Maximum Force Capacity of Back Extensor Muscles in Healthy Women and Men: Not Different, If Anthropometrically Normalized

1. Introduction

For a long time, diagnostic data, therapeutic recommendations, and, for example, reference values for laboratory parameters were applied uniformly to all patients, regardless of sex. The recognition that such generalizations are not appropriate first emerged in the context of laboratory values—sex-specific reference ranges, particularly for hematological parameters, have been established for approximately 100 years. However, other sex-specific laboratory parameters have only become established over the past 30–40 years [1]. In clinical diagnostics, sex-specific symptoms are now also taken into account; a prominent example is the differing presentation of myocardial infarction symptoms between men and women [1].

Today, sex-specific approaches in diagnosis and therapy have become the rule rather than the exception in medicine.

Sex-specific differences are particularly relevant for parameters related to anthropometric characteristics [2] as well as physical performance [3]. Interestingly, according to the literature on this topic, no systematic sex differences can be identified in the fiber type composition of skeletal muscle [4,5]. Instead, individual factors—primarily genetic and psychosocial—play a decisive role.

However, within this generally comparable distribution of muscle fiber types, sex differences do exist: men typically exhibit a larger muscle fiber cross-sectional area than women [5]. This is particularly evident in type II fibers, whose diameters are significantly greater than those of type I fibers in men, whereas such differences are not observed in women [4].

A similar pattern is observed in anthropometric data of healthy individuals: women are, on average, shorter and lighter than men [2]; nevertheless—or perhaps precisely for this reason—no systematic sex differences in body mass index (BMI) can be identified [6]. However, sex-specific reference values are increasingly being recommended for this parameter as well, since body composition differs between women and men [7] and is additionally influenced by age [7].

The function of the back muscles, in conjunction with the abdominal musculature, is to provide an adequate balance between stability and mobility of the spine in order to prevent injury. In this context, three functional domains can be identified: maximal strength capacity, endurance capacity, and appropriate coordination with respect to the timing of muscular activation [8,9], as well as the aspect of intermuscular coordination.

Particularly with regard to maximal strength capacity, women appear to be at a disadvantage, as their values are estimated to be approximately 30% lower than those of men [10].

Therefore, the aim of the present study was to investigate the extent to which such differences can be observed when strength values are normalized for anthropometric characteristics. The primary outcome parameter was the maximal strength capacity of the trunk extensors normalized to upper body weight.

2. Methods

The present study includes data from a total of seven prior investigations in which the same parameters were assessed. All studies were conducted on healthy voluntary participants of both sexes and received approval from the respective ethics committee (4348-02/15, 2021-2373-BO, 2021-2373_1-BO, 2024-3462-BO, 2024-3634-BO-A). Participation was based on written informed consent provided by all subjects.

In total, data from 115 women and 124 men aged 18–52 years were included in the present analysis. All participants were healthy, reported no acute back pain or back pain within the previous three months, and were either physically inactive or only moderately active. Individuals engaging in regular physical exercise more than twice per week were excluded, as were individuals with a BMI greater than 32 kg/m². The complete anthropometric characteristics of the cohort are presented in Table 1.

Investigation

All measurements were conducted using a computer-assisted testing and training device (CTT Centaur, BfmC, Germany). In the device participants stood in an upright position, with the lower body fixed while the upper body remained mobile. The system is equipped with a shoulder harness that is lowered onto the participants’ shoulders. Integrated force sensors within this harness are capable of measuring forces in both the sagittal and frontal planes.

Maximal trunk extension force was assessed following a standardized warm-up protocol with submaximal loads. The MVC measurement itself was performed three consecutive times, each lasting five seconds, with rest intervals of five seconds between trials. Participants were instructed to reach maximal force within approximately one second, maintain this level for two to three seconds, and then relax. Verbal encouragement was provided during each trial [11,12,13].

To familiarize participants with the procedure, the first trial was conducted as a practice run at approximately 50% of maximal effort. During the measurements, participants crossed their arms in front of their chest (Figure 1).

Additionally, upper body weight was measured by tilting the CTT Centaur forward by 90° and having participants rest relaxed in the shoulder harness, with arms and head hanging freely. In this relaxed posture, upper body weight was recorded. The procedure was repeated three times, and the highest plausible value was used as the valid upper body weight (UBW) for further analysis. According to the procedure described, the UBW measured here comprises the torso, arms, and head.

The measured force values were converted into torques to account for differences in body size and to provide directly comparable values for analysis.

Analyzed Parameters

The highest force value of the three extension trials was taken as the MVC (maximum voluntary contraction) value. This value was analyzed both as force and as torque. The measured maximal force values were then related to the upper body weight, yielding the upper body torque ratio (UBTR).

Statistics

Sex comparisons were conducted using the independent-samples t-test. In addition, effect sizes (ES) were always calculated [14].

As the aim was to provide an overview of strength performance values based on data from multiple studies, both sex groups featured a large sample size. A post hoc power analysis indicated that for an ES of 0.5, a power of 0.8, and an assumed significance threshold of 0.05, a sample size of 64 per group would have been sufficient. Our group sizes were approximately twice as large.

To avoid Type II errors due to the large sample sizes, the minimal important difference (MID) was additionally calculated as a mathematically defined measure (SD/2) to identify differences considered relevant relative to the difference in means. A detected difference between compared values was only considered relevant if the mean difference exceeded the MID. Furthermore, regarding the ES differences were only considered relevant if the ES was at least 0.5. Accordingly, only results meeting all three criteria: p < 0.05, ES > 0.5, and mean difference > MID were reported as significant.

3. Results

In the sex comparison, men exhibited significantly and relevantly higher values for both UBW and MVC (Figure 2). No significant differences were detected for the UBTR (Figure 3). Detailed results of the statistical analyses are presented in Table 2.

4. Discussion

With respect to the MVC data, the results are not surprising, as men are generally expected to exhibit considerably higher maximal strength than women [15]. This is reflected in the maximal strength data, which in the present analysis were approximately 45% higher in men than in women. Corresponding findings have been consistently reported in the literature [10].

What is rarely measured, however, is the upper body weight of the study population. There is virtually no standardized solution available that is cost-effective, practical, and reliable [16,17]. By implementing specific modifications to the device used, it was possible to capture force measurements in the tilting direction, which in combination with the lower-body-restricted fixation of participants allowed us to determine upper body weight for all subjects. In general, reliable measurements can be expected from the device employed [18].

Using this approach, we were able to determine individual upper body weight directly, not by modelling. To our surprise, relevant sex differences were also observed, with men's upper body weighing approximately 35% more than women. As we found no normative data in the literature, these measurements already represent novel findings.

The findings for the UBTR were somewhat surprising: although the p-value indicated significance, this was ultimately deemed not relevant when considering the effect size (0.421 – low [14]) and the MID (larger than the mean difference). Thus, both sexes exhibit a comparable ratio of maximal trunk extension force to upper body weight. The existing differences in maximal force are compensated by the considerably lighter upper body in women. Consequently, the physiological reserve - that is, the remaining strength surplus - remains at a comparable level for both sexes. Similar observations have been reported in the literature, where values normalized to body weight also eliminated the differences in MVC [19].

The presented data also provide reference values for the UBTR, which equals approximately to 2.0 for women and 2.1 for men. This indicates that a physiological strength reserve of roughly 100% of upper body weight can be assumed in both sexes. The calculated MID of approximately 0.2 further allows to identify deviations from the norm, enabling evidence-based recommendations for intervention if values fall below the UBTR’s margin of variability.

Limitations

The investigations were conducted in a specialized testing environment. The used device (CTT Centaur) is not part of the standard equipment in diagnostic or therapeutic settings. Therefore, the results should be interpreted with caution, as they may be device-specific. On the other hand, all 258 participants were tested using the same device, which supports the robustness of the observed sex differences. The provided approach to assess the results of the statistical analyses may be subject to debate. Further studies are warranted to verify the drawn conclusions.