Engaging Older Adults in Fall Prevention Using Motivational Interviewing (MI)

A 24-item, 1-4 self-report rating to measure frequency of protective strategies to prevent falling and risky behaviors that facilitate a fall. The average value is calculated with a minimum value of 1 and maximum value of 4. Higher FAB score is a better value since it indicates more consistent fall protective behaviors.

Time frame: 3-, 6-, and 12-months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below looks different from the participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

A 7-item scale, 1-4 self-report rating, widely used to measure levels of concern related to preventing falls. Minimum value is 7 and maximum value is 28. Higher FES-I scores is a worse value since it has been positively correlated with decreased activities, future falls, and injury.

Time frame: 3-, 6-, and 12-months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below looks different from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

Instances that participants can stand from a sitting position will be measured. Minimum value is 0 times. Higher value is a better value since it is indicative of participants' ability to sit-to-stand.

Time frame: 6-months

Population: Some participants were unable to complete the physical assessment due to physical limitations, access to a video platform, or a support person.

Duration that participants can hold feet side-by-side stand (heel to toe standing positions) will be measured up to 10 seconds. Minimum value is 0 seconds and maximum score is 10 seconds. Higher value is a better value since it is indicative of participants' ability to hold balance.

Time frame: 6- months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below differs from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

Individuals are asked to hold a balance position for 10 sections without moving their feet or needing support. This position require standing on one foot. The minimum is 0 and the maximum is 10. A higher number is a better outcome.

Time frame: 6- months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below differs from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

Duration that participants can hold semi-tandem stand (heel to toe standing positions) will be measured up to 10 seconds. Minimum value is 0 seconds and maximum score is 10 seconds. Higher value is a better value since it is indicative of participants' ability to hold balance.

Time frame: 6- months

Population: The number analyzed differed by time point because of the loss to follow-up and participant withdrawal as indicated in the participant flow data, and participants' willingness and ability to participate in the measurements.

Individuals are asked to hold a tandem stand (one foot in front of the other- heel touching toe). The minimum is 0 and the maximum is 10. A higher number is a better outcome.

Time frame: 6-months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below differs from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

1-item 10-point scale response. Often used in motivational interviewing sessions to gauge an individual's readiness to engage in certain health behaviors. Higher levels of confidence are known to be associated with an increased rate of behavior change.

Time frame: 3-, 6-, and 12-months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below differs from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

1-item 10-point scale response. Often used in motivational interviewing sessions to gauge an individual's readiness to engage in certain health behaviors. Higher levels of importance are known to be associated with an increased rate of behavior change.

Time frame: 3-, 6-, and 12-months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below differs from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)

Individuals were asked to mail-in reports of falls every month. Participants reporting a fall at least once are counted as a faller. Minimum value is 0. Higher value is worse value as it is indicative that there were more participants who fell during the year.

Time frame: 1-year

Population: One participant in the MI group had an unreliable reporting of falls due to his unstable memory (ID 1277). Based on the discussion with the Safety Monitoring Council, his data was considered an outlier and removed from the analysis.

100-point, quantifiable scale determining patient engagement in healthcare. Minimum value is 0 and the maximum value is 100. Higher value is a better value since it is indicative of participant's activation with their healthcare.

Time frame: 3-, 6-, and 12-months

Population: We used mixed-effect modeling to measure changes over time, which includes all participants who have outcome measure in at least one time-point. Thus, the analytic sample in the table below differs from participant flow data. We conducted intent-to-treat analysis (including participants regardless of drop out. No list-wise deletion.)