The accurate identification of the cause of a fall can be challenging in older people, especially in case of atypical disease presentation. Diseases may manifest with atypical features in older adults, such as falls or reduced mobility, especially in case of frailty, multimorbidity and cognitive disorders. Atypical disease presentation is defined as lack of signs and symptoms or vague, unusual signs and symptoms, unrelated or even opposite of what is usually expected (Limpawattana, 2016; Jarrett, 1995).

The majority of falls are of multifactorial origin, and by now many modifiable fall-risk factors have been identified (see guideline module risk factors for falling in older persons). In case of a high fall risk, a comprehensive multifactorial fall risk assessment along with a multidomain intervention is recommended. Additionally, if a patient presents with an acute fall, a sudden increase in fall frequency, or a change in the nature of the falls, the possibility of an atypical disease presentation should be evaluated. Also, cardiovascular risk factors (see guideline module cardiovascular fall risk factors) and underlying acute or chronic conditions, such as infections or neurological disorders, may also be present and should be evaluated  The key question is which underlying acute or chronic diseases, beyond the commonly known risk factors, are identified in older adults who present acutely with an atypical disease manifestation, such as a fall?

Description of studies

COVID-19

Blain (2021) performed a retrospective longitudinal study to determine whether atypical symptoms may predict COVID-19 in nursing home residents. The study was conducted in a French nursing home during a COVID-19 outbreak. In total, 456 residents were included in the study. These residents were PCR-tested for COVID-19. In total, 295 residents tested COVID-19 negative (mean age 86.2 years, 71.5% female), and 161 residents tested COVID-19 positive (mean age 87.4 years, 72% female). Typical and atypical symptoms were recorded in residents’ files during the fourteen days before and after real-time reverse-transcriptase (rRT)-PCR testing for COVID-19. One of the atypical symptoms that was recorded were falls.

Clifford (2021) performed a retrospective study to investigate the association between different triage chief complaints and COVID-19 status. This multicenter study was conducted in multiple hospitals in New York (USA). Patients visiting the emergency department were tested for COVID-19. The positive and negative cohorts (respectively n = 6524 and n = 5468) were then assessed for different chief complaints, which were obtained from structured triage data. Per complaint, the percentage of patients tested positive for COVID-19 were presented. A sub-analysis was performed for patients older than 65 years old.

Gan (2021) performed a retrospective community-based study to evaluate the incidence, characteristics, and clinical outcome of older patients with atypical presentations of COVID-19. Adult patients ≥ 65 years with confirmed COVID-19 admitted to an acute geriatric medicine unit in the UK were included in the analysis. In total 122 were included in the study. These patients were subdivided in a typical group (n = 73) and an atypical group (n = 49) based on their primary presenting complaint. Typical presentations of COVID-19 were defined as fever, dry cough and dyspnoea. The presence or absence of these symptoms, and the presence of unusual, e.g. atypical, symptoms, including falls, were recorded.

Hunt (2021) performed a retrospective study to characterize the clinical features of older people with COVID-19. Patients ≥ 70, presenting to a district hospital in the UK with PCR-confirmed COVID-19 were eligible to participate. The study population consisted of 225 patients. Mean age of the study population was 83 years, 46.7% female. Of these patients, 30.2% did not present with typical symptoms (cough, fever, breathlessness) at admission. Clinical characteristics and complaints were reported.

Karlsson (2021) performed a retrospective study to investigate the presentation of symptoms at hospital admissions in older patients with COVID-19 and to evaluate its impact on disease outcome. This multicentre study was executed in hospitals in Denmark. All patients, aged ≥ 80, presenting with PCR-confirmed COVID-19 at one of the centres, were included in the study population. In total, 102 patients were included in the analysis, mean age 84 years, 52.9% female. The majority of patients (71.6%) lived at their own home, 28.4% lived in a nursing home or intermediate care facility, prior to hospitalization. Clinical characteristics and symptoms at admission were recorded. Additionally discharge destination and in-hospital and 30-day mortality were reported as an outcome.

Lafuente (2021) executed a retrospective, descriptive study to better define which symptoms were the most frequent symptoms of COVID-19 in older people and to determine if those symptoms followed any temporal pattern. The study was conducted in three geriatric rehabilitation and convalescence units located at two hospitals in France. The study population consisted of patients aged ≥ 75 years who were hospitalized at one of these units and were tested for COVID-19 (n = 128). Sixty-nine patients tested positive (mean age 86 years, 75.4% female) and 59 tested negative (mean age 82 years, 70.3% female). Clinical signs or symptoms that were present at time of testing, in the previous 14 days, or following 21 days, were recorded for both groups. In the PCR-negative group, outcomes were only available for 25 patients, as the others were in other geriatric units and therefore not followed-up.

Nabors (2019) retrospectively reported the patient characteristics, clinical presentation, and course of illness for patients aged 80 years and older, who required hospitalization in the USA. The study included 73 patients aged 80 years and older, who had a PCR-confirmed COVID-19 diagnosis. Mean age was 86 years (range 80 – 105). The primary indication for hospitalization was COVID-19 in 67% of the patients, the other 33% were hospitalized for another illness and were additionally tested positive for COVID-19. Atypical clinical representations were reported as an outcome.

Ohuabunwa (2022) performed a retrospective study to characterize the symptomatology of COVID-19. Hospital charts of all patients aged 65 years and older who tested PCR-positive for COVID-19 in a hospital in the USA, were reviewed (n = 134). Mean age of the study population was 76.4 years, 43% was female. The hospital largely serves African American patients with a low income. Demographic information and data regarding the symptoms and signs of the patients, laboratory findings and outcomes documented by the medical team were reviewed from electronic medical records and retrospectively analysed as an outcome.

Steinmeyer (2020) performed a retrospective cohort study to describe clinical characteristics and outcomes of older patients hospitalized with COVID-19. The study was conducted on three acute geriatric wards in France. Older patients, hospitalized in one of these centres and diagnosed with COVID-19 (PCR-positive and/or CT-scan positive) were included in the study (n = 94). Mean age of the study population was 85.5 years, 55% female. Of these patients, 68.1.% lived at home (private household) and 30.8% in a nursing home or assisted living. Clinical outcomes, including falls, were recorded at admission and during hospitalization.

Van Son (2023) performed a retrospective study to assess the association between atypical presentation of COVID-19, frailty, and adverse outcomes, as well as the incidence of atypical presentation. The multicentre study was performed at eight hospitals in the Netherlands. Patients aged ≥ 70 years, admitted at one of the centres with PCR and/or CT-confirmed COVID-19 were eligible to participate. In total 780 patients were included in the study. These patients were divided in two groups: patients with typical symptoms (n = 706; mean age 79 years, 41.1% female) and patients with atypical symptoms (n = 74; mean age 80 years, 48.6% female). Typical symptoms were defined as fever, cough and dyspnea. Clinical characteristics and symptoms at presentation were recorded.

Systemic infections

Blair (2017) performed a retrospective study to investigate the characteristics of patients presenting to hospital because of a fall and who were diagnosed with a coexisting systemic infection (CSI). In total, 161 patients were included in the study, of whom 52.5% were female and the mean age was 75.2 years. In total, 120 out of the 161 patients who were included in the study population were 70 years or older. Patients were included only if there was documentation of a chief complaint of a recent fall less than seven days prior to their presentation, were admitted to the hospital, and were found to have a CSI within 48 hours of their presentation. In-hospital falls were excluded.

Idiopathic normal pressure hydrocephalus

Petrella (2023) retrospectively reviewed hospital records to investigate the incidence of idiopathic normal pressure hydrocephalus (iNPH), amongst patients presenting to an Italian hospital. CT-scans from patients presenting at the emergency department were reviewed on radiological criteria typical for iNPH (i.e. obvious increase in size of the ventricles, other intracranial pathologies excluded). These patients were identified as ‘suspected iNPH’. Next, the CT-scans were compared with the clinical records from the emergency department.

The cohort included 192 patients, mean age 82.1 ± 10.9 years, 45% female. Primary diagnoses made at the emergency department, and reason for admission to the emergency department (including falls), were reported as an outcome.

Frailty

Jarrett (1995) performed a retrospective study to estimate the prevalence of atypical disease presentation in older persons and to determine whether the prevalence differs between frail and non-frail older adults. Data was retrieved from older patients aged 65 years and over, presenting to the emergency department of a tertiary care hospital in Canada (n = 247). Mean age of the study population was 78.3 ± 7.6 years, 54% female. Demographic and diagnostic data were recorded, including a score on the Barthel Index (BI) to determine frailty. Based on their score on the BI patients were identified as “fit older adults” (n = 76; score of 95 or higher) or as “frail older adults” (n = 117; score of less than 95). Additionally, information on presenting symptoms was collected. Presenting symptoms were classified as either typical symptoms (typical symptoms of a certain disease, e.g. chest pain for unstable angina) or atypical symptoms. The atypical symptoms were defined as delirium, falls, immobility, incontinence, functional decline, and breakdown of social supports. As an outcome, characteristics, and symptoms of the “fit older adults” and the “frail older adults” were reported.

Atypical presentation of illness

Hofman (2017) performed a retrospective study to determine the prevalence and clinical outcome of older patients with atypical disease presentations in the emergency department. The study was conducted in a Dutch hospital. Patients aged 80 years and older, admitted to the emergency department were eligible to participate. The total study population consisted of 355 patients, mean age: 85.7 years, 60% female, before admission 80% of the patients lived at home. The study population was divided in two groups: patients with typical presentation of illness (n = 167, 47%) and patients with atypical presentation of illness (n = 188, 53%). A typical presentation of illness was defined as “medical history in the emergency department which includes the usual symptoms of the underlying disease”. If the medical history did not reveal any of the usual symptoms known for the underlying disease, and/or if the presentation was preceded by a geriatric syndrome, the presentation was defined as atypical illness presentation. The clinical characteristics of both groups were recorded.

Results

Falls as an atypical symptom

Ten studies reported fall episodes in patients presenting with any disease in the acute setting. Consequently, patients were asked whether they experienced any fall (episodes). The number of patients experiencing falls was reported as an outcome. Diseases that were reported were COVID-19 and normal pressure hydrocephalus. Also, studies reporting frailty as presenting problem were included, see Table 1.

Table 1. Patients with fall episodes as atypical symptom of any disease

Underlying diseases in patients presenting with a fall

Three studies analyzed underlying diseases of patients presenting with a fall. Patients presented at a hospital (emergency department), with a fall as main reason for presentation. Consequently, underlying diseases in these patients were identified. Prevalence of disease was reported as the outcome. A comparison of disease prevalence amongst patients not presenting with a fall could not be made, as this was not presented in the studies, see Table 2.

Table 2. prevalence of (underlying) disease in patients presenting with a fall in the acute setting

^{CSI = coexisting systemic infection}

Additional information on falling and prevalence of disease

Narrative reviews, case-series and case-reports were reviewed for relevant information on fall-episodes as atypical presentation of disease. See Table 3.

Table 3. results from narrative reviews, case-series and case-reports

A systematic review of the literature was performed to answer the following question:

“Which diseases are found in older adults presenting with a fall, particularly those that present atypically in the acute setting, or in individuals at high risk of falling who are undergoing a multifactorial fall assessment?”

As the aim of the guideline development group was to provide an overview of the possible underlying diseases amongst older persons presenting with a fall, i.e. atypical disease presentation, a PICO was not defined. A search was performed to identify studies reporting prevalence of disease amongst adults (65 years and older) who presented with a fall in an acute setting (e.g. emergency department). Additionally, articles were searched that reported fall episodes identified as atypical disease presentation in hospitalized adults (65 years and older), or the history of falls identified as atypical disease presentation in patients (65 years and older) who underwent a multifactorial fall risk assessment.

Relevant outcome measures

As no PICO was defined, the guideline development group did not define critical and important outcome measures.

Search and select (Methods)

The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms until 03-06-2024. Search terms for ‘older adults’, ‘falls’ and ‘atypical’ were used. The detailed search strategy is depicted under the tab Methods. The systematic literature search resulted in 1019 hits. Studies were selected based on the following criteria:

• Studies describing the prevalence of (underlying) diseases of patients presenting acutely with falls in different care settings were included.
• Studies reporting falls as the (atypical) symptom or outcome of a disease were included.
• Studies describing well known (cardiovascular) fall risk factors, as described in Risicofactoren voor valincidenten bij ouderen, Cardiovasculaire oorzaken voor vallen were excluded.
• Intervention studies and conference papers were excluded.
• Narrative reviews and case-reports were only included if they reported diseases not identified in (larger) observational studies.

In addition to the primary search, relevant articles were identified by reviewing (systematic) reviews. Studies from these reviews that met the selection criteria were included.

Forty studies were initially selected based on title and abstract screening. After reading the full text, sixteen studies were excluded (see the table with reasons for exclusion under the tab Methods), and twenty-four studies were included.

Results

Twenty-four studies were included in the analysis of the literature, of which fourteen studies were retrospective (cohort) studies, and 10 were narrative reviews or case-reports. As the GRADE-approach was not used, evidence tables and risk of bias tables were not created, and GRADE-conclusions were not formulated. The relevant information from the retrospective studies is provided under ‘description of studies’ and ‘results’. A summary of the narrative reviews and case-reports is presented in Table 3 (under ‘results’).