Harnessing innovation to help meet the needs of elders: field testing an electronic tool to streamline geriatric assessments across healthcare settings

Introduction

As human life expectancy increases, older adults face progressive declines in physical and cognitive abilities and tend to have multiple medical conditions and critical events concurrently.1 The rising elderly population paired with fewer caregivers puts great pressure on healthcare systems regarding costs and resources. By 2050, over 1600 million people worldwide will be aged 60 years or older. This demographic shift poses challenges for sustainable health systems and impacts all facets of society.1

In particular, the growing senior population strains primary care delivery models that are typically not optimised for complex geriatric patients.2 Older adults have multidimensional health needs spanning medical, functional, mental health and social domains. Comprehensive geriatric assessment (CGA) evaluates an elder’s medical, psychosocial and functional status to develop individualised care plans. First developed in the 1960s, CGA addresses the complexity of geriatric patients who often have multiple cooccurring issue.3

Over 50 years of research has demonstrated CGA’s effectiveness through reduced mortality, functional and cognitive decline, nursing home placement and healthcare costs.3 4 A standard CGA involves a multidisciplinary team assessing key domains like medical diagnoses/medications, functional ability, nutrition, mental health and social supports using standardised tools.5 These domains provide a holistic understanding of an older adult’s health in order to develop a coordinated care plan targeting common geriatric challenges such as multimorbidity, falls risk and polypharmacy.6

Following a visit that can take 1–2 hours, the team synthesises a plan addressing all areas of need. Long-term studies have found that CGA reduces mortality, functional and cognitive decline, nursing home placement rates and healthcare costs.7 However, the intensive resource requirements limit its availability. Also, CGAs are time-intensive and primary care visits rarely allow comprehensive assessment and care planning across all relevant areas. International studies estimate that only 10%–25% of older adults receive a CGA.8 Israel is a relatively young country compared with other developed nations, mainly attributable to its high fertility rates (see appendix 1 for more details regarding Israeli healthcare system).9 However, projections indicate the number of Israelis aged 70 and over will double by 2040. During this period, the proportion of those 70+ is expected to rise from 8.5% to 11% of the total population. Furthermore, among those aged 65 and over, rates of chronic diseases, acute morbidity risk, functional decline, polypharmacy and health service utilisation are also increasing.10

Concurrently, issues affecting seniors such as multiple long-term illnesses, hospitalisation probability, self-care limitations, polypharmacy concerns and healthcare visits are progressively worsening within the 65+ cohort as well. These shifting national demographic trends pose looming challenges for the Israeli health system to comprehensively meet geriatric needs over the coming decades.11,9

Despite Israel’s rapid ageing of its population, there are only approximately 400 geriatricians in the country.12 This limited specialist workforce does not appear capable of providing comprehensive care to all elderly Israelis. Thus, primary care physicians must play a larger role in addressing the distinct needs of older adults, including through the application of geriatric assessment.13 With just 400 geriatricians for its growing senior cohort, Israel requires increased involvement from primary providers to deliver solutions tailored to this vulnerable population.14 Implementing standardised geriatric screening and care planning tools in primary care settings could help optimise currently fragmented care for seniors across settings.8 Harnessing primary care in this manner may be necessary to fill gaps as the proportion of older Israelis continues expanding faster than available specialist geriatric resources.

Geriatric assessment evaluates medical, social and environmental factors that affect overall well-being, and addresses functional status, fall risk, medication review, nutrition, vision, hearing, cognition, mood and services.15 CGA can lead to early recognition of problems that affect the patient’s quality of life by identifying areas for focused intervention, and/or subtle or hidden problems.15 For example, in a study conducted by Tak et al16 among 180 homebound elderly, a geriatric assessment tool was used by primary care physicians to identify a wide range of mental and social problems among them, such as anxiety, depression, cognitive impairment, suspicion, loneliness and somatisation. Researchers found that using this tool helped family physicians to detect social and/or psychological problems early among their elderly patients and provide tailored solutions.

Following this, Maccabi Healthcare Services (MHS) developed a computerised tool called ‘Golden Age Visit’, for primary care physicians to detect geriatric symptoms at early stages, with the goal of promoting health and initiating preventive medicine for patients aged 65 and over. Accordingly, a smart structured visit sheet (SMARTSET) was built which includes the following components: physical/clinical, mental, functional, social, environmental/safety and nutritional.

While CGA is established as the gold standard for geriatric assessment, barriers around resource intensiveness and limited accessibility persist. Our study presents a novel approach that seeks to address this gap by automating core CGA components for use in primary care. To our knowledge, this is the first study to implement and evaluate an electronic comprehensive geriatric screening tool designed specifically for the primary care setting. By streamlining CGA into a structured e-form, our research explores whether proactive geriatric screening can be enhanced within existing healthcare structures. The automated assessment captures multiple domains known to be predictive of senior health outcomes, but streamlines the process for feasibility in routine clinical visits. In so doing, our study has potential to significantly expand access to proactive geriatric screening, care coordination and preventive services for growing senior populations worldwide.

Positive results could support widespread adoption of our geriatric screening approach to optimise healthcare quality, costs and efficiency as populations age. Given scarce research on technology-enabled CGA models, our study also addresses an important knowledge gap with implications for future care delivery innovation. Overall, this project emphasises an original strategic shift towards making CGA scalable within primary care.

Methods

Instrument development

Golden Age Visit and SMARTEST

Figure 1 illustrates the process of developing the SMARTEST for Golden Age Visit.

Request permission

Figure 1

The process of developing the SMARTEST for Golden Age Visit.

The tool developed at MHS is a smart system tool called ‘Golden Age Visit’, which is found in the patient’s computerised medical file (see appendix 1) and allows the family physician to assess six areas: physical/clinical, mental, functional, social, environmental/safety and nutritional. The duration of the visit is about 45 min. Accordingly, a smart structured visit sheet (SMARTSET) was built which includes the following components:

1. Cognitive assessment: the Sweet 16 is a brief cognitive assessment tool that was developed and validated in two cohorts as an alternative to the Mini-Mental State Examination (MMSE).17 The Sweet 16 demonstrated high correlations with the MMSE and Modified Blessed Dementia Rating Scale in both cohorts and performed equivalently to or better than the MMSE when assessed against the independent formant Questionnaire on Cognitive Decline (IQCODE) reference standard, with areas under the receiver operating characteristic curve ranging from 0.84 to 0.97 depending on the analysis. Sensitivity of the Sweet 16 was equal to or higher than the MMSE across validation analyses, meeting the study goal of maximising sensitivity for use as a brief cognitive screening tool.

2. Assessment of affective state using the two depression screening questions from the Patient Health Questionnaire-2 (PHQ-2). The PHQ-2 consists of only the first two items from the larger Patient Health Questionnaire-9 (PHQ-9) scale, assessing depressed mood and anhedonia. With just two items, internal consistency statistics like Cronbach’s alpha cannot be calculated for the PHQ-2. However, the PHQ-9 from which the PHQ-2 is derived has demonstrated excellent internal reliability, with a Cronbach’s alpha of 0.89 reported in the large validation study by Kroenke et al.18

3. The 15-item Geriatric Depression Scale (GDS) is a commonly used screening tool to assess depression in older adult populations. A validation study conducted by Hoyl et al in 1999 found the 15-item GDS to have high internal reliability. Specifically, the study reported a Cronbach’s alpha of 0.86 for the full scale, demonstrating good internal consistency between the 15 items on the GDS questionnaire.19

4. Interview regarding support systems and identification of primary caregiver distress (with a family member of the patient present at the visit).

5. Functioning assessment: Barthel Index for Activities of Daily Living questionnaire. The Barthel Index is a validated tool used to assess functional independence after a stroke or other disabling disease/injury. The two primary version of the Barthel Index—the 10-item and the 15-item—both received high mark reliability ratings. The 15-item version has a 0.89 test–retest reliability and 0.95 internal reliability calculated by Granger et al.20 The 10-item has an alpha internal consistency coefficient of 0.87–0.92 reported by Shah et al.21

6. Nutritional status—to comprehensively assess nutritional status, we evaluated multiple factors beyond only body weight changes and meal receipt. Participants underwent screening for risk of malnutrition using the validated Short Nutritional Assessment Questionnaire.22 Through interview, we documented each individual’s appetite level, recent eating habits and ability to independently manage food shopping and preparation. Dental status and symptoms of difficulty in swallowing were recorded due to their potential impact on nutrition.

7. Functional status was evaluated using components from standardised tools. Gait speed was measured via the 'Get Up and Go' test, where participants rise from a chair, walk 3 metres, turn and return to sit down. This assesses mobility and risk of falls. Fall history in the past year was determined through structured inquiry about the number, circumstances and injuries related to any falls.

At the end of the visit, the system assists the family physician by suggesting relevant referrals to additional care providers as needed. This may include referrals to a social worker, multidisciplinary geriatric clinic, geriatrician, physiotherapist, dietician, nurse or services from the National Insurance Institute for the elderly. The system can also facilitate virtual consultations. In addition, informational leaflets on preventive healthcare topics are available to print for the patient and their family. Example topics include home and street safety procedures, healthy longevity and smart nutrition in old age.

The duration of a regular visit using this system is approximately 45 min, allowing adequate time for consultation, review of the patient’s needs and history, referrals, and health education resources.

Developing an application

Prior to this study, IT team at MHS developed a clinical support application to operationalise the SMARTEST geriatric assessment for family physicians. Key application features were informed by previous studies23 24 on best practices for clinical decision support systems and virtual patient simulation tools. These included explanatory videos summarising common geriatric conditions, a digital handbook with assessment protocols, and a virtual simulation of the ‘Golden Age Visit’ modelled after descriptions. Since its public launch, administrative data indicate over 6000 geriatric assessments have been facilitated through the SMARTEST application by primary care clinicians in our healthcare system. As successful adoption occurred, it became important to formally evaluate users' experiences through their perspectives.

Result

A total of 9022 older adults were included in the analysis, of which 1421 (15.8%) visited the Golden Age Visit. There were no significant differences between Golden Age visitors and non-visitors in mean age (84.1 vs 83.8 years, p=0.121), gender distribution (55.9% vs 54.5% female, p=0.31), residing in the periphery (6.4% vs 6%, p=0.629), distribution of socioeconomic status (p=0.114), prevalence of oncological disease (36.5% vs 38.9%, p=0.096), hypertension (93.1% vs 92.2%, p=0.254), ischaemic heart disease (69% vs 71.1%, p=0.112), chronic obstructive pulmonary disease (16.5% vs 15.1%, p=0.186), diabetes (53.8% vs 53.3%, p=0.75) and antidepressant use (44.3% vs 42.3%, p=0.161). In addition, Golden Age visitors had higher rates of history of falls (96.2% vs 91.7%, p<0.001), osteoporosis (54.8% vs 49.8%, p<0.001) and cognitive problems (53.3% vs 28.9%, p<0.001) (table 1).

Table 2 presents the comparison of outcomes between patients who received a golden visit and those who did not, stratified by cognitive status. Among older adults with cognitive problems (n=1279), those who received a Golden Age Visit were more likely to have subsequent geriatric visits (69% vs 76.3%, p<0.001), physical therapy visits (31.2% vs 18%, p<0.001), dietitian visits (33.4% vs 15.4%, p<0.001) and occupational therapy visits (22.3% vs 9.6%, p<0.001) compared with those who did not receive a Golden Age Visit. There were no significant differences in rates of hip fracture at 3 months (1.2% vs 1.6%, p=0.491) or relocation (11.1% vs 8%, p=0.011) between the groups.

Among older adults without cognitive problems (n=2189), those who received a Golden Age Visit were also more likely to have physical therapy visits (27.3% vs 14.2%, p<0.001), dietitian visits (25.3% vs 11%, p<0.001) and occupational therapy visits (16.3% vs 4.8%, p<0.001) compared with non-visitors. However, there was no significant difference in subsequent geriatric visits (37.7% vs 35.9%, p=0.369) between the two groups. No differences were observed in hip fracture rates (1.5% vs 1%, p=0.22) or relocation (10.4% vs 9%, p=0.253).

Table 3 presents the number of patients who received initial diagnoses for the first time in the 3 months following the index date, stratified by those who received a Golden Age Visit and those who did not. Among patients who had a Golden Age Visit, the most common initial diagnosis was mild cognitive impairment, affecting 483 patients (19.75% of the group). The second most frequent diagnosis in this group was functional deterioration, received by 409 (16.72%) patients.

Comparatively, for patients who did not receive a Golden Age Visit, mild cognitive impairment remained the leading initial diagnosis but was substantially lower at 20 patients (0.82% of the group). This difference in rates of mild cognitive impairment between the two groups was statistically significant (p<0.001).

Depression and falling were the next most prevalent initial diagnoses within the no visit group, received by 13 (0.53%) patients and 32 (1.31%) patients, respectively. These rates were significantly lower than that in the group who had a Golden Age Visit, where depression affected (3.73%) and falling (2.96%) (p<0.01 and p<0.05, respectively).

In general, initial diagnostic rates tended to be higher in patients who received a Golden Age Visit. For instance, dementia/Alzheimer’s was diagnosed for the first time in 37 (1.51%) patients with a visit compared with 20 (0.82%) patients without. Exceptions were falling and walking difficulty, which showed similar initial diagnosis rates between the groups. Rare diagnoses such as cerumen impaction and weight loss of 10% or more affected small percentages overall.

A total of 151 physicians participated in the pilot programme by conducting at least one golden visit. Of these, 67 physicians (44% response rate) completed the postpilot questionnaire. The majority were female (61%) and specialised in family medicine (64%). Physicians aged 48 years or younger constituted the largest age group that conducted golden visits (42%) and responded to the questionnaire (48%). Regional response rates varied, with 34% of respondents practicing in the Sharon district.

Among physicians who performed golden visits, satisfaction with the programme was high. Nearly two-thirds (65%) reported being greatly to very greatly satisfied overall. Satisfaction was even higher (82%) for the largest participating age group of physicians under 48 years old (figure 2).

Request permission

The main perceived contribution of golden visits was in diagnosis and geriatric assessment (69%). Over half felt it helped identify new previously unknown geriatric symptoms (53%) and monitor existing symptoms (48%), while nearly half said it expanded their knowledge of caring for older patients (45%). Approximately 70% believed the supporting training was helpful.

For physicians who did not conduct golden visits (n=1100), the questionnaire response rate was lower at 13% (n=148). Respondents were predominantly female (53%). The Jerusalem and Lowland districts exhibited the highest response rates (25%). Despite not directly participating, most (80%) were aware of the pilot programme. The primary deterrent cited was concerns about time investment required per visit (41%). A similar proportion indicated the pilot closed before they could enrol (41%).

Discussion

This comprehensive evaluation provides valuable insights into deploying geriatric assessment programmes on a large scale within a major healthcare system. By analysing outcomes for both community-dwelling older adult patients and their primary care physicians across Israel’s second largest health organisation, this study evaluated the real-world impact of the Golden Age Visit programme implemented as a digital tool, on a significant population level.

Conclusion

In conclusion, this study found that the Golden Age Visits CGA programme was associated with increased diagnoses, referrals and physician satisfaction after 1 year. However, longer-term impacts on health, function and costs require further examination.

Sustainable models for routinely incorporating comprehensive assessments into varied primary care systems deserve additional exploration. Future research should evaluate scalability across settings and populations. Standardised screening tools could potentially be integrated across multiple primary care clinics through training, shared care plans and outcomes monitoring.

Proactively managing common geriatric conditions via coordinated screening may benefit older adults. As populations age, optimising primary care delivery for their needs is increasingly important. Additional high-quality studies can provide guidance on best practices for geriatric care delivery and integration into healthcare systems.