Sex differences in physical performance by age, educational level, ethnic groups and birth cohort: The Longitudinal Aging Study Amsterdam

Lena D Sialino; Laura A Schaap; Sandra H van Oostrom; Astrid C J Nooyens; Hendrika S J Picavet; Johannes W R Twisk; W M Monique Verschuren; Marjolein Visser; Hanneke A H Wijnhoven

doi:10.1371/journal.pone.0226342

. 2019 Dec 18;14(12):e0226342. doi: 10.1371/journal.pone.0226342

Sex differences in physical performance by age, educational level, ethnic groups and birth cohort: The Longitudinal Aging Study Amsterdam

Lena D Sialino ^1,^*, Laura A Schaap ¹, Sandra H van Oostrom ², Astrid C J Nooyens ², Hendrika S J Picavet ², Johannes W R Twisk ³, W M Monique Verschuren ^2,⁴, Marjolein Visser ¹, Hanneke A H Wijnhoven ¹

Editor: Stephen D Ginsberg⁵

PMCID: PMC6919600 PMID: 31851709

Abstract

Background

Older women perform consistently poorer on physical performance tests compared to men. Risk groups for this “female disadvantage” in physical performance and it’s development over successive birth cohorts are unknown. This is important information for preventive strategies aimed to enhance healthy aging in all older women. This study aims to longitudinal investigate whether there are risk groups for a more apparent female disadvantage and study its trend over successive birth cohorts.

Methods

Data of the Longitudinal Aging Study Amsterdam (LASA) were used. All participants were aged 55–65 years at baseline. Longitudinal data of two birth cohorts with baseline measurements in 1992/1993 (n = 966, 24 year follow-up) and 2002/2003 (n = 1002, 12 year follow-up) were included. Follow-up measurements were repeated every three/four years. Cross-sectional data of two additional cohorts were included to compare ethnic groups: a Dutch cohort (2012/2013, n = 1023) and a Migration cohort (2013/2014, n = 478) consisting of migrants with a Turkish/Moroccan ethnicity.

Results

Mixed model analysis showed that women aged 55 years and older had a lower age- and height-adjusted gait speed (-0.03 m/s; -0.063–0.001), chair stand speed (-0.05 stand/s; -0.071–-0.033), handgrip strength (-14,8 kg; -15.69–-13.84) and balance (OR = 0.71; 0.547–0.916) compared to men. The sex difference in handgrip strength diminished with increasing age, but remained stable for gait speed, chair stand speed and balance. In general, results were consistent across different, educational levels and Turkish/Moroccan ethnic groups and birth cohorts.

Conclusions

There is a consistent “female disadvantage” in physical performance among older adults, which remains stable with increasing age (except for handgrip strength) and is consistent across different educational levels, ethnic groups and successive birth cohorts. So, no specific risk groups for the female disadvantage in physical performance were identified. Preventive strategies aimed to enhance healthy aging in older women are needed and should target all older women.

Introduction

Physical performance is an important indicator of healthy aging, since it is associated with a wide range of adverse health outcomes among older adults [1–4]. Women have a higher life expectancy but, paradoxically, older women perform poorer on physical performance tests than men [5–7]. This “female disadvantage” is well established, but its longitudinal course by age remains unclear. Furthermore, risk groups with regard to educational level and ethnic groups and trends over birth cohorts are unknown.

Most studies investigating the sex difference in physical performance by age are limited by a cross-sectional design or a short-term longitudinal follow-up period (five years) [5,8,9]. In exception, the well-studied sex difference in handgrip strength diminishes by age due to a more rapid decline in men compared with women [7,10,11]. There is some evidence from longitudinal data (ten years follow-up time) that the sex difference in gait speed remains stable during aging [12]. However, longitudinal research with a longer follow-up time is needed to confirm this finding, which will determine whether there is a specific age group where the female disadvantage is most apparent.

The influence of other sociodemographic factors on the sex difference in physical performance have not been studied so far. Low educated subgroups have a poorer physical performance [13], which is mainly explained by a higher prevalence of common chronic disease, obesity, smoking and though workload [14]. Since men and women differ in their lifestyle and incidence of chronic diseases [15], the influence of education on physical performance may differ between men and women. Indeed, more years of education seems to be a stronger and longer protective factor at higher ages, for a decline in gait speed for men compared with women [12]. More longitudinal research and other physical performance indications are needed to confirm this finding.

Ethnicity also influences physical performance, where persons with a European-American ethnicity have a higher gait and chair stand speed and better balance compared with persons with an African-American ethnicity [16,17]. This was largely explained by differences in education, obesity and diabetes [16]. Since men and women also differ in these factors [18], sex differences may differ per ethnic group. So far, this has not been investigated, but it was demonstrated that the sex difference in a physical performance summary score differs between some ethnic groups [16]. However, this research was limited to Hispanic- and Mexican-Americans ethnic groups only. The three largest ethnic groups in the Netherlands; Turkish, Moroccan and Dutch have not been studied so far. Since we know Turkish en Moroccan ethnic groups in the Netherlands have a lower education, this might result in a larger sex difference compared to a Dutch ethnic group [19].

Since a female disadvantage in physical performance in older adults has been reported by different studies over the past three decades, it may be a consistent phenomenon throughout time. However, factors affecting healthy aging change over time, were recent cohorts of older adults have a healthier lifestyle, higher socio-economic status, self-reported health and life expectancy but more chronic diseases [20–22]. These birth cohort effects may differ between men and women, since older women of a recent cohort have a better health profile related to cardiovascular diseases and diabetes mellitus and their average educational level becomes higher compared to men [23]. Since these are known protective factors for a decline in physical performance, we expect a decrease in the female disadvantage over successive birth cohorts. However, longitudinal research is needed to confirm this hypothesis.

A more detailed understanding of the sex difference in physical performance with regard to the longitudinal course by age, educational level and ethnic groups may identify potential risk groups for a more pronounced female disadvantage. This will help determine target groups for preventive strategies to enhance healthy aging in all older women. Also, knowledge with regard to the course of the sex difference over successive birth cohorts will identify the current trend over time, informing us whether this is an increasing or decreasing phenomenon. This study aimed to longitudinally investigate risk groups for a more apparent female disadvantage and study its trend over successive birth cohorts, which was achieved.

Methods

Study population

Data from the prospective Longitudinal Aging Study Amsterdam (LASA) were used. This study was initiated to determine predictors and consequences of aging and contains a nationally representative sample from three culturally distinct regions in the Netherlands (Amsterdam, Zwolle and Oss). Measurements and interviews are performed by trained interviewers, who visit respondents at home [24]. For a more detailed description of LASA and the used questionnaires, see Huisman et al. (2011). Ethical approval for the LASA study was given by the Medical Ethics Committee of the VU University Medical Centre Amsterdam, and all participants provided written informed consent.

(Birth) cohort populations

Longitudinal data of the birth cohort 1927–1937 with baseline measurements in 1992/1993 (n = 966, 25 year follow-up) and the birth cohort 1937–1947 with baseline measurements in 2002/2003 (n = 1002, 12 year follow up) were used. Follow-up measurements were performed every three to four years. Furthermore, cross-sectional data of an additional birth cohort (n = 1023) and a migration cohort including Turkish and Moroccan migrants (n = 478) both measured in 2012–2014 were used. All participants were aged 55–65 years at baseline.

Physical performance

Four measures of physical performance were included in this study: gait speed, chair stand speed, handgrip strength and balance. Gait speed was measured by asking participants to walk 3 meters, turn around and walk 3 meters back as quickly as possible, recorded by trained staff using a stopwatch. Gait speed was expressed in meters per second. Chair stand speed was measured by asking participants to fold their arms across their chest and to stand up and sit down from a sitting position five times at usual pace. Chair stand speed was defined as the number of chair rises per second. Handgrip strength was measured by asking participants to perform two maximum strength measurements for both hands in standing position with their arms along their body, recorded in kg by a Takei TKK 5001 dynamometer. Handgrip strength was defined as the average of the maximum measurement of each hand. In the migration cohort only the right hand was measured twice, where handgrip strength was defined as the maximum measurement. Balance was measured by asking participants to maintain their feet in the tandem position (heel of one foot directly in front of and touching the toes of the other foot) for ten seconds. Balance was defined as able if the test was performed correctly for at least 10 seconds, or unable if the test could not be performed due to physical inability or if balance was kept for less than 10 seconds.

Birth cohort, educational level and ethnic groups

Two birth cohorts were included in the longitudinal analysis: birth cohort 1927–1937 and birth cohort 1937–1947. Educational level was categorized into: low (elementary education or less), middle (lower vocational education and general intermediate education) and high (intermediate vocational education, general secondary education, higher vocational education, college education and university). Ethnic groups were categorized as Dutch or Turkish/Moroccan, using data of birth cohort 1947–1957 and the migration cohort. These two cohorts are comparable with regard to age (mean, SD), year of measurement and design (cross-sectional).

Statistical analyses

To examine the longitudinal data of birth cohorts 1927–1937 and 1937–1947, mixed model analysis with a random intercept for the individual was used for the continuous outcomes: gait speed, chair stand speed and handgrip strength. Generalized estimating equations (GEE) with an exchangeable correlation structure and robust variance estimator were used for the dichotomous outcome: ability to perform the tandem balance test. Here, GEE was used because the regression coefficients are calculated as “population averaged”, which fits best for our population study [25]. To examine the cross-sectional data of birth cohort 1947–1957 and the migration cohort, linear and logistic regression analyses were used.

The percentage missing values is similar between men and women and is relatively stable over different measurements and outcome measures (approximately 15% on average). S1 and S2 Tables show the number of participants included in each analysis. Mixed model analysis and GEE include all longitudinal data of the outcome, thereby preventing the loss of data and power [25]. Taken this into account and the fact that there are no further missing values in sex, age and baseline height, we assume no implications due to missing values in our analysis and a robust regression coefficient estimate.

Several analysis steps were taken: (1) The overall age- and height-adjusted sex difference in physical performance was investigated for all longitudinal and cross-sectional cohort populations separately (“sex” in a model with “age”). (2) The longitudinal course of the sex difference by age was investigated using the longitudinal data of the birth cohorts 1927–1937 and 1937–1947. Therefore, the interaction term age and sex: “age*sex” was used. If “age*age” or “age*age*sex” was significant, the quadratic regression coefficient in addition to the linear regression coefficient was included in the models throughout the rest of the analysis steps. This allowed a better prediction of the decline in physical performance measures during aging. (3) The modification of the age- and height-adjusted sex difference by educational level, ethnic groups or birth cohort was investigated for all longitudinal and cross-sectional cohort populations separately (interaction term with sex, in a model with “age” and “sex”: “birth cohort*sex”, “education*sex” or “ethnicity*sex”). (4) The modification of the course of the sex difference by age, by educational level or birth cohort was investigated using the longitudinal data of birth cohorts 1927–1937 and 1937–1947 (interaction term with “sex*age”: “birth cohort*sex*age” and “education*sex*age”). In other words, whether the course of the sex difference by age was different for different birth cohorts or educational levels. All analyses were adjusted for baseline height to estimate the true unadjusted sex difference in physical performance, since it was demonstrated that height partly explains the sex difference in physical performance measures, but is not part of the proposed causal pathway between physical performance and various health outcomes [26]. Body height was measured to the nearest 0.001 m using a stadiometer by a trained interviewer.

To visually depict the longitudinal course of the sex difference by age, a trend line for gait speed, chair rise speed and handgrip strength was plotted by age and sex, based on the regression coefficients of the full model. To depict balance, the probability to be able to perform the tandem balance test was predicted by age and sex. All analyses were performed using SPSS (version 26.0, SPSS inc.).

Results

(Birth) cohort populations

All (birth) cohort populations had an equal amount of men and women, with a similar mean age at baseline (Table 1). Overall, women were lower educated, had a poorer physical performance and were more often unable to perform the physical performance tests due to physical problems compared to men.

Table 1. Baseline characteristics for men and women of the different (birth) cohort populations.

	Birth cohort 1927–37		Birth cohort 1937–47		Birth cohort 1947–57		Migration cohort 1948–58
Characteristic variables	Men	Women	Men	Women	Men	Women	Men	Women
Sociodemographic
Study population	467 [48]	499 [52]	475 [47]	527 [53]	496 [48]	527 [52]	275 [58]	203 [42]
Age in years	60 (2.8)	60 (2.8)	60 (2.9)	60 (3.0)	60 (2.9)	60 (3.0)	61 (3.1)	61 (2.9)
Education
- Low education	104 [22]	202 [41]	89 [19]	124 [24]	50 [9.3]	55 [10]	179 [65]	164 [81]
- Middle education	148 [32]	191 [38]	239 [50]	334 [63]	266 [54]	328 [62]	64 [23]	26 [13]
- High education	215 [46]	106 [21]	147 [31]	69 [13]	180 [36]	144 [27]	31 [11]	8 [3]
Physical performance
Gait speed (m/s) Physically unable^a	0.96 (0.28) 5 [1.0]	0.93 (0.28) 12 [2.4]	1.0 (0.28) 6 [1.3]	0.91 (0.26) 10 [1.9]	1.1 (2.40) 2 [0.4]	1.0 (0.24) 4 [0.8]	0.76 (0.27) 3 [1.1]	0.66 (0.27) 8 [3.9]
Chair stand speed (rise/s) Physically unable^a	0.45 (0.14) 19 [4.1]	0.47 (0.45) 22 [4.4]	0.49 (0.14) 11 [2.3]	0.46 (0.13) 29 [5.5]	0.43 (0.11) 8 [1.6]	0.44 (0.11) 15 [2.8]	0.39 (0.16) 20 [7.1]	0.32 (0.12) 19 [9.4]
Handgrip strength (kg) Physically unable^a	40 (6.9) 4 [0.9]	24 (4.9) 5 [1.0]	45 (7.9) 13 [2.7]	27 (5.8) 14 [2.7]	44 (8.5) 1 [0.2]	24 (6.2) 4 [0.8]	36 (12.3) 3 [1.1]	20 (8.2) 2 [1.0]
Balance^b Physically unable^a	339 [73] 58 [12.4]	350 [70] 100 [20.0]	366 [77] 48 [10.1]	376 [71] 80 [15.2]	469 [95] 24 [4.8]	472 [90] 51 [9.5]	232 [84] 38 [13.8]	136 [67] 52 [25.6]

Open in a new tab

Birth cohort 1927–37 and 1937–47 contain longitudinal data, birth cohort 1947–57 and migration cohort contain cross-sectional data. Explanation of data: mean (SD), n [%].

^aParticipants who were unable to perform the test correctly/completely or refused due to being physical unable (for example; participants in a wheelchair).

^bNumber of participants able to successfully complete the tandem balance test (percentage of total).

Age- and height-adjusted sex difference in physical performance

Adjusted for age and height, women performed in general poorer than men in gait speed, chair stand speed, handgrip strength and balance in both longitudinal cohorts, except for gait speed in birth cohort 1927–1937 where no significant sex difference was observed (Table 2). To illustrate, mean gait speed of women in birth cohort 1937–1947 was 0.05 m/s (95% confidence interval:-0.069–-0.021) lower than men. Also, women had about half the odds on the ability to successfully complete the tandem balance test for 10 seconds compared to men in birth cohort 1937–1947 (OR = 0.59, 95% confidence interval: 0.416–0.844). For birth cohort 1947–1957, women performed only significantly poorer in handgrip strength compared to men and for the migration cohort women performed significantly poorer in gait speed, chair stand and handgrip strength compared to men (Table 2).

Table 2. Age- and height-adjusted mean difference in physical performance between women and men.

	Longitudinal data		Cross-sectional data
Physical performance	Birth cohort 1927–1937	Birth cohort 1937–1947	Birth cohort 1947–1957	Migration cohort 1948–1958
Gait speed (m/s) B	-0.03 (-0.063–0.001)	-0.05 (-0.069–-0.021)	-0.01 (-0.054–0.036)	-0.08 (-0.140–-0.014)
Chair stand speed (rise/s) B	-0.05 (-0.071–-0.033)	-0.04 (-0.057–-0.028)	-0.02 (-0.039–0.001)	-0.06 (-0.100–-0.027)
Handgrip strength (kg) B	-14.8 (-15.69–-13.84)	-16.3 (-17.01–-15.49)	-14.8 (-16.10–-13.49)	-15.5 (-18.01–-13.08)
Balance (ability) OR	0.71 (0.547–0.916)	0.59 (0.416–0.844)	0.97 (0.929–1.022)	0.96 (0.872–1.058)

Open in a new tab

Abbreviations: B = regression coefficient, OR = odds ratio, both with 95% confidence interval. For example: the value of -0.05 means that the mean gait speed of women in birth cohort 1937–1947 is 0.05 m/s lower than men.

Longitudinal course of sex differences in physical performance by age

The sex difference in gait speed, chair stand speed and balance remained stable during aging for both birth cohort 1927–1937 and 1937–1947 (Fig 1) (Table 3). The first exception was handgrip strength, where the decline by age was larger in men compared to women for both birth cohorts (age*sex; p<0.001), resulting in a decrease of the sex difference with increasing age (Fig 1D) (Table 3). In addition, the sex difference in balance for birth cohort 1937–1947 slightly decreased by age in the exponential part of the decline in balance (sex*age*age; p = 0.001) (Table 3).

Table 3. Multivariate model of the longitudinal course of the height-adjusted sex difference in phsyical performance.

	Cohort	Gait speed	Chair stand	Handgrip strength	Balance^a
Age (y)	1927–37	-0.013 [-0.0139–-0.1203], p<0.001	-0.005 [-0.0052–-0.0045], p<0.001	-0.466 [-0.4971–-0.4357], p<0.001	0.914 [0.902–0.926], p<0.001
Age (y)	1937–47	0.009 [-0.0102–-0.0073], p<0.001	-0.004 [-0.0043–-0.0030], p<0.001	-0.581 [-0.6136–-0.5548], p<0.001	0.953 [0.930–0.977], p<0.001
+ Age * Age (y²)	1927–37	-0.001 [-0.0003–-0.0001], p = 0.004	-0.001 [-0.0001–-0.0001], p = 0.003	-0.018 [-0.0220–-0.0131], p<0.001	0.967 [0.995–0.998], p<0.001
+ Age * Age (y²)	1937–47	-0.001 [-0.0009–-0.0005], p<0.001	-0.001 [-0.0002–-0.0001], p = 0.037	-0.017 [-0.0210–-0.0128], p<0.001	1.002 [0.994–1.002], p = 0.245
+ Sex (female)	1927–37	-0.028 [-0.0629–0.0007], p = 0.116	-0.052 [-0.0711–-0.0330], p<0.001	-14.77 [-15.689–-13.841], p<0.001	0.708 [0.547–0.916], p<0.001
+ Sex (female)	1937–47	-0.050 [-0.0781–-0.0212], p = 0.001	-0.042 [-0.0569–-0.0271], p<0.001	-16.29 [-17.079–-15.492], p<0.001	0.592 [0.416–0.844], p = 0.004
+ Sex (female) * Age (y)	1927–37	-0.001 [-0.0034–-0.0011], p = 0.505	-0.001 [-0.0010–0.0005], p = 0.573	0.156 [0.0959–0.2160], p<0.001	1.011 [0.986–1.037], p = 0.951
+ Sex (female) * Age (y)	1937–47	0.001 [-0.0021–0.0036], p = 0.616	-0.001 [-0.0023–0.0003], p = 0.115	0.201 [0.1443–0.2571], p<0.001	1.018 [0.931–1.036], p = 0.516
+ Sex (female) * Age * Age (y²)	1927–37	-0.001 [-0.0003–-0.0001], p = 0.660	-0.001 [-0.0035–0.0001], p = 0.476	0.006 [-0.0026–0.0151], p = 0.166	1.002 [0.999–1.006], p = 0.095
+ Sex (female) * Age * Age (y²)	1937–47	0.001 [-0.0003–0.0005], p = 0.555	0.001 [-0.00017–0.0002], p = 0.870	0.005 [-0.0035–0.0127], p = 0.263	1.010 [1.002–1.018], p = 0.011

Open in a new tab

Longitudinal data of LASA birth cohort 1927–1937 (light grey) and LASA birth cohort 1937–1947. Beta [95% confidence interval], p-value.

^aOR [95% confidence interval], p-value.

Modification by educational level

For each educational level, women had a poorer physical performance compared with men (data not shown). In general, there was no significant difference between the educational levels in age- and height-adjusted sex difference or its course by age (S3 Table). In exception, for birth cohort 1937–1947 the sex difference in gait speed was lower in high educated participants compared to middle education persons (p = 0.017) but chair stand speed was higher in high and middle educated participants compared to low educated persons (respectively; p = 0.046 and p = 0.001).

Modification by ethnic groups

Among older adults with a Turkish/Moroccan or Dutch ethnicity, women had a poorer physical performance compared to men (Table 2). The difference in age- and height-adjusted sex difference between the two ethnic groups was not consistent for different physical performance measures; the sex difference in chair stand speed was larger but in handgrip strength was smaller and for gait speed or balance similar in persons with a Turkish/Moroccan ethnicity compared to persons with a Dutch ethnicity. Noteworthy, the physical performance of both men and women is poorer in Turkish/Moroccan ethnicities compared to persons with a Dutch ethnicity.

Modification by birth cohort

There was no significant difference between the birth cohorts 1927–1937 and 1937–1947 in age- and height-adjusted sex difference in physical performance or its course by age (Fig 1) (S3 Table). The only exception was handgrip strength, where the age- and height-adjusted sex difference was slightly smaller for birth cohort 1927–1937 (-1.36 95% confidence interval; -0.166–-0.003 -, p = 0.026) and the sex difference decreased more rapidly by age compared with birth cohort 1937–1947 (p = 0.037) (Fig 1C) (S3 Table).

Discussion

The present study confirmed a consistent age- and height- adjusted sex difference in physical performance in persons aged 55 years and older, were women perform worse compared to men [1,7,27]. The sex differences for gait speed and handgrip strength are clinically relevant, because it reaches the minimally clinical significant individual difference estimates for gait speed (0.03–0.05 m/s) [28] and handgrip strength (5.0–6.5 kg) [29].

The female disadvantage in gait speed, chair stand speed and balance remained stable during aging, except for handgrip strength where it reduced by increasing age, confirming previous longitudinal findings for gait speed and handgrip strength [12,30]. In contrast, previous cross-sectional studies suggested an increase in the female disadvantage in gait speed and balance with age [7,8], demonstrating the importance of longitudinal studies.

The present study observed no consistent risk groups where the female disadvantage is most apparent, with regard to educational level and ethnic groups. The results of this study showed a better physical performance by participants with a Dutch ethnicity or higher education compared with other ethnicities (Moroccan and Turkish) or lower education, which is in line with previous research [13,16,17]. The current study demonstrated that these differences are similar for men and women.

This study is the first to show that the sex difference in physical performance and its longitudinal course by age does not significantly differ between the two investigated birth cohorts. Although the first birth cohort 1927–1937 was born before and birth cohort 1937–1947 born during world war two, this did not influence the sex difference in physical performance. Other factors that changed between the two birth cohorts, also did not influence the sex difference. This suggests that the female disadvantage is a robust phenomenon.

The observed findings raise the question whether the female disadvantage in physical performance during aging is due to a sex difference in physiology or in unhealthy aging. When the female disadvantage is present throughout the adult life course, it may suggest physiological differences between men and women, which may provide all women with a disadvantage in physical performance. On a different note, men and women might perform at different physical performance levels, although they have the same health status. This would suggest the use of sex-specific cut-off points and remains to be investigated. Next to a physiological difference, the sex difference in physical performance may also arise due to differences in (un)healthy aging such as lifestyle and chronic diseases. The present study did not identify at which age the female disadvantage arises since it was already present at age 55, suggesting it could be a physiological sex difference. However, there was no sex difference demonstrated in similar physical performance tests among healthy adults aged 20 to 39 years [5,31,32], except for handgrip strength [33]. This may point towards sex differences in (un)healthy aging or to a ceiling effect of these tests. Previous research supports both theories or the combination [7]. For example, the higher body fat percentage in women has been suggested to put them at a significant biomechanical disadvantage for greater disability at older age [34], suggesting a sex difference in (un)healthy aging. In addition, sex differences in exposure to lifestyle factors and chronic health conditions in older adults were shown to be explanatory factors [7]. For example, women have a higher prevalence and perceived disease burden of osteoarthritis, which may explain the observed sex difference in physical performance [35,36]. On the other hand, sex differences in muscle strength and exposure to sex hormones were also identified as possible explanatory factors [7,34], suggesting a sex difference in physiology. Future longitudinal research across the adult life span investigating physiological and (un)healthy aging explanations is recommended.

The present study has several strengths. First, data from a large prospective longitudinal cohort study were used, which represents the older adult life course of the Dutch population [37]. To note, response rate was high and drop-out was low; approximately 3% per follow-up. Secondly, the longitudinal design made it possible to study the longitudinal course by age. Thirdly, the use of four different objectively physical performance measures provided a broad image but allowed the study of various aspects of physical performance separately, in contrast to a combined physical performance score [38]. Furthermore, all measures are strongly associated with negative health outcomes [1–4] and demonstrated to be of high quality [39]. At last, this is the first time that the influence of various sociodemographic factors are investigated in one large longitudinal study. A possible limitation is the loss to follow-up due to mortality, since healthy people live longer and women have a higher life expectancy then men (S2 Table) [11]. This could have caused an overestimation of the mean age- and height-adjusted sex difference, since more unhealthy men compared with unhealthy women are lost. In contrast, more women than men were physically unable to perform the physical performance tests, which may have caused an underestimation. Secondly, the influence of ethnic groups could only be tested in a cross-sectional setting.

In conclusion, this large prospective longitudinal cohort study showed that in persons aged 55 years and older there is a consistent and stable female disadvantage in physical performance which pertains by increasing age. There are no indications for specific risk groups, with regard to educational level and ethnic groups or for a change of this female disadvantage in the coming decades. Therefore, future research on underlying mechanisms, explanations and preventive healthy aging strategies aimed to reduce the female disadvantage, should target all older adults. This novel information forms a solid basis for future research and strategies regarding the female disadvantage in physical performance.

Supporting information

S1 Table. Number participants per follow-up (FU) measurement for longitudinal birth cohorts 1927–1937 and 1937–1947.

Percentage and number of participants for each physical performance measurement (percentage of total at same follow-up measurement).

(DOCX)

Click here for additional data file.^{(16.1KB, docx)}

S2 Table. Number participants for each physical performance measurement (percentage of total) for cross-sectional birth cohorts 1947–1957 and migration cohort.

(DOCX)

Click here for additional data file.^{(13.3KB, docx)}

S3 Table. Effect modification of sex differences in physical performance by education, ethnic groups and birth cohort.

(DOCX)

Click here for additional data file.^{(15KB, docx)}

Acknowledgments

The authors are grateful to all LASA participants for their valued contributions.

Data Availability

Data cannot be shared publicly because of confidentiality. Data are available from the LASA Institutional Data Access / Ethics Committee (contact via https://www.lasa-vu.nl/index.htm) for researchers who meet the criteria for access to confidential data. The data underlying the results presented in the study are available from the Longitudinal Aging Study Amsterdam (https://www.lasa-vu.nl/index.htm). In the analysis proposal the following variables with file codes should be included: Gait speed, chair stand and tandem balance test in file LASA034, handgrip strength and height in file LASA161, education and sex in file Z004 and age in file Z008. For each variable all the waves should be included, from all baseline waves (B) to the wave (I). The LASA Steering Group will review all requests for data to ensure that proposals for the use of LASA data do not violate privacy regulations and are in keeping with informed consent that is provided by all LASA participants. The authors of this study do not have any special access privileges to the data underlying this study that other researchers would not have.

Funding Statement

This work was supported by the Netherlands Organization for Health Research and Development (ZonMw) [project number 849200005] (https://www.zonmw.nl/nl/). The Longitudinal Aging Study Amsterdam is supported by a grant from the Netherlands Ministry of Health Welfare and Sports, Directorate of Long-Term Care (https://www.government.nl/ministries/ministry-of-health-welfare-and-sport). The data collection in 2012-2013 and 2013-2014 was financially supported by the Netherlands Organization for Scientific Research (NWO) in the framework of the project “New Cohorts of young old in the 21st century” [file number 480-10-014] (https://www.nwo.nl/). The authors are grateful to all LASA participants for their valued contributions and have no conflict of interest to declare. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Cooper R, Kuh D, Cooper C, Gale CR, Lawlor DA, Matthews F, et al. Objective measures of physical capability and subsequent health: A systematic review. Age and Ageing. 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Den Ouden MEM, Schuurmans MJ, Arts IEMA, Van Der Schouw YT. Physical performance characteristics related to disability in older persons: A systematic review. Maturitas. 2011. [DOI] [PubMed] [Google Scholar]
3.Cooper R, Kuh D, Hardy R. Objectively measured physical capability levels and mortality: Systematic review and meta-analysis. BMJ (Online). 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Vermeulen J, Neyens JC, Van Rossum E, Spreeuwenberg MD, De Witte LP. Predicting ADL disability in community-dwelling elderly people using physical frailty indicators: A systematic review. BMC Geriatr. 2011; [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Butler AA, Menant JC, Tiedemann AC, Lord SR. Age and gender differences in seven tests of functional mobility. J Neuroeng Rehabil. 2009;6(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Wheaton F V, Crimmins EM. Female disability disadvantage: a global perspective on sex differences in physical function and disability. Ageing Soc. 2016;36(6):1136 10.1017/S0144686X15000227 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Cooper R, Hardy R, Sayer A, Ben-Shlomo Y, Birnie K, Cooper C, et al. Age and gender differences in physical capability levels from mid-life onwards: The harmonisation and meta-analysis of data from eight UK cohort studies. PLoS One. 2011; [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Seino S, Shinkai S, Fujiwara Y, Obuchi S, Yoshida H, Hirano H, et al. Reference values and age and sex differences in physical performance measures for community-dwelling older Japanese: A pooled analysis of six cohort studies. PLoS One. 2014;9(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Crimmins EM, Kim JK, Solé-Auró A. Gender differences in health: Results from SHARE, ELSA and HRS. Eur J Public Health. 2011;21(1):81–91. 10.1093/eurpub/ckq022 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Nahhas RW, Choh AC, Lee M, Cameron Chumlea WM, Duren DL, Siervogel RM, et al. Bayesian longitudinal plateau model of adult grip strength. Am J Hum Biol. 2010; [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Oksuzyan A, Crimmins E, Saito Y, O’Rand A, Vaupel JW, Christensen K. Cross-national comparison of sex differences in health and mortality in Denmark, Japan and the US. Eur J Epidemiol. 2010;25(7):471–80. 10.1007/s10654-010-9460-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Weber D. Differences in physical aging measured by walking speed: Evidence from the English Longitudinal Study of Ageing Physical functioning, physical health and activity. BMC Geriatr. 2016; [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Welmer A-K, Kareholt I, Rydwik E, Angleman S, Wang H-X. Education-related differences in physical performance after age 60: a cross-sectional study assessing variation by age, gender and occupation. BMC Public Health. 2013; [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Sainio P, Martelin T, Koskinen S, Heliövaara M. Educational differences in mobility: The contribution of physical workload, obesity, smoking and chronic conditions. J Epidemiol Community Health. 2007; [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Abad-Díez JM, Calderón-Larrañaga A, Poncel-Falcó A, Poblador-Plou B, Calderón-Meza JM, Sicras-Mainar A, et al. Age and gender differences in the prevalence and patterns of multimorbidity in the older population. BMC Geriatr. 2014; [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Al Snih S, Kaushik V, Eschbach K, Markides K. Ethnic differences in physical performance in older Americans: data from the Third National Health and Nutrition Examination Survey (1988–1994). Aging Clin Exp Res. 2008;20(2):139–44. 10.1007/bf03324760 [DOI] [PubMed] [Google Scholar]
17.Mossop H, Granic A, Sayer AA, Engstrom G, Davies K, Dodds R, et al. 41Differences in Physical Performance Across a Multi-Ethnic Cohort of Older Adults: Insights From the Healthy Aging Research Initiative. Age Ageing [Internet]. 2017;46(suppl_2):ii11–3. Available from: https://academic.oup.com/ageing/article-lookup/doi/10.1093/ageing/afx110.41 [Google Scholar]
18.Mauvais-Jarvis F. Epidemiology of gender differences in diabetes and obesity. In: Advances in Experimental Medicine and Biology. 2017. [DOI] [PubMed] [Google Scholar]
19.Fokkema T, Conkova N. Turkse en Marokkaanse ouderen in Nederland en België: een sociaal-demografisch profiel. Geron. 2018; [Google Scholar]
20.Deeg DJH, Huisman M. Cohort differences in 3-year adaptation to health problems among Dutch middle-aged, 1992–1995 and 2002–2005. Eur J Ageing. 2010; [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Heikkinen E, Kauppinen M, Rantanen T, Leinonen R, Lyyra TM, Suutama T, et al. Cohort differences in health, functioning and physical activity in the young-old Finnish population. Aging Clin Exp Res. 2011; [DOI] [PubMed] [Google Scholar]
22.Della Rovere F, Granata A, Familiari D, Arrigo GD, Mondello B, Basile G, et al. Mast Cells in Invasive Ductal Breast Cancer: Different Behavior in High and Minimum Hormone-receptive Cancers. 2007;2472:2465–71. [PubMed] [Google Scholar]
23.Cabrera C, Wilhelmson K, Allebeck P, Wedell H, Steen B, Lissner L. Cohort differences in obesity-related health indicators among 70-year olds with special reference to gender and education. Eur J Epidemiol. 2003; [DOI] [PubMed] [Google Scholar]
24.Huisman M, Poppelaars J, van der Horst M, Beekman ATF, Brug J, van Tilburg TG, et al. Cohort profile: The longitudinal aging study Amsterdam. Int J Epidemiol. 2011;40(4):868–76. 10.1093/ije/dyq219 [DOI] [PubMed] [Google Scholar]
25.Twisk JWR. Applied longitudinal data analysis for epidemiology: A practical guide, second edition Applied Longitudinal Data Analysis for Epidemiology: A Practical Guide. 2011. [Google Scholar]
26.Samson M. Relationships between physical performance measures, age, height and body weight in healthy adults. Age Ageing. 2002; [DOI] [PubMed] [Google Scholar]
27.Musselman K, Brouwer B. Gender-related differences in physical performance among seniors. J Aging Phys Act. 2005; [DOI] [PubMed] [Google Scholar]
28.Kwon S, Perera S, Pahor M, Katula JA, King AC, Groessl EJ, et al. What is a meaningful change in physical performance? Findings from a clinical trial in older adults (The LIFE-P study). J Nutr Heal Aging. 2009; [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Kim JK, Park MG, Shin SJ. What is the minimum clinically important difference in grip strength? Clin Orthop Relat Res. 2014; [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Oksuzyan A, Maier H, McGue M, Vaupel JW, Christensen K. Sex differences in the level and rate of change of physical function and grip strength in the danish 1905-cohort study. J Aging Health. 2010; [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Torres SF, Reis JG, de Abreu DCC. Influence of gender and physical exercise on balance of healthy young adults. Fisioter em Mov. 2014; [Google Scholar]
32.Hall KS, Cohen HJ, Pieper CF, Fillenbaum GG, Kraus WE, Huffman KM, et al. Physical performance across the adult life Span: Correlates with age and physical activity. Journals Gerontol—Ser A Biol Sci Med Sci. 2017; [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Steiber N. Strong or weak handgrip?normative reference values for the German population across the life course stratified by sex, age, and body height. PLoS One. 2016; [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Tseng LA, Delmonico MJ, Visser M, Boudreau RM, Goodpaster BH, Schwartz A V., et al. Body composition explains sex differential in physical performance among older adults. Journals Gerontol—Ser A Biol Sci Med Sci. 2014; [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Dunlop DD, Manheim LM, Sohn MW, Liu X, Chang RW. Incidence of functional limitation in older adults: The impact of gender, race, and chronic conditions. Arch Phys Med Rehabil. 2002; [DOI] [PubMed] [Google Scholar]
36.Boyan BD, Tosi L, Coutts R, Enoka R, Hart DA, Nicolella DP, et al. Sex differences in osteoarthritis of the knee. Journal of the American Academy of Orthopaedic Surgeons. 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.LASA. Mortality in LASA compared to the Dutch population, ages 60–85 [Internet]. LASA. 2012. http://www.lasa-vu.nl/data/lasa/documents/mortality-in-lasa-compared-to-the-dutch-population.pdf
38.Balachandran A, Signorile JF. How to improve reporting of the short physical performance battery scores. Journals of Gerontology—Series A Biological Sciences and Medical Sciences. 2015. [DOI] [PubMed] [Google Scholar]
39.Freiberger E, De vreede P, Schoene D, Rydwik E, Mueller V, Frändin K, et al. Performance-based physical function in older community-dwelling persons: A systematic review of instruments. Age Ageing. 2012; [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0226342.r001

Decision Letter 0

Stephen D Ginsberg

20 Aug 2019

PONE-D-19-18300

Sex differences in physical performance by age, birth cohort, educational level and ethnic groups; The Longitudinal Aging Study Amsterdam

PLOS ONE

Dear Dr. Sialino,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration by 2 Reviewers and an Academic Editor, all of the critiques of both Reviewers must be addressed in detail in a revision to determine publication status. If you are prepared to undertake the work required, I would be pleased to reconsider my decision, but revision of the original submission without directly addressing the critiques of the two Reviewers does not guarantee acceptance for publication in PLOS ONE. If the authors do not feel that the queries can be addressed, please consider submitting to another publication medium. A revised submission will be sent out for re-review. The authors are urged to have the manuscript given a hard copyedit for syntax and grammar.

==============================

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: General comments:

I think the manuscript is well written and the topic of great interest across different fields.

Yet, there are a few issues that in my view require further explanations.

Specific comments:

1. Abstract

I suggest that study aims be more precisely written because they are “somehow lost” in the “background objectives” section.

I did not clearly understand with there are cross-sectional samples in a longitudinal study. But I will come to this issue later on.

Results are somehow misleading, and not clearly enough to show their clinical relevance, although I know it is not always easy to write this in the abstract. Yet, I think authors can do a better job.

Finally, the “discussion conclusion” section ends up with what may be considered “common knowledge”. Can you please re-phrase it differently?

2. Introduction

It is well written, and apparently clear. I think it would benefit from a more substantive approach of the subject.

What do authors mean by “longitudinal stability”. This expression is often used in the draft, but no definition is given nor which statistic best captures this expression.

In paragraph two, in the last sentence, no clear explanation is given why further research is needed. Please explain.

The following paragraph also “suffers”, in my view, on too much descriptives instead of going a little deeper and suggest/explain/describe putative mechanism behind your correlates – socio-demographics and ethnicity. The same can be said in the last paragraph about cohort effects.

Finally, what are exactly the study aims? Are there any reasons why you did not formally posit hypotheses to be tested? If you provide these, please include also some substance linked to them.

Again, what is meant by longitudinal stability, and “would inform us on future trends“ (line 88). Are you thinking about prediction? If so, be more precise about its meaning and “clinical/intervention” implications.

3. Methods

Please provide more information about the LASA study, especially its aims and goals.

What about missing data and its putative implications in data analyses done?

Why do you include cross-sectional data in a longitudinal data set? Please explain, especially why include subjects with only one observation when you are interested in longitudinal stability!

At baseline having subjects with a 10 year lag (55-65) is problematic. Can you please explain why you use a decade lag.

In the “birth cohort, educational level and ethnic groups” entry birth cohorts were born before or during the 2nd World War with devastation consequences in the Netherlands. Yet, you did not mention this and its putative consequences, or is this issue of no relevance?

Statistical analyses are OK, although: (1) why do you adjust your analysis for height? Please explain; (2) what is this “longitudinal stability”?

Further down you wrote (line 159) “the modification of the stability, …” – what do you mean by this?

4. Results

Much of your results are written down, but the potential reader of the paper has no direct access to your results. I suggest, whenever needed, to include supplementary Tables so that the reader may “judge” by her/himself about them.

On the entre of “Longitudinal stability …” please be more precise. How do we know about it? Have all subjects the same “stability” or are there substantial differences? Please comment on this.

In the Modification by birth cohort entre, you write “there was no clinically relevant,…”. Yet, you never mention how can one decide about what is or is not clinical relevant. Are there cut-points for a reader to judge by her/himself?

In the Modification by educational level entry, I wonder if providing more detail (maybe a supplementary table) would help the reader to gain more insight about the results.

The very same suggestion goes to the Modification by ethnic groups entry.

5. Discussion

My main concern is that you do not provide substantive reasons/mechanisms/explanations in clinical terms about your findings. I think that this adding would increase the quality of the paper.

Reviewer #2: The authors investigate the sex differences in physical aging, which is addressed by the four measures gait speed, chair stand speed, handgrip strength, and balance. Several studies investigated the sex differences in different physical aging but only a few analyzed several measures and longitudinal data. This paper addresses the interesting question of potentially diminishing sex differences in physical aging. Overall, even though I see potential for this work, some issues need to be addressed.

Major issues:

The authors hypothesize that sex differences in physical performance diminish with increasing age, but what’s the expected mechanism? Why should the sex difference diminish and why should men and women perform at the same level? Several medical studies show differences in the body composition which is also related to muscle power relevant for e.g handgrip strength (next to hand size). The authors need to elaborate their motivation and the potential mechanism a bit more precise. Further why should a higher value in physical performance automatically go together with better health? There are also shown differences in body composition across continents (e.g. Europe and Asia).

Many studies on physical performance also showed body weight to be a relevant factor next to and in addition to body height. This measure should be included.

The authors have chosen a random intercept model to analyze the panel data. They call it random intercept for age, which is a very misleading term. Why was this model applied what is the advantage of this model here? Further, the statistical models should be formalized or at least the results should be provided within proper tables (either within the manuscript or as supplementary material). Right now, the model specification in steps 1 and 2 are not clear.

Another important issue: the description of the results is sparse at the moment. The authors refer quite often to Figure 1 when describing their results but they mis to provide numbers to support their visual interpretation. (In addition, the quality of Figure 1 makes it impossible to support all their interpretation). The interaction effect of age and sex is very relevant for this study, therefore I am quite surprised that there is no statistical result provided to show or disprove that sex differences are diminishing at higher ages.

Minor issues:

Was age centered within the analyses? (e.g around 55?)

The authors refer to other studies on younger adults showing no sex differences. This does not represent the full literature e.g https://doi.org/10.1371/journal.pone.0163917 shows sex differences in handgrip strength at young adulthood.

About the sample population:

An overview of dropouts per cohort group by sex might be relevant for readers to fully understand the sample. Some information is provided within the discussion section, but this needs to be done much earlier.

The high education subpopulation among the migration cohort is very small, authors need to consider dropping this group or interpret it with caution.

Table 1 could/should also include some information about participants refusing to perform the test or not willing to perform the test. Can these two categories be distinguished within the sample? How are participants treated who were willing but failed to perform the test or used the arms for the chair stand test….?

A limitation that should be considered when comparing the results: The chair stand test was not a maximum performance test as were the other three tests.

Was body height self-reported or measured?

References 11 and 24 are equal.

The quality of Figure 1 is poor .

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

==============================

We would appreciate receiving your revised manuscript by February, 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Stephen D. Ginsberg, Ph.D.

Section Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. During your revisions, please note that a simple title correction is required: the semicolon should be replaced by a colon. Please ensure this is updated both in the manuscript file and the online submission information.

3. Please provide additional details regarding participant consent.

In the ethics statement in the Methods and online submission information, please ensure that you have specified what type of consent you obtained (for instance, written or verbal).

If you obtained verbal consent, please state why it was not possible to obtain written consent, how verbal consent was recorded and whether the ethics committee approved this consent procedure.

4. Please include additional information regarding the survey or questionnaires used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

5. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

6. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information

PLoS One. 2019 Dec 18;14(12):e0226342. doi: 10.1371/journal.pone.0226342.r002

Author response to Decision Letter 0

23 Oct 2019

PONE-D-19-18300 L.D. Sialino et al. Response to Reviewers

We thank the reviewers for their comments. We made the suggested revisions in the manuscript. All revisions in the manuscript are highlighted by track-changes (Revised Manuscript with Track Changes). Our responses to the comments are written below, with references to lines in the final revised manuscript (Manuscript).

Reviewer #1

I think the manuscript is well written and the topic of great interest across different fields. Yet, there are a few issues that in my view require further explanations. Specific comments:

1. Abstract: I suggest that study aims be more precisely written because they are “somehow lost” in the “background objectives” section. Results are somehow misleading, and not clearly enough to show their clinical relevance, although I know it is not always easy to write this in the abstract. Yet, I think authors can do a better job. Finally, the “discussion conclusion” section ends up with what may be considered “common knowledge”. Can you please re-phrase it differently?

We rewrote the Background and Objectives and Discussion and Conclusion part of the abstract. We included the study aims (lines 26 to 28) and wrote a more detailed conclusion with extra information on implications for practice to emphasize the clinical relevance (lines 43 to 48).

2. Introduction: It is well written, and apparently clear. I think it would benefit from a more substantive approach of the subject. What do authors mean by “longitudinal stability”. This expression is often used in the draft, but no definition is given nor which statistic best captures this expression. Again, what is meant by longitudinal stability, and “would inform us on future trends“ (line 88). Are you thinking about prediction? If so, be more precise about its meaning and “clinical/intervention” implications.

Longitudinal stability referred to the stability by age. We used this term to refer to the course of the sex difference by age, also the stability of the sex difference. To prevent confusion, we changed the term into “longitudinal course by age” (line 55). The statistics behind this term is the sex*age and sex*age*age interaction, which is explained in an extra sentence added to the method (lines 179 to 181). In general, trends in health or disease in birth cohorts or generations in the past are often predictive for future trends. The two birth cohorts will indicate a stable, decrease or increase of the sex difference in physical performance. This provides an indication of whether the female disadvantage is an increasing or decreasing phenomenon within our changing society. It may also inform us whether this needs intervention in the future. A more detailed explanation was added (lines 94 to 99).

3. Introduction: In paragraph two, in the last sentence, no clear explanation is given why further research is needed. Please explain.

We added more information on the follow-up time of the described study (line 62 to 65). Further research with a longer follow-up time is needed to confirm the longitudinal finding (so far only a follow-up of 10 years) that the sex difference in gait speed remains stable during aging. In addition, we added a sentence on the practical implications of investigating this longitudinal course by age to better emphasize the relevance (line 64 to 66).

4. Introduction: The following paragraph also “suffers”, in my view, on too much descriptives instead of going a little deeper and suggest/explain/describe putative mechanism behind your correlates – socio-demographics and ethnicity. The same can be said in the last paragraph about cohort effects.

We split the paragraph on socio-demographics into two separate paragraphs (education and ethnicity). We now discussed in greater with more detailed the suggested putative mechanism behind the association between education/ethnicity and physical performance and our hypothesis why this might influence men and women differently. Please see the revised paragraph three and four of the introduction. The last paragraph about cohort effects was also extended with explanatory sentences (lines 90 to 94).

5. Introduction: Finally, what are exactly the study aims? Are there any reasons why you did not formally posit hypotheses to be tested? If you provide these, please include also some substance linked to them.

We added the study aims (lines 107 to 108). Furthermore, we posit sub-hypothesis based on previous literature within each paragraph of the introduction as was suggested by the reviewer. See the revised introduction.

6. Methods: Please provide more information about the LASA study, especially its aims and goals.

We revised the paragraph “Study population” of the methods accordingly and included a reference to the cohort paper of LASA (line 117).

7. Methods: What about missing data and its putative implications in data analyses done?

In our models we include sex, age and the outcome (physical performance). There are no missing values for sex and very limited for age (since it can be calculated by the date of the interview and birth date). So, there is only a missing when the whole interview was not carried out. Furthermore, LASA has a very low drop-out, the latter was a maximum of only 3% per wave. Both mixed models and Generalized Estimated Equations (GEE) use all longitudinal data of the outcome, not only the full case models, so this gives no problem for missing values for physical performance. Together with the fact that the missing values are limited (see Supplementary Table 1) we expect no implications due to missing values in our analysis. We agree that this should be made more clear in the method section and therefore added a paragraph on missing with reference to the paragraph statistical analyses (lines 169 to 175). Furthermore, to provide more insight into the missing values we included a supplementary table 1 and 2. These show the number of participants and its percentage of valid measurements versus the total, for each cohort, for each follow-up measurement, for each outcome measurement and for both men and women.

8. Methods: Why do you include cross-sectional data in a longitudinal data set? Please explain, especially why include subjects with only one observation when you are interested in longitudinal stability!

The migration cohort including participants of Turkish and Moroccan ethnicity started in 2012/2013, with one measurement so far. These is unique data to study the influence of ethnicity on the sex difference in physical performance, since this is the first time such extensive data was collected in these ethnic groups in the Netherlands. Although it is cross-sectional data, it was the only way to investigate the influence of ethnicity, so we decided to use this cross-sectional data for a sub-analysis. We agree that this is not useful for studying a longitudinal course. Therefore, we only studied whether the sex difference was significantly different between different ethnic groups (sex*ethnicity) and not study the longitudinal course by age (sex*age*ethnicity). To explain this more clearly, we revised the sentences explaining the cross-sectional analyses throughout the analyses steps.

9. Methods: At baseline having subjects with a 10 year lag (55-65) is problematic. Can you please explain why you use a decade lag.

We agree that having subjects aged 55-65 years at baseline creates a cross-sectional part in a longitudinal analysis. We however do not see this as problematic per se. Our analyses are adjusted for age, the distribution of age in our participants over these 10 years is homogeneous and the follow-up time is longer than this baseline lag (24 years for birth cohort 1927-37 and 12 years for 1937-47). LASA contains data with a 10 year lag and we decided to include all these participants to ensure a high power (n ~ 1000) and comparability between LASA cohorts (all have a baseline of 55-65 years).

10. Methods: In the “birth cohort, educational level and ethnic groups” entry birth cohorts were born before or during the 2nd World War with devastation consequences in the Netherlands. Yet, you did not mention this and its putative consequences, or is this issue of no relevance?

We hypothesized that the war might be of influence on the access to education. Indeed, we see that the birth cohort born during World War 2 has a lower percentage of high educated persons for both men and women. However, we do not expect that it affected men and women differently. Also, we showed in this study that education does not influence the sex difference in physical performance. It was investigated using the LASA data if the famine that took place during the war had an influence on the height and chance of developing diabetes and cardiovascular disease as older adults (F.R.M. Portrait et al. 2011 & 2017). It was demonstrated that these associations are significant for women, but not for men, which was suggested to be related to the fact that female health is more influenced by childhood events compared to men. We however, in our study, found no difference in effect on physical performance measures. We added a sentence relating to this point to the discussion (lines 303 to 306).

11. Methods: Statistical analyses are OK, although: (1) why do you adjust your analysis for height?

It has been demonstrated that sex differences in physical performance are partly explained by the height differences between men and women on a population level (see reference 65). Height affects the performance on the physical performance tests but is not related to the causal pathways between physical performance and various health outcomes. Therefore, we adjust for height to estimate the sex difference in physical performance without the bias of height in the performance measurement. Although body weight was also shown to partly explain the sex difference in physical performance, this is usually a consequence of life style patterns (diet and physical activity), which is in the causal pathway between physical performance and health outcomes. Therefore, we only adjusted our analyses for height. We added a sentence to the method section “statistical analyses” to explain this more elaborately (lines 192 to 197).

12. Methods: Please explain; (2) what is this “longitudinal stability”?

Longitudinal stability referred to the stability by age. We used this term to refer to the course of the sex difference by age, the stability of the sex difference. To prevent confusion, we changed the term into “longitudinal course by age” (line 55).

13. Methods: Further down you wrote (line 159) “the modification of the stability, …” – what do you mean by this?

We meant whether the course of the sex difference by age (increase, decrease or stable sex difference by age) was modified by birth cohort or educational level. In other words, whether the course of the sex difference was different for different birth cohorts or educational levels. We revised the sentence and added more explanation (lines 191 to 192).

14. Results: Much of your results are written down, but the potential reader of the paper has no direct access to your results. I suggest, whenever needed, to include supplementary Tables so that the reader may “judge” by her/himself about them.

In agreement with this comment, we provided an extra Table 3 which shows the multivariate model of the longitudinal course of the height-adjusted sex difference per physical performance measure by age, for both longitudinal birth cohorts 1927-1937 and 1937-1947. In addition, an extra Supplementary Table 3 was provided including all the effect-modification results.

15. Results: On the entre of “Longitudinal stability …” please be more precise. How do we know about it? Have all subjects the same “stability” or are there substantial differences? Please comment on this.

We tested the longitudinal course of the sex difference in physical performance by age by testing the interaction terms sex*age and sex*age*age (methods). The current study is a population study so we can only conclude that on a population level, the sex difference in average physical performance remains stable (or decreases for handgrip strength). It was not our objective to study and report on the individual differences in longitudinal stability.

16. Results: In the Modification by birth cohort entre, you write “there was no clinically relevant,…”. Yet, you never mention how can one decide about what is or is not clinical relevant. Are there cut-points for a reader to judge by her/himself?

We discuss the cut-off points for clinical relevance for gait speed and handgrip strength in the discussion section. We agree that this should not be mentioned in the result section but only in the discussion section. We revised it accordingly.

17. Results: In the Modification by educational level entry, I wonder if providing more detail (maybe a supplementary table) would help the reader to gain more insight about the results. The very same suggestion goes to the Modification by ethnic groups entry.

We provided Supplementary Table 3 which shows all tested effect modification by education, birth cohort and ethnicity with regression coefficient, 95% confidence interval and p-value.

18. Discussion: My main concern is that you do not provide substantive reasons/mechanisms/explanations in clinical terms about your findings. I think that this adding would increase the quality of the paper.

The objective of this study was not to study the explanatory factors of the sex difference in physical performance, but provide a detailed understanding of the sex difference. In our discussion we do mention two possible explanatory theories (physiological or healthy aging). Since we do not test explanatory factors, we only briefly introduce the main factors from literature so far. We added an example of a sex difference in chronic disease osteoarthritis, that has been suggested to explain the observed sex difference in physical performance (lines 326 to 328). In future studies of our project “Sex and Aging” we will investigate explanatory factors.

Reviewer #2

The authors investigate the sex differences in physical aging, which is addressed by the four measures gait speed, chair stand speed, handgrip strength, and balance. Several studies investigated the sex differences in different physical aging but only a few analyzed several measures and longitudinal data. This paper addresses the interesting question of potentially diminishing sex differences in physical aging. Overall, even though I see potential for this work, some issues need to be addressed.

Major issues:

1. The authors hypothesize that sex differences in physical performance diminish with increasing age, but what’s the expected mechanism? Why should the sex difference diminish and why should men and women perform at the same level? Several medical studies show differences in the body composition which is also related to muscle power relevant for e.g handgrip strength (next to hand size). There are also shown differences in body composition across continents (e.g. Europe and Asia). The authors need to elaborate their motivation and the potential mechanism a bit more precise.

We understand the questions mentioned by the reviewer. We hypothesize that if the sex difference is stable by age, it suggests a physiological difference. We think the body composition is a potential explanatory factor here. This indeed does not mean that men and women have the same health at the same performance level, as was questioned by the reviewer. If the same health by men and women is measured at different physical performances, sex-specific cut-offs for the relationship between physical performance and health are needed. However, this needs to be assessed in future research. To discuss these points, we added two sentences on this topic (lines 312 to 314). Next to an physiological difference, the sex difference might be due to difference in (un)healthy aging life style factors. These factors might change of time, resulting in a change of the sex difference by age. However, all discussed points are not addressed in our research question and are therefore only discussed as possible theories in the discussion (lines 315 to 321)

2. Further why should a higher value in physical performance automatically go together with better health?

There has been extensive research performed investigating the relationship between physical performance and various health outcomes. A strong association for handgrip strength, gait speed and lower body strength (party measured by chair stand) with probability of disability was observed. Also, gait speed and mortality were found to be associated. Although this does not mean that a person with higher physical performance always has a better health, it means that on average good physical performance is an indicator for health. To emphasize this, we only used the word “associated” and not “affects” or “means” in the introduction (lines 57 to 58).

3. Many studies on physical performance also showed body weight to be a relevant factor next to and in addition to body height. This measure should be included.

Height affects the performance on the physical performance tests but is not incorporated in the (hypothesized) causal pathway between physical performance and various health outcomes. Therefore, we adjust for height to estimate the sex difference in physical performance without the bias of height in the measurement of performance. Although body weight was also shown to partly explain the sex difference in physical performance, this is usually affected by life style (diet and physical activity), which may be part of the causal pathway between physical performance and health outcomes. Therefore, to show the unadjusted sex difference in physical performance, we only adjusted our analyses for height. We added a sentence to the method section “statistical analysis” to explain our choice more elaborately (lines 193 to 196).

4. The authors have chosen a random intercept model to analyze the panel data. They call it random intercept for age, which is a very misleading term. Why was this model applied what is the advantage of this model here? Further, the statistical models should be formalized or at least the results should be provided within proper tables (either within the manuscript or as supplementary material). Right now, the model specification in steps 1 and 2 are not clear.

We meant a random intercept for the individual, we revised the sentence accordingly. The advantage of this model is that it takes the correlation between the measurements within one person into account. To provide more insight in our statistical models we added Table 3, which shows our multivariable model and its build-up step by step as described in the method section.

5. Another important issue: the description of the results is sparse at the moment. The authors refer quite often to Figure 1 when describing their results but they mis to provide numbers to support their visual interpretation. (In addition, the quality of Figure 1 makes it impossible to support all their interpretation). The interaction effect of age and sex is very relevant for this study, therefore I am quite surprised that there is no statistical result provided to show or disprove that sex differences are diminishing at higher ages.

We added Table 2 to provide the statistical method used to produce figure 1 and show the interaction effects between age and sex. In addition, we also added Supplementary Table 3 to show the results of the effect modification analyses.

Minor issues

1. Was age centered within the analyses? (e.g around 55?)

Age was not centered in our analysis. We decided to not center within the analysis because we only take age and sex as covariates in our analysis (where the value zero is not necessarily arbitrary). Also, we are mostly interested in the interaction terms, where age centered analysis do not have an beneficial effect.

2. The authors refer to other studies on younger adults showing no sex differences. This does not represent the full literature e.g https://doi.org/10.1371/journal.pone.0163917 shows sex differences in handgrip strength at young adulthood.

We adjusted the sentence in the discussion and added the reference advised by the reviewer (line 320).

3. An overview of dropouts per cohort group by sex might be relevant for readers to fully understand the sample. Some information is provided within the discussion section, but this needs to be done much earlier.

We included two supplementary tables (1 and 2), which show the number of participants and its percentage of valid measurements versus the total, for each cohort, for each follow-up measurement, for each outcome measurement and for both men and women.

4. The high education subpopulation among the migration cohort is very small, authors need to consider dropping this group or interpret it with caution.

We only analyze the migration cohort as a whole group together to estimate the sex difference and compare the sex difference with the Dutch cohort 1947-1957. Since we do not group by education level, we did not encounter a low power problem.

5. Table 1 could/should also include some information about participants refusing to perform the test or not willing to perform the test. Can these two categories be distinguished within the sample? How are participants treated who were willing but failed to perform the test or used the arms for the chair stand test….?

We included supplementary tables 1 and 2 for an overview of the participants who refused, were not willing to, did not perform any interview or unknown missing values. Table 1 “unable” shows the participants who were unable to perform the test correctly or completely or refused due to being physically unable to perform the test (for example; participants in a wheelchair). To make this point more clear, we added an extra footnote to Table 1.

6. A limitation that should be considered when comparing the results: The chair stand test was not a maximum performance test as were the other three tests.

The gait speed and handgrip strength test are maximum performance tests, but the balance and chair stand test are not. We do not see this as a limitation when comparing results. Since we are interested in the functioning these tests represent and not the maximum performance, both test are applicable. In addition, our goal is to detect a differentiation between participants and within in a participant by age, which is achieved by both our maximum and non-maximum performance tests. We however agree that the point made by the reviewer is something to take into account when interpreting the results. We therefore explained all measures in detail in the method section.

7. Was body height self-reported or measured?

Body height was measured to the nearest 0.001 m using a stadiometer by a trained interviewer. We added this information to the method section (lines 196 to 197).

8. References 11 and 24 are equal.

We removed the duplicate reference.

9. The quality of Figure 1 is poor .

We used another program to make the matrix graph to ensure high quality after submission. Please see revised Figure 1.

Concluding author remarks

- All analyses were rerun in a newer version of SPSS (26) to create the additional tables. Therefore, the figures and values in tables and text slightly changed after the revision.

- During the rerun of all analysis, we discovered a minor error in the gait speed analysis of LASA birth cohort 1927-1937. As a result, the effect sizes for the sex difference in gait speed were revised (see Table 1). This did not influence our final conclusions.

- The graph for depicting the balance outcome measure was changed using the full model estimated regression coefficients instead of the raw data. This ensured a line that better fits to the model as was previously used.

PLoS One. doi: 10.1371/journal.pone.0226342.r003

Decision Letter 1

Stephen D Ginsberg

26 Nov 2019

Sex differences in physical performance by age, birth cohort, educational level and ethnic groups: The Longitudinal Aging Study Amsterdam

PONE-D-19-18300R1

Dear Dr. Sialino,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Stephen D. Ginsberg, Ph.D.

Section Editor

PLOS ONE

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: I am satisfied with author´s answers to my comments/questions/suggestions. Further, the new draft is richer and more reliable.

I have no further questions.

Reviewer #2: Dear authors,

My raised issues were fully addressed. The manuscript approved a lot and reads well, in particular the methods and results section are now much clearer.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

PLoS One. doi: 10.1371/journal.pone.0226342.r004

Acceptance letter

Stephen D Ginsberg

10 Dec 2019

PONE-D-19-18300R1

Sex differences in physical performance by age, educational level, ethnic groups and birth cohort: The Longitudinal Aging Study Amsterdam

Dear Dr. Sialino:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Stephen D Ginsberg

Section Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Number participants per follow-up (FU) measurement for longitudinal birth cohorts 1927–1937 and 1937–1947.

Percentage and number of participants for each physical performance measurement (percentage of total at same follow-up measurement).

(DOCX)

Click here for additional data file.^{(16.1KB, docx)}

S2 Table. Number participants for each physical performance measurement (percentage of total) for cross-sectional birth cohorts 1947–1957 and migration cohort.

(DOCX)

Click here for additional data file.^{(13.3KB, docx)}

S3 Table. Effect modification of sex differences in physical performance by education, ethnic groups and birth cohort.

(DOCX)

Click here for additional data file.^{(15KB, docx)}

Data Availability Statement

[pone.0226342.ref001] 1.Cooper R, Kuh D, Cooper C, Gale CR, Lawlor DA, Matthews F, et al. Objective measures of physical capability and subsequent health: A systematic review. Age and Ageing. 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref002] 2.Den Ouden MEM, Schuurmans MJ, Arts IEMA, Van Der Schouw YT. Physical performance characteristics related to disability in older persons: A systematic review. Maturitas. 2011. [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref003] 3.Cooper R, Kuh D, Hardy R. Objectively measured physical capability levels and mortality: Systematic review and meta-analysis. BMJ (Online). 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref004] 4.Vermeulen J, Neyens JC, Van Rossum E, Spreeuwenberg MD, De Witte LP. Predicting ADL disability in community-dwelling elderly people using physical frailty indicators: A systematic review. BMC Geriatr. 2011; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref005] 5.Butler AA, Menant JC, Tiedemann AC, Lord SR. Age and gender differences in seven tests of functional mobility. J Neuroeng Rehabil. 2009;6(1). [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref006] 6.Wheaton F V, Crimmins EM. Female disability disadvantage: a global perspective on sex differences in physical function and disability. Ageing Soc. 2016;36(6):1136 10.1017/S0144686X15000227 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref007] 7.Cooper R, Hardy R, Sayer A, Ben-Shlomo Y, Birnie K, Cooper C, et al. Age and gender differences in physical capability levels from mid-life onwards: The harmonisation and meta-analysis of data from eight UK cohort studies. PLoS One. 2011; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref008] 8.Seino S, Shinkai S, Fujiwara Y, Obuchi S, Yoshida H, Hirano H, et al. Reference values and age and sex differences in physical performance measures for community-dwelling older Japanese: A pooled analysis of six cohort studies. PLoS One. 2014;9(6). [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref009] 9.Crimmins EM, Kim JK, Solé-Auró A. Gender differences in health: Results from SHARE, ELSA and HRS. Eur J Public Health. 2011;21(1):81–91. 10.1093/eurpub/ckq022 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref010] 10.Nahhas RW, Choh AC, Lee M, Cameron Chumlea WM, Duren DL, Siervogel RM, et al. Bayesian longitudinal plateau model of adult grip strength. Am J Hum Biol. 2010; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref011] 11.Oksuzyan A, Crimmins E, Saito Y, O’Rand A, Vaupel JW, Christensen K. Cross-national comparison of sex differences in health and mortality in Denmark, Japan and the US. Eur J Epidemiol. 2010;25(7):471–80. 10.1007/s10654-010-9460-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref012] 12.Weber D. Differences in physical aging measured by walking speed: Evidence from the English Longitudinal Study of Ageing Physical functioning, physical health and activity. BMC Geriatr. 2016; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref013] 13.Welmer A-K, Kareholt I, Rydwik E, Angleman S, Wang H-X. Education-related differences in physical performance after age 60: a cross-sectional study assessing variation by age, gender and occupation. BMC Public Health. 2013; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref014] 14.Sainio P, Martelin T, Koskinen S, Heliövaara M. Educational differences in mobility: The contribution of physical workload, obesity, smoking and chronic conditions. J Epidemiol Community Health. 2007; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref015] 15.Abad-Díez JM, Calderón-Larrañaga A, Poncel-Falcó A, Poblador-Plou B, Calderón-Meza JM, Sicras-Mainar A, et al. Age and gender differences in the prevalence and patterns of multimorbidity in the older population. BMC Geriatr. 2014; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref016] 16.Al Snih S, Kaushik V, Eschbach K, Markides K. Ethnic differences in physical performance in older Americans: data from the Third National Health and Nutrition Examination Survey (1988–1994). Aging Clin Exp Res. 2008;20(2):139–44. 10.1007/bf03324760 [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref017] 17.Mossop H, Granic A, Sayer AA, Engstrom G, Davies K, Dodds R, et al. 41Differences in Physical Performance Across a Multi-Ethnic Cohort of Older Adults: Insights From the Healthy Aging Research Initiative. Age Ageing [Internet]. 2017;46(suppl_2):ii11–3. Available from: https://academic.oup.com/ageing/article-lookup/doi/10.1093/ageing/afx110.41 [Google Scholar]

[pone.0226342.ref018] 18.Mauvais-Jarvis F. Epidemiology of gender differences in diabetes and obesity. In: Advances in Experimental Medicine and Biology. 2017. [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref019] 19.Fokkema T, Conkova N. Turkse en Marokkaanse ouderen in Nederland en België: een sociaal-demografisch profiel. Geron. 2018; [Google Scholar]

[pone.0226342.ref020] 20.Deeg DJH, Huisman M. Cohort differences in 3-year adaptation to health problems among Dutch middle-aged, 1992–1995 and 2002–2005. Eur J Ageing. 2010; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref021] 21.Heikkinen E, Kauppinen M, Rantanen T, Leinonen R, Lyyra TM, Suutama T, et al. Cohort differences in health, functioning and physical activity in the young-old Finnish population. Aging Clin Exp Res. 2011; [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref022] 22.Della Rovere F, Granata A, Familiari D, Arrigo GD, Mondello B, Basile G, et al. Mast Cells in Invasive Ductal Breast Cancer: Different Behavior in High and Minimum Hormone-receptive Cancers. 2007;2472:2465–71. [PubMed] [Google Scholar]

[pone.0226342.ref023] 23.Cabrera C, Wilhelmson K, Allebeck P, Wedell H, Steen B, Lissner L. Cohort differences in obesity-related health indicators among 70-year olds with special reference to gender and education. Eur J Epidemiol. 2003; [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref024] 24.Huisman M, Poppelaars J, van der Horst M, Beekman ATF, Brug J, van Tilburg TG, et al. Cohort profile: The longitudinal aging study Amsterdam. Int J Epidemiol. 2011;40(4):868–76. 10.1093/ije/dyq219 [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref025] 25.Twisk JWR. Applied longitudinal data analysis for epidemiology: A practical guide, second edition Applied Longitudinal Data Analysis for Epidemiology: A Practical Guide. 2011. [Google Scholar]

[pone.0226342.ref026] 26.Samson M. Relationships between physical performance measures, age, height and body weight in healthy adults. Age Ageing. 2002; [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref027] 27.Musselman K, Brouwer B. Gender-related differences in physical performance among seniors. J Aging Phys Act. 2005; [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref028] 28.Kwon S, Perera S, Pahor M, Katula JA, King AC, Groessl EJ, et al. What is a meaningful change in physical performance? Findings from a clinical trial in older adults (The LIFE-P study). J Nutr Heal Aging. 2009; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref029] 29.Kim JK, Park MG, Shin SJ. What is the minimum clinically important difference in grip strength? Clin Orthop Relat Res. 2014; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref030] 30.Oksuzyan A, Maier H, McGue M, Vaupel JW, Christensen K. Sex differences in the level and rate of change of physical function and grip strength in the danish 1905-cohort study. J Aging Health. 2010; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref031] 31.Torres SF, Reis JG, de Abreu DCC. Influence of gender and physical exercise on balance of healthy young adults. Fisioter em Mov. 2014; [Google Scholar]

[pone.0226342.ref032] 32.Hall KS, Cohen HJ, Pieper CF, Fillenbaum GG, Kraus WE, Huffman KM, et al. Physical performance across the adult life Span: Correlates with age and physical activity. Journals Gerontol—Ser A Biol Sci Med Sci. 2017; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref033] 33.Steiber N. Strong or weak handgrip?normative reference values for the German population across the life course stratified by sex, age, and body height. PLoS One. 2016; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref034] 34.Tseng LA, Delmonico MJ, Visser M, Boudreau RM, Goodpaster BH, Schwartz A V., et al. Body composition explains sex differential in physical performance among older adults. Journals Gerontol—Ser A Biol Sci Med Sci. 2014; [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref035] 35.Dunlop DD, Manheim LM, Sohn MW, Liu X, Chang RW. Incidence of functional limitation in older adults: The impact of gender, race, and chronic conditions. Arch Phys Med Rehabil. 2002; [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref036] 36.Boyan BD, Tosi L, Coutts R, Enoka R, Hart DA, Nicolella DP, et al. Sex differences in osteoarthritis of the knee. Journal of the American Academy of Orthopaedic Surgeons. 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0226342.ref037] 37.LASA. Mortality in LASA compared to the Dutch population, ages 60–85 [Internet]. LASA. 2012. http://www.lasa-vu.nl/data/lasa/documents/mortality-in-lasa-compared-to-the-dutch-population.pdf

[pone.0226342.ref038] 38.Balachandran A, Signorile JF. How to improve reporting of the short physical performance battery scores. Journals of Gerontology—Series A Biological Sciences and Medical Sciences. 2015. [DOI] [PubMed] [Google Scholar]

[pone.0226342.ref039] 39.Freiberger E, De vreede P, Schoene D, Rydwik E, Mueller V, Frändin K, et al. Performance-based physical function in older community-dwelling persons: A systematic review of instruments. Age Ageing. 2012; [DOI] [PubMed] [Google Scholar]

PERMALINK

Sex differences in physical performance by age, educational level, ethnic groups and birth cohort: The Longitudinal Aging Study Amsterdam

Lena D Sialino

Laura A Schaap

Sandra H van Oostrom

Astrid C J Nooyens

Hendrika S J Picavet

Johannes W R Twisk

W M Monique Verschuren

Marjolein Visser

Hanneke A H Wijnhoven

Roles

Abstract

Background

Methods

Results

Conclusions

Introduction

Methods

Study population

(Birth) cohort populations

Physical performance

Birth cohort, educational level and ethnic groups

Statistical analyses

Results

(Birth) cohort populations

Table 1. Baseline characteristics for men and women of the different (birth) cohort populations.

Age- and height-adjusted sex difference in physical performance

Table 2. Age- and height-adjusted mean difference in physical performance between women and men.

Longitudinal course of sex differences in physical performance by age

Fig 1. Four physical performance measures by age and sex for two longitudinal birth cohorts 1927–1937 and 1937–1947.

Table 3. Multivariate model of the longitudinal course of the height-adjusted sex difference in phsyical performance.

Modification by educational level

Modification by ethnic groups

Modification by birth cohort

Discussion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Stephen D Ginsberg

Roles

Author response to Decision Letter 0

Decision Letter 1

Stephen D Ginsberg

Roles

Acceptance letter

Stephen D Ginsberg

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases