OBJECTIVE: To describe functional limitations in obese women.
DESIGN: Comparisons of functional limitations in obese women and in a normal-weight reference group regarding mobility, balance and housework transport.
SETTING: A large room with a corridor and staircase nearby.
SUBJECTS: Fifty-seven consecutively selected obese female outpatients, mean age 44 y, body mass index (BMI) 37 kg·m−2, and 22 voluntary references, mean age 49 y, BMI 22 kg·m−2.
MAIN OUTCOME MEASURES: Pain questionnaire (for background data); a new valid and reliable test protocol measuring activities of daily life (ADL), pain and exertion, heart rate, balance and speed.
RESULTS: The patients had problems in performing 13 of 16 actions compared with the references (P-values between 0.05 and 0.001). They moved more slowly and had difficulties and /or pain and exertion in flexibility tasks, balancing, activities at floor level (bending and kneeling), climbing stairs and carrying groceries. BMI (P=0.002), age (P=0.02) and current pain (P=0.02) from the lower body predicted functional limitation.
CONCLUSIONS: The functional test reflected functional limitations in obese women, revealing difficulties and perceived pain in ADL. The test protocol can be used to evaluate the effects of training or dieting. For clinical use we recommend considering BMI, age and current pain for use when individualizing advice on training.
Obesity, defined as a body mass index (BMI) of >30 kg·m−2, is an increasing health problem in developed countries.1,2,3,4,5 The cause is multifactorial but the increase is regarded as a result of today's sedentary lifestyle.2,6,7 The negative influence of obesity on health status with higher mortality, cardiovascular disease, diabetes, osteoarthritis, poorer quality of life etc is well documented.6,8,9 Being unfit is considered even more dangerous to health than being obese.10 Being female, in poor physical condition and obese not only increases medical risks,10,11 it also decreases social and physical activity.12,13,14,15
Since losing weight and maintaining the loss are very difficult in practice, maybe advice should focus on fitness training to decrease medical risks, increase energy and preserve physical functions.16 The most simple form of fitness training is to be physically active in everyday life. It is not necessary to visit a gym three times a week; 30 minutes of activity a day, such as walking, climbing stairs, doing housework, gardening etc at an intensity level of 40–60% VO2max (maximum oxygen capacity) are recommended for acquiring and maintaining moderate fitness.17
Discrimination against obese persons is common in work settings and employers are reluctant to hire obese women,18 who are generally considered to have low efficiency, tire easily and be prone to illness.6,13,14,18,19 Physicians in the health care system often have negative attitudes toward the obese.4,20,21 They advise their obese patients to lose weight by eating less and with low-intensity physical training such as brisk walking, before other measures can be taken.15,22 Such advice is naturally not easy to follow since most obese patients are already on a diet, and brisk walking can have negative side effects in obese women.23 A previous study showed that obese women used 56% (range 31–98%) of their VO2max and experienced heavy exertion while walking indoors for about 5 min at their own pace.23 This could explain why many obese women find walking too strenuous for keeping fit or losing weight.
Many obese women are unfit and physically inactive.13,23 Apart from self-reports where current BMI was associated with poor physical function in women,7 the impact of obesity on functional health and disability has not been closely investigated.24 We lack suitable instruments, that are quick and easy to administer in the clinic, to measure or judge the grade of disability in these patients. Our clinical experience is that the drop-out rate is high and compliance concerning physical activity is low in obese subjects. We therefore need more knowledge of common barriers and functional limitations in obese women before giving further advice on fitness training and physical activity.
The aim of this study was to describe functional limitations in obese women and in a normal-weight reference group and to find functional and /or medical predictors of low compliance with ordinary fitness training.
As used in this study, according to Verbrugge and Jette,25 functional limitations refer to restrictions in performing fundamental physical actions used in daily life by one's own age—sex group. The intention of including some activities of daily life (ADL) in the functional test was primarily to measure capability to reach etc, not to reflect degree of disability.25
Obese female outpatients applying for participation in weight-loss intervention programmes at the Karolinska Hospital, Stockholm, Sweden were, consecutively during a year, invited to take part in this study. Fifty-seven patients fulfilled two simple criteria for participating: age between 20 and 65 y and BMI≥30 kg·m−2.
Normal-weight women of the same age working in education, administration or dentistry were also invited to participate. Twenty-five women answered a local advert for reference persons. Twenty-two of them had BMI<25 kg·m−2. All were asked to fill in a self-administered questionnaire about musculoskeletal pain (Örebroformulär 007D, FSF-METOD AB, Örebro).26 The reference group reported good health and had normal joint problems compared with reference material including 17 701 women,26 whereas the patients had experienced more pain from neck, hands, back, knees and feet (χ2; d.f.=1; P=0.04–<0.001) during the current week.26 The two groups also differed in age; the references being older than the patients (Student's unpaired t-test; P=0.04). Descriptive data on patients and references are given in Table 1.
Patients and references were informed about the tests and gave their consent to participate. The study was approved by the local ethical committee.
Development and validity.
Existing tests25,27,28 were judged to lack sensitivity to register functional limitations in obese adults and accordingly a special test protocol was developed (Tables 2 and 3). This contained 16 different activities of basic ADL (eg cutting toenails) and instrumental ADL (eg walking up stairs carrying two grocery bags). To make a valid test the choice of actions was based on interviews by an experienced physiotherapist with the staff of the obesity unit and with obese women. The women were also observed by the same physiotherapist during household activities, locomotion and exercising. In order to validate the constructed test protocol the women gave their points of view and helped to choose the most relevant actions. Attempted performance of all actions but nos 2, 8 and 14–16 was graded ‘no difficulty’ (grade 1), ‘some difficulty’ (grade 2), ‘great difficulty’ (grade 3) or ‘impossible’ (grade 4). The need for a handrail for support (yes/no) and perceived pain (yes/no) were also noted (see Table 2). A detailed manual describing each different grade in the test protocol was constructed to facilitate the grading/scoring.
After testing the first 20 patients a few actions were replaced by more difficult ones (nos 3, 11 and 16) or developed (an exchange of a score for pain and exertion and reliable heart rate measures in nos 14–16) to improve the test. This explains why the number of patients was ≤37 in the actions above (Tables 2 and 3).
The reliability tests concerned this final version of the test protocol and embraced the four-graded actions. Timekeeping (actions nos 2, 12 and 14) by means of a stop-watch, measuring distance between hands in cm (no 8) with a tape measure or reading the number of heart beats (nos 14 and 16) from a pulsometer (double-checked by manual testing for 15 s) were regarded as highly reliable and thus not involved in the reliability test.
Three physiotherapists with different skills and experience and 13 obese male and female patients participated in the test of inter-rater reliability of the remaining actions in the protocol. When the raters were compared, pairwise agreement was good, the coefficient of weighted kappa being 0.65.29 The inter-rater reliability and the proportional concordance between all three raters are shown in Table 4. Coefficients of kappa between 0.21 and 0.40 are considered fair.29
The intra-rater reliability coefficient of weighted kappa checked with test–retest on 14 male and female obese patients was 0.80, which is good.29
All the tests were performed indoors at the hospital. A large room was used for most of the actions and for the others the corridor and a nearby staircase. Actions such as stepping up onto a kitchen stool took place close to a set of wall bars to allow hand support. The different actions were performed in numerical order at the participants′ own pace (Tables 2 and 3). Before each action the test-leader informed or showed the participants what they were expected to do. They were free to omit the most straining actions considering potential pain or risk. The results were scored on the test protocol according to the manual. Time (s) was noted from a stop-watch and heart rate (bpm) from a pulsometer (Sport Tester TM PE-3000, Polar Electro, Kempele, Finland). Perceived pain and exertion were graded by the participants using the Category Ratio Scale, CR10, which is reliable and valid,30 (see Figure 1).
Owing to extensive repairs in the building where the patients climbed the stairs, the reference group had to walk up similar, but not identical, stairs. The difference favoured the patients due to the height of the steps: the patients climbed 12.8 m (four floors=88 steps; 14.5 cm high) and 6.4 m (two floors=44 steps), while the references climbed 15 m (88 steps; 17 cm high) and 7.5 m (44 steps).
Results are presented as mean±standard deviation (s.d.) and /or median (Md) and range and /or number of patients (n) able to perform the tasks. For comparisons between the obese patients and the reference group Student's unpaired t-test was used for parametric data. The Mann–Whitney U-test was used for nonparametric data, ordinal values and for data lacking a normal distribution. The χ2 test was used for nominal values. P-values≤0.05 were accepted as significant. Multiple regression was used to determine which of the variables were the best predictors of functional limitation.
The results from the testing of obese patients and normal-weight references are presented in Tables 2 and 3. Some individual exclusions were made action-wise (Tables 2 and 3). One patient had recently had a fat biopsy and was not allowed to lift or carry (actions nos 13 and 16). Two patients used beta-blockers and their heart rate values were excluded from actions nos 14 and 16. One patient had a frozen right shoulder and could not reach behind her back (action no 8). One of the reference persons did not carry the two grocery bags up stairs (action no 16), as she was afraid to provoke low-back pain from her prolapsed intervertebral disc.
In 13 of the 16 actions tested there were significant differences (Mann–Whitney U-test; P-values between 0.05 and 0.001) in results between the patients and the references. See Tables 2 and 3. Five patients of 57 could not climb four floors. Fifteen patients of 36 (42%) were unable to walk 50 m and carry groceries two floors upstairs, (see Table 3). Three tasks, stepping up onto a low stool, rising from a chair and lifting a heavy box from the floor (P=0.06), did not offer any significant difficulties for the patients compared with the reference group.
In contrast to the reference group, pain was more often reported from the patients: in a one-legged stance by 20 patients (χ2; d.f.=1; P=0.008); in squatting by 17 patients (of 37) (χ2; d.f.=1; P=0.01), in rising from squat by 19 patients (χ2; d.f.=1; P=0.009), in stepping up onto a kitchen stool by 13 patients (χ2; d.f.=1; P=0.03) and in rising from a low stool by 14 (χ2; d.f.=1; P=0.02). A handrail was needed in rising from squat by 11 patients (χ2; d.f.=1; P=0.04), in stepping up onto a kitchen stool by 37 patients (χ2; d.f.=1; P<0.001), in rising from a low stool by 21 patients (χ2; d.f.=1; P=0.006) and in rising from supine lying on the floor by 14 (χ2; d.f.=1; P=0.02). One of the reference persons with knee problems found squatting for 10 seconds painful and another with a prolapsed disc could not lift the box. None of the other references reported pain or needed a handrail to perform any of the tasks reported in Table 2. In stair-climbing the patients walked more slowly up stairs (Mann–Whitney U-test; P=0.005), had higher heart rates (Mann–Whitney U-test; P<0.001), and experienced more pain (Mann–Whitney U-test; P<0.001) and exertion (Mann–Whitney U-test; P<0.001) than the references, carrying grocery bags or not. The patients also experienced pain while walking down stairs (Mann–Whitney U-test; P=0.05 see Table 3).
Owing to these highly significant results, a short version for judging the functional limitations was constructed. It included the summed up results of four disparate actions (nos 5, 7, 9 and exertion in no 14) that were easy to test in the clinic. This sum was between 7 and 27 (Md 15) for the patients and between 5 and 9 (Md 8) for the references, which was significantly lower (Mann–Whitney U-test; P<0.001). Three patients had a sum ≤9, which was the highest individual value in the reference group, (see Figure 2). A pain sum was constructed by adding the values from the patients' data on current pain in lower back, hips, knees and feet (yes=1; no=0), (see Table 1).
Multiple regression revealed that the sum of the functional limitations (actions nos 5, 7, 9 and exertion in no 14) was highly influenced by BMI (P=0.002), and also by age (P=0.02) and pain sum(P=0.02), but not by height (NS; see Table 5).
Although fairly well acquainted with the problems in physical function that obese women complain of, we did not quite expect our patients to have difficulties as great as those we found. In 10 of the 16 tested actions there were very strong significant differences between patients and references (P<0.001; see Tables 2 and 3).
The results showed that the obese women, but not the normal-weight references, had flexibility problems, eg reaching their own feet or back. They also had difficulties and sometimes pain in balancing, in performing tasks at floor level such as picking things up, squatting or kneeling, rising from low levels, in stepping up onto high steps or walking up and down stairs and walking indoors while carrying something heavy. They moved more slowly, had higher heart rate and experienced more exertion than the references, even though they climbed a staircase with lower steps and less total height thus doing less work than the references.
The general results concerning speed, perceived exertion and pain are in agreement with a previous study where fitness and walking ability of the same women were found to be low compared to normals.23 Our subjects had during the current week experienced significantly more pain from their backs, knees and feet than the reference group of 17 701 women had.26 The findings of others on bodily pain and osteoarthritis in obese persons confirm our results.8,9,24 Poor physical functioning as shown by the SF-36 questionnaire has previously been reported in obese women.7,24 The questions in the Swedish version of SF-3631 are fairly similar to the actions in our own test protocol. Unfortunately the earlier reported results were given in means and not shown question-by-question, which prevents further comparisons, but the women with the poorest function reported severe limitations in stair-climbing. They also reported some limitations in bending, kneeling, lifting and carrying.7
All our patients were urban dwellers with a fairly high educational level and many lived in a family and had jobs (Table 1). With a mixed or rural sample from the obese female population the study results might have been different.2,32 The reference group was recruited from workplaces near Huddinge Hospital in Stockholm. Those who answered the advert volunteered to participate because they felt the study was important. They had a high educational level and were all working, compared to 88% of the patients. However, since the references were older than the patients and reported musculoskeletal pain to a normal extent,26 we considered them a good group for comparison (see Table 1).
The time for testing one patient with the new test protocol depended on the patient's condition. In a normal-weight and healthy person it took less than 15 min and was easy to administer, since it only required common hospital or primary care centre equipment. Preferably we should have used an existing and well-evaluated functional test, where the actions could easily be scored. It would have been interesting to compare and express the rate of disability in our obese females with disability in patients with other diagnoses.
Unfortunately the tests available at the start of the study did not meet our requirements. Tests such as the Barthel Index, the Katz ADL Index, the Linn Rapid Disability Rating Scale and the Performance ADL Test assess very basic ADL (feeding, continence, need of human assistance); others, such as the PULSES Profile, the Sickness Impact Profile and the Instrumental ADL Scale also assess mental and /or social functions.25,27,28 Many of the tests have ceiling effects and thus lack sufficient challenge for obese persons. The SF-3631 partly corresponded to our requirements, but was rejected since we sought an objective test and not a questionnaire. The actions summing up the functional limitations from the test protocols (actions nos 5, 7, 9 and exertion in no 14) were chosen to consider easy performance and reflect important ADL functions with demands on strength, flexibility and fitness. They all involved lower back, hips, knees and feet. The pain sum was made up from reports on current pain in the corresponding joints.
Multiple regression revealed that a high BMI value was an outstanding predictor of functional limitation, but increasing age and lower-body pain also predicted reduced function in the patients. Not only these but also confounding factors such as family and job situation, educational level, sleep problems (Table 1) or being a smoker could influence the functional limitation. References with current pain had no difficulty in performing the test actions, though, which makes the impact of the BMI value even stronger. According to Stafford et al7 the odds ratios of having low self-reported function increase with increasing BMI. Fontaine et al24 have reported functional limitation due to bodily pain, especially among morbidly obese patients. Age was an expected predictor since the frequency of osteoarthritis and disability increases with age.9 Also, individual height might have been important since short patients had difficulties in stepping up onto the kitchen stool, but this turned out to be of minor importance. It is interesting that the three patients with normal function had no current pain problems, they were all working and active with golf, dancing and /or gardening several times a week (Figure 1).
The functional test very well reflected what problems in physical function the obese women in this study had compared to normal-weight women. The test protocol can be used to evaluate important effects of exercise or weight-loss interventions, but may be clinically impractical and time-consuming given the large number of actions. Since functional limitations were closely linked to high BMI, and also to age and current pain, we recommend for clinical use simply considering the patient's BMI value, age and current back and lower extremity pain before giving individual advice on exercise.
Our present and also our previous results on walking ability and maximum oxygen uptake (VO2max/kg) in obese women23 show that obese women frequently experience problems such as pain and exertion in everyday activities, which might explain why many of our patients were inactive in scheduled exercises and ordinary physical activity (Table 1). Still we can use our results and encourage obese, sedentary patients with pain to become fitter to counterpromote disease and to preserve physical function and good health. They should avoid protracted weight-bearing on the lower extremities in combination with kneeling, squatting, rising, climbing or extensive staircase training. This includes walking for those who find it painful,23 and indoor training activities such as step-up, low-impact aerobics and also many activities such as ball games and winter sports. A varied training programme combining brief nonweightbearing and weightbearing exercises will probably be better tolerated. Suitable options left for obesity fitness training could be ergometer bicycling, pool exercising/swimming and training in a gym (resistance training plus aerobic exercising such as bicycling or rowing),33 which should not provoke pain from the lower limbs. Pool exercising has the extra advantage of including balance training.
The functional test reflected functional limitations in the obese women tested. They had difficulties in reaching (flexibility), balancing, squatting, kneeling, rising from low furniture, stepping up onto high steps, staircase-climbing and carrying grocery bags. They were also slower and experienced more pain and exertion than normal-weight women in the reference group. A high BMI value, age and current lower-body pain could predict their functional limitation.
The functional test can be used to evaluate the effects of intervention on functional limitation in obese women, but in the clinic it would suffice to investigate the patient's BMI, age and current lower-body pain to be able to give suitable advice on physical activities. Our study results support recommendations on non-weightbearing training for obese women with lower-body pain.
World Health Organization . Obesity preventing and managing the global epidemic World Health Organization: Geneva 1998.
Deurenberg P, Hautvast JG . Prevalence of overweight and obesity in the Netherlands in relation to sociodemographic variables, lifestyle and eating behavior: starting points for the prevention and treatment of obesity Bibl Nutr Dieta 1989 44: 8–21.
Kuskowska-Wolk A, Rössner S . Body mass distribution of a representative adult population in Sweden Diabetes Res Clin Pract 1990 10: 37–41.
Rippe M . The case for medical management of obesity: a call for increased physician involvement Obes Res 1998 6 (Suppl 1): 23–33.
Rissanen A, Heliövaara M, Knekt P, Reunan A, Aromaa A, Maatela J . Risk of disability and mortality due to overweight in a Finnish population Br Med J 1990 301: 835–837.
Buchowski MS, Sun M . Energy expenditure, television viewing and obesity Int J Obes Relat Metab Disord 1996 20: 236–244.
Stafford M, Hemingway H, Mannot M . Current obesity, steady weight change and weight fluctuation as predictors of physical functioning in middle aged office workers: the Whitehall II study Int J Obes Relat Metab Disord 1998 22: 23–31.
van Itallie TB, Lew EA . Health implications of overweight in the elderly Prog Clin Biol Res 1990 326: 89–108.
Felson DT, Anderson JJ, Naimark A, Walker AM, Meenan RF . Obesity and knee osteoarthritis. The Framingham study Ann Intern Med 1988 109: 18–24.
Blair SN . Evidence for success of exercise in weight control Ann Intern Med 1993 119: 702–706.
Blair SN, Kampert JB, Kohl HW, Barlow CE, Macera CA, Paffenbarger RS et al. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women JAMA 1996 276: 205–210.
Kuskowska-Wolk A, Rössner S . Decreased social activity in obese adults Diabetes Res Clin Pract 1990 10: 265–269.
de Groot LC, van Staveren WA . Reduced physical activity and its association with obesity Nutr Rev 1995 53: 11–13.
Tryon WW, Goldberg JL, Morrison DF . Activity decreases as percentage overweight increases Int J Obes Relat Metab Disord 1992 16: 591–595.
Maddox GL, Back KW, Liberman VL . Overweight as social deviance and disability J Health Soc Behav 1968 9: 287–298.
Rippe JM . Overweight and health: communications challenges and opportunities Am J Clin Nutr 1996 63 (Suppl 3): 470–473.
Blair SN, Horton E, Leon AS, Lee IM, Drinkwater BL, Dishman RK et al. Physical activity, nutrition and chronic disease Med Sci Sports Exerc 1996 28: 335–349.
Wadden TA, Stunkard AJ . Social and psychological consequences of obesity Ann Intern Med 1985 103: 1062–1067.
Maiman LA, Wang VL, Becker MH, Finlay J, Simonsson M . Attitudes toward obesity and the obese among professionals J Am Diet Assoc 1979 74: 331–336.
Price JH, Desmond SH, Krol RA, Snyder FF, O'Connell JK . Family practician's beliefs, attitudes, and practices regarding obesity Am J Prev Med 1987 3: 339–345.
Simkin Silverman LR, Wing RR . Management of obesity in primary care Obes Res 1997 5: 603–612.
Spector TD . The fat on the joint: osteoarthritis and obesity J Rheumatol 1990 17: 283–284.
Mattsson E, Evers Larsson U, Rössner S . Is walking too exhausting for obese women? Int J Obes Relat Metab Disord 1997 21: 380–386.
Fontaine KR, Cheskin LJ, Barofsky I . Health-related quality of life in obese persons seeking treatment J Fam Pract 1996 43: 265–270.
Verbrugge LM, Jette AM . The disablement process Soc Sci Med 1994 38: 1–14.
Örebroformulär 007D, FSF-METOD AB, Örebro, Sweden
Basmajian J (ed) . Physical rehabilitation outcome measures Williams & Wilkins: Baltimore MD 1995.
Feinstein AR, Josephy BR, Wells CK . Scientific and clinical problems in indexes of functional disability Ann Intern Med 1986 105: 413–420.
Altman DG . Practical statistics for medical research, 1st edn Chapman & Hall: London 1991.
Borg GAV . Borg's perceived exertion and pain scales Human Kinetics: Leeds 1998.
Sullivan M, Karlsson J, Ware JE . The Swedish SF-36 Health Survey I. Evaluation of data quality, scaling assumptions, reliability, and construct validity across general populations in Sweden Soc Sci Med 1995 41: 1349–1358.
Kuskowska-Wolk A, Rössner S . Inter-relationships between socio-demographic factors and body mass index in a representative Swedish adult population Diabete Res Clin Pract 1990 10: 271–275.
Ashutosh K, Methrotra K, Fragale Jackson J . Effects of sustained weight loss and aerobic fitness in obese women J Sports Med Phys Fitness 1997 37: 252–257.
This study was supported by the Department of Physical Therapy and the Committee for the Health and Caring Sciences at the Karolinska Institutet, and carried out at the Departments of Physical Therapy and Medicine at the Karolinska Institutet.
We thank the patients and reference group who volunteered to participate in this study. We also thank Professor Stephan Rössner and his staff for generously providing patients and valuable points of view. Physiotherapists Pia Walander, Robin Kihlbaum and Mattias Röjås helped to perform the reliability tests in an excellent way. We are most grateful for expert help from Jan Kowalski, statistician and Tim Crosfield regarding the English language.
About this article
Cite this article
Larsson, U., Mattsson, E. Functional limitations linked to high body mass index, age and current pain in obese women. Int J Obes 25, 893–899 (2001). https://doi.org/10.1038/sj.ijo.0801553
- functional limitations
- obese women
- test protocol
Acute Effects of Whole-Body Vibration Exercises at 2 Different Frequencies Versus an Aerobic Exercise on Some Cardiovascular, Neuromotor and Musculoskeletal Parameters in Adult Patients With Obesity
Obesity Surgery (2019)
Applied Ergonomics (2019)
Predictors of Adherence to Different Volumes of Exercise in the Breast Cancer and Exercise Trial in Alberta
Annals of Behavioral Medicine (2019)
Fear of falling in obese women under 50 years of age: a cross-sectional study with exploration of the relationship with physical activity
BMC Obesity (2019)