Secondary Logo

Journal Logo

Original Research

Force-Time Curve Characteristics and Hormonal Alterations During an Eleven-Week Training Period in Elite Women Weightlifters

Haff, G Gregory1; Jackson, Janna R1; Kawamori, Naoki2; Carlock, Jon M3; Hartman, Michael J4; Kilgore, J Lon4; Morris, Robert T5; Ramsey, Michael W6; Sands, William A3; Stone, Michael H2,6

Author Information
Journal of Strength and Conditioning Research: March 2008 - Volume 22 - Issue 2 - p 433-446
doi: 10.1519/JSC.0b013e31816191be
  • Free



To date, most of the research concerning the muscular and hormonal responses to competitive weightlifting training has been carried out by using elite male weightlifters (2,6,11,16-19). Conversely, very few data exist in the scientific literature that explore these effects in competitive female weightlifters (33). However, several investigations have been conducted to examine the hormonal and neuromuscular adaptations to various durations of strength and power training in women (20,21,26,28).

When examining the anabolic and catabolic effects of resistance training, the acute and chronic responses of testosterone and cortisol are often examined (20,21,26,28). Generally, women have a circulating testosterone level that is approximately 9.5% of that found in men, which is a result of the primary sources of testosterone in women (7). The primary sources of testosterone in women occur as a result of the peripheral conversion of androstenedione and dehydroepiandrosterone, with the remainder of the circulating testosterone being secreted from the ovaries and the adrenal cortex (8). It may seem intuitive to suggest that hormonal fluctuations during the menstrual cycle affect the response of testosterone and cortisol to exercise, but several studies suggest no significant phase effect during resistance training (9,20) or aerobic training (4,12,23,24).

Recently, Elliott et al. (9) reported that resting testosterone concentration does not significantly change in response to the menstrual cycle, and as a result, it was indicated that the menstrual cycle phase exerts a minimal affect on strength performance. Similarly, Häkkinen et al. (20) reported that physically active women showed no statistically significant differences in mean serum concentrations of testosterone, free testosterone, or cortisol in response to 3 weeks of training, which crossed several phases of the menstrual cycle, although an 8.4% increase in maximal force generating capacity was reported. However, it is noteworthy that Häkkinen et al. (20) also suggested that resting serum testosterone concentrations have a large interindividual difference in women, which, in turn, may explain interindividual differences in hypertrophy and strength development. Additionally, Marx et al. (28) reported that after 12 weeks of training, women have higher testosterone levels when compared to a control group. Therefore, it may be hypothesized that elite female strength athletes may have higher serum testosterone concentrations.

In one of the few studies that have examined female weightlifters, Stossell et al. (33) reported that there was no significant difference in resting testosterone levels of elite female weightlifters and untrained controls. More recently, Cardinale and Stone (7) reported that the basal concentrations of testosterone differ between female athletes, with elite level sprinters having higher testosterone concentrations than elite female volleyball players. Basal testosterone concentrations were reported to be higher in those athletes who participated in sports that required high power outputs. Because weightlifting requires athletes to generate high power outputs (13), one may speculate that elite female weightlifters would have higher basal testosterone levels, which may be a result of the type of training in which they participate, but the authors know of no investigations that have directly assessed this notion.

Several studies have examined the hormonal responses of physically active women to resistance training and have reported that the training cycle itself can exert an effect on the serum concentrations of testosterone (26), which may be related to force- and power-generating capacity (21,28).

Häkkinen et al. (21) examined the effects of 16 weeks of power training on hormonal and neuromuscular adaptations in 7 physically active women. The results of this investigation suggested that there are positive relationships between individual mean resting serum testosterone concentrations and selected force-time curve characteristics, such as the ability to produce maximal force. It is noteworthy that the resting serum testosterone concentrations were significantly related to force- and power-generating capacity during the training period, whereas cortisol levels appeared to be largely unaffected by the training stimulus.

In another study, Kraemer et al. (26) examined the effects of 8 weeks of resistance training on the endocrine functions of active women. Levels of testosterone before exercise were significantly higher than those in the first week of training, when examined after 6 and 8 weeks of training. Additionally, it was shown that cortisol levels decreased across the training cycle. Taken as a whole, these observations (9,20,21,26) suggest that the training stimulus exerts a stronger effect on resting serum testosterone concentrations than the menstrual cycle phases do.

Different training stimuli have also been reported to produce different hormonal alterations in untrained physically active women (28). Marx et al. (28) showed that in a 36-week periodized training program, resting testosterone levels increase significantly. When compared to a lower-volume training regimen, the high-volume periodized training program produced significantly greater increases in resting testosterone levels and strength power performance.

It is noteworthy that most of the research performed to examine the hormonal responses to resistance training has used untrained or recreationally trained women. To the authors' knowledge, no studies have evaluated the hormonal fluctuations and muscular performance characteristics of elite female weightlifters during periods of prolonged training. Therefore, the purpose of this study was to examine the effects of 11 weeks of training on neuromuscular performance capacity, preceding a major national competition and performed by elite female weightlifters. This type of observation is particularly important because the female weightlifters were studied during real-life training. Thus, there were no artificial adaptations in the training program variables or training environment to fit the menstrual cycle. It was hypothesized that fluctuations in volume load and training intensities would result in concomitant alterations in force-time curve parameters collected during dynamic and isometric muscle actions. Additionally, it was hypothesized that alterations in the anabolic and catabolic hormones would correspond to these fluctuations.


Experimental Approach to the Problem

This study was designed to evaluate the hormonal and neuromuscular changes in response to 11 weeks of training leading up to the 2002 American Weightlifting Championships. The basic design of this investigation was based on previously published research looking at the seasonal changes in performance and hormonal concentrations in elite female rowers (40). Six elite female weightlifters volunteered to be part of this investigation. The six athletes were part of the USA Weightlifting Resident Women's Weightlifting Program based at the U.S. Olympic Training Center in Colorado Springs, Colorado. Subjects were assessed twice for baseline hormonal concentrations before the study began. The baseline period consisted of 2 weeks of active rest training, which was designed to give a stable hormonal baseline. During the 11 weeks of training, the subjects were assessed for neuromuscular performance characteristics and hormonal concentrations on a biweekly basis. Subject characteristics were also collected on a biweekly basis throughout the course of the investigation. Figure 1 shows the research design.

Figure 1:
Experimental design.


Six elite eumenorrihic female weightlifters participated in the study. All subjects were members of USA Weightlifting's Resident Women's Weightlifting Program and were living and training at the Olympic Training Center in Colorado Springs, Colorado. Descriptive data for the subjects are shown in Table 1. All participants had competed in national- and international-level weightlifting competitions before this study. Table 2 shows a summary of the athletes' national senior or junior rankings and the percentage of the national record that the athletes' maxima in the snatch, clean and jerk, and total represent. Because of the methodologies used in the selection of the athlete sample in the current investigation, it was not possible to influence the use of oral contraceptives by the subjects or control the phase of the menstrual cycle when testing was undertaken. In a similar study, Vervoorn et al. (40) did not control for oral contraceptive use or phase of menstrual cycle in a study in which the seasonal changes of the testosterone:cortisol ratio in female rowers were monitored. However, based on a questionnaire and careful independent questioning during the screening before the study and throughout the investigation, it was determined that none of the 6 subjects participating in the study used oral contraceptives during the study period. Additionally, all subjects were currently in the U.S. Anti-Doping Association random testing pool and were not taking any anabolic androgenic steroids or other doping products; at least one of the athletes was selected for random testing each month. Before participating in the study, all subjects read and signed an informed consent form and completed a health history questionnaire in accordance with guidelines set forth by the Human Subjects Review Committee at Midwestern State University and the U.S. Olympic Committee.

Subject characteristics at the beginning of the study (n = 6).
Summary of athletes' maxima in the competitive lifts (n = 6).

Training Program

The study was performed during 11 weeks of a 12-week training period leading to the 2002 American Weightlifting Championships. The training program consisted of core exercises (i.e., clean, clean and jerk, and the snatch) and other supplemental exercises (i.e., clean pull, snatch pull, squat, and the front squat), as prescribed by the national coach. All core and supplemental exercises were used in the calculation of the volume load, repetitions encountered, and the training intensity.

Volume of training can be estimated in 2 ways (36): by determining the total repetitions performed or by calculating the volume load (repetitions × mass lifted). During the 11 weeks of training, the subjects accomplished a total of 3,593 ± 606 repetitions for a total volume load of 270,221 ± 42,345 kg. The weekly volume load trends for the training program are shown in Figure 2. The training intensity, which estimates the rate at which training proceeds, was calculated from the following equation: training intensity = total volume load/total repetitions (35,36). The weekly training intensity trends are shown in Figure 3.

Figure 2:
Volume load distribution for 11-week mesocycle.
Figure 3:
Relationship of weekly training intensities to testosterone:cortisol ratio. *Significant difference between baseline testosterone:cortisol ratio and week 9 (P < 0.02). **Significant difference between testosterone:cortisol ratio of week 1 and week 3 and week 1 and week 11 (P < 0.01).

Force-Time Curve Analysis

For the analysis of force-time curve characteristics, the primary factors analyzed for their important relationship with sporting performance are peak force (PF) (37) and peak rate of force development (PRFD) (30).

All force-time curve analyses were performed on a customized rack system that allows the bar to be fixed for isometric or dynamic muscle actions at any height above the floor by using a combination of pins and hydraulic jacks (Figures 4 and 5). The rack system was positioned over a 61 × 121.9 cm force plate (Advanced Mechanical Technologies, Newton, MA), which was sampled at 600 Hz. All data were calculated from the vertical force component of the ground reaction forces by previously reported methods (15).

Figure 4:
Isometric mid-thigh clean pull testing apparatus. Schematic (A) and photograph (B) of testing apparatus set-up.
Figure 5:
Dynamic mid-thigh clean pull testing apparatus. Schematic (A) and photograph (B) of testing apparatus set-up.

Three types of mid-thigh pulls were tested in this investigation: isometric, dynamic pull at 30% of isometric PF (IPF), and dynamic pull at 100 kg. The mid-thigh pull was selected as a methodology for evaluating performance based on previously published research, which suggests a strong relationship to lifting performance (15). The position selected for testing was at the mid-thigh because it corresponds to where the highest forces and velocities are generated in the snatch and clean (14). The mid-thigh position during a clean corresponds to the knee angle just before the bar leaves the thigh, with the trunk upright, during the start of the second pull (14). The positions for the isometric and dynamic trials were set based on knee angles calculated from a 2-dimensional video analysis (Peak Performance, Englewood, CO) of each subject's actual weightlifting performance. The knee angles established for this subject population ranged from 127° to 145°.

One week before the study, each subject went through a familiarization session in which she performed several practice isometric and dynamic pulls. On the day of testing, each subject performed a standardized warm-up based on previous literature (15). Before each testing session, the subjects' lifting position was established by goniometry. Subjects were then strapped to the bar according to previously reported methods (15). With each trial, subjects were instructed to pull as hard and as fast as possible. Previous research has suggested that the use of these instructions produces optimal results for maximal force and the rate of force development (3). Each subject performed 4 isometric trials and 2 trials of each of the 2 dynamic tests (i.e., 8 total mid-thigh pulls). Subjects were given 2 minutes of rest between each trial. Test-retest reliability for PF and PRFD was determined to be an intraclass correlation of 0.98 and 0.81, respectively. All PF values were scaled by using Sinclair coefficients for women's weightlifting (31).

Body Composition

Body composition was evaluated on a biweekly basis after the morning blood draw. Subjects' body mass was measured on a calibrated digital scale (Toledo Scales, Columbus, OH) to the nearest 0.1 kg, and their height was measured to the nearest 0.1 cm on a stadiometer. The same tester performed skinfold analyses for each of the testing sessions. Test-retest reliability has consistently produced high intraclass correlations (i.e., 0.99) for this procedure in the authors' laboratories. Seven skinfold sites were tested by using Lange skinfold calipers (Cambridge Scientific Industries, Cambridge, MA). The Siri equation was used to calculate percentage body fat for each subject (32).

Hormonal Measurements

Approximately 10 ml of blood was collected after an 8-hour fast to assess the resting concentrations of testosterone and cortisol on a biweekly basis. All blood samples were collected at the same time of day (6:45-7:45 am) by the same investigator. Blood samples were obtained by a venipuncture procedure from an antecubital arm vein. A trained phlebotomist using a vacutainer assembly, which consisted of a 21-gauge needle and a vacutainer, sampled all blood.

After the blood collection was completed, the vacutainer was centrifuged for 15 minutes in a refrigerated centrifuge (MSE Mistral 1000; Sanyo Scientific, Bensenville, IL) at 1,400g. The resulting serum was frozen at −80°C and later analyzed in duplicate for testosterone and cortisol concentrations. The serum testosterone and cortisol concentrations were analyzed by using a solid-phase radioimmunoassay technique (ICN Biomedicals, Inc., Carson, CA). The intra-assay variances were calculated to be less than 3% for both cortisol (intraclass correlation, 0.99) and testosterone (intraclass correlation, 0.98) assays. Testosterone:cortisol ratios were calculated to examine the anabolic-to-catabolic changes during the 11 weeks of training (19). The testosterone:cortisol ratios were evaluated with the relative criterion of more than a 30% decrease in comparison to a reference value obtained from a period of relative rest or a preceding value as being indicative of a decreased level of athletic preparedness (1, 21, 40). Serum hormone levels were not corrected for plasma volume shifts in this investigation because previous research performed on elite weightlifters has shown only small, nonstatistically significant alterations in resting plasma volume (<10%) in response to training while at the Olympic Training Center in Colorado Springs (25).

Statistical Analyses

All data were analyzed according to methods by Vervoorn et al. (39), in which 2-tailed Student's t-tests for paired comparisons were used to determine whether differences between data at a given time point and the time point immediately preceding the measure (tn−1−tn) and between the baseline measure and each additional measure (1-t or base−t) existed. Effect size was calculated for all comparisons. Statistical significance was set at P ≤ 0.05. Percentage change was also calculated [% Δ = 100 × (after - before)/before] for the hormone data, force-time curve data, and training variables. Pearson product moment correlations were performed to provide a descriptive view of the relationships between selected force-time curve, training, and hormonal variables. Statistical power was determined for the subject size in this investigation to be between 0.69 and 0.80 for the paired comparisons and 0.31 and 1.00 for the correlation analyses depending on the specific variable tested. All data are reported as mean ± SD and were analyzed with SPSS (version 13.0; SPSS, Inc., Chicago, IL).


Subject Characteristics

During the 11 weeks of the training cycle in this investigation, no significant alterations in body mass were found. No significant differences were found in the percentage body fat when comparing week 1 to all other weeks. However, a statistically significant difference (P = 0.047; η2 = 0.58) was found between body fat at week 3 (27.5% ± 6.7%) and week 5 (27.9% ± 6.9%). In terms of the changes in lean body mass during the 11 weeks of training, only week 11 (57.7 ± 5.5 kg) was statistically different (P = 0.045; η2 = 0.59) from week 1 (58.8 ± 6.3 kg). When lean body mass values from immediately preceding weeks were compared, only week 11 (57.7 ± 5.5 kg) was statistically different (P = 0.01; η2 = 0.76) from week 9 (59.1 ± 6.3 kg). Changes in subject characteristics are shown in Table 3.

Changes in body mass, lean body mass, and body composition during the eleven weeks of training (n = 6).

Testosterone:Cortisol Ratio, Testosterone, and Cortisol

The mean values for the testosterone:cortisol ratio are shown in Table 4. The testosterone:cortisol ratio at the baseline (1.19 ± 0.64) was statistically different from that at week 1 (0.67 ± 0.36) (P = 0.03; η2 = 0.63) and week 9 (0.94 ± 0.66) (P = 0.01; η2 = 0.75). When the immediately preceding values were compared, week 1 (0.67 ± 0.36) was statistically different (P = 0.004; η2 = 0.84) from week 3 (1.06 ± 0.54).

Anabolic-to-catabolic hormone alterations over the 11-week mesocycle of weightlifting training.

The data for the fluctuations in testosterone and cortisol concentrations during the 11-week training period are shown in Table 4. The mean values for the testosterone measurements did not show any significant differences, whether compared to the baseline measurement or the measurement directly preceding it. Conversely, the baseline cortisol measurement (797.35 ± 514.14 nmol·L−1) was statistically different from that at week 1 (11,184.07 ± 575.20 nmol·L−1) (P = 0.001; η2 = 0.75) and week 9 (1,250.75 ± 842.23 nmol·L−1) (P = 0.03; η2 = 0.65). When the immediately preceding values were compared, week 3 (834.6 ± 465.01 nmol·L−1) was determined to be statistically different (P = 0.007; η2 = 0.80) from week 1 (11,184.07 ± 575.20 nmol·L−1).

Testosterone:Cortisol Ratio Criteria

An evaluation of the occurrence or indication of decreased athlete preparedness as indicated by changes in the testosterone:cortisol ratio being greater than 30% when compared to baseline, as defined by Adlercreutz et al. (1), suggested that a decrease in athlete preparedness occurred in 13 of 36 trials. Considering the overall data, week 9 produced the most indications of suppressed athlete preparedness, occurring with 67% of the subjects experiencing more than a 30% decrease in testosterone:cortisol as compared to baseline. Only week 1 showed an overall trend of decreased athlete preparedness (−43%) with all the athletes tested when compared to baseline.

Performance Variables

The data for the isometric mid-thigh pull during the 11-week training period are shown in Tables 5-7. Week 9 was shown to be statistically different from week 7 for the absolute IPF (P = 0.025; η2 = 0.66), relative IPF (P = 0.006; η2 = 0.80), absolute PF during 30% IPF trial (P = 0.023; η2 = 0.68), relative PF during 30% IPF trial (P = 0.016; η2 = 0.72), absolute PF during the 100-kg trial (P = 0.035; η2 = 0.62), and relative PF during the 100-kg trial (P = 0.031; η2 = 0.64). The relative PF during the 30% IPF trial (P = 0.05; η2 = 0.57) and 100-kg trial (P = 0.027; η2 = 0.66) during week 7 was statistically different from week 5.

Isometric force-time curve characteristics during the 11-week mesocycle.
Dynamic mid-thigh clean pull peak force characteristics during the 11-week mesocycle.
Dynamic mid-thigh clean pull peak rate of force development characteristics during the 11-week mesocycle.

When examining the isometric PRFD (IPRFD) responses to the training stimulus during the 11 weeks of training, no statistical differences were noted between any of the weeks tested. Additionally, the PRFD during the 30% of IPF showed a similar response to that of the IPRFD, whereby no statistical differences were noted.

Conversely, when comparing the PRFD during the 100-kg trial, week 1 was statistically higher (P = 0.02; η2 = 0.72) than week 5 and week 11 (P = 0.02; η2 = 0.64). Week 3 was statistically higher (P = 0.02; η2 = 0.69) than week 5.

Hormonal Values, Training Volume, and Performance Tests

Overall, the percentage change in testosterone:cortisol ratio from weeks 1 to 11 (+71.3%) showed very large negative correlations with the percentage change in volume load (77.8%) (r = −0.83; r2 = 0.69) and percentage change in the training intensity (−14.7%) (r = −0.72; r2 = 0.51) during this period. When examining weeks 1 to 5, it was noted that the testosterone:cortisol ratio was increased by 72.5%, whereas the volume load decreased by 37%. Additionally, the increase in testosterone:cortisol ratio for this period corresponded to a 16.3% decrease in the overall training intensity. An 8.4% increase in the testosterone:cortisol ratio, which corresponded to a 57.5% decrease in volume load, was noted when examining weeks 5 and 11. A summary of the testosterone:cortisol ratio to volume load relationship is shown in Figure 6.

Figure 6:
Relationship between the testosterone:cortisol ratio and volume load.

An increase in the IPF (+1.0%) between weeks 1 and 5 corresponded to a concomitant increase in the testosterone:cortisol ratio (+72.5%) across the same period. Additionally, a 3.5% increase in the IPF corresponded to the 8.4% increase in the testosterone:cortisol ratio when examining weeks 5 and 11. Overall, the 71.3% increase in the testosterone:cortisol ratio noted when examining weeks 1 and 11 corresponded to a 4.2% increase in IPF.

The changes noted in volume load corresponded to changes in the IPF during the 11-week training cycle. These relationships are noted by the large negative correlations between the percentage change in IPF and the percentage change in volume load from weeks 1 to 5 (IPF, +1.0%; volume load, −37.0%) (r = −0.58; r2 = 0.34), 5 to 11 (IPF, +3.5%; volume load, −57.5%) (r = −0.50; r2 = 0.25), and 1 to 11 (IPF, +4.2%; volume load, −77.8%) (r = −0.66; r2 = 0.44).

Overall, the 77.8% decrease in volume load noted when comparing weeks 1 and 11 corresponded to a 2.2% increase in PF during the 30% trial. Additionally, there was a 1.2% increase in PF that corresponded with a 37.0% decrease in volume load when examining weeks 1 and 5. This trend continued with a 1.1% increase in PF that corresponded to a 57.5% decrease in volume load across weeks 5 and 11.

The percentage change in IPRFD (+2.7%) was found to be inversely related to the volume load of training (−77.8%) when examining weeks 1 and 11. A 5.1% increase in the IRFD, which corresponded to a 57.5% decrease in volume load, was noted when examining weeks 5 and 11. The 37% decrease in volume load and a 0.03% decrease in IPRFD were noted between weeks 1 and 5.


Häkkinen et al. (21) indicated that total testosterone levels are an important indicator of the individual trainability among physically active women who are undertaking a strength power-based resistance training program. Although no significant changes in mean testosterone levels were noted during a 16-week training mesocycle, the mean total testosterone levels during the mesocycle showed a strong correlation (r = 0.83; n = 6) to the athletes' force- and power-generating capabilities (21). Similarly, in the current study, the change in peak power during the 30% IPF trial during the 11-week training period and the mean total testosterone levels during the training period showed a strong correlation (r = 0.71; r2 = 0.50). A very strong correlation (r = 0.70; r2 = 0.49) was also noted between baseline testosterone levels and change in peak power during the 30% IPF trial during the 11-week mesocycle.

Additionally, Häkkinen et al. (21) suggested that women who possessed higher resting concentrations of total testosterone were able to increase their strength and power to a greater degree than those who possessed lower resting total testosterone concentrations. In the current study, however, this relationship was not observed. The women in the current study had lower resting serum testosterone levels than those reported by Stoessel et al. (33) and Häkkinen et al. (21). These findings are similar to data in the literature regarding elite male weightlifters (19), which may indicate that elite weightlifters of both genders experience similar hormonal fluctuations in response to strenuous weightlifting training. Additionally, this occurrence may partially be explained by the fact that there were very little interindividual differences in basal testosterone concentrations in the current population, as all the female weightlifters tested were considered to be elite and were of a similar training status (Table 4).

Resting cortisol concentrations have been reported to increase in response to stressful weightlifting training and to decrease when the training stimulus is reduced (16). Conversely, in physically active women undertaking strength power training, no changes in the serum cortisol levels were noted during a 16-week training period (21). In the current study, during week 1, the volume load was the highest, and this week corresponded to a significant increase (+48.4%) in the resting cortisol concentration when compared to baseline levels. Additionally, after completing 2 microcycles in which the training volume load was increased (i.e., weeks 7 and 8), a significant increase in the resting cortisol levels (+56.7%) was noted (week 9). After 2 weeks of reduced volume load, the resting serum cortisol levels were reduced to baseline levels (Table 4). These findings are very similar to the data reported by Häkkinen et al. (16) on elite male weightlifters; 2 weeks of reduced training volume load resulted in cortisol returning to baseline values. The data from the current study suggest that the overall training stress experienced is the primary mediating factor with regard to the cortisol response, regardless of the gender of the athlete tested. Furthermore, the alterations in cortisol appear to be the predominating factor altering the testosterone:cortisol ratio among these female weightlifters.

Although the free testosterone:cortisol ratio was not determined in the current study, research by Häkkinen et al. (21) indicates that the free testosterone and total testosterone concentrations respond in similar fashions to alterations in volume load and training intensity in women undertaking strength power training. A reduction of the free testosterone:cortisol ratio of more than 30% has been suggested as an indicator of the occurrence of overtraining (1). Additionally, the use of the testosterone:cortisol ratio has been suggested by some authors to be a valuable tool for the evaluation of athlete preparedness (1,5,29,40,41). Preparedness deals with the athletes' potential to perform well. Whether more than a 30% decrease in the total testosterone:cortisol ratio is indicative of overtraining is debatable, but the results of the current study indicate that a decrease in the testosterone:cortisol ratio of this magnitude can affect performance requiring high forces and power output.

Häkkinen et al. (16) reported that highly trained men who are undertaking periods of stressful weightlifting training consisting of a volume load of approximately 48,000 kg can express significant decreases in the testosterone:cortisol ratio (approximately −24%). These decreases in the testosterone:cortisol ratio are often linked with impairments in neuromuscular performance characteristics, which are indicative of a reduction in the athlete preparedness (19). Week 1 of the current training period consisted of the highest volume load with athletes undertaking 35,014.3 ± 9,025.5 kg. Overall, an average 44% reduction in the testosterone:cortisol ratio, when compared to baseline measures, was noted during this week of training with a high volume load. Week 9 represented the second training microcycle in which a significant reduction (−24%) in the testosterone:cortisol ratio was noted when compared to baseline, although the 19,670.9-kg volume load was significantly less than that of week 1. The 24% reduction in the testosterone:cortisol ratio was not enough to be classified as being indicative of a meaningful reduction in preparedness based on the criteria established by Adlercreutz et al. (1).

One explanation for the suppression in the testosterone:cortisol ratio during week 9 could be that the athletes had not yet recovered from the increased training volume encountered during weeks 7 and 8. The overall volume load was 32,367.9 kg during week 7 and 30,962.9 kg during week 8. The combination of these 2 weeks of high volume loads most likely resulted in an accumulated fatigue response that required a longer duration of unloading to stimulate an increase in the testosterone:cortisol ratio. Häkkinen et al. (16) reported that it can take 2 to 4 weeks of decreased volume load until the testosterone:cortisol ratio increases after encountering 2 weeks of increased training volume. This hypothesis is partially supported by the fact that the testosterone:cortisol ratio increased at the week 11 assessments, after 3 weeks of reduced training volume.

In the current study, 13 of 36 incidents showed more than a 30% decrease in the testosterone:cortisol ratio. In particular, week 9 had the most individual incidences (+67%) in which the testosterone:cortisol ratio decreased by more than 30%. Similarly, Vervoorn et al. (40) reported that there was an increased incidence (+62%) of testosterone:cortisol ratio reduction after a 2-week intensive training camp. The results of the current study may be expected as the increase in the frequency of the testosterone:cortisol ratio reduction in the current study occurred after 2 weeks of increased volume load.

In terms of the relationship between the testosterone:cortisol ratio and training volume, it has previously been reported that increases in volume load result in significant decreases in the testosterone:cortisol ratio (16). The results of the current study support this contention with a general trend that suggests that as volume load increases, the testosterone:cortisol ratio decreases (Figure 6). The observed percentage change in volume load (−77.8%) vs. the percentage change in the testosterone:cortisol ratio (+71.3%) from weeks 1 to 11 showed a very large correlation (r = −0.83; r2 = 0.69), which suggests that as the volume load decreases, an increase in the testosterone:cortisol ratio occurs. A similar inverse relationship was also depicted by the 37% decrease in the volume load, which corresponded to the 72.5% percent increase in the testosterone:cortisol ratio from weeks 1 to 5 and the 57.5% decrease in volume load and an 8.4% increase in the testosterone:cortisol ratio determined across weeks 5 to 11. These relationships may be of particular importance because it has been previously reported that the increased stress associated with microcycles in which volume load is significantly increased and the testosterone:cortisol ratio is suppressed stimulates a concomitant decrease in mode-specific strength in elite strength power athletes (10,19).

This contention is supported in this study by the relationships that were noted in the percentage changes among volume load, IPF, PF during the 30% IPF trial, and PF during the 100-kg trial. Additionally, this study indicates that the fluctuations in the testosterone:cortisol ratio that are associated with changes in volume load are also related to changes in the PF-generating capacity of female weightlifters. This can be noted by the 4.2% increase in IPF and the 71.3% increase in the testosterone:cortisol ratio when comparing weeks 1 and 11. Additionally, increases in IPF corresponded to increases in the testosterone:cortisol ratio by examining weeks 1 to 5 (IPF, +1.0%; testosterone:cortisol ratio, +72.5%) and weeks 5 to 11 (IPF, +3.5%; testosterone:cortisol, +8.4%). These results suggest that as the testosterone:cortisol ratio increases, so does the ability to generate IPFs. Häkkinen et al. (19) also reported that changes in the testosterone:cortisol ratio are strongly related to the ability to generate IPFs. The current study determined that when volume load was highest, the testosterone:cortisol ratio and ability to generate IPF were suppressed. Häkkinen et al. (16) reported that 2 weeks of reduced training after 2 weeks of increased volume load resulted in an increase in the testosterone:cortisol ratio and ability to generate maximal force. The current study supports this finding, in that the testosterone:cortisol ratio and IPF were suppressed during week 9, after 2 weeks of increased volume load and then rebounded at week 11.

In the available scientific literature, the relationship between changes in PF-generating capacity and volume load is understandable. First, there is a relationship between maximum strength and rate of force development (RFD). For example, Haff et al. (15) found strong correlations between IPF and IPRFD (r = 0.68; r2 = 0.46). Thus, one may speculate that the IRFD may have similar relationships to substantial changes in volume load. Based on previous observations, there can be considerable fatigue developed during high-volume strength training protocols (16,34). It is also known that acute and chronic fatigue can markedly reduce RFD (22,27) and that PRFD may be affected to a greater extent than PF production is (27). Furthermore, Stone et al. (38) observed that isometric and dynamic PRFDs, using mid-thigh pulls, were lower after a concentrated load (i.e., high-volume multiple sets of 10) compared to subsequent lower-volume training emphasizing strength and power among collegiate throwers. Therefore, one may assume that IRFD would also be affected by alterations in training variables. However, the authors know of no direct data that have been presented which directly consider the relationships between the PRFD and volume load.

In the current study, the idea that the PRFD is affected by volume load alterations is partially supported by the 2.7% increase in the IRFD that occurred when volume load was decreased by 77.8%. Additional support can be garnered by examining the results of the 30% IPF trial, in which a 13.8% increase in PRFD was noted when volume load was decreased by 57.5% across weeks 5 to 11. Similarly, for the 100-kg trial, it was determined that the PRFD was reduced to a lesser extent when comparing larger to smaller volume loads of training across the 11 weeks of training. Indeed, based on the current data, PRFD was affected more often and to a somewhat greater extent than were PF values. Based on these findings, it appears that volume load exerts an effect on the PF-generating capacity and PRFD of strength power athletes, thus giving insight into the overall preparedness of the athlete.

In summary, alterations in the training volume load appear to result in concomitant alterations in the testosterone:cortisol ratio, the ability to generate maximal forces, and the rate at which forces are generated. Of particular interest is that periods of decreased volume load appear to result in a rebound effect in which the testosterone:cortisol ratio increases and markers of performance also improve.

Practical Applications

The results of this study suggest that there are relationships among the volume load, the testosterone:cortisol ratio, and the ability to generate maximal isometric force in elite female weightlifters undertaking an 11 week training period. These results support the findings of several studies that have found a similar relationship in men (16,19). The combination of the current study of female weightlifters and the current body of scientific knowledge suggests that acute programming variables, particularly the volume load, can have a distinct impact on the ability to generate maximal force and PRFD. This ability may be linked to the alterations in the testosterone:cortisol ratio, which is affected by the training stimulus. Although it may be warranted to evaluate the hormonal status of female weightlifters throughout their training cycles, it is likely that this process is impractical in the typical training setting. A noninvasive assessment of maximal force-generating characteristics, such as the mid-thigh pull, may serve as a more convenient tool for establishing the preparedness of athletes.

Additionally, it appears that particular attention must be paid to appropriately designing and implementing a periodized training plan. Altering the volume of training can markedly alter the testosterone:cortisol and force-power capabilities. Obviously, minor alterations in volume would not necessarily affect these variables; however, it appears that alterations in volume load of more than 30%, especially when sustained for several weeks, can have a considerable impact on performance. Furthermore, this impact may be delayed. In particular, increases in volume load and decreases in the testosterone:cortisol ratio appear to result in a decrease in force-generating capacity and should be considered as an indicator of decreased preparedness in female weightlifters. It appears that a 2- to 3-week period of decreased training volume load results in an increase in the testosterone:cortisol ratio and markers of maximal force-generating capacity. Because the testosterone:cortisol ratio can be considered a measure of anabolic status and of performance preparedness, altering training variables to maximize the testosterone:cortisol ratio would be advantageous during certain periods of training and at competitions. Based on these findings, it appears that the strength and conditioning professional should pay particular attention to the individual fluctuations in volume load when designing and implementing any periodized model of training.


1. Adlercreutz, H, Harkonen, M, Kuoppasalmi, K, Naveri, H, Huhtaniemi, I, Tikkanen, H, Remes, K, Dessypris, A, and Karvonen, J. Effect of training on plasma anabolic and catabolic steroid hormones and their response during physical exercise. Int J Sports Med 7(Suppl): 27-28, 1986.
2. Alen, M, Pakarinen, A, and Häkkinen, K. Effects of prolonged training on serum thyrotropin and thyroid hormones in elite strength athletes. J Sports Sci 11: 493-497, 1993.
3. Bemben, MG, Clasey, JL, and Massey, BH. The effect of the rate of contraction on the force-time curve parameters of male and female subjects. Res Q 61: 96-99, 1990.
4. Bonen, A, Haynes, FJ, Watson-Wright, W, Sopper, MM, Pierce, GN, Low, MP, and Graham, TE. Effects of menstrual cycle on metabolic responses to exercise. J Appl Physiol 55: 1506-1513, 1983.
5. Booth, A, Shelley, G, Mazur, A, Tharp, G, and Kittok, R. Testosterone, and winning and losing in human competition. Horm Behav 23: 556-571, 1989.
6. Busso, T, Häkkinen, K, Pakarinen, A, Carasso, C, Lacour, JR, Komi, PV, and Kauhanen, H. A systems model of training responses and its relationship to hormonal responses in elite weight-lifters. Eur J Appl Physiol 61: 48-54, 1990.
7. Cardinale, M and Stone, MH. Is testosterone influencing explosive performance? J Strength Cond Res 20: 103-107, 2006.
8. Consitt, LA, Copeland, JL, and Tremblay, MS. Endogenous anabolic hormone responses to endurance versus resistance exercise and training in women. Sports Med 32: 1-22, 2002.
9. Elliott, KJ, Cable, NT, Reilly, T, and Diver, MJ. Effect of menstrual cycle phase on the concentration of bioavailable 17-beta oestradiol and testosterone and muscle strength. Clin Sci (Lond) 105: 663-669, 2003.
10. Fry, AC and Kraemer, WJ. Resistance exercise overtraining and overreaching. Neuroendocrine responses. Sports Med 23: 106-129, 1997.
11. Fry, AC, Kraemer, WJ, Stone, MH, Koziris, LP, Thrush, JT, and Fleck, SJ. Relationships between serum testosterone, cortisol, and weightlifting performance. J Strength Cond Res 14: 338-343, 2000.
12. Galliven, EA, Singh, A, Michelson, D, Bina, S, Gold, PW, and Deuster, PA. Hormonal and metabolic responses to exercise across time of day and menstrual cycle phase. J Appl Physiol 83: 1822-1831, 1997.
13. Garhammer, J. Power production by Olympic weightlifters. Med Sci Sports Exerc 12: 54-60, 1980.
14. Garhammer, J. A review of power output studies of Olympic and powerlifting: methodology, performance prediction, and evaluation tests. J Strength Cond Res 7: 76-78, 1993.
15. Haff, GG, Stone, MH, O'Bryant, HS, Harman, E, Dinan, CN, Johnson, R, and Han, KH. Force-time dependent characteristics of dynamic and isometric muscle actions. J Strength Cond Res 11: 269-272, 1997.
16. Häkkinen, K, Pakarinen, A, Alen, M, Kauhanen, H, and Komi, PV. Relationships between training volume, physical performance capacity, and serum hormone concentrations during prolonged training in elite weight lifters. Int J Sports Med 8(Suppl): 61-65, 1987.
17. Häkkinen, K, Pakarinen, A, Alen, M, Kauhanen, H, and Komi, PV. Daily hormonal and neuromuscular responses to intensive strength training in 1 week. Int J Sports Med 9: 422-428, 1988.
18. Häkkinen, K, Pakarinen, A, Alen, M, Kauhanen, H, and Komi, PV. Neuromuscular and hormonal adaptations in athletes to strength training in two years. J Appl Physiol 65: 2406-2412, 1988.
19. Häkkinen, K, Pakarinen, A, Alen, M, and Komi, PV. Serum hormones during prolonged training of neuromuscular performance. Eur J Appl Physiol 53: 287-293, 1985.
20. Häkkinen, K, Pakarinen, A, and Kallinen, M. Neuromuscular adaptations and serum hormones in women during short-term intensive strength training. Eur J Appl Physiol Occup Physiol 64: 106-111, 1992.
21. Häkkinen, K, Pakarinen, A, Kyrolainen, H, Cheng, S, Kim, DH, and Komi, PV. Neuromuscular adaptations and serum hormones in females during prolonged power training. Int J Sports Med 11: 91-98, 1990.
22. Kallenberg, LA, Schulte, E, Disselhorst-Klug, C, and Hermens, HJ. Myoelectric manifestations of fatigue at low contraction levels in subjects with and without chronic pain. J Electromyogr Kinesiol 17: 267-274, 2007.
23. Kanaley, JA, Boileau, RA, Bahr, JA, Misner, JE, and Nelson, RA. Substrate oxidation and GH responses to exercise are independent of menstrual phase and status. Med Sci Sports Exerc 24: 873-880, 1992.
24. Kanaley, JA, Boileau, RA, Bahr, JM, Misner, JE, and Nelson, RA. Cortisol levels during prolonged exercise: the influence of menstrual phase and menstrual status. Int J Sports Med 13: 332-336, 1992.
25. Kraemer, WJ, Fry, AC, Warren, BJ, Stone, MH, Fleck, SJ, Kearney, JT, Conroy, BP, Maresh, CM, Weseman, CA, Triplett, NT, and Gordon, SE. Acute hormonal responses in elite junior weightlifters. Int J Sports Med 13:103-109. 1992.
26. Kraemer, WJ, Staron, RS, Hagerman, FC, Hikida, RS, Fry, AC, Gordon, SE, Nindl, BC, Gothshalk, LA, Volek, JS, Marx, JO, Newton, RU, and Häkkinen, K. The effects of short-term resistance training on endocrine function in men and women. Eur J Appl Physiol 78: 69-76, 1998.
27. Lattier, G, Millet, GY, Martin, A, and Martin, V. Fatigue and recovery after high-intensity exercise part I: neuromuscular fatigue. Int J Sports Med 25: 450-456, 2004.
28. Marx, JO, Ratamess, NA, Nindl, BC, Gotshalk, LA, Volek, JS, Dohi, K, Bush, JA, Gomez, AL, Mazzetti, SA, Fleck, SJ, Häkkinen, K, Newton, RU, and Kraemer, WJ. Low-volume circuit versus high-volume periodized resistance training in women. Med Sci Sports Exerc 33: 635-643, 2001.
29. Mazur, A and Booth, A. Testosterone and dominance in men. Behav Brain Sci 21: 353-363, 1998.
30. Schmidtbleicher, D. Training for power events. In: Strength and Power in Sport. Komi, PV, ed. Oxford: Blackwell, 1992. pp. 381-385.
31. Sinclair, RG. Normalizing the performances of athletes in Olympic weightlifting. Can J Appl Sport Sci 10: 94-98, 1985.
32. Siri, WE. Body composition from fluid spaces and density: analysis of methods. In: Techniques for Measuring Body Composition. Brozek, J and Henschel, A, eds. Washington, DC: National Academy of Sciences, 1961. pp. 223-244.
33. Stoessel, L, Stone, MH, Keith, R, Marple, D, and Johnson, R. Selected physiological, psychological and performance characteristics of national-caliber United States women weightlifters. J Appl Sport Sci Res 5: 87-95, 1991.
34. Stone, MH and Fry, AC. Increased training volume in strength/power athletes. In: Overtraining in Sport. Kreider, RB, Fry, AC, and O'Toole, ML, eds. Champaign, IL: Human Kinetics, 1998. pp. 87-106.
35. Stone, MH, O'Bryant, HS, Schilling, BK, Johnson, RL, Pierce, KC, Haff, GG, Koch, AJ, and Stone, M. Periodization: effects of manipulating volume and intensity. Part 1. Strength Cond J 21: 56-62, 1999.
36. Stone, MH, O'Bryant, HS, Schilling, BK, Johnson, RL, Pierce, KC, Haff, GG, Koch, AJ, and Stone, M. Periodization: effects of manipulating volume and intensity. Part 2. Strength Cond J 21: 54-60, 1999.
37. Stone, MH, Plisk, S, and Collins, D. Training principles: evaluation of modes and methods of resistance-training-a coaching perspective. Sport Biomech 1: 79-104, 2002.
38. Stone, MH, Sanborn, K, O'Bryant, HS, Hartman, M, Stone, ME, Proulx, C, Ward, B, and Hruby, J. Maximum strength-power-performance relationships in collegiate throwers. J Strength Cond Res 17: 739-745, 2003.
39. Vervoorn, C, Quist, AM, Vermulst, LJ, Erich, WB, De Vries, WR, and Thijssen, JH. The behaviour of the plasma free testosterone/cortisol ratio during a season of elite rowing training. Int J Sports Med 12: 257-263, 1991.
40. Vervoorn, C, Vermulst, LJ, Boelens-Quist, AM, Koppeschaar, HP, Erich, WB, Thijssen, JH, and De Vries, WR. Seasonal changes in performance and free testosterone:cortisol ratio of elite female rowers. Eur J Appl Physiol Occup Physiol 64: 14-21, 1992.
41. Viru, A and Viru, M. Preconditioning of the performance in power events by endogenous testosterone: in memory of Professor Carmelo Bosco. J Strength Cond Res 19: 6-8, 2005.

rate of force development; peak power; peak isometric force; testosterone; cortisol

© 2008 National Strength and Conditioning Association