Content » Vol 46, Issue 7

Original report

Self-reported fatigue and physical function in late mid-life

Han Boter, PhD1,2, Minna Mänty, PhD1,3, Åse Marie Hansen, PhD1,4, Tibor Hortobágyi, PhD5 and Kirsten Avlund, PhD1,3,6†

From the 1Section of Social Medicine, Department of Public Health, University of Copenhagen, Copenhagen, Denmark, 2University of Groningen, University Medical Center Groningen, Department of Epidemiology, Groningen, The Netherlands, 3Centre for Healthy Aging, University of Copenhagen, 4National Research Centre for the Working Environment, Copenhagen, Denmark, 5University of Groningen, University Medical Center Groningen, Center for Human Movement Sciences, Groningen, The Netherlands and 6Danish Centre for Aging Research, Universities of Southern Denmark, Aarhus and Copenhagen, Denmark. Deceased 31 August 2013

OBJECTIVE: To determine the association between the 5 subscales of the Multidimensional Fatigue Inventory (MFI-20) and physical function in late mid-life.

DESIGN: Cross-sectional study.

SUBJECTS: A population-based sample of adults who participated in the Copenhagen Aging and Midlife Biobank population cohort (n = 4,964; age 49–63 years).

METHODS: Self-reported fatigue was measured using the MFI-20 comprising: general fatigue, physical fatigue, reduced activity, reduced motivation, and mental fatigue. Handgrip strength and chair rise tests were used as measures of physical function. Multiple logistic regression analyses were used to determine the associations between handgrip strength and the chair rise test with the MFI-20 subscales, adjusted for potential confounders.

RESULTS: After adjustments for potential confounders, handgrip strength was associated with physical fatigue (adjusted odds ratio (OR) 0.75 (95% confidence interval (CI) 0.66–0.86); p ≤ 0.001) and reduced motivation (adjusted OR 0.85 (95% CI 0.75–0.96); p ≤ 0.05), but not with the other subscales. After these adjustments, the chair rise test was associated with physical fatigue (adjusted OR 0.61 (0.53–0.69); p ≤ 0.001), general fatigue (adjusted OR 0.72 (0.62–0.84); p ≤ 0.001), reduced activity (adjusted OR 0.79 (0.70–0.90); p ≤ 0.001) and reduced motivation (adjusted OR 0.84 (0.74–0.95); p ≤ 0.01), but not with mental fatigue. Subgroup analyses for sex did not show statistically significant different associations between physical function and fatigue.

CONCLUSION: The present study supports the physiological basis of 4 subscales of the MFI-20. The association between fatigue and function was independent of gender.

Key words: hand strength; fatigue; questionnaires; cross-sectional studies; self-reported fatigue; chair rise test.

J Rehabil Med 2014; 46: 00–00

Correspondence address: Han Boter, University of Groningen, University Medical Centre Groningen, Department of Epidemiology (HPC: FA40), PO Box 30001, NL-9700 RB Groningen, The Netherlands. E-mail:

Accepted Feb 4, 2014; Epub ahead of print May 8, 2014


Fatigue can compromise quality of life. In population studies up to 46% of individuals reported fatigue and, in cancer, multiple sclerosis, and stroke patients, the incidence of fatigue can reach 92% (1–4). After adjustments for confounding factors, such as age, sex, weight, chronic diseases, and depressive symptoms, longitudinal primary care and population studies showed that fatigue predicts outcomes such as poor self-rated health (5); use of home help (6); hospitalization (6); angina pectoris (7); non-fatal myocardial infarction (7); up to 15-year mortality (5, 8); mobility disability (8–10); and disability with regard to activities of daily living (5, 10). These poor health outcomes associated with fatigue and its disabling nature underline the importance of preventing and treating fatigue.

However, fatigue is an ambiguous concept. It is a summed reaction to biological processes after intensive or prolonged work because metabolic demands exceed energy resources. In addition, fatigued individuals report low energy, vitality, cognitive slowness, depression, anxiety and a strong desire to sleep (3). Population studies show that middle-aged and old individuals report fatigue at a significantly higher rate than young individuals (11, 12). Yet, most fatigue measures are designed to quantify fatigue in patients who have a specific condition such as cancer or stroke. However, when measuring fatigue in the general population, as in the present study, the idea is to measure critical aspects of fatigue relevant to each individual, regardless of whether a specific disease, medication or a psychological problem is the cause of fatigue.

To assess fatigue both unidimensional and multidimensional self-report instruments have been developed that include disease-specific tools. Examples of unidimensional fatigue measures are the Fatigue Severity Scale (13) and the Mobility-Tiredness Scale (6); examples of multidimensional instruments are the Fatigue Scale (14) and the Multidimensional Fatigue Inventory (MFI-20) (15). Unlike unidimensional measures, multidimensional scales of fatigue facilitate the characterization of subjects with respect to the nature of their fatigue, and thus support condition-specific interventions. Of the self-reported multidimensional fatigue measures, the MFI-20 is probably the one most frequently used in Europe, especially in cancer patients (12).

The objective of the present study was to determine the association between MFI-20 subscales and physical function in a subset of 4,964 late mid-life participants in the Copenhagen Aging and Midlife Biobank (CAMB) cohort. An examination of the associations between the MFI-20 subscales and grip strength and chair rise performance at an early stage, such as late mid-life, is a logical strategy for the development and prescription of (cost-) effective preventive programmes to decrease fatigue and other poor outcomes in high-risk individuals.


Study sample and procedure

The CAMB cohort has been described previously (16). Briefly, it comprises late mid-life participants from 3 cohorts: (i) the Metropolit Study; 7,750 men born in Copenhagen in 1953 (17); (ii) The Copenhagen Perinatal Cohort; 5,282 men and women born in the National University Hospital in Copenhagen in 1959–1961 (18); and (iii) The Danish Longitudinal Study on Work, Unemployment and Health; a random sample of 4,906 men and women born in 1949 and in 1959 (19). Subjects living outside Eastern Denmark were excluded. For safety reasons participants with high blood pressure (BP) (i.e. diastolic blood pressure > 100 mmHg or systolic blood pressure > 160 mmHg) did not perform the chair rise assessment since that requires high physical exertion that may lead to a further rise in BP.

From 2009 to 2011, eligible subjects received a postal questionnaire and a letter with information about the CAMB survey. In addition, subjects were asked to visit the National Research Centre for the Working Environment and participate in a health examination and physical assessments. During the visit the subjects received oral information about the study, and both oral and written informed consent was obtained. Non-respondents were sent a reminder 4 weeks after the first invitation, which was repeated at the end of the data collection period if subjects had not responded. The study protocol was approved by the local ethics committee (number H-A-2008-126) and the Danish Data Protection Agency (number 2008-41-2938).

Of the 17,937 invited subjects, 4,964 (28%) completed the postal questionnaire, the chair rise, and the handgrip test. Compared with non-participants, participants did not differ much with regard to educational levels, but higher proportions of participants were employed (16). However, the number of general practitioner visits in 2009 did not differ between participants and non-participants, which suggested that participants do not have better health (16).


Self-reported fatigue. Fatigue was assessed with the self-reported MFI-20, comprising 5 subscales: general fatigue, physical fatigue, reduced activity, reduced motivation, and mental fatigue (15). Each subscale consists of 4 items rated on a 5-point Likert-scale (1 = yes, it is true; 5 = no, it is not true) with positively and negatively phrased items. The items are summed to calculate a subscale score with a theoretical range of 4–20 (higher scores indicate more fatigue). Examples of the items include: “I feel tired” (general fatigue); “Physically I feel I am in a bad condition” (physical fatigue); “I get little done” (reduced activity); “I dread having to do things” (reduced motivation); and “It takes a lot of effort to concentrate on things” (mental fatigue). Several studies report the MFI-20 as a feasible, reliable, and valid measure (12, 15, 20).

Physical function. Physical function was measured by two tests: (i) the maximal number of chair rises the subject was able to perform in 30 s, with their arms crossed against their chest and their knees fully extended before starting to descend (21). During the test, the subject’s back did not need to touch the backrest. The test was performed using a chair with a seat-height of 45 cm. A mechanical contact was connected to a signal-conditioning interface that automatically recorded the number of transitions. (ii) Maximal handgrip strength, in kg, of the dominant hand was measured with a dynamometer, as well as being recorded automatically (22). Subjects were instructed to squeeze their hand as hard and quickly as they could for about 3 s. Subsequently, participants could rest for 30 s until the next trial. Subjects completed 3–5 trials according to the following rule: If the force of the third handgrip trial exceeded both preceding trials by more than 5%, subjects performed a fourth trial. A fifth trial was done if the force of the fourth trial exceeded the third trial by more than 5%. The highest score on the 3–5 trials was selected for the analyses.

Health examination and assessment of covariates. The health examination included the measurement of height, body mass, and blood pressure. Blood pressure was measured to assess the subject’s eligibility for participation in the chair rise test. In addition to sociodemographic data (age, sex, education, and living situation), subjects’ occupational social class was derived by classifying occupation into social classes I–VI, according to the Danish occupational social class classification standard. Accordingly, social classes I–V comprise economically active individuals ranging from professional occupation in social class I to unskilled occupation in social class V. Social class VI represents people on public transfer income, including sickness benefits and disability pension. Depressive symptoms were assessed by the Major Depression Inventory (MDI), which comprises 10 items (including 2 sub-items) with a 6-point Likert-scale to measure symptom frequency during the previous 2 weeks (0 = at no time; 5 = all the time; theoretical score range 0–50, higher scores indicate higher levels of depressive symptoms) (23). To assess present morbidity and other health problems, such as stroke and osteoarthritis, 19 items from the Danish National Health Interview Survey were used and 2 questions were added on kidney stones and gallstones (a total of 21 items) (23, 24).

Statistical analyses

Before calculating sum scores of the MFI-20 subscales, per participant missing responses were replaced by the respondent’s mean scores for completed items of the same subscale when at least half of the subscale responses were valid. Likewise, total scores were calculated for the MDI. No other data were imputed.

The independent physical function measures were dichotomized at their sex-specific medians (see notes in Tables II and III for more details). Since MFI-20 subscale scores as the outcome variables did not meet the linear regression assumptions, it was decided to perform multivariate logistic regression analyses and use sex-specific medians as cut-offs (see notes in Tables II and III for more details). In addition to analyses of the total group, analyses were stratified by sex and adjusted for age, height, body mass and, in the total group only, sex (model 1). In subsequent models the following covariates were added: present morbidity (model 2) and present morbidity and depressive symptoms (model 3). The selection of the covariates age, height, body mass, sex, present morbidity, and depressive symptoms was based on previous studies (5–10, 25, 26). In post-hoc analyses gender differences were studied by entering the Sex × Physical function interaction terms in models 1–3.

All statistical analyses were performed with SPSS 20. The level of significance was set at p = 0.05.



Table I shows that the 4,964 participants comprise 2 age-groups: 2,175 subjects, age range 49–53 years (44%) and 2,789 subjects, age range 56–63 years (56%); with a median age of 50 years for women and 57 years for men. Of the subjects, 90% had a job, 55% were overweight (BMI ≥ 25) and two-thirds had one or more diseases. Concerning subscales of self-reported fatigue, higher median levels were reported on general fatigue and physical fatigue (9.0) in comparison with the other 3 subscales (7.0). The mean value for handgrip strength was lower in women (31.3 kg) than men (49.6 kg), but the results on the chair rise test appeared similar (21.2 and 22.0).

Table I. Characteristics of participants, by sex


(n = 4,964)


(n = 3,331)


(n = 1,633)


49–53 years, n (%)

2,175 (44)

904 (27)

1,271 (78)

56–63 years, n (%)

2,789 (56)

2,427 (73)

362 (22)

Median age (IQR; range)

56 (50–57; 49–63)

57 (51–57; 49–63)

50 (50–52; 49–63)

Living alone, n (%)


4,180 (84)

2,818 (85)

1,362 (83)


765 (15)

498 (15)

267 (16)


19 (< 1)

15 (< 1)

4 (< 1)

Occupational social class, n (%)a

Class 1 and 2

2,073 (42)

1,466 (44)

607 (37%)

Class 3 to 5

2,367 (48)

1,522 (46)

845 (52%)

Class 6

469 (9)

315 (9)

154 (9%)


55 (1)

28 (1)

27 (2%)

Present smoker, n (%)b


3,812 (77)

2,548 (76)

1,264 (77%)


1,147 (23)

779 (23)

368 (23%)


5 (< 1)

4 (< 1)

1 (< 1%)

Height, cm, mean (SD)

175.2 (8.9)

179.6 (6.5)

166.3 (6.2)

Body mass, kg, mean (SD)

80.0 (15.3)

85.0 (13.6)

69.6 (13.3)

BMI (kg/m2):

Mean (SD)

26.0 (4.2)

26.3 (3.8)

25.2 (4.7)

Overweight (≥ 25), n (%)

2,747 (55)

2,029 (61)

718 (44)

Present morbidity: n (%)c

3,344 (67)

2,222 (67)

1,122 (69)


1,045 (21)

644 (19)

401 (25)


994 (20)

634 (19)

360 (22)


863 (17)

638 (19)

225 (14)

Depressive symptoms – MDI, median (IQR; range)d

5.0 (3.0–9.0; 0–48)

5.0 (2.0–8.0; 0–48)

6.0 (3.0–10.0; 0–48)

Fatigue – MFI-20e

General fatigue

Mean (SD)

Median (IQR; range)

9.6 (3.9)

9.0 (6.0–12.0; 4–20)

9.3 (3.8)

9.0 (6.0–12.0 ; 4–20)

10.1 (4.2)

10.0 (7.0–13.0; 4–20)

Physical fatigue

Mean (SD)

Median (IQR; range)

9.1 (3.8)

9.0 (6.0–12.0; 4–20)

8.9 (3.7)

8.0 (6.0–11.0; 4–20)

9.6 (4.0)

9.0 (6.0–12.0; 4–20)

Reduced activity

Mean (SD)

Median (IQR; range)

7.9 (3.3)

7.0 (5.0–10.0; 4–20)

8.0 (3.3)

7.0 (6.0–10.0; 4–20)

7.9 (3.5)

7.0 (5.0–10.0; 4–20)

Reduced motivation

Mean (SD)

Median (IQR; range)

7.2 (2.5)

7.0 (5.0–9.0; 4–20)

7.3 (2.5)

7.0 (5.0–9.0; 4–20)

6.9 (2.5)

6.0 (5.0–8.0; 4–18)

Mental fatigue

Mean (SD)

Median (IQR; range)

7.7 (3.3)

7.0 (5.0–9.0; 4–20)

7.6 (3.2)

7.0 (5.0–9.0; 4–20)

7.8 (3.5)

7.0 (5.0–10.0; 4–20)

Physical function, mean (SD)

Handgrip strength, kg

43.5 (11.4)

49.6 (8.4)

31.3 (5.3)

Chair rise, n/30s

21.7 (5.7)

22.0 (5.7)

21.2 (5.7)

aClass 1–2 – job requires ≥ 3 years of theoretical or university training or management of ≥ 11 subordinates; class 3–5 – job requires no to 1.5 years of theoretical training or management control of ≤ 10 subordinates; class 6 – economically inactive (transfer income); other – students, housewives/husbands, and individuals with no or insufficient job information.

bDaily and not daily smoking.

c≥ 1 self-reported somatic or psychiatric disease or health problem; including the 3 most prevalent diseases or health problems.

dMajor Depression Inventory (MDI, theoretical range 0–50, higher scores indicate higher levels of depressive symptoms).

eMultidimensional Fatigue Inventory (MFI-20, higher scores indicate more self-reported fatigue).

IQR: interquartile range; SD: standard deviation.

Handgrip strength

In the total group the associations were the strongest between handgrip strength and self-reported physical fatigue (Table II). In contrast to associations with other subscales of self-reported fatigue, the associations of handgrip strength with physical fatigue and reduced motivation were still statistically significant after adjustments for age, sex, body mass, height, present morbidity, and depressive symptoms in the total group (model 3 – physical fatigue; odds ratio (OR) 0.75 (95% confidence interval (CI) 0.66–0.86); p ≤ 0.001; – reduced motivation; OR 0.85 (95% CI 0.75–0.96); p ≤ 0.05). This suggests that participants with a stronger handgrip were 25% less likely to report physical fatigue and 15% less likely to report reduced motivation than participants with a weaker handgrip. Without adjustments for depressive symptoms (models 1 and 2), handgrip strength, in the total group and in men and women separately, was statistically significantly associated with all subscales of self-reported fatigue, except for mental fatigue (total group, men, and women) and reduced activity in women.

Table II. Odds ratios (OR) and 95% confidence intervals (95% CI) from multiple logistic regression analyses of handgrip strength with dimensions of self-reported fatigue, by sex

General fatigue

OR (95% CI)

Physical fatigue

OR (95% CI)

Reduced activity

OR (95% CI)

Reduced motivation

OR (95% CI)

Mental fatigue

OR (95% CI)

Total (n = 4,964)

Model 1

0.79 (0.70–0.89)***

0.70 (0.62–0.79)***

0.85 (0.75–0.95)**

0.78 (0.69–0.87)***

0.90 (0.80–1.01)

Model 2

0.79 (0.70–0.89)***

0.70 (0.62–0.79)***

0.85 (0.75–0.95)**

0.78 (0.69–0.88)***

0.90 (0.80–1.01)

Model 3

0.89 (0.77–1.04)

0.75 (0.66–0.86)***

0.94 (0.83–1.07)

0.85 (0.75–0.96)*

1.03 (0.90–1.18)

Men (n = 3,331)

Model 1

0.82 (0.71–0.95)**

0.72 (0.62–0.83)***

0.86 (0.74–0.99)*

0.78 (0.67–0.90)***

0.88 (0.76–1.01)

Model 2

0.82 (0.70–0.94)**

0.71 (0.61–0.83)***

0.86 (0.74–0.99)*

0.78 (0.67–0.90)***

0.88 (0.76–1.01)

Model 3

0.94 (0.79–1.13)

0.77 (0.66–0.91)**

0.96 (0.82–1.12)

0.85 (0.73–0.99)*

1.00 (0.85–1.17)

Women (n = 1,633)

Model 1

0.71 (0.57–0.88)**

0.66 (0.53–0.82)***

0.81 (0.66–1.01)

0.75 (0.61–0.93)**

0.93 (0.76–1.15)

Model 2

0.71 (0.57–0.89)**

0.66 (0.53–0.83)***

0.82 (0.66–1.02)

0.75 (0.61–0.93)**

0.94 (0.76–1.16)

Model 3

0.77 (0.58–1.01)

0.71 (0.56–0.90)**

0.91 (0.72–1.16)

0.82 (0.65–1.03)

1.10 (0.87–1.40)

*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001.

Sex-specific cut-offs (medians) for: General fatigue ≤ 10 (= ”low”) vs > 10 (= ”high”) (females), ≤ 9 (= ”low”) vs > 9 (= ”high”) (males); Physical fatigue ≤ 9 (= ”low”) vs > 9 (= ”high”) (females), ≤ 8 (= ”low”) vs > 8 (= ”high”) (males); Reduced activity ≤ 7 (= ”low”) vs > 7 (= ”high”) (females/males); Reduced motivation ≤ 6 (= ”low”) vs > 6 (= ”high”) (females), ≤ 7 (= ”low”) vs > 7 (= ”high”) (males); Mental fatigue ≤ 7 (= ”low”) vs > 7 (= ”high”) (females/males); Handgrip strength ≤ 31.1 (= ”weak”) vs > 31.1 (= ”strong”) (females), ≤ 49.4 (= ”weak”) vs > 49.4 (= ”strong”) (males).

Model 1: adjusted for age, body mass, height, and – in total group only – sex; Model 2: adjusted for age, body mass, height, sex (total group only), and present morbidity; Model 3: adjusted for age, body mass, height, sex (total group only), present morbidity, and depressive symptoms.

Chair rise test

Table III shows that in the total group, after adjustments for age, sex, body mass, height, present morbidity, and depressive symptoms (model 3), results on the chair rise test were statistically significantly associated with self-reported physical fatigue (OR 0.61 (95% CI 0.53–0.69); p ≤ 0.001), general fatigue (OR 0.72 (95% CI 0.62–0.84); p ≤ 0.001), reduced activity (OR 0.79 (95% CI 0.70–0.90); p ≤ 0.001) and reduced motivation (OR 0.84 (95% CI 0.74–0.95); p ≤ 0.01). For example, in contrast to subjects with a poor performance on the chair rise test, subjects with a good performance had a 39% lower risk to report physical fatigue, 28% lower risk to report general fatigue, 21% lower risk to have a reduced activity level, and 16% lower risk to report reduced motivation. Without adjustments for depressive symptoms (models 1 and 2), all associations between chair rise performance and subscales of self-reported fatigue were statistically significant in the total group and in men and women, except for mental fatigue in women.

Table III. Odds ratios (OR) and 95% confidence intervals (95% CI) from multiple logistic regression analyses of the chair rise test with dimensions of self-reported fatigue, by sex

General fatigue

OR 95% CI

Physical fatigue

OR 95% CI

Reduced activity

OR 95% CI

Reduced motivation

OR 95% CI

Mental fatigue

OR 95% CI

Total (n = 4,964)

Model 1

0.62 (0.55–0.70)***

0.54 (0.48–0.61)***

0.68 (0.61–0.77)***

0.73 (0.65–0.82)***

0.83 (0.74–0.93)**

Model 2

0.64 (0.57–0.72)***

0.55 (0.49–0.62)***

0.70 (0.62–0.78)***

0.74 (0.66–0.83)***

0.85 (0.76–0.95)**

Model 3

0.72 (0.62–0.84)***

0.61 (0.53–0.69)***

0.79 (0.70–0.90)***

0.84 (0.74–0.95)**

1.02 (0.90–1.16)

Men (n = 3,331)

Model 1

0.64 (0.55–0.74)***

0.55 (0.47–0.63)***

0.71 (0.62–0.82)***

0.76 (0.66–0.88)***

0.89 (0.77–1.02)

Model 2

0.66 (0.57–0.76)***

0.56 (0.48–0.65)***

0.72 (0.63–0.83)***

0.77 (0.67–0.89)***

0.90 (0.78–1.04)

Model 3

0.71 (0.60–0.85)***

0.60 (0.52–0.71)***

0.81 (0.69–0.94)**

0.86 (0.74–1.00)

1.07 (0.91–1.25)

Women (n = 1,633)

Model 1

0.59 (0.48–0.73)***

0.53 (0.43–0.65)***

0.64 (0.52–0.79)***

0.66 (0.54–0.82)***

0.72 (0.59–0.89)**

Model 2

0.61 (0.50–0.76)***

0.54 (0.44–0.67)***

0.65 (0.53–0.81)***

0.68 (0.55–0.83)***

0.74 (0.60–0.91)**

Model 3

0.74 (0.56–0.96)*

0.61 (0.49–0.77)***

0.77 (0.61–0.97)*

0.80 (0.63–1.00)*

0.91 (0.72–1.15)

*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001.

Sex-specific cut-offs (medians) for: General fatigue ≤ 10 (= ”low”) vs >10 (= ”high”) (females), ≤ 9 (= ”low”) vs > 9 (= ”high”) (males); Physical fatigue ≤ 9 (= ”low”) vs > 9 (= ”high”) (females), ≤ 8 (= ”low”) vs > 8 (= ”high”) (males); Reduced activity ≤ 7 (= ”low”) vs > 7 (= ”high”) (females/males); Reduced motivation ≤ 6 (=”low”) vs > 6 (= ”high”) (females), ≤ 7 (=”low”) vs > 7 (= ”high”) (males); Mental fatigue ≤ 7 (= ”low”) vs > 7 (= ”high”) (females/males); Chair rise test ≤ 20.6 (= ”poor function”) vs > 20.6 (= ”good function”) (females), ≤ 21.8 (= ”poor function”) vs > 21.8 (= ”good function”) (males).

Model 1: adjusted for age, body mass, height, and – in total group only – sex; Model 2: adjusted for age, body mass, height, sex (total group only), and present morbidity; Model 3: adjusted for age, body mass, height, sex (total group only), present morbidity, and depressive symptoms.

Post-hoc analyses

Post-hoc analyses including interaction terms for sex by physical function measures showed that all terms were statistically non-significant.


The main finding in this population study in late mid-life men and women is that, after adjusting for potential confounding variables, physical function, as measured by handgrip strength and the chair rise test, was most strongly associated with the MFI-20 physical fatigue subscale. Apparently the self-reported physical fatigue subscale reflects both muscle strength as measured by the handgrip test and also a broader concept of physical function, i.e. the chair rise test, including muscle strength, balance and mobility (27, 28), and maybe also aerobic capacity. In addition, the association between function and self-reported fatigue is independent of gender in this late mid-life cohort.

The present study expands our knowledge of the physiological basis of the MFI-20 subscales, since, except for a study in patients with fibromyalgia, to our knowledge no population studies have examined the associations between these subscales and measures of physical function (29). In addition, it complements previous cross-sectional population studies of other self-reported fatigue measures that are significantly associated with measures of physical function, such as handgrip strength (30, 31) and the chair rise test (32). To represent physical function, we selected the widely used handgrip strength and chair rise tests that predict cognitive decline, disability, and mortality (25, 26, 33, 34). Handgrip strength is strongly associated with other measures of muscle strength and can be considered as a marker of physiological reserve in middle-aged and older adults (25, 26). We also used the chair rise test because it is a robust measure of physical function and quantifies the function of physiological systems associated with muscle strength and aerobic capacity. Performance in chair rise is also a strong indicator of mobility (27, 28).

This study shows that, after adjustments, the chair rise compared with the grip strength test was not only more strongly associated with the self-reported MFI-20 physical fatigue subscale, but also with general fatigue, reduced activity, and reduced motivation. These data suggest that these 4 subscales of self-reported fatigue also involve physical function in a broader sense. Rising from a chair compared with gripping requires the activation of a substantially larger muscle mass. In older adults especially, the activation of large muscle volume necessitates an effort that represents a relatively high proportion of the maximum capacity (35). In contrast with single, brief efforts of gripping, chair rise performance includes the repetitive activation of a larger muscle volume and is a more sensitive and comprehensive measure of physical function, including physical dimensions of fatigue. In total, the association between physical function as measured by grip strength and chair rise provides a physiological basis for MFI-20, especially for its physical dimensions.

Furthermore, adding morbidity to the first model did not affect the odds ratios. This is unexpected because previous studies in (older) adults showed that morbidity is associated with both self-reported fatigue and physical function (11, 36, 37). However, this finding might be explained by the included adjustments for age, body mass, height, and sex.

After the adjustments for confounders and adding depressive symptoms to model 3, the associations of the chair rise test with self-reported mental fatigue were no longer statistically significant. This suggests that depressive symptoms are an important factor in the association and that this subscale reflects mental components of perceived fatigue. Also, many previous population studies, with data from children and (older) adults, found depressive symptoms to be significantly associated with all MFI-20 subscales (11, 12, 20, 38, 39). The present findings are also congruent with previous data, showing that in older adults depressive symptoms, together with morbidity and sleep quality, substantially weakened the association between handgrip strength and a self-reported fatigue measure (31).

The last finding reflects the results of the post-hoc analyses of the interaction between sex and physical function. These analyses were performed because (i) previous studies reported gender differences on the MFI-20 subscales, handgrip strength, and chair rise test, and (ii) stratification was done in studies on the association between handgrip strength and self-reported fatigue (11, 12, 20, 31, 34, 38, 39). However, the present study showed that the association of physical function with subscales of self-reported fatigue is not different between men and women.

The present study has some advantages compared with previous studies on (self-reported) fatigue and physical function. To our knowledge, this is the first population study that linked measures of physical function with subscales of fatigue as measured by the MFI-20: it showed the difference between the associations of handgrip strength and the chair rise test vs more physical or psychological subscales of self-reported fatigue. Also, support for this cohort to be representative for the Danish population is found in another Danish population study that had a comparable mean age and similar mean scores of BMI and handgrip strength (40). Compared with men, women in the present study had much weaker handgrip strength but only slightly poorer performance on the chair rise test. This is congruent with previous studies, including the above-mentioned Danish population study and a study of a representative middle-aged group in the UK (37, 40, 41). However, a previous population study reported lower mean MFI-20 subscale scores for the age group of 40–59 years (interquartile range (IQR) 7.1–8.0 in men and 7.8–8.7 in women), than found in the present study (IQR 7.3–9.3 in men and 6.9–10.1 in women), which might be explained by the age-discrepancy (12). Other strengths of the present study are the large sample sizes for men and women, the age span limited to late mid-life, and the use of sex-specific cut-offs for each subscale that increased the statistical power of the analyses.

A limitation of the study is the quite low response rate. However, non-participants did not differ from participants on education level and the number of general practitioner visits in 2009, suggesting that, to some extent, it is possible to generalize the results to the general population. To delineate the direction of the association between physical function and self-reported fatigue, future follow-up studies should assess whether non-fatigued subjects with poor physical function have a higher incidence of becoming fatigued, than non-fatigued subjects with good physical function.

In conclusion, physical function, as measured by handgrip strength and the chair rise test, is associated with self-reported physical fatigue, general fatigue, reduced activity, and reduced motivation. Therefore this study supports the validity of the MFI-20. Knowledge about the features of physical functioning that are associated with subtypes of self-reported fatigue may contribute to the development of specific (preventive) interventions.


The authors thank the staff at the Department of Public Health and the National Research Center for the Working Environment who undertook the data collection. Further thanks to Helle Bruunsgaard, Nils-Erik Fiehn, Poul Holm-Pedersen, Rikke Lund, Erik Lykke Mortensen and Merete Osler who initiated and established the Copenhagen Aging and Midlife Biobank 2009–2011 together with Kirsten Avlund and Åse Marie Hansen. The authors acknowledge the crucial role of the initiators and steering groups of the Metropolit Cohort, The Copenhagen Perinatal Cohort and The Danish Longitudinal Study on Work Unemployment and Health. The authors also thank Glyn Howatson, PhD, Faculty of Health and Life Sciences, Northumbria University, Newcastle upon Tyne, UK and Joanne Smith, PhD, Center for Human Movement Sciences, University of Groningen, Groningen, The Netherlands for checking the text for English.

The Copenhagen Aging and Midlife Biobank has been supported by a generous grant from the Velux Foundation. The present study was supported by a grant from the Nordea Foundation.

The authors declare no conflicts of interest.


1. Lewis G, Wessely S. The epidemiology of fatigue: more questions than answers. J Epidemiol Community Health 1992; 46: 92–97.

2. Duncan F, Wu S, Mead GE. Frequency and natural history of fatigue after stroke: a systematic review of longitudinal studies. J Psychosom Res 2012; 73: 18–27.

3. Alexander NB, Taffet GE, Horne FM, Eldadah BA, Ferrucci L, Nayfield S, et al. Bedside-to-Bench conference: research agenda for idiopathic fatigue and aging. J Am Geriatr Soc 2010; 58: 967–975.

4. Chaudhuri A, Behan PO. Fatigue in neurological disorders. Lancet 2004; 363: 978–988.

5. Moreh E, Jacobs JM, Stessman J. Fatigue, function, and mortality in older adults. J Gerontol A Biol Sci Med Sci 2010; 65: 887–895.

6. Avlund K, Damsgaard MT, Schroll M. Tiredness as determinant of subsequent use of health and social services among nondisabled elderly people. J Aging Health 2001; 13: 267–286.

7. Appels A, Mulder P. Fatigue and heart disease. The association between ‘vital exhaustion’ and past, present and future coronary heart disease. J Psychosom Res 1989; 33: 727–738.

8. Schultz-Larsen K, Avlund K. Tiredness in daily activities: a subjective measure for the identification of frailty among non-disabled community-living older adults. Arch Gerontol Geriatr 2007; 44: 83–93.

9. Avlund K, Sakari-Rantala R, Rantanen T, Pedersen AN, Frandin K, Schroll M. Tiredness and onset of walking limitations in older adults. J Am Geriatr Soc 2004; 52: 1963–1965.

10. Avlund K, Damsgaard MT, Sakari-Rantala R, Laukkanen P, Schroll M. Tiredness in daily activities among nondisabled old people as determinant of onset of disability. J Clin Epidemiol 2002; 55: 965–973.

11. Watt T, Groenvold M, Bjorner JB, Noerholm V, Rasmussen NA, Bech P. Fatigue in the Danish general population. Influence of sociodemographic factors and disease. J Epidemiol Community Health 2000; 54: 827–833.

12. Schwarz R, Krauss O, Hinz A. Fatigue in the general population. Onkologie 2003; 26: 140–144.

13. Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 1989; 46: 1121–1123.

14. Chalder T, Berelowitz G, Pawlikowska T, Watts L, Wessely S, Wright D, et al. Development of a fatigue scale. J Psychosom Res 1993; 37: 147–153.

15. Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995; 39: 315–325.

16. Avlund K, Osler M, Mortensen EL, Christensen U, Bruunsgaard H, Holm-Pedersen P, et al. Copenhagen Aging and Midlife Biobank (CAMB). An introduction. J Aging Health 2014; 26: 5–20.

17. Osler M, Lund R, Kriegbaum M, Christensen U, Andersen AM. Cohort profile: the Metropolit 1953 Danish male birth cohort. Int J Epidemiol 2006; 35: 541–545.

18. Zachau-Christiansen B. Development during the first year of life. Helsingor: Poul Andersens Forlag; 1972.

19. Christensen U, Lund R, Damsgaard MT, Holstein BE, Ditlevsen S, Diderichsen F, et al. Cynical hostility, socioeconomic position, health behaviors, and symptom load: a cross-sectional analysis in a Danish population-based study. Psychosom Med 2004; 66: 572–577.

20. Lin JM, Brimmer DJ, Maloney EM, Nyarko E, Belue R, Reeves WC. Further validation of the Multidimensional Fatigue Inventory in a US adult population sample. Popul Health Metr 2009; 7: 18.

21. Jones CJ, Rikli RE, Beam WC. A 30-s chair-stand test as a measure of lower body strength in community-residing older adults. Res Q Exerc Sport 1999; 70: 113–119.

22. Schaubert KL, Bohannon RW. Reliability and validity of three strength measures obtained from community-dwelling elderly persons. J Strength Cond Res 2005; 19: 717–720.

23. Bech P, Rasmussen NA, Olsen LR, Noerholm V, Abildgaard W. The sensitivity and specificity of the Major Depression Inventory, using the Present State Examination as the index of diagnostic validity. J Affect Disord 2001; 66: 159–164.

24. Ekholm O, Hesse U, Davidsen M, Kjoller M. The study design and characteristics of the Danish national health interview surveys. Scand J Public Health 2009; 37: 758–765.

25. Rantanen T, Guralnik JM, Foley D, Masaki K, Leveille S, Curb JD, et al. Midlife hand grip strength as a predictor of old age disability. JAMA 1999; 281: 558–560.

26. Rantanen T, Masaki K, He Q, Ross GW, Willcox BJ, White L. Midlife muscle strength and human longevity up to age 100 years: a 44-year prospective study among a decedent cohort. Age (Dordr) 2012; 34: 563–570.

27. Bean JF, Herman S, Kiely DK, Frey IC, Leveille SG, Fielding RA, et al. Increased Velocity Exercise Specific to Task (InVEST) training: a pilot study exploring effects on leg power, balance, and mobility in community-dwelling older women. J Am Geriatr Soc 2004; 52: 799–804.

28. Bean JF, Leveille SG, Kiely DK, Bandinelli S, Guralnik JM, Ferrucci L. A comparison of leg power and leg strength within the InCHIANTI study: which influences mobility more? J Gerontol A Biol Sci Med Sci 2003; 58: 728–733.

29. Ericsson A, Bremell T, Mannerkorpi K. Usefulness of multiple dimensions of fatigue in fibromyalgia. J Rehabil Med 2013; 45: 685–693.

30. Manty M, Ekmann A, Thinggaard M, Christensen K, Avlund K. Fatigability in basic indoor mobility in nonagenarians. J Am Geriatr Soc 2012; 60: 1279–1285.

31. Vestergaard S, Nayfield SG, Patel KV, Eldadah B, Cesari M, Ferrucci L, et al. Fatigue in a representative population of older persons and its association with functional impairment, functional limitation, and disability. J Gerontol A Biol Sci Med Sci 2009; 64: 76–82.

32. Silva JP, Pereira DS, Coelho FM, Lustosa LP, Dias JM, Pereira LS. Clinical, functional and inflammatory factors associated with muscle fatigue and self-perceived fatigue in elderly community-dwelling women. Rev Bras Fisioter 2011; 15: 241–248.

33. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med 1995; 332: 556–561.

34. Auyeung TW, Lee JS, Kwok T, Woo J. Physical frailty predicts future cognitive decline – a four-year prospective study in 2737 cognitively normal older adults. J Nutr Health Aging 2011; 15: 690–694.

35. Hortobagyi T, Mizelle C, Beam S, DeVita P. Old adults perform activities of daily living near their maximal capabilities. J Gerontol A Biol Sci Med Sci 2003; 58: M453–M460.

36. Avlund K, Rantanen T, Schroll M. Factors underlying tiredness in older adults. Aging Clin Exp Res 2007; 19: 16–25.

37. Kuh D, Bassey EJ, Butterworth S, Hardy R, Wadsworth ME, Musculoskeletal Study Team. Grip strength, postural control, and functional leg power in a representative cohort of British men and women: associations with physical activity, health status, and socioeconomic conditions. J Gerontol A Biol Sci Med Sci 2005; 60: 224–231.

38. Valentine RJ, Woods JA, McAuley E, Dantzer R, Evans EM. The associations of adiposity, physical activity and inflammation with fatigue in older adults. Brain Behav Immun 2011; 25: 1482–1490.

39. de Rijk AE, Schreurs KM, Bensing JM. What is behind “I’m so tired”? Fatigue experiences and their relations to the quality and quantity of external stimulation. J Psychosom Res 1999; 47: 509–523.

40. Aadahl M, Beyer N, Linneberg A, Thuesen BH, Jorgensen T. Grip strength and lower limb extension power in 19–72-year-old Danish men and women: the Health2006 study. BMJ Open 2011; 1: e000192.

41. Okamoto N, Nakatani T, Morita N, Saeki K, Kurumatani N. Home-based walking improves cardiopulmonary function and health-related QOL in community-dwelling adults. Int J Sports Med 2007; 28: 1040–1045.


Do you want to comment on this paper? The comments will show up here and if appropriate the comments will also separately be forwarded to the authors. You need to login/create an account to comment on articles. Click here to login/create an account.