When climbing stairs, there are 2 practical strategies, contact each step with alternating feet (single) or contact every other step (double) with alternating feet (Figure 1). Neither the exercising individual, the conditioning coach, nor the scientific researcher can definitively answer which stair-climbing method is optimal in terms of metabolic cost and muscle function. This lack of a consensus is surprising considering that stairs are encountered on a daily basis and are often the focus of various publications. For the individual, stairs are selected as a novel form of exercise for improved fitness. For coaches, stairs are used as a training method for enhanced strength. For researchers, stairs pose unique questions as they attempt to understand the basic mechanisms of human gait. Thus, our purpose was to evaluate the metabolic cost and muscular activity of the single and double stair-climbing strategies in healthy college-age adults. We aim to provide a recommendation about which stair-climbing method would provide the greatest cardiovascular benefit and determine if there is a difference in muscle activity patterns.
Previous studies have demonstrated that stair climbing is a beneficial form of physical activity, providing numerous physiological advantages compared to level walking. Most commonly, these studies have focused on the improvement in cardiovascular fitness after the completion of a stair-climbing regimen (3,5,8,10). To add, Teh and Aziz (19) stated that oxygen consumption during preferred speed stair ascent was 83% of the corresponding maximal oxygen consumption values, equivalent to 10 METS. These values exceed the recommended vigorous physical activity guidelines of the American College of Sports Medicine and the American Heart Association (7). Aziz and Teh (2) performed a subsequent investigation, which focused on the physiological response to both single and double stair-climbing strategies. They prescribed stepping frequencies to ensure equal amounts of mechanical power output and concluded that there was not a practical difference between the 2 strategies. However, the step frequencies were not the preferred frequencies of the subjects. Regardless, these studies demonstrate the advantages of stair climbing as a form of physical activity with the potential to improve cardiovascular fitness.
Past research has also established the typical muscle activity patterns of the legs during stair climbing and how these patterns differ from level walking (1,9,11,13,18,20). For example, Shinno (18) reported that the knee extensors, in particular the vastus lateralis (VL), demonstrate the greatest activity during single support, whereas the knee flexors do not demonstrate significant activity until swing. McFadyen and Winter (13) concurred and added that the ankle extensors are the predominant group in the progression from step to step. More specifically, although both the medial gastrocnemius and soleus (SOL) were active during the second half of stance for propulsion, the soleus was also active earlier, during the first half of stance, possibly for anterior-posterior stability and body weight support. Additionally, Loy et al. concluded that a 12-week stair-climbing regimen increased leg strength (10), which is consistent with the evidence of increased muscle activity of both the knee and ankle extensors during the stance phase of stair climbing compared to level walking.
It is clear that stair climbing is beneficial in terms of cardiovascular and muscular fitness, but it is less clear during preferred frequencies if single or double-stepping strategies differ. Thus, our first objective was to evaluate the natural speed and stride frequency of individuals as they climbed up stairs with both a single and double strategy. Our second objective was to mimic the speed and stride frequency combinations in the laboratory on an incline trainer treadmill to assess the metabolic cost and muscular activity required to complete these 2 strategies. We hypothesized that the speed would not differ between the 2 stepping pattern strategies and therefore the step frequency during the single-step pattern would be greater than the double-step pattern. We also hypothesized that the knee extensor and flexor activity during stance would not differ between the 2 stepping strategies but that the ankle extensor activity, specifically the SOL, would be greater during the double-step pattern for stability and support. In the end, we aim to provide evidence about which strategy to recommend for maximum cardiovascular and muscular benefit to exercising individuals and athletic coaches. We also intend to provide information regarding the muscular activity pattern differences in each strategy for therapists and trainers working with individuals attempting to improve mobility.
Experimental Approach to the Problem
Each subject completed baseline and experimental protocols. For the baseline protocol, the subjects walked up a stairwell with a single-step and double-step strategy at their preferred speed and step frequency. For the experimental protocol, each subject walked on a treadmill inclined to the same degree as the stairs at the speed and step frequency determined from the baseline protocol.
Both the baseline and experimental protocols were completed indoors within temperature and light regulated areas. We instructed the subjects to wear the same athletic footwear for both of the protocols.
Before this baseline testing, each subject completed a 10-minute warm-up on a level treadmill set to 3 mi·h−1. Next, the subjects performed, in random order, a single-step (contacting every step with alternating feet) and a double-step (contacting every other step with alternating feet) baseline protocol on a 10.6 m, 32° set of stairs, traversing 3 floors. The subjects walked up and down the stairwell for 7 minutes with the single-step strategy and 7 minutes with the double-step strategy with 10 minutes of rest between the 2 bouts. We instructed the participants to complete each protocol at their self-selected pace and encouraged them to select a pace that could be sustained for the entire 7-minute protocol. We recorded the average speed and step frequency for each strategy and each subject during the upstair portion only. Each subject completed this baseline protocol 2 days before the experimental protocol.
Before the experimental testing, each subject completed a 7-minute standing trial and a 10-minute warm-up on a level treadmill set to 3 mi·h−1. Next, the subjects completed 4 randomly assigned 7-minute incline walking conditions on a treadmill (FreeMotion Fitness, Logan, UT, USA). More specifically, the subjects walked on the treadmill inclined to the angle of the stairs with the speed (3.33 mi·h−1) and frequency (109 steps·min−1) of the single-step strategy (Sspd_Sfreq), the speed (4.27 mi·hr−1) and step frequency (83 steps·min−1) of the double-step strategy (Dspd_Dfreq), the speed (3.33 mi·h−1) of the single-step strategy and step frequency (83 steps·min−1) of the double-step strategy (Sspd_Dfreq), and the speed (4.27 mi·h−1) of the double-step strategy and step frequency (109 steps·min−1) of the single-step strategy (Dspd_Sfreq). The treadmill was set to the proper speed (mi·h−1), and the subjects were asked to match their step frequency to a metronome that was set to frequency from the baseline protocol (steps·min−1).
We measured the rates of oxygen consumption (O2) and carbon dioxide production (CO2) using an open circuit respirometry system (ParvoMedics, Sandy, UT, USA). Before beginning the experimental trials, we measured standing metabolic rate. For all trials, we allowed 3 minutes for the subjects to reach steady state and then calculated the average O2 (mL O2 per second) for the subsequent 4 minutes. Our rationale for measuring metabolic cost was to understand which strategy, double or single, would maximize oxygen consumption.
We collected electromyography (EMG) signals at 1,000 Hz using a wired amplifier system (Bortec Octopus AMT-8, Calgary, AB, Canada) with a bandpass filter setting of 5-500 Hz. Before electrode placement, we prepared the skin with fine sandpaper and rubbing alcohol. We placed 1- × 1.5-cm2 bipolar, silver-silver chloride, surface electrodes (Vermed, A10041, Bellows Falls, VT, USA) over 6 muscles of the left leg according to the recommendations by Cram and Kasman (4). The interelectrode distance was 2 cm. We verified that the position of the electrodes was functionally correct and that crosstalk between the muscles was negligible with a series of flexion and extension exercises suggested by Winter et al. (22) and Cram and Kasman (4).
For the muscle activity analysis, we generated a linear envelope (21) obtained by low pass filtering (dual-pass, fourth order, Butterworth) the rectified EMG data at 10 Hz. For the amplitude analysis, we full-wave rectified the band-pass filtered signals, calculated the mean EMG amplitude (mEMG) for 5% segments of the stride, and averaged 5 strides for each condition. Our rational for measuring muscle activity was to evaluate how the activities of lower limb muscles influence metabolic cost.
Six men and 6 women completed the experiment (age = 20.90 [0.99] years, mass 69.57 [16.23] kg, and height 1.72 [0.11] m, mean [SD]). We included any subject that could complete the baseline testing session and did not screen for fitness level or activity background. Subjects with any current bone or muscular injuries and lower body operations within the last 2 years were excluded. All of these healthy subjects gave written, informed consent according to The Pennsylvania State University Human Research Committee.
Metabolic and muscle activity data were analyzed across all conditions using a 1-way repeated-measures analysis of variance. We performed Newman-Keuls post hoc tests to analyze the difference between each significantly different experimental condition. Significance was defined as p ≤ 0.05.
Contrary to our hypothesis, during the baseline testing, every subject completed the double-step protocol with a speed (3.33 mi·h−1 or 1.91 m·s−1) that was 22% faster than the single-step protocol (4.27 min·h−1 or 1.49 m·s−1, p < 0.01). This faster speed during the double-step protocol was achieved with a step frequency (83 steps·min−1) that was on average 24% less than the single speed (109 steps·min−1).
The metabolic data indicate that faster speeds affected metabolic cost more than a quicker step frequency (Table 1). The results of our post hoc analysis illustrated that each experimental condition was significantly different from each other (p < 0.05). During the experimental session, oxygen consumption during Dspd_Dfreq was 13% greater than Sspd_Sfreq (p < 0.001) because of the faster speed. For an hour of exercise, the calculated difference between the 2 activities is 87 kcal for a 70-kg adult.
The muscle activity of the ankle muscles was also dependent upon both speed and step frequency (Figure 2). The lateral gastrocnemius (LG) muscle activity during the entire stance phase of the faster speed conditions was 21% greater than both the slower speed conditions (p < 0.05). But during terminal stance, LG activity during the single-step frequency conditions was at least 29% greater than the double-step frequency conditions (p < 0.05). The SOL activity demonstrated analogous patterns during terminal stance of the faster speed conditions. In fact, at the extreme, SOL activity during terminal stance of the Dspd_Dfreq condition was 62% greater than during terminal stance of the Sspd_Sfreq condition (p < 0.01). To add, SOL activity during initial stance of the double-frequency conditions was 44% greater than the single-frequency conditions (p < 0.05). As for the tibialis anterior (TA) activity, the only significant difference was between the slower speed conditions. During terminal swing, TA activity was 36% less during the Sspd_Sfreq condition than the Sspd_Dfreq condition (p < 0.05).
At the knee, the differences in speed and step frequency also elicited respective changes in muscle activity (Figure 3). The VL muscle activity during the stance phase of the faster speed and double-step frequency condition was significantly greater than the other conditions (p < 0.05). This overall difference in the Dspd_Dfreq condition was largely because of an additional VL activity during the second half of stance, which was 52% greater than the Sspd_Sfreq condition (p < 0.01). However, the peak VL activity during the first half of stance was on average 31% greater for the single-step frequency conditions (p < 0.05). In contrast, rectus femoris (RF) activity during stance was not statistically different between conditions during stance until the initiation of swing. The RF activity of Sspd_Sfreq and Dspd_Sfreq was 18% greater than Sspd_Dfreq (p < 0.05) and 36% greater than Dspd_Dfreq (p < 0.01) during the transition between stance and swing. Although the knee extensors, VL and RF, demonstrated significant activity changes during stance and early swing, the knee flexor, biceps femoris (BF) demonstrated significant changes during swing. In particular, the BF activity, during swing of the faster speed conditions was on average, 26% greater than the slower speed conditions (p < 0.05).
Every subject completed the baseline testing with a faster average speed during the double-step strategy. After mimicking each single and double stair-climbing strategy with our experimental methods, we calculated that the double-step strategy would yield a greater use of metabolic energy, equal to approximately 1.0-1.3 kcal·kg−1·h−1, on average 70-90 kcal·h−1 (Table 1). This double-step strategy required a greater activity for propulsion during stance for the LG, SOL, VL, and RF. Further, the SOL activity was also greater during the first half of stance, potentially to aid in anterior-posterior stability and body weight support, whereas the RF activity was also greater during the second half of stance, potentially to aid in hip flexion. Therefore, to maximize caloric output and muscle activity, we would recommend a double-step strategy to individuals completing a stair-climbing workout.
Our experimental protocol included the baseline speed and step frequency combinations for the single (Sspd_Sfreq) and double (Dspd_Dfreq) step strategies and the cross combinations of single speed with double frequency (Sspd_Dfreq) and double speed with single frequency (Dspd_Sfreq). The cross combinations provide insight regarding the proportional influence of speed and step frequency on total metabolic cost. During level walking, as speed increases, the rate of oxygen consumption increases curvilinearly (12). The metabolic cost during both of the faster conditions (Dspd_Dfreq and Dspd_Sfreq), was significantly greater than the slower conditions (Sspd_Sfreq and Sspd_Dfreq). Although we tested only 2 speeds, it appears that the metabolic cost points for this steep walking grade were located on the positive portion of the oxygen consumption curve.
This curvilinear relationship between speed and rate of oxygen consumption demonstrates that at a particular speed, humans select a walking pattern, or step frequency, that minimizes metabolic cost (15,17). So we would assume that oxygen consumption during the cross combination conditions would be greater than the baseline combinations for a given speed because the subjects did not choose these combinations of speed and step frequency. In concert with this principle, during the faster speed conditions, oxygen consumption was significantly greater during the cross combination Dspd_Sfreq condition than the baseline Dspd_Dfreq condition with a slower step frequency. Contrary to this principle, during the slower speed conditions, oxygen consumption was significantly less during the cross combination Sspd_Dfreq condition than the baseline Sspd_Sfreq condition with a faster step frequency. Minetti et al. reported that total mechanical power is minimized at stride rates 20-30% below the preferred rate (14). So perhaps mechanical power was less during the Sspd_Dfreq condition in comparison to the Sspd_Sfreq condition.
The muscle activity data of both the ankle and the knee also reveal patterns that are unique for speed and step frequency independently. Because the LG is one of the primary muscles responsible for propulsion (6,16,21), the activity during stance was significantly greater during the faster conditions. There was also a significant difference between the step frequency conditions. During terminal stance, the LG activity during the single-frequency conditions was greater than the double frequency conditions. Neptune et al. (16) stated that during this segment of stance, the LG aids in the initiation of swing. It is possible therefore that the single frequency conditions require additional activity to maintain the quicker frequency.
Although the LG and SOL share activity patterns during midstance, the SOL activity during initial stance did not correspond to the LG. Gottschall and Kram (6) also recognized this discrepancy in firing within the triceps surae and concluded that the soleus aids in propulsion during scenarios that require supplementary propulsion. During the first 10% of stance, the SOL activity during the double-frequency conditions was greater than the single-frequency conditions. McFadyen and Winter (13) attribute this activity to the necessity of providing vertical displacement for the center of mass and increasing the vertical position of the knee. Thus, it is feasible that the SOL provides supplementary support of body weight and anterior-posterior stability during the first half of stance.
There were also unique activity patterns between the double- and single-step conditions at the knee. The most significant difference was found for the VL during the faster speed conditions. Similar to previous research (17) and to our data for the ankle extensors, the VL activity increases with the demand for propulsion during stance. Interestingly, there was no similar increase for RF, also a knee extensor. Instead, the greatest difference for RF was during swing of the double-step conditions. This increased activation has not been previously reported potentially because of the unique requirements of skipping a stair during the double-step condition. The BF also elicited increased activity during swing of the faster speed conditions perhaps to aid in the control of the leg before footstrike.
Our protocol enabled us to compare 2 different natural frequencies of walking on an extreme grade that mimics stair walking. However, stair walking and steep grade walking are not identical. The most obvious difference is the position of the ankle during stance. This angle is more consistent during stair walking as the support surface is flat and limits our conclusions regarding the functional role of the ankle extensors. Likewise, the ankle function would also differ during the swing phase because of the different demand for toe clearance on a slope of a treadmill in contrast to the steps on a flight of stairs. Despite these variations, our protocol provides insight into how these 2 methods of stair climbing differ and the unique relationship between speed and stride frequency.
In summary, to maximize metabolic cost and muscular activity, we recommend a double stair-climbing strategy. For the exercising individual, the double-stair strategy will result in a greater amount of energy expenditure during a stair-climbing routine. For the conditioning coach, we also suggest recommending the double-stair strategy to athletes. This method will maximize cardiovascular benefits and muscle recruitment.
We thank Andrew J. Michael for developing the motion analysis calibration protocol and Nori Okita for creating the muscle activity analysis program.
1. Andriacchi, TP, Andersson, GB, Fermier, RW, Stern, D, and Galante, JO. A study of lower-limb mechanics during stair-climbing. J Bone Joint Surg Am
62: 749-757, 1980.
2. Aziz, AR and Teh, KC. Physiological responses to single versus double stepping pattern of ascending the stairs. J Physiol Anthropol Appl Human Sci
24: 253-257, 2005.
3. Boreham, CA, Wallace, WF, and Nevill, A. Training effects of accumulated daily stair-climbing exercise in previously sedentary young women. Prev Med
30: 277-281, 2000.
4. Cram, JR and Kasman, GS, Electrode Placement. Introduction to Surface Electromyography
. Cram, JR, Kasman, GS, and Holz, J, eds. Gaithersburg, MD: Aspen, 1998. pp. 371-375.
5. Fardy, PS and Ilmarinen, J. Evaluating the effects and feasibility of an at work stairclimbing intervention program for men. Med Sci Sports
7: 91-93, 1975.
6. Gottschall, JS and Kram, R. Energy cost and muscular activity required for propulsion during walking
. J Appl Physiol
94: 1766-1772, 2003.
7. Haskell, WL, Lee, IM, Pate, RR, Powell, KE, Blair, SN, Franklin, BA, Macera, CA, Heath, GW, Thompson, PD, and Bauman, A. Physical activity and public health: Updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc
39: 1423-1434, 2007.
8. Ilmarinen, J, Ilmarinen, R, Koskela, A, Korhonen, O, Fardy, P, Partanen, T, and Rutenfranz, J. Training effects of stair-climbing during office hours on female employees. Ergonomics
22: 507-516, 1979.
9. Joseph, J and Watson, R. Telemetering electromyography
of muscles used in walking
up and down stairs. J Bone Joint Surg Br
49: 774-780, 1967.
10. Loy, SF, Conley, LM, Sacco, ER, Vincent, WJ, Holland, GJ, Sletten, EG, and Trueblood, PR. Effects of stairclimbing on O2max and quadriceps strength in middle-aged females. Med Sci Sports Exerc
26: 241-247, 1994.
11. Mann, R and Inman, VT. Phasic activity of intrinsic muscles of the foot. J Bone Joint Surg Am
46: 469-481, 1964.
12. Margaria, R, Sulla fisiologica e specialmente sul consumo energetico della corsa a varie velocita ed inclinazioni del terreno. Atti Accad Naz Lincei Menorie
7: 299-368, 1938.
13. McFadyen, BJ and Winter, DA. An integrated biomechanical analysis of normal stair ascent and descent. J Biomech
21: 733-744, 1988.
14. Minetti, AE, Capelli, C, Zamparo, P, di Prampero, PE, and Saibene, F. Effects of stride frequency on mechanical power and energy expenditure of walking
. Med Sci Sports Exerc
27: 1194-1202, 1995.
15. Molen, NH, Rozendal, RH, and Boon, W. Graphic representation of the relationship between oxygen-consumption and characteristics of normal gait of the human male. Proc K Ned Akad Wet C
75: 305-314, 1972.
16. Neptune, RR, Kautz, SA, and Zajac, FE. Contributions of the individual ankle plantar flexors to support, forward progression and swing initiation during walking
. J Biomech
34: 1387-1398, 2001.
17. Ralston, HJ. Energy-speed relation and optimal speed during level walking
. Int Z Angew Physiol
17: 277-283, 1958.
18. Shinno, N. Statico-dynamic analysis of movement of the knee. V. Analysis of knee function in ascending and descending stairs. Tokushima J Exp Med
15: 53-57, 1968.
19. Teh, KC and Aziz, AR. Heart rate, oxygen uptake, and energy cost of ascending and descending the stairs. Med Sci Sports Exerc
34: 695-699, 2002.
20. Townsend, MA, Lainhart, SP, Shiavi, R, and Caylor, J. Variability and biomechanics of synergy patterns of some lower-limb muscles during ascending and descending stairs and level walking
. Med Biol Eng Comput
16: 681-688, 1978.
21. Winter, DA. The Biomechanics and Motor Control of Human Gait
. Waterloo, Canada: Waterloo Biomechanics, 1991.
22. Winter, DA, Fuglevand, AJ, and Archer, SE. Crosstalk in surface elctromyography: Theoretical and practical estimates. J Electromyogr Kinesiol
4: 15-26, 1994.