Grip strength reference values for Canadians aged 6 to 79: Canadian Health Measures Survey, 2007 to 2013
by Suzy L. Wong
Grip strength is a simple, fast and reliable measure of the maximum voluntary force of the hand.Note 1Note 2 It is used to assess hand injuriesNote 2 and is an indicator of overall muscular strength, nutritional status, muscle mass and walking performance.Note 1 Grip strength is also a marker of hypertension and type 2 diabetesNote 3 and a predictor of all-cause, cardiovascular and non-cardiovascular mortality, heart attack, stroke,Note 4disability and surgical complications.Note 5
Based on normative data, the grip strength of an individual can be assessed relative to a reference population. Grip strength norms, or reference values, have been established to evaluate hand injuries, set treatment goals, evaluate surgical outcomes, and assess a patient’s ability to return to employment.Note 6 Such norms are presented separately for each hand―left and right, or dominant and non-dominant. More recently, as interest in grip strength has extended beyond the evaluation of hand function to overall muscular strength, nutritional status and disability,Note 1Note 5 norms have been presented as the maximum grip strength measured from either hand.Note 1
Grip strength norms are stratified by age and sex; height and weight may also be taken into account to provide more refined norms.Note 1 Generally, the reference population is healthy, although norms have been published for specific populations, such as older adults with chronic conditions.Note 7
Numerous studies have provided grip strength norms, but some studies were based on samples that were small and/or not nationally representative.Note 1 In addition, most studies focussed on adults, particularly older adults; relatively few examined childhood, adolescence and early adulthood.Note 1
National variations in grip strength suggest that norms developed for one country are not applicable to other countries.Note 2 In Canada, the grip strength of people aged 15 to 69 has been interpreted using the Canadian Physical Activity, Fitness and Lifestyle Approach (CPAFLA) musculoskeletal fitness norms, which categorize total grip strength values into five Health Benefit Ratings ranging from “Excellent” to “Needs Improvement.”Note 8 However, the methods and reference population used to derive the CPAFLA norms are not documented. Further, the use of total grip strength (sum of right- and left-hand grip strength) is inconsistent with other norms and does not allow comparisons with previous studies.
The purpose of this analysis was to use data from the 2007-to-2013 Canadian Health Measures Survey to develop reference equations for maximum, right-hand and left-hand grip strength for Canadians aged 6 to79, based on a healthy, nationally representative population. These equations can be used to determine reference values against which to assess an individual’s grip strength.
Data and methods
Canadian Health Measures Survey
Data are from the first three cycles of the Canadian Health Measures Survey (CHMS), an ongoing survey conducted by Statistics Canada in partnership with the Public Health Agency of Canada and Health Canada. The CHMS provides comprehensive direct health measures at the national level for the household population. It involves an in-person household interview and a subsequent visit to a mobile examination centre (MEC). The household interview gathers general demographic and socioeconomic data and detailed health, nutrition and lifestyle information. At the MEC, direct physical measurements are taken, including grip strength, height and weight. Ethics approval for the CHMS was obtained from Health Canada’s Research Ethics Board.
Cycle 1 collected information from respondents aged 6 to 79 from March 2007 through February 2009. Cycles 2 and 3 collected information from respondents aged 3 to 79 from August 2009 through November 2011, and from January 2012 through December 2013, respectively. For the three cycles combined, 16,606 respondents completed the MEC component. After adjustments for the sampling strategy, the final response rate for 6- to 79-year-olds was 52.9%. Details about the survey are available at www.statcan.gc.ca/chms.
Grip strength measurement
Grip strength was measured using handgrip dynamometry. Respondents were not eligible for testing if they were younger than 6, or had an acute or chronic condition that would likely make grip strength measurement unsafe or the results unreliable or unrepresentative of their usual grip strength.
Grip strength was measured to the nearest kilogram (kg) twice on each hand (alternating) using a Smedley III handgrip dynamometer (Takei Scientific Instruments, Japan). The testing procedure was based on the Canadian Physical Activity Fitness and Lifestyle Approach (CPAFLA), Third Edition.Note 8 Respondents stood with feet slightly apart and the dynamometer held in line with the forearm away from the body at the level of the thigh. They were asked to squeeze the dynamometer as hard as possible, exhaling while squeezing. The highest values attained by each hand were considered the right-hand and left-hand grip strength scores. Maximum grip strength was the highest value attained from either hand, which is less likely than the mean to be affected by the number of trials.Note 9
Age was self-reported age at the MEC visit. Height was measured to the nearest 0.01 cm using a ProScale M150 digital stadiometer (Accurate Technology Inc., Fletcher, USA). Weight was measured to the nearest 0.01 kg using a Mettler Toldedo VLC with a Panther Plus terminal scale (Mettler Toledo Canada, Mississauga, Canada). Body mass index (BMI) was calculated as kg/m2.
Initially, 16,572 respondents were eligible for grip strength testing. Respondents were excluded from the statistical analysis if they were 80 years old at the time of the MEC visit (n = 2); their grip strength results were outliers based on visual inspection of scatterplots (n = 8); or test results were not obtained for both hands (n = 135). According to previous studiesNote 3Note 7Note 10 and preliminary analyses, people with some chronic conditions have significantly lower grip strength compared with their healthy peers. Therefore, respondents were excluded if they had asthma, fibromyalgia, arthritis, high blood pressure (or took medication for high blood pressure in the past month), chronic bronchitis, emphysema, chronic obstructive pulmonary disease, type 2 diabetes or heart disease; ever had a heart attack; had cancer or ever had cancer; suffered from the effects of a stroke; usually walked with difficulty and mechanical support; or were unable to grasp and handle small objects (n = 5,319). This left a final sample of 11,108 (5,438 male; 5,670 female).
All analyses were based on weighted data using the CHMS sample weights for combining cycles 1, 2 and 3.Note 11 Descriptive statistics were calculated with SAS version 9.3 and SUDAAN version 11. Standard errors, coefficients of variation and 95% confidence intervals were calculated with the bootstrap technique.Note 12Note 13The number of degrees of freedom was specified as 35 to account for the CHMS sample design.Note 11
Quantile regression was used to derive reference equations for the 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles of maximum, right-hand and left-hand grip strength for males and females aged 6 to 79. Quantile regression enables estimation of the conditional median and other percentiles, rather than the conditional mean,Note 14 and produces similar estimates to the LMS method when used to develop growth charts.Note 15 An advantage of quantile regression is that it does not rely on distributional assumptions, such as normality, and is, therefore, robust to outliers and skewness.Note 15 Polynomial regression models using different combinations of integer powers of age, height and weight were evaluated using Wald tests, the sum of squared residuals, graphical representation of fitted values, and comparison of fitted and observed percentile values.
Based on the results of these preliminary analyses, grip strength was modeled as a function of age, age2, height, height2 and weight for males, and as a function of age, age2, height2 and weight for females. Quantile regression was conducted using the QUANTREG procedure in SAS version 9.3 with the method of confidence interval estimation specified as resampling with 500 repeats.
To examine model fit, graphs were created using three sets of values of the 50th percentile of maximum grip strength at each age. One set of values employed an extension to the World Health Organization (WHO) Growth Charts Adapted for CanadaNote 16 to determine median height and weight by age and sex for 6- to 19-year-olds; for 20- to 79-year-olds, median height and weight were assumed to stay constant. These height and weight values were used in the reference equations to obtain values of grip strength (WHO-fitted values). Another set of values was obtained by calculating the median height and weight, by age and sex, of the study population and using them in the reference equations to obtain grip strength values (CHMS-fitted values). The last set of values was obtained by calculating the median maximum grip strength of the study population (observed values). The three sets of values were then plotted by age.
To compare maximum, right-hand and left-hand grip strength, values for the 50th percentile were fitted using the WHO growth chartsNote 16 values for median height and weight by age and sex for 6- to 19-year-olds; median height and weight were assumed to stay constant for people aged 20 to 79. Values for maximum, right-hand and left-hand grip strength were plotted by age.
If precision is not required and ease of use is important, a table of reference values may be preferred over calculating person-specific values based on reference equations. Therefore, 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentile reference values for maximum grip strength were calculated using the WHO growth chartsNote 16 values for median height and weight by age and sex for 6- to 19-year-olds, and median height and weight were assumed to stay constant for 20- to 79-year-olds. Reference values were calculated for each year for those aged 6 to 19, and for five-year age groups for those aged 20 to 79.
To compare reference values from this study with those from other countries, values for the 50th percentile for maximum grip strength were fitted using the WHO growth chartsNote 16 as described above. Although reference values for other countries were available from a large number of studies, only four were selected for comparison,Note 17Note 18Note 19Note 20 based on sample size, a wide age range, representativeness, measurement of grip strength in kilograms or pounds, and date of publication. Multiple studies were not selected from the same country. For studies presenting reference values by right and left hand, the right-hand values were selected for comparison. For studies presenting reference values by age group, the age in the middle of the range was selected as the data point. Values were then plotted by age.
The mean age of the study population was 35.1 years (95% CI: 34.6 to 35.6) for males and 36.0 years (95% CI: 35.5 to 36.6) for females. Mean BMI was 25.7 (95% CI: 25.4 to 26.0) for males and 26.0 (95% CI: 25.5 to 26.5) for females. Males’ maximum grip strength significantly exceeded that of females: 42.8 versus 26.2 kg (Table 1).
Coefficients for the reference equations for the 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles of maximum, right-hand and left-hand grip strength are shown in Table 2. Reference values based on these equations are presented by age in Table 3. A comparison of reference values based on fitted values using the WHO growth charts,Note 16 and the CHMS study population with observed median values of maximum grip strength is shown in Figure 1.
Grip strength increased from childhood through adolescence, peaked in the 35-to-45 age range for men and in the 30-to-50 age range for women, and then decreased. Differences in grip strength between males and females were smaller during childhood than in adolescence and adulthood.
Fitted values for the 50th percentile were compared for maximum, right-hand and left-hand grip strength (Figure 2). Among women, the right hand tended to be stronger; among men, this tendency was less pronounced. Among children, right- and left-hand grip strength were similar.
WHO-fitted values for the 50th percentile of maximum grip strength were compared with the results for four other countriesNote 17Note 18Note 19Note 20 (Figure 3). Norms for males in the United StatesNote 7 and Great BritainNote 18 increased from childhood to peak around age 30, and then decreased. The CHMS norms increased more steeply through adolescence, but thereafter, rose less sharply and peaked around age 40. The CHMS norms were among the lowest for 20- to 30-year-olds, but among the highest for people aged 40 or older. Trends for females were generally similar, except the CHMS norms did not increase more sharply through adolescence than did norms from other studies.sNote 17Note 18Note 19Note 20
With data from the 2007-to-2013 CHMS, grip strength reference equations were developed for Canadians aged 6 to 79. These equations can be used to compare an individual’s measured grip strength with the predicted 5th, 10th, 25th, 50th, 75th, 90th and 95th percentiles of grip strength for healthy individuals of the same age, sex, height and weight. Unlike previous reference values, these equations range from childhood to older adulthood and were based on a large, healthy, nationally representative sample.
Patterns of grip strength norms through the life course for males and females were consistent with other studies.Note 2Note 17Note 18 Males were stronger than females, although differences were smaller during childhood than in adolescence and adulthood.
The relationship with age was curvilinear. Grip strength increased from childhood through adolescence, peaked in mid-adulthood, and then declined. The age at which grip strength peaked varied among studies. The CHMS norms for men were highest at ages 40 to 44, which is within the range of 20 to 49 reported by Massey-Westropp et al.,Note 19 and slightly older than the ages of 30,Note 17 30 to 39Note 20 and 29 to 39Note 18 reported elsewhere. For women, the CHMS norms were highest at ages 30 to 49, which is similar to studies reporting peaks between ages 30 to 39Note 19Note 20 and ages 26 to 42,Note 18 but slightly older than the 25-to-30 age range reported by others.Note 18
WHO-fitted values and CHMS-fitted values were close to observed median values of maximum grip strength. The values were most similar for children, and generally more similar for females than for males. Observed values varied more from one age to the next for males than for females. For male adolescents, the WHO-fitted values were slightly lower than the CHMS-fitted and observed values. This suggests that the median height and weight of the healthy adolescents in this study slightly exceeded the median height and weight of the WHO growth charts.Note 16 For men, the WHO-fitted values were somewhat lower than observed values from ages 20 to 40 and somewhat higher than observed values from ages 40 to 60. The WHO-fitted values were also a little higher than the CHMS-fitted values for those aged 70 or older. These differences may be partly due to the use of a fixed value for height and weight throughout adulthood for the WHO-fitted values, and suggest that median height and weight varied by age for the adults in this study. However, overall, the reference equations fit the observed grip strength measurements relatively well.
Results of the comparison between norms for maximum, right-hand and left-hand grip strength were consistent with earlier research.Note 21 For the vast majority of right-hand-dominant people, the right hand tends to be stronger. By contrast, for a substantial percentage of left-hand-dominant people, the right hand is stronger than the left.Note 21 The difference in strength between hands tends to be greater for those who are right-handed.Note 22 Men are also slightly more likely than women to be left-handed.Note 23 Thus, it would be expected that maximum grip strength norms more closely resemble those for the right hand than the left, particularly among women.
While it may be more accurate to present norms by both side and hand dominance, because approximately 90% of people are right-handed,Note 24 sample sizes have tended to be insufficient to do so.Note 19Note 25Note 26 Therefore, norms have been published for the right hand and the left hand, or for dominant and non-dominant hand, but not for side and dominance together. Hand dominance was not determined in the CHMS, so norms are presented by side. If a specific hand is of interest, such as assessing the outcome of hand surgery, separate norms for each hand may be helpful. However, in most cases, use of maximum grip strength norms avoids the inaccuracy associated with norms presented by side or hand dominance.
When the 50th percentile reference equation was fitted with data from the WHO growth charts,Note 16 the values were within the range of other norms.Note 17Note 18Note 19Note 20 Differences may be partly attributable to the norms being based on data from the United States,Note 17 Great Britain,Note 18 Australia,Note 19 and Brazil.Note 20 Previous researchNote 2 has also noted differences in norms between countries, which supports the notion that norms should be country-specific.
Differences among norms may also reflect the composition of study samples. The present study excluded respondents who had chronic or other conditions that would affect grip strength. By contrast, Peterson and KrishnanNote 17 and Dodds et al.Note 18 did not exclude respondents to obtain a healthy sample. Massy-Westropp et al.Note 19 and Schlussel et al.Note 20 excluded respondents with some conditions, such as hand pain and osteoarthritis, but the exclusion criteria were not as strict as those applied in this study. The higher prevalence of chronic conditions such as heart disease, type 2 diabetes and chronic obstructive pulmonary disease at older ages may explain why the WHO-fitted values for the CHMS were higher than for the other norms at age 40 or older.
Another difference is the manner in which the norms were derived. Peterson and KrishnanNote 17 and Dodds et al.Note 18 modelled grip strength as a function of age, whereas Massy-Westropp et al.Note 19 and Schlussel et al.Note 20 calculated mean grip strength from the sample population. The CHMS reference equations were a function of age, height and weight. Median height and weight values from the WHO growth chartsNote 16 were used to fit values that could be graphed for comparison with other norms. Thus, the resulting values do not represent the 50th percentile of the entire Canadian population, but rather, the 50th percentile of a healthy Canadian population with the median height and weight from the WHO growth charts.Note 16 Median grip strength values of the entire Canadian population would vary to the extent that height and weight differ from these values, and from the inclusion of people with chronic and other conditions that influence grip strength.
Testing position affects the results of grip strength tests. The position recommended by the American Society of Hand Therapists (ASHT) requires being seated with the elbow flexed to 90o.Note 27 By contrast, the CHMS measured grip strength according to the CPAFLA protocol that specifies standing with the elbow being extended. Standing results in higher grip strengths than does sitting, but studies of the effect of elbow position have been inconsistent.Note 2 Although many studies adopted the ASHT’s recommended testing position, many others did not.Note 2 The testing position in the CHMS is consistent with that used in previous national surveysNote 17Note 28 and is recommended by the Canadian Society for Exercise Physiology for fitness assessments.Note 8
The reference equations developed in this study apply only to people aged 6 to 79. This is a limitation, particularly given the interest in the clinical and prognostic value of grip strength for older adults.Note 1Note 5 Deriving reference equations that include children younger than 6 and seniors older than 79 would require additional grip strength data collected using similar dynamometers and test protocols.
Normative values enable comparisons of grip strength relative to a reference population. This study presents reference equations and values for percentiles ranging from the 5th to the 95th percentiles, which is consistent with other studies.Note 17Note 18 The 5th percentile has been proposed as a point of reference for abnormally low grip strength.Note 29 However, future studies are needed to establish its clinical relevance and prognostic value.
Data from the 2007-to-2013 CHMS made it possible to develop grip strength reference equations for Canadians aged 6 to 79. These equations can be used to compare an individual’s grip strength measurements to the predicted grip strength of a healthy individual of the same age, sex, height and weight. Unlike previous reference values, these reference equations were based on a large, nationally representative sample of healthy 6- to 79-year-olds.
- Date modified: