There is an increased emphasis on using subjective and objective outcome measures to characterize function in persons with shoulder disorders [1–4]. In particular, outcome assessments increasingly encompass terminology and concept from the International Classification of Functioning, Disability and Health [5]. For shoulder disorders, impairments most commonly measured include range of motion and strength. Disability includes activity limitations and participation restrictions that are often best measured by self-report questionnaires. Activity limitations can also be assessed by performance-based tests. It is important to understand the relationship between impairments and measure of function/disability in persons without shoulder problems, as well as the effects of age and gender on these different aspects of shoulder function.

The importance of objective measurements of strength and range of motion when evaluating shoulders is evidenced by the frequency with which they are reported in the literature [2, 4, 6–9]. Objective assessment of strength provides information on integrity of the rotator cuff [10] and is used to gauge the recovery of muscle function following intervention [11, 12]. Normative data is required for comparison to grade the extent of recovery. Otis et al. discussed the value of assessing strength relative to a matched population, especially in the presence of a shoulder deficit, in order to characterize these deficits [13]. Moreover, Hughes et al. noted the importance of having unaffected shoulders data for comparison in patients whose shoulder deficits manifest with bilateral involvement [14]. Kuhlman et al. acknowledged that while many studies have attempted to define normal values for the strength of shoulder muscles, there have been varying degrees of success because of a lack of standardization (plane of motion, position of the shoulder or stabilization of the body) [15].

Gender related differences in strength have been reported [14, 16]. More specifically, Hughes et al. have shown that men are stronger than women when controlling for age and weight [14]. The effects of age and dominance, however, are less well known. It has been suggested that, in the normal population, age is negatively associated with isometric shoulder strength and that some shoulder rotational strength measurements differ between dominant and nondominant sides [14, 16]. Studies examining range of motion and its relationships with age, gender and dominance, unfortunately, have reported varied results [17–19]. Most studies have reported that only some shoulder motions decrease with age [17, 18], however, the specific shoulder ranges of motion affected by age are inconsistent between studies. As for gender-related effects, minimal differences between genders have been described by Murray et al. [18], while Barnes et al. observed greater range of motion in women as compared to men [17].

Measures of function provide a broader view of patient status and are considered more patient centered. Several studies have examined the contribution of self-report questionnaire to disability assessment, or reported on their validity [20, 21]. Unfortunately, very few studies have established normative values for these self-report scales [21]. Furthermore, those that have presented normative values have not included a determination of how self-reported function is related to strength and range of motion [7, 13–16, 21, 22].

Recently, an upper-extremity functional performance test, the FIT-HaNSA, has been proposed by MacDermid et al [23]. This standardized test was developed to assess functional performance for sustained upper extremity activity. The reliability and concurrent validity of the FIT-HaNSA have been established in persons with shoulder disorders. It has been shown to discriminate between persons with and without shoulder disorders. Normative data for the FIT-HaNSA has yet to be published. Furthermore, the relation between the FIT-HaNSA and strength, range of motion and self-reported function has not been established for individuals without shoulder pathology.

The purpose of this investigation is to enhance existing comparative data on normal shoulder range of motion and strength, and to add information on functional performance and self-reported function. A secondary purpose is to establish the inter-relationships between these different types of assessments and to evaluate the impact of dominance, age and gender. The hypotheses are that men will be stronger than women in all age categories. Furthermore, rotational strength, range of motion, functional performance and upper extremity function will be decreased in the over-60 age category compared to the 18–39 and 40–59 age categories.

This study provides comparative data for strength, range of motion, functional performance and self-reported function in persons who state they have no shoulder pain or disorders. The data is sparse for the over-60 group since it was difficult to find people in this age group who could fit the criteria. This reflects the high prevalence of shoulder problems in advancing age groups [33–35]. Random sampling to provide normative data from the population would be expected to include a high rate of people with shoulder pathology [35], therefore, the difference between unaffected and normative data should widen as the age of interest increases.

Results for shoulder rotation strength were mostly consistent with the findings of several previous papers that tested isometric strength, showing that on average, men are stronger than women [11, 13–15]. Some differences, however, were observed. In men, strength significantly decreased in the over-60 age category. This differs slightly from the findings of Hughes et al. who suggest that strength declines linearly with age [14]. One possible cause of the decline in strength with aging is a decreased muscle mass, which may affect men (who have typically more muscle mass) more than women. Our data shows that there is a drop-off in strength around 60 years of age. Since retirement age is generally between 60 and 65, this might suggest that activity level is a contributing cause. Women appear to be strongest in their childbearing years (ages 18–39), which may be related to the type and quantity of activities required at this portion of their lifespan. Kim et al. [35] also found less age related decline in women as compared to men when evaluating four subgroups aged 40 years and older. It is clear that there is wide variation in strength in the younger categories, and that this variation decreases considerably in the eldest category; for the 18–39 category, the standard deviation for strength measurements ranges from 11.0 to 17.9, while in the over-60 category, the range is only from 9.5 to 12.4. This reduced variability may reflect a reduced contribution of occupation related inter-individual differences in the over-60 group.

Kim et al. [35] noted that 20% of subjects aged between 60–69 years, and 41% of subjects older than 70 years, claiming to be asymptomatic had a rotator cuff tear on ultrasonography. Although the numbers in the study subgroupps were small, the authors reported no significant loss of external rotation or abduction strength in partial or full-thickness rotator cuff tear; whereas a loss of abduction strength was observed with a large-to-massive full thickness tear. We did not perform imaging to define cuff status but recruited individuals reporting no shoulder pain or disability.

Muscle bulk may contribute to differences in range of motion between men and women. There were moderate negative correlations (-0.28 to -0.32) between strength and sitting range of motion. Due to testing position, muscle bulk may limit flexibility in stronger individuals. This interference would be eliminated in the supine testing due to the abduction of the humerus to 90°. Overall, there was a reduction in external rotation range of motion in the over-60 age category, while there was a negative correlation between age and external rotation range of motion. Our results are consistent with other studies that have shown that some shoulder range of motion decreases with aging [17, 18]. However, in the study performed by Barnes et al., passive and active internal rotation at 90° of abduction was not affected by aging [17].

It has been suggested that isometric strength testing may provide more useful data in older patients with rotator cuff repairs than isokinetic tests [36]. Therefore, we elected to only perform isometric testing. Only few studies have reported normative data for external and internal rotational strength in the plane of the scapula [14, 15]. This position has been described by Greenfield et al. as being clinically significant because it optimizes the length-tension relationship of the shoulder abductors and rotators and creates a significant degree of conformity between the humeral head and the glenoid [37]. Position can affect strength scores obtained so data comparisons should be made across similar test procedures.

The findings on the relationships between self-report scales in this study are counter-intuitive and incongruent with the results obtained in patient samples. One would expect there to be strong correlations between the self-report instruments because they all measure similar constructs and because strong correlations have been reported across different shoulder pathologies (0.88 <r < 0.91) [20]. Weak correlations, however, were observed (0.01 <r < 0.19). This may be due to attribute and scale factors. Firstly, the self-report scales were not designed to detect differences across a range of "normal" ability. Since a small range of the spectrum (minimal disability) was sampled in asymptomatic subjects, correlations obtained would not accurately portray the relationship across the full spectrum. Similarly, whereas a moderate relationship has been observed between self-reported function and measured of physical impairments (strength or range of motion) in studies assessing pathological groups, we found a weak relationship (0.18 < r < 0.56) [20]. This suggests that there is a wide variation in strength and range of motion that allows normal people to perform their normal tasks of daily life.

Unlike the DASH or WORC, a significant correlation between age and the SST was observed (r = -0.202). The SST contains a wider range of items, including more difficult items like questions on activities requiring muscle strength and power, such as lifting 3.6 kg (8 pounds), carrying 9.1 kg (20 pounds) or tossing a softball over-hand 18.3 m (20 yards). It also requires a yes/no response which may be more discriminative. The DASH asks about some higher level activities in a more indirect way saying "recreational activities which you take some force or impact through your arm, shoulder or hand". Perceptions or expectations around what this means to an individual may change overtime (response shift). Furthermore, most items on the DASH are less demanding activities of daily life tasks and the response is a 5-point scale so that changes in capability may be measured differently than on the SST. Similarly, the WORC also contains few specific difficult tasks and it measures on a 10-cm VAS scale. Thus, the weak relationships between the SST and the DASH or WORC may reflect their different approaches to measuring shoulder function.

The FIT-HaNSA is an alternative means to assess functional status that evaluates task performed in a timed test [23]. The FIT-HaNSA requires that subjects possess enough shoulder muscle endurance and joint range of motion to perform three different types of task. As with self-reported function, no significant correlations were observed between strength/range of motion and functional performance. This may be related to the spectrum issue. However, the FIT-HaNSA contains tasks more related to muscle endurance than to muscle strength. Therefore, future studies should look at the relationships between muscle endurance and the FIT-HaNSA. It is not surprising that no correlation with rotational range of motion and FIT-HaNSA were established as the test requires most of the movement in the frontal plane. The ranges of motion observed in the subjects recruited (the minimum was 32° in internal and 38° in external rotation) were more than enough to enable the subjects to perform the three tasks of the FIT-HaNSA. Unfortunately, shoulder elevation (flexion, scaption or abduction) ranges of motion were not evaluated in this study, but might be expected to have a bigger impact on the FIT-HaNSA scores.

The mean scores on the DASH index in this study fall below the normative values quoted in the study published by Hunsaker et al. [21]. For example, DASH normal scores were 10.1 ± 14.7 in the study by Hunsaker et al. [21], and 1.8 ± 2.9 in the present study. This suggests that we were successful in obtaining subjects with no upper extremity pathology for our study, and confirms our assumption that "normative" data is influenced by the high rates of shoulder pathology at older ages. This is important when considering "recovery" for older persons who previously were unaffected by shoulder pathology. While their scores may end up close to population norms, this might still represent a substantial loss from their preinjury level. We were easily able to recruit younger age cohorts, and there were large numbers of older people who were willing to be tested but did not qualify for the study criteria. In fact, the smaller size of this subgroup is a limitation in this study and suggests that future research on pain free shoulder in the elderly will be a challenge. Thus, we called our data unaffected comparison data, not normative data, since it reflects a non-pathologic state that is not "normal" in older age groups. The small number of older subjects could have also been detrimental in detecting differences across age categories.

This paper yields several important facts about the relationship among rotational strength, range of motion, functional performance and self-reported function in people with unaffected shoulders and can serve as a valuable clinical resource for comparison with a patient population. First, there is indeed a decline in strength with aging, especially for the men. This decline, however, does not appear linear, but is more notable over the age of 60. There is more variation in strength within younger age subgroups than within the older subgroups. Second, there is a significant negative correlation between strength and sitting range of motion. In the sitting position, women are more flexible than men in external rotation for the 40–59 age categories. Third, there is no significant relationship between self-reported function and strength/range of motion over a narrow range of normal function in asymptomatic individuals. Finally, measured functional performances also not correlated with strength/range of motion in asymptomatic individuals; however, it is correlated with self-reported function (DASH). The SST may be more discriminative to age related changes in shoulder function than the DASH or WORC. Further research is needed to expand upon the interactions delineated in this paper, and to further understand the relationship between different impairment and disability outcome measures.

This study emphasizes that both young and older individuals who present as asymptomatic experience very low levels of disability; but the older individuals scores are much better than population "norms". Selection of comparative data should consider that scores from patients with similar procedures/injuries, or individuals without pathology are different, both conceptually and quantitatively, from a "normal" score which reflects the overall rate of functional disability existing in a population or subgroup.