In the year 2010, 35% (223 patients) of adult cardiac surgeries were performed on patients 70 years of age and older at Carilion Roanoke Memorial Hospital (CRMH). Considering mortality at old age, it is observed that patient age is an independent risk factor for postoperative complications including major morbidity and death [1]. With aging there is decline in physical functioning [2, 3], and this leads to higher rates of disability, institutionalization and mortality [4–6]. Clinical researchers have suggested that there is significant need for a more accurate prediction of surgical risk in elderly patients considering cardiac surgery [7]. Frailty has been investigated as a predictor of disability and adverse outcomes in older adults, and it has recently been investigated as a risk factor for patients undergoing cardiac surgery [7].

Gait speed, as a clinical indicator of frailty, has been identified as the strongest predictor of mortality in individuals with coronary artery disease [8, 9]. Gait speed is a simple, but also a potential clinical indicator of functional status in older adults [10, 11] and is known to be associated with key health outcomes in elderly [12]. Gait speed is found to be associated with survival among older adults [13–17] with 12% higher mortality with every 0.1-m/s slower gait speed [10]. Some researchers have recommended gait speed as “sixth vital sign” from its ability to predict about one’s health status in future, with its ease of administration, grading, interpretation and minimal cost involved in its determination[18]. Studenski et. al. considered gait speed as a simple and accessible summary indicator of vitality which integrates disturbances in multiple organ systems such as heart, lungs, circulatory, nervous, and musculoskeletal systems which affect survival[10]. An objective gait speed measurement in clinical environments may refine survival estimates in frail cardiac patients because it is quite informative [19]. Afilalo et al. (2010) found that elderly patients who walked 5-meter and took more than 6 seconds were associated with a 2- to 3-fold increase in risk of major morbidity and mortality in post-cardiac surgery [20]. In addition, patients with slower gait speed were more likely to have longer lengths of stay in the hospital and did required discharge to a rehabilitation facility postoperatively [20, 21]. Although there being considerable variation in gait speed testing procedures including distance of walk [22]. Bohannon compared gait speed over 2.43 meters (8 feet) and 6.09 meters (20 feet) and found that it was feasible to measure gait speed in non-laboratory setting[23] for distance less than 20 feet with one step before and after the start and stop mark[23–25]. In the gait speed assessment protocol defined over short distances (such as 5 meter) there is an acceleration phase and a deceleration phase as part of gait speed [18, 26]. Inclusion of acceleration and deceleration phases in gait speed estimation is highly correlated to that with its exclusion in estimating gait speed. Inertial sensors have been found to be highly useful in finding temporal gait events of gait [27, 28].

Thus, there is a need to establish objective and reliable gait speed testing protocols in clinics. We hereby evaluated and established smartphone based technique of measuring gait speed which could feasibly be implemented as a routine measurement in outpatient clinics.

It was found that slow walking speed had excellent reliability (ICC=0.82) and fast (ICC=0.70) and normal walking (ICC=0.66) speed had good reliability as well. The absolute agreement reliability coefficient (measure of systematic errors in measurement systems) among three velocities was also found to be fairly good as seen in Table 1. It was also found that systematic difference for all three walking speed differed with the ground truth velocity as determined by camera system in laboratory (figure 5). This was evident when velocities were fitted through the regression line (figure 4).

It was evident from clinical feasibility experiment that velocity estimation using stopwatch was not found to be accurate when compared to the ground truth velocity (through light sensors) whereas smartphones always under-estimated the walking velocities in patients (figure 3 and figure 6).

Gait speed is a commonly used measure in health care research, particularly among preoperative cardiac patients [19, 21, 32]. Some studies have previously used inertial sensors in fall risk assessment in hemodialysis clinics [33, 34]. This study demonstrates how gait speed (related to major health related outcomes in frail population) can be measured feasibly using smartphone based methodology in clinical environments. There may be important losses of information when measurement of gait velocity is prone to human timing errors. Thus in clinical practice where gait speed is important predictive of severe health-outcomes such as mortality and a subsequent physical disability, an objective, accurate and reliable way is required for gait speed measurement. Five meter gait speed does not introduce fatigue in patients with cardiovascular impairment awaiting surgery[32]. Thus predictive validity of gait speed through an objective method of assessment could be used as a prescreening tool for many different kinds of surgical procedures and in turn help clinicians to identify cardiac patients who may need intensive rehab or preplan their longer stay in hospital with specialized nursing care before return to home.

Some limitations of current study need to be mentioned. Smartphone based assessments required participants to stand-still before and after 5-m walk. As automatic algorithms developed in smartphone app determined velocity by start and stop times of movement. If any other movement artifact is followed after or prior to the walking task, the movement time may get increased than the actual walking time and thus resulting in reporting of lower velocities by the app.

The accurate measure of gait speed can improve the clinical evaluation of cardiac patients, providing an earlier detection of individuals at higher risk of major health-related events such as physical disability and mortality. Our study demonstrated that 5-m gait speed measurement using smartphone is a reliable objective measure, however adhering to certain protocol is suggested for using smartphone app. Evidence of concurrent validity was demonstrated by showing associations with the speed measurement using stopwatch and motion capture system in laboratory environment and with light sensors in clinical environment. Five meter gait speed was easily implemented in cardiovascular clinical practice with high acceptability by the patients and clinical research staff. Participants started with standing still posture and walked at their usual pace, as if they were walking in their own home, and given no further encouragement or instructions. This data can be readily collected in non-laboratory environments and can be used to help interpret the results for health related events.

Although different methods have been used previously to measure gait speed and these have affected clinical interpretation and implementation of the gait speed [22, 35]. By providing a smartphone based clinically useful gait speed assessment method with a well-defined protocol which is simple, quick, easy to perform in clinics, it is hoped that smartphone for gait speed assessment will be promoted and encouraged in clinical and research settings.