This was a cross-sectional study. Participants were drawn from the baseline data obtained from two ongoing falls studies, the Life After Falls (LiAF) and the Obesity, Sarcopenia and Falls in Older Persons (OSFOP) studies. The sample population comprised individuals aged 60 years and over with a history of at least one fall in the past 12 months recruited via word of mouth, community health promotion events, and from the primary care department, outpatient clinics, or emergency department at the University of Malaya Medical Centre. In addition, participants were recruited from wave three follow-up interviews of the Malaysian Elders Longitudinal Research study who reported falls in the past 12 months . Control participants were primarily recruited through spouses, siblings, accompanying persons and acquaintances of the participants who meet the age criteria and did not have any falls in the past 12 months. The intended recruitment ratio for fallers to non-fallers was 3:1. Individuals with significant fractures such as hip or femur fractures and head injuries were excluded. Data collection commenced in January 2019 through hospital-based, face-to-face assessments which came to an abrupt halt when movement control orders were enforced on 18th March 2020. The study immediately switched to hybrid data collection methods, and data collection through this alternative method continued up to December 2020. Virtual interviews were conducted using one or more of the virtual communication devices of a smartphone, computer tablet or personal computer (laptop or desktop) using telephone calls, social media messaging or video calls and video conferencing (MeetTM, Google Inc., USA), according to participants choice and availability of technology to the participant. Written informed consent was obtained from all participants, though virtual interviews were first conducted following verbal consent and written consent sought at a subsequent face-to-face visit, as per approval from the ethics committee. Whenever movement control orders were permissive, participants were invited to a satellite research centre, 1.5km away from the hospital, with ground-floor disabled access and off-street open-air parking, and only physical assessments that could not be performed virtually were conducted during the visits to minimise time of exposure, with a maximal visit time of 30 minutes compared 1.5 hours when the assessment was conducted using face-to-face methods exclusively. The study had obtained approval from the Institutional Ethical Review Board (MECID: 2019525-7445) prior to commencement and a subsequent application for amendment to hybrid methods and accelerated approval was obtained.
Baseline data collected included falls history, medications, cognitive testing, postural blood pressure, physical performance, quality of life (QoL), psychological status, and social network and participation. Immediately after the announcement of lockdown measures, researchers comprising geriatricians, psychologists, an ophthalmologist, a rehabilitation physician, an emergency physician, a primary care physician and a gerontologist changed the original assessments to hybrid assessments within social media messaging (WhatsAppTM, USA) chat group. Study data were collected and managed using REDCap electronic data capture tools [8, 9].
Location of falls, the total number of falls in the past year, any injuries sustained, and any medical treatment received post-fall were also recorded. No alteration in fall history data collected was made apart from transitioning from face-to-face to hybrid assessments.
Physical performance was conducted face-to-face for both methods, with the inclusion of standard operating procedures for infection control when data collection was switched to hybrid. This was assessed using grip strength and the timed-up and go test (TUG). Grip strength was measured using the Jamar hydraulic hand dynamometer (Sammons Preston, Illinois, USA). Beginning with the dominant hand, the participant was asked to grip as hard as possible with their elbow flexed at 90 degrees in the seated position. Three measurements were obtained for each hand. For the TUG test the participant was instructed to rise from a standard chair with arms, walk at their normal speed using their usual walking-aid and regular footwear, to a marker at three meters away from the front legs of the chair, turn around and walk back to the chair and sit back down again. The TUG time was considered the time between the participant's back leaving and touching the back of the chair. Functional reach (FR) was the maximal forward reach in centimetres from the upright position measured from the tip of the middle finger with the participant standing with the left arm outstretched, parallel and left shoulder adjacent to a wall with a metre rule attached.
Cognitive performances were assessed face-to-face for both face-to-face and hybrid participants. Methods and determined using the Visual Cognitive Assessment Tool (VCAT) . The VCAT is a non-language dependent tool evaluating the cognitive domains memory, executive function, visuospatial function, attention, and semantic knowledge with minimum and maximum scores of 0 and 30 respectively. A higher score indicates better cognitive ability.
Blood pressure responses to posture change were assessed using a continuous non-invasive monitoring machine (Task Force Monitor, CNSystem, Austria). Synchronized physiological signals (ECG and beat-to-beat arterial blood pressure) are monitored throughout the experiment to determine the profile of blood pressure change during 10-minutes’ supine rest followed by 3-minutes’ active stand. Beat-to-beat blood pressure measurements were calibrated against oscillometric measurements obtained at the start of the recording. With virtual assessments, measurements were delayed to a later suitable date.
Medications were initially assessed using face-to-face methods, and this was switched to virtual interviews after the initial lockdown. Participants were asked to show the researcher all their medications in their original packaging, as well as prescription orders for their medications. Comparisons were made with hospital electronic records of prescriptions if available.
The mode of administration for all questionnaires was switched from face-to-face to virtual assessments whenever possible. For face-to-face assessments, the interviewer would sit next to the participant with a printed version of the questionnaire in front of them, and the interviewer would assist the participant in the completion of the questionnaire by reading out the questions and answers and marking the selected answers. The questionnaires were administered always in the same order, starting with social network and participation, followed by activities of daily living, physical activity, quality of life and ending with psychological status. During the virtual interviews, the questions and responses would be read out verbatim. If the participant's attention waned during the virtual interviews, the researchers would discontinue the interview and complete the questionnaires during the face-to-face visit.
Social Network and Participation
Lubben’s social network scale-6 (LSNS-6) and the Keele’s assessment of participation (KAP) were used to assess social networks and participation. The LSNS-6 measures the size of active and intimate networks of family and friends with whom respondents can talk or call on for help. Scores range from 0 to 30, with higher scores indicating stronger networks. The KAP is intended to measure an individual’s level of participation in various activities such as work, education, social activities, and activities of daily living. A minimum score of 0 indicates no participation restrictions (a score of 1-11 indicates participation restriction in at least one activity).
Instrumental Activities of Daily Living
Functional ability was evaluated using the Lawton Instrumental Activities of Daily Living (IADL) scale . The Lawton scale was scored dichotomously on eight items enquiring about telephone use, shopping, food preparation, housekeeping, laundering, use of transportation, medication use and managing money. The maximum total score was therefore eight, with a higher score indicating a higher level of independence.
Physical activity was assessed with the Physical Activity Scale for the Elderly (PASE). Information on leisure, household, and occupational activity are included. The PASE assesses the types of activities typically chosen by older adults, for example, recreational activities, exercise, housework, gardening, and caring for others. The score is calculated based on the frequency, duration, and intensity level of activity over the previous week, ranging from scores 0 to 793. A higher score indicates greater physical activity. Physical activity level immediately prior to the most recent fall would be recorded.
Quality of Life
QoL was assessed with the locally validated 12-item Control, Autonomy, Self-realization and Pleasure questionnaire (CASP-12). The CASP-12 is a shortened version of CASP-19. It is a 12-item Likert-scaled index, composed of the items pertinent to the subscales control or autonomy, participation, and self-realization, intending to capture quality of life in older adults. Higher scores indicate better quality of life. The minimum and maximum scores are 12 to 48 respectively .
Depression, Anxiety and Stress were evaluated using the 21-item Depression, Anxiety and Stress Scale (DASS-21). This is a self-reported measure in which participants rate the frequency and severity of the negative emotions of depression, anxiety, and stress over the previous week. Frequency and severity ratings were made on a series of 4-point scales, with 0 indicating “did not apply to me at all” and 3 indicating “applied to me very much, or most of the time.” The scores were calculated individually for the three components: depression, anxiety, and stress. The total score for each component was dichotomized using median values as the cut-offs, depression ≥2, anxiety ≥2, and stress ≥2, respectively .
Fear-of-falling was assessed with the 7-item Falls Efficacy Scale-International (short FES-I) The short FES-I consists of seven items on a 4-point Likert scale, with 1 indicating no concern and 4 indicating severe concern. The minimum and maximum scores for the short FES-I are therefore 7 and 28, respectively. To allow for adjustment for potential confounders, cut-offs were developed using the median values; subjects with a score of ≥10 were considered to have greater fear of falling .