Normative reference values of the handgrip strength for the Portuguese workers.

Methodology
Sample:A cross-sectional study was conducted in the final assembly area of an automobile industry. All workers were invited to participate in the study (n=1225).
For sample size calculation participants were stratified by gender and age. The women's workers were divided into five age groups and the men's workers in six age groups. Height measures were restricted to 146-177 cm for women and 150-190 cm for men.
In order to ensure normative values for healthy workers, the following exclusion criteria were applied: a) minimum value of 10 kg of gripping force; b) present medical restriction and / or occupational disease; c) SF-12 criteria (Short Form Health Survey) with a score lower than 5% of the physical component score of the quality of life scale (24).
The scientific committee of the Faculty of Human Kinetics from the University of Lisbon approved the study protocol. All workers were informed about the purpose and procedures of the study and gave their written informed consent.

Methods:
SF-12 (Short Form Health Survey)
The SF-12 is composed by 12 items organized according to a Likert scale and includes physical components score (PCS) and mental components scores (MCS). The physical dimension comprises items related to physical function, physical performance, pain, and health in general, and the mental dimension covers mental health, emotional performance, social function, and vitality (24,25).
Anthropometry
For the height measurement was obtained with an upright position, participant-centred position on the tape, in relation to the stadiometer, footwear, arms extended along the body, feet joined or slightly separated. The head was oriented according to the Frankfurt plane, parallel to the ground, regardless of the worker’s posture (26). Due to the European safety norms industry setting, to height measure were removed 3 centimetres to compensate the height of the work shoes sole.
Handgrip Strength
For the manual grip strength test, the standard position was used for all participants (24,27,28). The grip strength was measured in kilograms (kg) using the Jamar digital dynamometer (29). Two measurements were performed in both hands. Position 2 of the Jamar dynamometer was used, because is the most appropriate position to measure the handgrip strength (27,29). The maximum value obtained with either hand is used as a summary measure of a person’s isometric strength of the hand and forearm muscles (30,31). The Jamar dynamometer is extremely used, which is validated as a gold standard, with high test-retest reproducibility (r> 0.8) and excellent reliability (r = 0.98) (16).

Data analyses:
In order to perform the final sample, the eligibility criterion will be applied, based on the SF-12 PCS score, being standardized from the z-standardized with an average value of 36 and SD 2.35, defining the criterion for 5% below the PCS average.
Descriptive statistical analysis was used to determine mean (M), standard deviation (SD) and median values (ME) of the handgrip strength, stratified by five age groups of the women and the six age groups of the men. Sex-specific profiles of handgrip strength were designed by Ordinary Least Square regression (OLS) analysis, where height, age, age squared and height squared entered in the models as determinant factors of the maximum grip strength in the women and men groups. The objective of this step was to verify the mean peak values for women and men.
After, the cut-off values were calculated in order to identify workers with weak grip strength. These values were defined as 1SD and 2SD below the mean peak value and stratified by gender. The resulting values were plotted from the age groups defined by height.
Finally, handgrip strength measurements were first standardized for age and body height. It was determined the risk threshold, subtracting 1 SD from the mean value of height-adjusted handgrip strength and age group. The standardized measures of handgrip strength values were the z-standardized residuals derived from sex-specific OLS regressions of handgrip strength values in kg on age (in years) and body height (in cm).