead to early fall risk evaluation and intervention when needed. Intervention programmes should be based on a multi-disciplinary approach and apply models such as the ICF, which are relevant for fall rehabilitation (Saverino et al., 2015). Furthermore, interventions should apply performance-based ecological assessments to identify body dysfunction that predict falls in a real-life context. Thus, interventions will be directed more to the individual’s specific needs, interests, resilience and vulnerability and will yield better results in terms of enhanced function and better QoL.
Key Findings
Fall risk in older adults plays a significant role in mediating the associations among depression, executive dysfunction and daily life, as reflected by reduced performance of BADL/IADL and lower HRQoL.
What the study has added
Fall risk evaluation and intervention among older adults should include emotional-cognitive body dysfunctions as expressed during the performance of daily activities. Occupational therapists play a significant part in this.
Patient and public involvement data
During the development, progress, and reporting of the submitted research, Patient and Public was included in the conduct of the research.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed no receipt of financial support for the research, authorship, and/or publication of this article.
Contributorship
EYB and ZY researched literature and conducted data analysis conceived the study. ZY and OT students and Keren Ravitz-Ron were involved in data collection. were involved in data collection. All authors reviewed and edited the manuscript and approved the final version of the manuscript.
Research Ethics
The study was authorised by the Ethics Committee of the Faculty of Social Welfare and Health Sciences, University of Haifa, Israel. Approval number: 007/18, year of approval 12/2017.
Informed consent
Written informed consent was obtained.
