Anna K Kuramoto and V. Gregory Payne

Predicting Muscular Strength in Women : A Preliminary Study

The scientific and medical community recommend resistance training as an adjunct to adult fitness programs (American College of Sports Medicine [ACSM], 1990; Fiatarone et al., 1990). Proper training intensity, is essential to optimizing gains in strength and muscle mass. Using a training intensity of 80% of a one repetition maximum (1-RM), Frontera, Meredith, O'Reill),, Knuttger, and Evans (1988) found greater increases in strength and muscular hypertrophy in older men than did Moritani and deVries (1980), who used a training regimen at 66% of a 1-RM and yielded no significant increase in muscle area. Fiatarone et al. (1990) studied nonagenarians and found intensity of training to be an important factor in the large strength gains (M= 174%, SD = 31 %) and muscle mass gains (M= 9.0%, SD = 4.5%) of the subjects. An inexperienced weight trainer can easily under or overestimate a specific RM or a 1-RM, leading to poor training stimulus or injury. Measuring 1-RM requires a trial-and-error method of manipulating weight loads until a maximal effort is achieved. This can be time-consuming and inaccurate. In addition, maximal lifts can cause fractures, spondylolisthesis, and torn ligaments (Matheson, MacIntyre, Taunton, Clement, 8: Lloyd-Smith, 1989). Using a muscular endurance test procedure without the use of a trial-and-error method to predict muscular strength would be a safe and simple alternative. Muscular endurance can be assessed by performing repetitions to fatigue using a set percentage of body weight. However, insufficient information currently exists to establish the efficacy of such a technique. Recent investigations have focused on dynamic muscular endurance as a predictor of muscular strength. A strong correlation was found between the YMCA bench press protocol (an absolute endurance measure) and bench press strength using college-aged male subjects, r=.92 (Invergo, Ball, & Looney, 1991), and female subjects, r=.85 (Ball & Rose, 1991). Relative muscular endurance (%1-RM) has also been shown to correlate with bench press strength, r= .98 (Mayhew, Ball, Arnold, & Bowen, 1992) and knee extension strength, r=.94 (Braith, Graves, Leggett, & Pollock, 1993). Over time, muscular strength appears to deteriorate at a greater rate in the lower body than does grip or arm strength (Grimby & Saltin, 1983). For women, weakened lower body musculature and inadequate upper body strength can contribute to an increase in falls, hip fractures, and difficulty in performing daily activities (Heyward, Johannes-Ellis, & Romer, 1986; Whipple, Wolfson, & Amerman, 1987). Strength training modalities have been shown to be beneficial in maintaining musculoskeletal integrity (Charette et al., 1991). Defining methods for establishing safe, simple, and effective intensity levels would be useful. Determining strength training workloads based on strength prediction equations is one such measure. To date, investigations have been limited to college aged subjects focusing on prediction equations for the bench press (Ball & Rose, 1991; Invergo et al., 1991; Mayhew et al., 1992) and knee extension (Braith et al., 1993). Studies have not included older women for strength prediction. Furthermore, investigations have neglected regions of the back. Scapulothoracic and shoulder joint actions upon the midback are influenced by several important muscles used in many daily tasks, sports, and recreation and in the maintenance of good upper body posture. Weakness in this area can have serious health and functional consequences for women. Therefore, the purpose of this study was to determine the value of a relative muscular endurance field test in predicting the midback strength of women using the lat pulldown machine. The development of accurate strength prediction equations using submaximal weight loads would be valuable in community-based adult fitness programs for establishing strength training routines. Method Subjects Seventy-three women were tested for muscular endurance and strength measures of the midback. Subjects were recruited at large from the community. Three age groups were formed: (a) early adulthood (20-30 years), (b) middle adulthood (40-50 years), and (c) late adulthood (60-70 years). Refer to Table 1 for descriptive characteristics of each age group. Subjects completed an informed consent and a medical/training history questionnaire for screening of musculoskeletal problems and limitations, previous weight training experience, past and present exercise habits, medications, and chronic and current illnesses. The criteria for being included in the study were less than or equal to 2 years of weight training experience and no participation in a weight training program for the past 6 months. The exclusionary criteria were hypertension(>140/90), hypotension (<90/60), medical diagnosis of cerebro- or cardiovascular disease, or musculoskeletal limitations to the neck, shoulder, and back. Apparatus Muscular strength and endurance testing was conducted on a constant resistance let pulldown machine (Icarian Fitness Equipment, Inc., San Fernando, CA). The subject was secured into position by raising the seat to allow the thighs to meet the stabilizing pads and to maintain the hip and knee joints in horizontal alignment Table 1. Means and standard deviations of physical and performance characteristics by age group Early adulthood Middle adulthood Late adulthood n=23 n=27 n=23 Variable M SD M SD M SD _____________________________________________________ Age 26.8 2.9 45.2 3.1 64.0 3.3 Weight 57.7 5.2 64.2 12.4 63.4 8.8 Height 162.0 4.4 164.8 8.2 162.3 5.0 BMI 21.9 2.3 23.6 4.3 24.0 3.0 Repetitions 10.7 3.8 8.8 4.0 4.6 4.3 ME WT 25.9 2.5 28.9 5.6 28.5 3.9 1-RM 35.8 4.8 36.0 6.4 31.4 4.7 Because the stabilizing pads were not adjustable, wooden platforms were placed under the subject's feet for proper alignment. The grip sites on the pulldown bar were marked with tape for consistent hand placement. Design A pilot study was conducted with 26 subjects to evaluate the feasibility of the protocol, appropriate cadence parameters, hand placement, body positioning on equipment, amount of muscular endurance weight load (ME WT), and amount of warm-up weight load and repetitions Heyward (1991) used a ME WT of 50% of body weight for muscular endurance testing in college-aged subjects However, in the pilot study, a ME WT of 45% of body weight was selected to allow comparisons of the three age groups. The late adulthood pilot values for repetitions at a ME WT of 45% of body weight ranged from 3 to 11. At the same percentage-based ME WT, the early adulthood pilot subjects performed a range of 5 to 18 repetitions, while the middle adulthood pilot subjects completed 3 to 14 repetitions. Increasing the weight load to 50% of body weight would risk performance failure by a portion of the older subjects. Conversely, lowering the weight load below 4% of body weight would inflate values for the younger subjects Subjects were briefed on the 1-RM and muscular endurance tests for the let pulldown using a computerized slide show introduction as well as videotaped instructions on testing procedures. Height and weight measurements were taken. Subjects reported for testing after abstaining from new or unusual physical activity for at least 48 hr. The 1-RM test was conducted first followed by the muscular endurance test. A 1- RM test period between the two tests reduced the effects of fatigue. Procedure 1-RM standards. The 1-RM let pulldown procedures were derived from pilot findings and research conducted by Beckett ( 1983) . A verbal cue and a light tap to the marker on the base of the subject's neck; were starting signals to initiate the movement down. Each trial was followed by a 3-min rest period. The 1-RM was recorded as the largest amount of weight pulled from full elbow extension overhead to the end point of elbow flexion and the bar touching the marker on the base of the subject's neck. Muscular endurance standards. The testing involved the coordination of movement to the established cadence. The audiotaped cadence ratio was two counts for the pulldown and four counts for the return to overhead elbow extension. The criterion for discontinuing the test was failure to complete the pulldown to the taped marker on the base of the subject's neck in time with the cadence. Data Analysis A one way analysis of variance was used to determine if significant differences existed among age groups for relative muscular endurance. The level of significance was .05. Scheffe post hoc analysis was employed to determine which specific age-group means were significantly different from one another. Significant differences would require separate nonstepwise multiple regression analyses to derive the appropriate prediction equation of strength. Possible predictor variables were age, repetitions to fatigue, weight, height, and body mass index. Muscular endurance weight load was also included as repetitions were dependent on the magnitude of the load. The dependent variable was muscular strength (1-RM). Prior to conducting regression statistics, a correlation matrix was examined to determine if multicollinearity existed among any of the predictor variables. Results The means and standard deviations for various physical and performance characteristics for all 73 subjects are summarized in Table 1. Height and weight values were greatest for the middle adulthood group. Muscular endurance and muscular strength of the midback were represented by repetitions to fatigue and 1-RM efforts. A general decline in muscular endurance performance (repetitions) was observed with increasing age, while muscular strength (1-RM) was reduced only in the oldest group (see Table 1). The ME WTs for 8 of the 23 subjects from the late adulthood group were close to or above 1-RM performances, ranging from 96% to 115% 1-RM. As a result, these subjects were not able to complete a single repetition during the muscular endurance test. The remaining 15 subjects had ME WTs equivalent to 75% to 94% 1-RM and repetitions averaging 7.1. A one way analysis of variance indicated that a significant difference among age groups was noted for repetitions, F (2, 70) = 3. 13, p < .05. Scheffe post hoc analysis indicated that the late adulthood group performed significantly fewer repetitions than did the two younger groups. The early and middle adulthood groups were not significantly different from each other. The proportion of total variance (w-square) in muscular endurance attributed to age was 39.7%. As a result of these findings, the early and middle adulthood subjects were combined for regression statistics too predict 1-RM strength, and the late adulthood group was analyzed separately. Table 2 contains a correlation matrix of predictor variables for early-middle adulthood subjects. weight, body mass index, and ME WT were correlated above .70, indicating multcollinearity. Body mass index, weight, and height were not included in the equation as the ME WT was based on an individual's body weight, and the body mass index is a function of weight and height. Comparable findings existed for the late adulthood group (see Table 3). Table 2. Correlation matrix for early-middle adulthood predictor variables (n=50) Variable Age Weight Height BMI Repetitions ME WT ____________________________________________________________ Age Weight .307 Height .189 .385 BMI .241 .849 -.150 Repetitions -.232 -.454 -.350 -.263 ME WT .310 .995 .377 .856 -.420 Table 3. Correlation matrix for early-middle adulthood predictor variables (n=23) Variable Weight Height BMI Repetitions ME WT ____________________________________________________________ Weight Height .494 BMI .901 .070 Repetitions -.443 -.197 -.142 ME WT .999 .486 .856 -.437 For the early-middle adulthood subjects, 1-RM was regressed on age, ME W[, and repetitions. The regression analysis was significant, F(3, 46) = 139.0, p<.001, and the standard error of estimate (SEE) revealed that 1-RM strength could be predicted with an accuracy of 1.85 kg. The multiple correlation was strong at .95 with an adjusted R-square of .89. The resulting prediction equation was 1-RM = -3.41 + (-.20 x Age) + (1.06 x ME WT) + (.58 x Repetitions). One-repetition maximum was regressed on ME WT and the repetitions for the late adulthood subjects (see Equation 2). A significant relationship was found, F (2, 20) = 48.9, p < .001, with a SEE of 2.04 kg. The R value of .91 was high, resulting in an adjusted R-square of .81. The resulting prediction equation was 1-RM = -3.73 + (.92 x ME WT) + (.79 x Repetitions). Discussion The use of a predetermined hIE WI at 45% of total body weight represented intensities of 73% 1-RhI for the early adulthood women and S0% 1-RSt for the middle adulthood women. Repetitions averaged 9.S for the two youngest groups and fell close to the repetition range commonly prescribed for strength training (ACSM, 1990). Therefore, the high Rvalue (.95) for the combined early-middle adulthood group is partially explained by this variable. Additionally, because the number of repetitions performed is dependent upon the ME \VT, this was considered another important component. Age was included in the prediction equation be- cause the early and middle adulthood women were separated by a decade. The inability of eight women from the oldest group to complete the muscular endurance test suggests that closer scrutiny for this potential problem during the pilot study should have been warranted. Several possible influences may have contributed to this performance disparity within the late adulthood women. First, body composition differences between the performing and nonperforming subgroups is one area of consideration. The body weight and body mass index of late adulthood nonperformers were significantly greater than the performers of the muscular endurance test. Height was not a factor influencing this difference. Had the higher body weight and body mass index of late adulthood nonperformers been attributed to greater muscle mass, the expectation would have been to observe at least comparable and possibly enhanced test performance from late adulthood performers However, the 1-RSI efforts of nonperformers were less than the performers. The lower 1-RM values consequently resulted in an inability to execute the muscular endurance test with the designated weight load. Age disparity was another possible limiting factor The potential for the nonperformer subgroup to be composed of a preponderance of women greater thar 65.5 years of age was considered. Yet, significant differences in age were not found between late adulthood performers and nonperformers. Additionally, psychological inhibitions can influence muscular strength and endurance performances Although precautions were taken to guard against in jury, some of the older women expressed concern for their safety and may not have exhibited their true abilities. An older woman's motivation to fully exhibit strength capabilities can be hindered by age-role expectations and perceived physical limitations (Ostrow & Dzewaltowski, 19S6; Prohaska, Leventhal, Leventhal, & Keller, 1985). Therefore, despite strict adherence to protocol conducive to a genuine maximal effort, mental barriers may have contributed to the disparity between muscular endurance performers and nonperformers. The lower strength values of the nonperforming muscular endurance subgroup may also in part be at- tributed to a higher degree of inactivity of the upper body. Three of the eight nonperformers were identified as sedentary and as a result could have had reduced exposure to upper body movements necessary for the maintenance of muscular integrity in this region. The muscles of the midback; support the scapula and are in- volved in maintaining upright posture. Disuse of these muscles is highly likely as the nature of most daily activities is oriented in a forward flexed position. The resulting muscular atrophy can result in rounded shoulders and reduced shoulder joint flexibility thereby hampering efforts to achieve proper biomechanical alignment for upper body exercises and optimal performance. Comparatively, the high predictive values for midback strength support findings of previous muscular strength studies of the chest and anterior thigh. However, the independent variables chosen for prediction purposes have varied from study to study (Ball & Rose, 1991; Braith et al., 1993; Mayhew et al., 1992). The characteristics of body weight, weight load, and repetitions to fatigue have been employed singularly in the past, whereas this investigation has made use of weight load, repetitions, and age in an integrated fashion. (Braith et al.. 1993: Invergo et al.. 1991: Mayhew et al., 1992). Age-group comparisons may not be possible using a single selected value for the ME WT. However, strong predictive R values for the early middle as well as the late adulthood women suggest the practical use of this method in the field setting utilizing age specific weight loads. Future investigations focusing on separate age groups may result in age appropriate ME WTs based on percentage of body weight. Reducing the ME WT of late adulthood participants from 45% to 40% of body weight may be appropriate for within group comparisons.