Introduction

The prevalence of obesity among US adults has been steadily increasing over the past 40 y.¹ The global trend of obesity drew interest to approaching the problem from multiple dimensions. Preventing and treating obesity calls for involvement and participation from many sectors of society, including national and local governments, educators and legislators, health officials and researchers, food industry, etc.^2,3 From the economic perspective, the direct costs associated with obesity in 1994 was estimated to account for 5.7% of national health expenditures in the United States,⁴ and 2–3.5% in other countries.⁵ From the health perspective, obesity has been linked as a risk factor for many chronic conditions, including diabetes, hypertension, hypercholesterolemia, stroke, coronary heart disease, certain cancers, arthritis, gallbladder disease, osteoarthritis, sleep apnea and respiratory problems, and endometrial, breast, prostate, and colon cancers.^6,7

According to the National Health and Nutrition Examination Survey (NHANES) 1999–2000 data, the trend of increased obesity is particularly worrisome in the elderly group (ages 60–79 y).¹ Owing to multiple factors, older people tend to be more sedentary. Obesity has been shown to impair physical ability,⁸ contributing further to morbidity and mortality of obese adults.^9,10 Weight loss in the elderly has been shown to reduce morbidity from arthritis-, diabetes-, and cardiovascular-related diseases. Increased physical activity in the elderly improved weight loss, muscle strength, endurance, and well being.¹¹

Ambulation is an important functional activity. It is also a readily available, easy to understand, inexpensive exercise to offer older patients who want to burn calories. Our current study examines the relationships of ambulatory activity to age, body mass index (BMI), and back pain in the elderly using an objective measurement of ambulatory activity, a 7-day pedometer test.

Methods

Study protocol

This study used the data collected from a blinded longitudinal study of spinal stenosis being conducted in a university spine clinic. Five groups of subjects older than 54 y and under 80 y old were recruited to volunteer to participate, including asymptomatic persons, persons with mechanical low back pain (LBP) not radiating below the knee, and persons with spinal stenosis of varying degrees, as diagnosed on magnetic resonance imaging (MRI).

Asymptomatic subjects were recruited from the community. Spinal stenosis subjects and most back pain subjects were recruited from a systematic review of all University MRI reports of persons of the appropriate age. Individuals who volunteered to participate in the study were initially screened for meeting the study criteria. The exclusion criteria included the following: persons who have diabetes, diseases of the nerves in the arms and legs, people who drink more than 10 alcoholic drinks per week, or who had low back surgery. Persons with contraindication to electrodiagnostic testing or to MRI scanning were not studied. Subjects could ambulate at least 20 ft (6 m) independently. The study protocol was approved by the university institutional review board and all qualified persons who agreed to participate in the study gave written informed consent.

The study protocol included a patient questionnaire, physical examination, a walking tolerance test with long latency nerve conduction studies (F- and H-waves) before and after walking, a 7-day pedometer test, MRI, and electromyography. The walking tolerance test required ambulation at a comfortable pace until claudication caused the subject to stop, or until 15 min had passed.

For the pedometer test, subjects were instructed to wear the device (DIGI-WALKER Model SW-701) for 7 days during all hours when they are awake except when they bathe, shower, swim or are at places where the unit may get wet. The subjects were shown how to wear the instrument on their belt or waistband and given a weekly log to record the starting time and date and stopping time and date of the test. They were also instructed not to open the pedometer at any time and to return the pedometer to the study coordinator at the end of the 7 days in a postage prepaid mailer. The pedometer records the number of steps taken, the distance walked, and energy expenditure. Bassey et al¹² assessed the Digi-walker in bench tests and field studies, and found the Digi-walker to be a suitable instrument for objective long-term assessment of walking activity. The difficulty in establishing step count guidelines that correspond with other public health guidelines limited the utility of the Digi-walker step counter in assessing daily physical activity.¹³

Data analysis

SPSS Version 10.0¹⁴ was used for statistical analysis. Ambulatory measurements include walking velocity, stride length, steps/day, distance/day, and calories/day, and were expressed in metric system. Data for steps, distance, and energy expenditure were assessed by the pedometer. The stride length, obtained during the first 48 m of a 15-min walk test, was calculated by dividing 48 m by the total number of steps taken. The walking velocity data were also obtained from the 15-min walk test. It was calculated as the total distance (km) walked 60 min then divided by the number of min walked, which gives the average KMPH. The ambulation measurements and BMI, height, weight, and pain severity were treated as continuous variables. Pain severity was measured on a 100-mm Visual Analog Scale (VAS) for pain.^15,16 VAS asked subjects to place a check mark on a line from 0 (no pain) to 10 (worst) for their average pain of the last week. The pain category was classified into three groups: no pain, mechanical LBP, and spinal stenosis. BMI was calculated from measured height and weight, as weight in kilograms divided by the squared height in meters. Subjects were grouped into three BMI categories defined as normal (BMI<25 kg/m²), overweight (25BMI<30 kg/m²), and obese (BMI30 kg/m²).⁶

The group mean comparison by BMI category was examined in a one-way analysis of variance (ANOVA). Post hoc analysis using Tukey's honest significant difference test was performed to determine which means differ. The relationship between ambulatory measures and BMI and between ambulation and pain severity was examined in Pearson's correlation analysis. For correlations of the ordinal variables, Spearman's correlation coefficient was used to represent the relationship between ambulation and pain category, coded as 1=no pain, 2=mechanical LBP, and 3=spinal stenosis. The statistic is appropriate for ordinal data of this nature. To examine the effect of BMI on ambulation, a multiple linear regression analysis was performed separately on distance/day and steps/day as dependent variables. BMI was included as an independent variable in the regression. The purpose of the regression analysis is to assess the relationship between the primary independent variable of interest (BMI) with the ambulation-dependent variable, while controlling for other known predictors of the ambulation variable. To control for the potential confounding effect of age and pain on ambulation, these two variables were included as independent variables in the regression analysis. P-values <0.05 in all of the analyses were considered as statistically significant.

Results

Table 1 presents the demographic characteristics of the study subjects. A total of 82 individuals participated in the study from August 2001 to mid-November 2002 with a mean age of 65.67.0 y and 91.5% non-Hispanic white subjects in the sample. Based on the pain category, 22 (26.8%) had no pain, 17 (20.7%) had mechanical back pain, and 43 (52.5%) had radiographic findings of spinal stenosis. The mean score for the average pain severity in the last 7 days was 2.72.5 measured on VAS from 0 to 10. With a mean BMI score of 28.35.8, the BMI category classified 30 (36.6%) in the group of normal weight, 27 (32.9%) overweight, and 25 (30.5%) obese. The proportion of the obese in our sample, 30.5%, corresponded with the estimated prevalence of 30.5% in all age groups and in all racial/ethnic groups among US adults, 1999–2000.¹

Table 1 - Subject demographics.

Full table

Table 2 shows the results of the ANOVA of group mean differences in age, ambulatory activity, and pain severity by BMI category. The obese people were slightly older, 67.26.4 y compared to 63.17.1 for the normal and 66.97.0 for the overweight (F=3.162, P=0.048). People in the obese group averaged significantly lower in walking velocity in a 15-min walk test and distance walked, and steps taken per day. The most noticeable mean differences were in daily walking activity (distance and steps). The normal group walked almost three times more in distance per day than the obese group, 6.13.3 vs 2.41.7 km (P<0.01). The normal people took an average of more than twice as many steps per day as the obese people, 73853680 vs 33252346 (P<0.01). The three groups did not differ significantly in stride length, daily energy expenditure, and self-reported pain severity measured on the VAS scale.

Table 2 - Comparison of demographic variables by BMI category.

Full table

Table 3 shows the strength of the relationship between ambulation and BMI and between ambulation and pain. With the exception of stride length and energy expenditure, BMI had a significant inverse correlation with steps/day (r=-0.426), distance/day (r=-0.455), and walking velocity (-0.271), all statistically significant at P<0.01. Pain severity and pain category also had a significant inverse relationship with the same ambulation measures. People who had a higher mean BMI walked significantly less, took significantly fewer steps, and ambulated at a lower speed. The results of the correlation analysis confirmed the findings presented in Table 2 from the ANOVA. The data supported our hypothesis that obese people walked less than the nonobese people. The data also showed that pain (severity and category) had a similar relationship with ambulation as observed for BMI.

Table 3 - Correlation between obesity and ambulation and between pain and ambulation.

Full table

The inverse relationship between pain and ambulation (number of steps taken, distance, and calories) appeared to hold up in normal, overweight, and obese subjects, as shown in Table 4. Pain could be measured in terms of severity or diagnosis. Both severity and diagnosis affected ambulation ability in all three weight categories. In addition, the diagnosis of spinal stenosis resulted in an expected decrease in ambulation velocity.

Table 4 - Correlation between ambulation and pain by BMI category.

Full table

To further examine the relationship between BMI and ambulation and between pain and ambulation, a multivariate regression analysis was performed separately on distance/day and steps/day as dependent variables. Age was also found to have a significant inverse correlation with ambulation (steps/day (r=-0.467, P<0.001) and distance/day (r=-0.508, P<0.001)). Therefore, it was included along with BMI and pain severity as independent variables in the analysis. The results in Table 5 further confirmed our hypothesis that BMI was a strong independent predictor of ambulation. In the first model, BMI (B=-0.216, s.e.=0.044, P<0.001), age (B=-0.160, s.e.=0.040, P<0.001), and pain (B=-0.399, s.e.=0.108, P<0.001) explained 48.4% (adjusted R²=0.461) of the variance in distance walked. In the second model, BMI (B=249.457, s.e.=55.825, P<0.001), age (B=-174.588, s.e.=50.783, P=0.001), and pain (B=-500.840, s.e.=135.221, P<0.001) explained 44.5% (adjusted R²=0.421) of the variance in the number of steps taken. Pain severity, measured by VAS, was also found to be a strong and significant predictor of ambulation. Our hypothesis that obesity and pain independently contributed to ambulation was thus supported in terms of daily walking.

Table 5 - Regression of obesity (BMI), age, and pain severity (VAS) on ambulation.

Full table

Discussion

The study results suggest that increased obesity relates to decreased functional ambulation in the community. It further shows that increased pain relates to decreased ambulation in these people.

This is the first study we know of that relates pain and obesity to actual community ambulation. Others have noted parts of this relationship with less direct measures of community ambulation. In a community-based cohort of 1655 men and women aged 55 y and older, Sternfeld et al¹⁷ found fat mass to be significantly lower in those with a faster walking speed. Their outcome measures were survey responses regarding ambulation and walking velocity for 60 s. Leveille et al¹⁸ selected a community-based population of severely disabled, poor women over the age of 65 y. This study used a survey and walking velocity on a 4 m trial. The results of these studies were consistent with ours.

In our study, as shown in Table 3, while we found a significant inverse relationship between BMI and walking velocity (r=-0.271, P=0.014), the result of correlation analysis revealed an even stronger association between BMI and ambulation measured by steps/day (r=-0.426) and distance/day (r=-0.455), all statistically significant at P<0.001. The result of the regression analysis further demonstrated that BMI was a strong independent predictor of distance/day (B=-0.216, P<0.001) and steps/day (B=-249.457, P<0.001). A decrease in BMI was associated with increasing number of steps taken and the distance covered in daily walking activity.

Numerous studies have been conducted to examine the validity and reliability of using pedometers for evaluating daily physical activity. These include the work of Bassey et al,¹² Saris and Binkhorst,^19,20 Mizuno et al,²¹ Sequeira et al,²² and Bassett et al.^23,24 Tudor-Locke et al²⁵ examined the relationship between ambulatory activity (steps/day) and BMI and percentage body fat, and confirmed an inverse relationship between steps/day and BMI in an adult sample. Kashiwazaki et al²⁶ examined the relationship of pedometer readings with 24-h energy expenditure as determined by the heart rate in clerical workers vs assembly workers during work, commuting, and at home. They found the pedometer suitable for use in a sedentary population as the instrument is designed to detect vertical movements in response to the impact of the foot, and its reading correlated best with the net energy expenditure in walking.²⁶ Despite the pedometer's limitations, which were found to be variation in sensitivity between pedometers¹² and variations among brands in terms of accuracy,²³ it is considered a simple, inexpensive, and objective tool for studying physical activity.

Modern electronic pedometers have been used to assess ambulatory activity in children and adults objectively. They have been found to be reliable especially in sedentary subjects. Our study population tends to be sedentary due to either obesity or LBP. The steps/day measured in our study population (54813629, n=82) appeared to be lower than those studied by Tudor-Locke et al (73703080, n=109). This is due, in part, to the fact that LBP also contributes to the reduction of ambulatory activity. Age difference is also a factor. The subjects in the Tudor-Locke study were younger (44.915.8 y) compared with 65.67.0 y of our sample. In our study, 22 subjects (26.8%) reported no pain, 17 (20.7%) subjects were diagnosed as having mechanical back pain, and 43 (52.5%) had radiographic finding of spinal stenosis. An inverse correlation between steps/day and BMI (r=-0.426, P<0.001) observed in our study confirms the Tudor-Locke et al²⁵ finding of an inverse relationship between steps/day, also the pedometer (Yamax, Digiwalker) assessed, and BMI (r=-0.30, P<0.01).

Lumbar spinal stenosis is a degenerative disorder that primarily affects older adults. This condition causes pain in the lower back that often radiates down the legs, impacting negatively on the ability of older adults to ambulate. Our study shows that this disorder does indeed impact ambulation in older persons, but pain from non-neurogenic back disorders is also a cause of decreased ambulation. Our previous study found that patients who reported higher pain also performed poorly in simple daily functional activities.²⁷

Walking is not only a life function, it is a treatment for disability. Walking exercise has been shown to be effective in increasing walk endurance capacity, physical activity level, and mobility. In residents of a nursing home who were assigned to a 12-week walking program, the experimental walking group improved distance by 22% after the program, whereas no significant changes were reported for the social visit control group.²⁸ The effects of walking have also been proved to be successful in a study of patients assigned to an experimental (diet and exercise) group, who were instructed to walk at least 10 000 steps/day on a flat field vs a diet alone control group, who maintained a normal routine (4500290 steps/day). The experimental group showed a significantly greater amount of weight reduction (7.80.8 vs 4.20.5 kg, P<0.01) and had significant correlations between metabolic clearance rate and average steps walked per day (r=0.7257, P<0.005).²⁹

In persons with LBP and with severe spinal stenosis, walking has been demonstrated to be a useful and quantifiable measure of assessing treatment and surgical outcome. Frost et al³⁰ reported that patients with chronic LBP who attended the fitness class increased their walking capacity by 25% compared with no change in the control group (P<0.005). In a prospective study of exercise tolerance on the treadmill before and after decompressive laminectomy in 50 patients with severe lumbar spinal stenosis, the mean total ambulation time increased significantly (P<0.001) from 6.91 min in the preoperative trial to 13.26 min in the postoperative trial.³¹

In a correlative study one may question the causal relations. For obesity and ambulation, we might expect that further study shows a cycle of causation—poor caloric expenditure leading to obesity, making ambulation more energy intensive, and uncomfortable, resulting in more obesity. For pain there may also be a cyclic relationship, but of importance to clinicians and patients is the probability that pain interferes with the attempts of older people to ambulate, thus making obesity harder to control. Until prospective studies are carried out, it is prudent for the clinician who hopes to increase the physical activity of older persons to look closely for reversible causes of musculoskeletal pain.

Conclusions

In older persons with and without back pain complaints, this study has demonstrated the significance of BMI and pain severity as strong predictors of actual ambulation in the community. The data on ambulatory measurements, walking velocity (laboratory controlled) and daily distance walked, steps taken, and energy expenditures (pedometer assessed), provide a benchmark for future research in a population such as in this study. Clinicians who wish their obese elderly patients to become more active should consider musculoskeletal barriers to exercise.

References

1. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA 2002; 288: 1723–1727. | Article | PubMed | ISI |
2. Nestle M, Jacobson MF. Halting the obesity epidemic: a public health policy approach. Public Health Rep 2000; 115: 12–24. | Article | PubMed | ChemPort |
3. Jacobson MF. Nutrition, physical activity, and obesity. Lancet 2002; 360: 1250. | Article | PubMed |
4. Wolf AM, Colditz GA. Current estimates of the economic cost of obesity in the United States. Obes Res 1998; 6: 97–106. | PubMed | ISI | ChemPort |
5. Thompson D, Wolf AM. The medical-care cost burden of obesity. Obes Rev 2001; 2: 189–197. | Article | PubMed | ChemPort |
6. National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report. Obes Res 1998; 6 (Suppl. 2): 51s–209s, (Published erratum appears in Obes Res 1998; 6: 464). | PubMed | ISI |
7. Rippe JM. The case for medical management of obesity: a call for increased physician involvement. Obes Res 1998; 6 (Suppl. 1): 23S–33S. | PubMed |
8. Roissner S. Obesity in the elderly—a future matter of concern? Obes Rev 2001; 2: 183–188. | Article | PubMed |
9. Stevens J, Cai J, Pamuk ER, Williamson DF, Thun MJ, Wood JL. The effect of age on the association between body-mass index and mortality. N Engl J Med 1998; 338: 1–7. | Article | PubMed | ISI | ChemPort |
10. Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath Jr CW. Body-mass index and mortality in a prospective cohort of US adults. N Engl J Med 1999; 341: 1097–1105. | Article | PubMed | ISI | ChemPort |
11. Elia M. Obesity in the elderly. Obes Res 2001; 9 (Suppl. 4): 244S–248S. | PubMed |
12. Bassey EJ, Dallosso HM, Fentem PH, Irving JM, Patrick JM. Validation of a simple mechanical accelerometer (pedometer) for the estimation of walking activity. Eur J Appl Physiol Occup Phys 1987; 56: 323–330.
13. Welk GJ, Differding JA, Thompson RW, Blair SN, Dziura J, Hart P. The utility of the Digi-walker step counter to assess daily physical activity patterns. Med Sci Sports Exerc 2000; 32 (Suppl.): S481–S488. | PubMed |
14. Norusis MJ. SPPS 10, Guide to Data Analysis. Prentice-Hall: Englewood Cliffs, NJ; 2000.
15. Bijur PE, Silver W, Gallagher EJ. Reliability of the visual analog scale for measurement of acute pain. Acad Emerg Med 2001; 8: 1153–1157. | PubMed |
16. Gallagher EJ, Liebman M, Bijur PE. Prospective validation of clinically important changes in pain severity measured on a visual analog scale. Ann Emerg Med 2001; 38: 633–638. | Article | PubMed |
17. Sternfeld B, Ngo L, Satariano WA, Tager IB. Associations of body composition with physical performance and self-reported functional limitation in elderly men and women. Am J Epidemiol 2002; 156: 110–121. | Article | PubMed | ISI |
18. Leveille SG, Guralnik JM, Hochberg M, Hirsch R, Ferrucci L, Langlois J, Rantanen T, Ling S. Low back pain and disability in older women: independent association with difficulty but not inability to perform daily activities. J Gerontol A Biol Sci Med Sci 1999; 54: M487–M493. | PubMed |
19. Saris WH, Binkhorst RA. The use of pedometer and actometer in studying daily physical activity in man. Part I: reliability of pedometer and actometer. Eur J Appl Physiol Occup Physiol 1977; 37: 219–228. | PubMed |
20. Saris WH, Binkhorst RA. The use of pedometer and actometer in studying daily physical activity in man. Part II: validity of pedometer and actometer measuring the daily physical activity. Eur J Appl Physiol Occup Physiol 1977; 37: 229–235. | PubMed |
21. Mizuno C, Yoshida T, Udo M. Estimation of energy expenditure during walking and jogging by using an electro-pedometer [Japanese]. Ann Physiol Anthropol 1990; 9: 283–289. | PubMed |
22. Sequeira MM, Rickenbach M, Wietlisbach V, Tullen B, Schutz Y. Physical activity assessment using a pedometer and its comparison with a questionnaire in a large population survey. Am J Epidemiol 1995; 142: 989–999. | PubMed | ISI | ChemPort |
23. Bassett Jr DR, Ainsworth BE, Leggett SR, Mathien CA, Main JA, Hunter DC, Duncan GE. Accuracy of five electronic pedometers for measuring distance walked. Med Sci Sports Exerc 1996; 28: 1071–1077. | PubMed | ISI |
24. Bassett Jr DR, Cureton AL, Ainsworth BE. Measurement of daily walking distance-questionnaire versus pedometer. Med Sci Sports Exerc 2000; 32: 1018–1023. | Article | PubMed |
25. Tudor-Locke C, Ainsworth BE, Whitt MC, Thompson RW, Addy CL, Jones DA. The relationship between pedometer-determined ambulatory activity and body composition variables. Int J Obes Relat Metab Disord 2001; 25: 1571–1578. | Article | PubMed | ChemPort |
26. Kashiwazaki H, Inaoka T, Suzuki T, Kondo Y. Correlations of pedometer readings with energy expenditure in workers during free-living daily activities. Eur J Appl Physiol Occup Physiol 1986; 54: 585–590. | Article | PubMed | ISI | ChemPort |
27. Ruan CM, Haig AJ, Geisser ME, Yamakawa K, Buchholz RL. Functional capacity evaluations in persons with spinal disorders: predicting poor outcomes on the Functional Assessment Screening Test (FAST). J Occup Rehabil 2001; 11: 119–132. | Article | PubMed |
28. MacRae PG, Asplund LA, Schnelle JF, Ouslander JG, Abrahamse A, Morris C. A walking program for nursing home residents: effects on walk endurance, physical activity, mobility, and quality of life. J Am Geriatr Soc 1996; 44: 175–180. | PubMed |
29. Yamanouchi K, Shinozaki T, Chikada K, Nishikawa T, Ito K, Shimizu S, Ozawa N, Suzuki Y, Maeno H, Kato K. Daily walking combined with diet therapy is a useful means for obese NIDDM patients not only to reduce body weight but also to improve insulin sensitivity. Diabetes Care 1995; 18: 775–778. | PubMed | ChemPort |
30. Frost H, Klaber Moffett JA, Moser JS, Fairbank JC. Randomised controlled trial for evaluation of fitness programme for patients with chronic low back pain. BMJ 1995; 310: 151–154. | PubMed |
31. Deen HG, Zimmerman RS, Lyons MK, McPhee MC, Verheijde JL, Lemens SM. Use of the exercise treadmill to measure baseline functional status and surgical outcome in patients with severe lumbar spinal stenosis. Spine 1998; 23: 244–248. | Article | PubMed |

Acknowledgements

This is a publication of the Spine Program of the University of Michigan, which is funded by the United States Department of Health and Human Services, National Institutes of Health under Grant #5 R01 NS41855-02. The opinions contained in this publication are those of the grantee and do not necessarily reflect those of the United States Department of Health and Human Services.