Relations of Sit-up and Sit-and-Reach Tests to Low Back Pain in Adults

Size: px

Start display at page:

Download "Relations of Sit-up and Sit-and-Reach Tests to Low Back Pain in Adults"

Evangeline Payne
6 years ago
Views:

1 Relations of Sit-up and Sit-and-Reach Tests to Low Back Pain in Adults Allen W. Jackson, EdD ' James R. Morrow, Jr., PhD ' Patricia A. Brill, P ~D* Harold W. Kohl, 111, phd3 Neil F. Gordon, M D ~ Steven N. Blair, PED5 P lowman (15) reviewed the literature on low back pain (LBP) and reported LBP is a common occurrence in adults, with estimates of 60-80% of the population experiencing LBP at some point in life. According to Plowman (15), "...the incidence of LBP syndrome is high and going higher in all industrialized societies in the world." The exact etiology of acute and/or chronic LBP is unknown as the pain arises suddenly and, in many cases, there is no anatomical or physical manifestation to indicate the source of pain. Possible risk factors associated with LBP include constitutional (eg., aerobic capacity, age, physical fitness, gender, muscular strength), posturalstructural, radiographic (eg., structural abnormalities), environmental (eg., smoking), occupational (eg., demanding jobs, lifting, vibration), recreational (eg., sports activities), and psychosocial (eg., anxiety, depression) characteristics (4,7,15). The economic impact of LBP is in billions of dollars in the U.S. (15). It has been suggested that high levels of abdominal strength/endurance, increased low back and hamstring flexibility, or both may provide protection against developing LBP (2,3). However, Nachemson (14) suggests there is a weak correlation be- The sit-up and sit-and-reach tests are found on nearly all youth and adult fitness tests because of the perceived relation between performance on these tests and low back pain. However, this relationship has not been well validated. Therefore, the purpose of the study was to examine the relationship between performance on these two common field tests of muscular strength and flexibility (the sit-up and the sit-and-reach tests) and self-reported low back pain (LBP). The sample included 2,747 adults with a mean age of 44.6? 9.8 years. The 1-minute sit-up (mean = ) and sit-and-reach tests (mean = t cm) were administered to participants as part of a voluntary clinical health and fitness evaluation between 1980 and Participants completed a mail-back survey in 1990 on musculoskeletal health problems. Low back pain was quantified by developing an ordinal variable from questionnaire responses which represented a range of severity of LBP from none (0) to LBP which required medical care (3). With an average of 6.1 (22.0) years of follow-up, LBP was reported by 54% of the study participants (men = 45%, women = 54%). Pearson correlations behveen sit-up (r =.002; p =.94), sit-andreach (r = -.043; p =.03), and LBP indicated poor LBP criterion-related validity from the sit-up and sit-and-reach tests. Partial correlations, where age, gender, percent of body fat, and time between testing and survey response were controlled, displayed no increase in the relationship. This study does not support the validity of sit-up and sit-and-reach test items for health-related fitness batteries because they were unrelated to LBP. Key Words: low back pain, health fitness, survey ' Professor, Department of KHPR, University of North Texas, Denton, TX Director of Gerontological Fitness, National Guest Homes, Houston, TX ' Director of Research, Baylor College of Medicine, Houston, TX ' Research Cardiologist, Heart and Lung Group of Savannah, Savannah, GA Director of Epidemiology and Clinical Applications, The Cooper Institute for Aerobics Research, Dallas, TX This research was supported in part by a research grant from the National Institutes of Health/National Institute on Aging (AG06945). tween physical fitness and LBP consequences. Nevertheless, because of this hypothesized association and the morbidity and physical limitations associated with LBP, many fitness testing protocols include measures of abdominal strength/endurance and low back flexibility. The most widely used youth fitness tests [eg., The Pru- den tial FITNESSGRAM (6), President's Challenge (l7), or AAU Fitness Program (I)] and adult fitness testing programs (8) all include measures of abdominal strength/endurance and lower back and hamstring flexibility. Various forms of the sit-up (eg., bent-leg, straight-leg, timed, curl-up, crunch) are used for the ab- Volume 27 Number 1 January 1998 JOSPT

2 dominal muscular strength and endurance measurement, and the sitand-reach is typically used to assess lower back and hamstring flexibility. Sit-up and sit-and-reach items are reportedly included because "...ab dominal strengthendurance and low back-posterior thigh flexibility is [sic] important for the prevention and rehabilitation of low back disorders." The purpose of including such tests is the perceived relationship between test performance on the measures and LBP. Thus, there is a theoretical relationship between measures of musculoskeletal strength and endurance and flexibility and LBP with some research findings to support that relationship (15). However, the primaly justification for the sit-up is content validity (3), whereas that for the sitand-reach is clinical in nature (ie., individuals with LBP have restricted range of motion in the hamstrings and low back) (3). The relation between musculoskeletal fitness and LBP may be limited as Plowman (15) reports "...there is no evidence that high levels of abdominal or back extensor strength and endurance or hamstring flexibility in any way predispose an individual to LBP." Thus, improving and/or measuring musculoskeletal fitness is not contraindicated and can be part of a conservative clinical treatment of LBP (14). There appears to be no criterionrelated evidence relating performance on the sit-up and sit-and-reach tests to reported LBP. Therefore, the purpose of this research is to examine the relationship between sit-up and sit-and-reach performance to selfreported LBP from both cross-sectional and cohort perspectives. Specifically, 1) did sit-up or sit-and-reach performance relate to LBP at any time in a person's life and 2) did sit-up or sit-and-reach performance predict future incidence of LBP? Additional variables shown (13) to be related to sit-up and sit-and-reach performance (eg., gender, age, percent of body fat) were controlled in the statistical analyses. A cross-sectional study was conducted to examine concurrent validity and a cohort study was conducted to examine the predictive validity of the sit-up and sit-and-reach tests. METHODS Participants The study sample included participants (age = 44.6? 9.8 years) in a health and fitness examination conducted at a preventive medicine facility between 1980 and The sample size for the cross-sectional analysis was 2,747 (men = 2,270, women = 477). The sample size for the cohort analysis (a subset of the cross-sectional analysis) was 1,661 (men = 1,327, women = 334). The sample was 99% white, and detailed procedures for the total health and fitness examination have been described in previous publications (5,12). The fitness examination included administration of the sit-up and sit-and-reach tests. Fitness Tests The sit-up protocol required the participant to perform as many bentknee sit-ups as possible in 1 minute. Participants laid in a supine position with the knees bent and feet flat on a mat. The hands were placed on the side of the head with fingers over the ears. The participants elevated the trunk until the elbows made contact with the legs. They reversed directions until the shoulder blades touched the mat. The feet were secured by the test examiner who counted the sit-ups during the 1-minute test. Participants were encouraged to give maximal effort during the test. The sit-and-reach was administered with a standard protocol with trained technicians using a sit-andreach box (2). The participant sat on the floor with the legs extended, shoulder width apart, and feet flat against the box. With one hand on top of the other, the participant slowly slid the hands across the top of a ruler attached to the top of the box until maximum reach was attained. The participant completed three reaches and the best reach was recorded. Measurements were recorded to the.25 inch and converted to centimeters for analysis. Numerous studies have been conducted with the sit-up and sit-andreach tests. Both tests have demonstrated high test-retest reliability. The typical reliability for the sit-up test is greater than.80 and that for the sitand-reach is greater than.90 (13). Musculoskeletal Health Survey A mail survey to assess selfreported musculoskeletal health was conducted in The survey was designed to address a variety of muscular and orthopaedic health issues. This study focused on LBP. The average length of time between the initial health and fitness examination and the mail survey was 6.1 (52.0) years. It was mailed to over 14,000 participants and 4,296 participants responded at least in part to the survey and had recorded sit-up and sit-andreach test scores. An intensive set of standardized survey procedures was used to maximize response to the survey. The follow-up procedures included sending a reminder card at 1 week after the initial mailing, a letter and second questionnaire at 3 weeks after the initial mailing to those who had not yet responded, and, finally, a third letter and questionnaire at 7 weeks after the initial mailing to nonrespondents. Low Back Pain Survey respondents answered specific questions regarding self-reported LBP. To assess LBP, an ordinal variable with a range of 0 to 3 was developed in response to answering yes or no to the following survey questions: 1) Have you ever had low back pain JOSPT Volume 27 Number 1 January 1998

Multivariate ana1vsi.c oi variance indicated a signiiicant niult~v,lr~~te dliibrrncc> rp <,0011 I~et~vccn the genders for sit-and-reach, sit-up, percent oi hodv iat, bodv weight, height, and age.

(Yes or No); and 3) Was medical treatment required for the low back pain? (Yes or No)?

all three questions. If the respondent answered "yes" to question 1, they also supplied the year of the first episode of that LBP.

Data Analyses Descriptive statistics including means and standard deviations were calculated for the sit-up and sit-andreach tests, percent of body fat (seven-site skin-fold estimate measured during

3 Multivariate ana1vsi.c oi variance indicated a signiiicant niult~v,lr~~te dliibrrncc> rp <,0011 I~et~vccn the genders for sit-and-reach, sit-up, percent oi hodv iat, bodv weight, height, and age. TABLE 1. Means and standard deviations. that persisted more than 3 days during a single year? (Yes or No); 2) Did you miss work due to the pain? (Yes or No); and 3) Was medical treatment required for the low back pain? (Yes or No)? Score: 0 if No on Question 1, 1 if Yes on Question 1 and No on Questions 2 and 3, 2 if Yes on Questions 1 and 2 and No on Question 3 or Yes on Questions 1 and 3 and No on Question 2, and 3 if Yes on all three questions. If the respondent answered "yes" to question 1, they also supplied the year of the first episode of that LBP. This allowed the determination of a cohort with no LBP at the time of the physical examination. Data Analyses Descriptive statistics including means and standard deviations were calculated for the sit-up and sit-andreach tests, percent of body fat (seven-site skin-fold estimate measured during the health and fitness examination), body weight, and height. Multivariate analysis of variance and post hoc analyses of variance were used to test for differences between genders over the measured variables. Percentages for the four levels of the variable representing LBP were determined for each gender and the Chisquare test of association (gamma) was used to determine if a relation existed between gender and LBP. Cross-sectional analysis This sample included (N = 2,747) subjects who had data on all of the following variables: age, gender, percent of body fat, time between fitness testing and survey response, self-reported LBP, sit-ups, and sit-and-reach test performance. We calculated bivariate correlations to determine the relationship between the sit-up and sitand-reach tests and LBP. Partial correlations were used to examine the relationship while controlling for age, gender, percent of body fat, and the time between fitness testing and survey response. Polynomial regression was used to determine if a nonlinear relation might exist between LBP and the sit-up and sit-and-reach tests. Cohort analysis Additional statistical analyses were conducted for participants (N = 1,661) who 1) never reported LBP (N = 1,275) or 2) reported LBP (N = 386) after the initial physical examination (ie., the 1,086 survey respondents who reported LBP at or prior to the date of the initial testing were not included in the cohort analyses). Logistic regression analysis was conducted to determine the odds ratio (OR) for sit-up and sit-and-reach performance at the initial testing and reported LBP from the survey. The model controlled for age, gender, percent of body fat, and the time between fitness testing and survey response. Lastly, the top and bottom quarters of sit-up and sit-and-reach performance were determined and related to the incidence of LBP to obtain the relative risk. Due to the number of significance tests conducted and the large sample size, caution should be used when interpreting significance levels. TABLE 2. Numbers (percentages) of men and women in the reported levels of low back pain. RESULTS Table 1 provides the means and standard deviations for the measured variables in the study for each gender. Table 2 presents the numbers and percentages for the LBP levels for each gender. Examination of Table 1 reveals a consistently reported physical and performance comparison of men and women. Women performed fewer sit-ups, were more flexible, higher in percent of body fat, lighter in body weight, and shorter in height than the men. This indicates that the sample of this study is typical of the general population. The men and women were different in reported LBP (p =.01), with women reporting no LBP more frequently than men. In the cross-sectional sample, the correlations between LBP and sit-up (r =.002; p =.94) and sit-and-reach (r = -.043; p =.03) indicated essentially no relationship between the variables. The partial correlations controlling for age, gender, percent of body fat, and time between survey and examination between LBP and sit-up (r = -.003; p =.89) and sitand-reach (r = -.012; p =.53) were not significant and in agreement with the bivariate correlations. Since all of the correlations approached zero, it was possible that a nonlinear relationship was being overlooked. Thus, a polynomial regression analysis was completed to see if nonlinearity was "masked" by the zero order or partial correlations. The polynomial regression analysis indicated no significant or meaningful quadratic relation between LBP and sit-up (p =.57) and sit-and-reach (p =.43) performance. Volume 27 Number 1 January 1998 JOSPT

4 RESEARCH STUDY The cohort results were similar to the cross-sectional findings (ie., no statistical significance). The logistic regression analysis, using a dichotomous variable (no LBP or reported LBP), found no relation (P =.47) between sit-up and sit-and-reach performance and LBP when controlling for age, gender, percent of body fat, and time between survey and examination. While using the lowest quarter (poorest performance) on each test as the referent group (OR = 1.0), the top fitness quarter odds ratio was not significantly lower than the lowest quarter for either the sit-up (OR =.98) nor the sit-andreach (OR =.99). DISCUSSION Prevalence of LBP in this selfreported assessment is similar to that reported elsewhere (ie., >50%) (15). Note that follow-up in the current study was for an average period of less than 7 years and did not follow up throughout lifetime. Thus, the prevalence throughout life could well approach the 80% reported by Plowman (15). The current analyses (cross-sectional and cohort) indicate no relation between sit-up and sit-and-reach performance and reported LBP. Zero-order, partial correlations, and regression analyses all indicate that poor performance on the sit-up and sit-and-reach tests does not predict self-reported LBP over the next several years of life. From concurrent validity (cross-sectional analyses) and predictive validity perspectives (cohort analysis), the sit-up and sit-andreach tests provide no criterion-related validity. The correlational analyses indicate sit-up was unrelated to LBP. The sit-up used in this study is one of a variety of sit-up protocols available in fitness batteries and has been challenged (18) because of the "...inability to differentiate between active hip flexor and spinal flexor muscles... when the feet are held down." Perfor- mance on the present sit-up involves musculature that theoretically would not be related to LBP. Therefore, it is not surprising to find little relation between sit-up performance and LBP. Thus, the sit-up should not be included in health-related fitness batteries based on its presumed relation to LBP. Robertson and Magnusdottir (18) suggested a modified curl-up test which requires a minimal involvement of the hip flexors. The Prudential FITNESSGRAM (6) has adopted the modified curl-up as a measure of abdominal muscular strength/endurance. At present, the modified curl-up test also lacks criterion-related validity with LBP (16). Our results also suggest the sitand-reach was not related to LBP. This finding is supported by Plowman (15), who suggests both the sit-up and sit-and-reach "...tests have serious anatomical shortcomings and together they may not be sufficiently comprehensive." Jackson and Baker (9) and Jackson and Langford (10) reported that the sit-and-reach performance had a weak relation with flexibility of the low back. Therefore, a plausible mechanism for the relation of the sit-and-reach to LBP does not exist. Thus, further inclusion of the sit-and-reach test in fitness batteries based on the perceived relation to LBP appears unjustified. Measures of flexibility are important in functional health batteries and the sit-and-reach is a moderately valid measure of hamstring flexibility (9.10). However, justification for the sit-and-reach from a LBP predictive validity perspective is unwarranted. These results do not suggest that being physically active has no effect on LBP symptomology. Muscles, bones, joints, tendons, and connective tissue deteriorate with physical inactivity and respond to appropriate physical activity (14). Indeed, conservative treatment of LBP patients often includes specific exercises and rehabilitative activities asociated with abdominal, lower back, and leg musculature (14). Physical activity to de- velop general muscular strength, endurance, and flexibility should be encouraged because of the positive effects on lean body mass, osteoporosis, and functional capacity. The sit-up and sit-and-reach tests were unrelated to LBP. Therefore, inclusion of alternative fitness items that are related to LBP is indicated. Plowman recommends a series of items, including partial curls, hip flexor flexibility, lateral trunk lift, trunk extension, and a straight leg raise (15). Unfortunately, these tests either lack sufficient criterion-related validity for LBP, are contraindicated for some individuals, or do not lend themselves to mass testing. Strength ' and endurance of the trunk extensor muscles are related to LBP (15). In recognition of this relationship, the Prudential FITNESSGRAM (6) includes a trunk lift test as a measure of trunk extensor strength and flexibility based on content validity. Recent evidence suggests that the FITNESSGRAM back extension test is reliable but information is lacking on the criterion-related validity with LBP (11). The nature of this work does not preclude conservative medical treatment that attempts to increase abdominal strength and endurance and hamstring flexibility in patients with chronic low back pain (14). Waddell et al (19) report differences in asymptomatic individuals and patients with chronic low back pain on eight measures of musculoskeletal performance. CONCLUSION In conclusion, we found the sit-up and sit-and-reach tests lack criterion-related validity with LBP for inclusion in health-related fitness tests. Inclusion of such items should not be based on a presumed relationship between test performance and LBP. Additional concurrent and predictive validity research is needed before including specific musculoskeletal test items in health-related JOSPT l Volume 27 l Number 1 ~ Janua~ 1998

5 fitness batteries if one of the test purposes is to identify variables related to LBP. Further research is needed to establish did criterion-referenced standards for musculoskeletal strength, endurance, and flexibility items. JOSPT ACKNOWLEDGMENTS We thank the physicians and technicians of The Cooper Clinic for administering the clinical examinations, Kia Vaandrager for supervising the fitness test assessments, and Carolyn E. Barlow for data management. REFERENCES 1. Amateur Athletic Union: Amateur Athletic Union Physical Fitness Program, Indianapolis, IN: AAU National Headquarters, 2. American Alliance for Health, Physical Education, Recreation and Dance: Health-Related Physical Fitness Test Manual, Reston, VA: American Alliance for Health, Physical Education, Recreation and Dance, American Alliance for Health, Physical Education, Recreation and Dance: Technical Manual: Health-Related Physical Fitness, Reston, VA: American Alliance for Health, Physical Education, Recreation and Dance, Andersson GB, Pope MH: The patient. In: Pope MH, Andersson GB, Fromoyer JW, Chaffin DB (eds), Occupational Low Back Pain: Assessment, Treatment and Prevention, pp St. Louis: Mosby, Blair SN, Kohl H W 111, Paffenbarger RS, Clark DG, Cooper KH, Gibbons LW: Physical fitness and all-cause mortality: A prospective study of healthy men and women. JAMA 262: , Cooper lnstitute for Aerobics Research: The Prudential FITNESSGRAM, Dallas, TX: The Cooper lnstitute for Aerobics Research, Frymoyer JW: Helping your patients avoid low back pain. J Musculoskel Med 1 :65-74, Golding LA, Myers CR, Sinning WE: Y's Way to Physical Fitness, Champaign, IL: YMCA Program Store, Jackson AW, Baker AA: The relationship of the sit-and-reach test to criterion measures of hamstring and back flexibility in young females. Res Q Exerc Sport 57: , I Jackson AW, Langford NJ: The criterion-related validity of the sit-and-reach test: Replication and extension of previous findings. Res Q Exerc Sport 60: , Jackson AW, Morrow JR Jr, Jensen RL, Jones NA, Schultes SS: Reliability of the Prudential FITNESSGRAM trunk lift test in young adults. Res Q Exerc Sport 67: , Kohl HW 111, Gordon NF, Scott CB, Vaandrager H, Blair SN: Musculoskel- etal strength and serum lipid levels in men and women. Med Sci Sports Exerc 24: , Morrow JR Jr, Falls HB, Kohl HW 111 (eds): The Prudential FITNESSGRAM Technical Reference Manual, Dallas, TX: Cooper lnstitute for Aerobics Research, 14. Nachemson AL: Exercise, fitness, and back pain. In: Bouchard C, Shepard RJ, Stephens T, Sutton JR, McPherson (eds), Exercise, Fitness, and Health: A Consensus of Current Knowledge, pp Champaign, IL: Human Kinetics Publishers, Plowman SA: Physical activity, physical fitness, and low back pain. In: Holloszy 10 (ed), Exercise and Sport Sciences Reviews, pp Baltimore, MD: Williams & Wilkins, Plowman SA, Corbin CB: Muscular strength, endurance, and flexibility. In: Morrow JR Jr, Falls HB, Kohl HW 111 (eds), The Prudential FITNESSGRAM Technical Reference Manual, pp Dallas, TX: The Cooper lnstitute for Aerobics Research, 17. The President's Council on Physical Fitness and Sports: President's Challenge, Washington, DC: The President's Council on Physical Fitness and Sports, 18. Robertson LD, Magnusdottir H: Evaluation of criteria associated with abdominal fitness testing. Res Q Exerc Sport 58: , Waddell G, Somerville D, Henderson I, Newton M: Objective clinical evaluation of physical impairment in chronic low back pain. Spine 17: , 1992 Volume 27 Number 1 January JOSPT

INTERPRETING FITNESSGRAM RESULTS

CHAPTER 9 INTERPRETING FITNESSGRAM RESULTS FITNESSGRAM uses criterion-referenced standards to evaluate fitness performance. These standards have been established to represent a level of fitness that offers