Validity, Reliability, and Responsiveness of the Fingertip-to-Floor Test
Caroline Perret, MD, Serge Poiraudeau, MD, PhD, Jacques Fermanian, MD, PhD,
Marie Martine Lefe`vre Colau, MD, Marie Anne Mayoux Benhamou, MD, PhD, Michel Revel, MD ABSTRACT. Perret C, Poiraudeau S, Fermanian J, Lefe`vre
Colau MM, Mayoux Benhamou MA, Revel M. Validity, reliability, and responsiveness of the fingertip-to-floor test.
Arch Phys Med Rehabil 2001;82:1566-70.
Objective:To evaluate the validity, reliability, and respon- siveness of the fingertip-to-floor test to assess total mobility when bending forward in standing position.
Design:Experimental, prospective, correlational.
Setting:Rehabilitation and radiology departments in a uni- versity hospital in France.
Participants: Ten patients (6 women, 4 men; mean age, 42yr) with chronic low back pain (LBP) in the validity study;
32 LBP patients (16 women, 16 men; mean age, 52yr) in the reliability study; and 72 LBP patients (22 women, 50 men;
mean age, 30yr) in the responsiveness study.
Interventions: Dynamic radiographs and fingertip-to-floor test.
Main Outcome Measures:For the validity study, 2 lateral radiographs of the upper dorsal spine, 1 in neutral position, and then 1 in full trunk flexion, were made. Validity was assessed by means of Spearman’s correlation coefficient. Reliability was studied by using intraclass correlation coefficient (ICC) and the Bland and Altman method. Responsiveness was assessed by the effect size and the standardized response mean (SRM).
Results: The Spearman’s correlation coefficient for trunk flexion assessed by the test and the radiologic measure was excellent (rs5 2.96). The intra- and interobserver reliability were excellent (ICC 5 .99). The Bland and Altman method showed no systematic trend. The values observed for the test were .97 for SRM and .87 for effect size.
Conclusions: Because the fingertip-to-floor test has excel- lent validity, reliability, and responsiveness, it can be used in clinical practice and therapeutic trials.
Key Words: Evaluation studies; Exercise therapy; Low back pain; Rehabilitation; Reproducibility of results; Spine motility.
©2001 by the American Congress of Rehabilitation Medi- cine and the American Academy of Physical Medicine and Rehabilitation
I
MPROVEMENT IN THE FLEXIBILITY of the spine and the hips is the goal of many programs of exercise therapy in musculoskeletal diseases, for example, in ankylosing spondy- litis and in chronic low back pain (LBP). The outcome assess- ment, however, requires a valid and relevant method of mea- surement usable in daily clinical practice.Lumbar flexion is easily evaluated by the Scho¨ber and mod- ified Scho¨ber test.1These 2 tests are performed with the subject standing erect. Marks on the skin are made with a pen. The first mark is placed at the lumbosacral junction, as indicated by the posterior superior iliac spines or dimples of Venus; a second mark is made 10cm above and, for the modified test, a third mark is made 5cm below the lumbosacral junction. The subject is asked to bend forward as far as possible and the new distance between marks 1 and 2 for the Scho¨ber test and between marks 2 and 3 for the modified Scho¨ber test is measured. Macrae and Wright1 validated this method by comparing lumbar flexion assessed by the Scho¨ber and modified Scho¨ber techniques with radiographic measurements. Pearson’s correlation coefficients of .97 and .90 were found for the modified Scho¨ber and the original Scho¨ber techniques, respectively.1The Scho¨ber tech- nique has also been analyzed by age, gender, height, and weight.2 The distribution of measurements follows a normal pattern; it is significantly lower for women (10%) and de- creases with age. With this technique, spinal mobility is inde- pendent of height and weight.2
Total mobility when bending forward in standing position, also called trunk flexion, corresponds to spine plus pelvic flexion. It is a more relevant outcome measure after intensive rehabilitation than the Scho¨ber test,3probably because patients increase both pelvic and lumbar flexion simultaneously after treatment. This mobility can be assessed by the fingertip-to- floor test and by the inclinometer method of Loeble.2,4,5The fingertip-to-floor test is performed with the subject standing erect, feet together. The subject is asked to bend forward as far as possible, while maintaining the knees, arms, and fingers fully extended. The vertical distance between the tip of the middle finger and the floor is measured with a supple tape measure. The inclinometer is a hand-held, circular, fluid-filled disk with a weighted gravity pendulum indicator that remains oriented in the vertical direction. With the inclinometer, trunk flexion is measured as follows: with the patient standing erect, recordings are made at S2 and at T12–L1. Holding the incli- nometer on T12–L1, the subject is asked to bend forward as far as possible, while maintaining the knees fully extended. Keep- ing the subject fully flexed, a third recording is taken at T12–L1 and a fourth recording at S2. These 4 readings permit simple calculation of total flexion, pelvic flexion, and, by subtraction, lumbar flexion. To be reliable, the inclinometer method requires experienced testers6while the measurement of finger-to-floor distance requires only a tape measure and is taught during basic medical training.
The fingertip-to-floor test is assumed to evaluate the mobility of both the whole spine and the pelvis in the overall motion of bending forward. Although inexpensive, safe, quick, and easy, the metric characteristics of this clinical test in current use
From the Hoˆpital Cochin, Service de re´e´ducation et de re´adaptation de l’appareil locomoteur et des pathologies du rachis, Assistance Publique-Hopitaux de Paris, Universite´ Rene´ Descartes (Perret, Poiraudeau, Lefe`vre Colau, Mayoux Benhamou, Revel); and Hoˆpital Necker-Enfants Malades, Service de Biostatistiques, Assistance Publique-Hopitaux de Paris, Universite´ Rene´ Descartes (Fermanian, Mayoux Ben- hamou), Paris, France.
Accepted November 13, 2000.
Supported by the Fondation pour la Recherche Me´dicale.
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.
Reprint requests to Michel Revel, MD, Service de re´e´ducation et de re´adaptation de l’appareil locomoteur et des pathologies du rachis, 27 Rue du Fbg St Jacques, 75014 Paris, France, e-mail:[email protected].
0003-9993/01/8211-6080$35.00/0 doi:10.1053/apmr.2001.26064
remain to be shown. The present study sought to determine the validity, reliability, and responsiveness of the fingertip-to-floor test in LBP patients.
METHODS Subjects
We included 3 groups of subjects in the present study. For validity assessment, patients with chronic LBP and isthmic spondylolisthesis, needing dynamic radiographs of the spine, were invited to participate in the study. The radiographs were indicated to detect intervertebral instability after medical treat- ment had failed and when a surgical procedure was discussed.
Intervertebral instability was defined as an anteroposterior mo- bility of more than 4mm between flexion and extension at 1 level. According to the inclusion criteria, subjects must have:
(1) pain on the posterior aspect of the trunk below the basilar costal margins or above the greater trochanter without leg pain;
(2) be at or over the age of 18 years; (3) be able to stand for the duration of the experiment; and (4) volunteer to be enrolled in the study. Patients were excluded if they had neurologic im- pairment attributable to lumbar pathology, or were pregnant.
For reliability assessment, patients with chronic LBP were evaluated. They were recruited in the rehabilitation department.
In addition to the general inclusion criteria, these subjects must have experienced LBP for at least 6 months in the study.
For responsiveness assessment, we identified a group of patients with chronic, disabling LBP. They were included in a functional restoration program according to the following cri- teria: (1) LBP for at least 6 months; (2) inability to work for at least 4 months; (3) failure of other treatments; (4) age 18 years of age or older; (5) no contraindication to perform exercises because of cardiovascular or respiratory diseases; and (6) will- ingness to participate in the program.
Measurement Procedure
Validity and reliability of the fingertip-to-floor test were assessed according to the American Psychological Association recommendations.7For the validity study (criterion validity), each patient underwent 2 radiographs and 1 fingertip-to-floor test. The 3 procedures were performed within 1 hour of each other. The radiographic measurement of the angle correspond- ing to the whole trunk flexion was compared with the measure of the fingertip-to-floor test. For the reliability study, each LBP patient underwent 4 fingertip-to-floor tests performed succes- sively within 1 hour. Two assessments were performed blindly by the 2 investigators to facilitate test-retest and interobserver reliability. The measurement sequence for the 4 tests was randomly determined to avoid influence of the measurement sequence. For the responsiveness study, the fingertip-to-floor test and the Scho¨ber test were assessed at baseline and after 5 weeks of intensive rehabilitation.
Fingertip-to-Floor Test
The subject stood erect on a platform 20-cm high with shoes removed and feet together. He/she was asked to bend forward as far as possible, while maintaining the knees, arms, and fingers fully extended. The vertical distance between the tip of the middle finger and the platform was measured with a supple tape measure and was expressed in centimeters. The vertical distance between the platform and tip of the middle finger was positive when the subject did not reach the platform and negative when he could go further. Both investigators were experienced with the measurement of the finger-to-floor dis- tance.
Radiographs
Two lateral radiographs of the upper dorsal spine were taken, first with the patient standing in neutral (upright) posi- tion, then bending forward as far as possible, with the knees fully extended. All radiographs were taken by using a horizon- tal beam and with the film parallel to the floor. The angle formed by a line tangential to the upper or lower vertebral plateau of T5 vertebrae with the horizontal was measured. The lower vertebral plateau of T5 was chosen when the upper plateau of the vertebrae could not be precisely determined on the radiograph. This angle was determined in both positions.
The difference between the angle obtained in neutral position and in full flexion gave the trunk flexion in degrees. For each patient and each investigator, the same plateau was taken into account for radiographs made in the neutral position and in the fully flexed position. Each radiograph was analyzed blindly by 2 investigators using different goniometers, and the interob- server reliability was assessed by means of the intraclass cor- relation coefficient (ICC) as described later.
Statistical Analysis
Data analysis was performed by using SYSTAT, version 5.2.1.aQuantitative variables were described by using means, standard deviation (SD), and range. Qualitative variables were expressed by using proportion and percentage. Comparisons between means of 2 quantitative variables were assessed by using the Mann-WhitneyUtest. The significance level of the test was alpha equal to .05.
Criterion validity. The correlation between trunk mobility in flexion assessed by radiograph analysis (in degrees) and the fingertip-to-floor test (in centimeters) was evaluated by using Spearman’s correlation coefficient (rs), and the data were de- scribed by scatter plot analysis.
Reliability. The intra- and interobserver reliability was studied by simultaneously applying the ICC8and the Bland and Altman method9; the 2 methods give complementary informa- tion as shown by Atkinson and Nevill10and I-Kuei Lin and Chinchilli.11
Responsiveness. Responsiveness was assessed by the ef- fect size12,13and the standardized response mean (SRM).14
RESULTS Criterion Validity
Ten patients (6 women, 4 men) were included for the valid- ity study of the fingertip-to-floor test. Their mean age was 42 years (range, 33–51yr), mean height 168cm (range, 152–
185cm), and weight 63kg (range, 50 – 80kg). Mean fingertip- to-floor test 6SD was 21.5 621.7cm (range, 0 – 48cm) for men and 19.6 6 22cm (range, 0 –50cm) for women. Mean trunk mobility in flexion assessed by radiograph analysis was 106°632.5° (range, 49°–157°).
The interobserver reliability of the radiologic analysis was excellent with an ICC of .99. Spearman’s correlation coeffi- cient of trunk flexion assessed by the fingertip-to-floor test and the radiologic measure was excellent (rs 5 2.96). Figure 1 shows the scatter plot corresponding to this coefficient.
Reliability
Thirty-two LBP patients (16 women, 16 men) were evalu- ated. Their mean age was 52 years (range, 21– 80yr), mean height 166cm (range, 152–190cm), and weight 69kg (range, 50 –97kg). Disease duration was 65 6 50 months (range, 6 –180mo). Mean fingertip-to-floor test was 27.7 6 17cm (range, 0 –58.5cm) for men and 12.3616cm (range,210 to
43cm) for women, with a statistically significant difference (p,.001).
The intra- and interobserver reliability of the fingertip-to- floor test was excellent with an ICC of .99. The Bland and Altman method9for interobserver reliability (fig 2) showed that the mean 6 SD of the difference was 20.6 6 2.8cm. The distribution of the differences was homogeneous with no sys- tematic trend (rs5 2.25).
Responsiveness
Seventy-two chronic disabling LBP patients (22 women, 50 men) were evaluated. Their mean age was 40 years (range, 19 –55yr), mean height 169cm (range, 149 –191cm), and weight 71kg (range, 45–104yr). Disease duration at baseline was 60656 months (range, 8 –240mo). The responsiveness of the fingertip-to-floor test and the Scho¨ber test evaluated by the SRM and the effect size is shown in table 1. We observed higher values (SRM5.97; effect size5.87) for the fingertip- to-floor test than for the Scho¨ber test (SRM 5 .69; effect size5.75).
DISCUSSION
The finger-to-floor distance might be used to assess spine stiffness and the effects of exercise on spine stiffness in persons with spine disorders such as ankylosing spondylitis and back pain. The results of the present study suggest that this test has very good metric properties.
We decided to use radiographic measurements as the crite- rion standard for the validity study because of their accuracy and their extensive use in other studies.1,15-24Also, the reliabil- ity of the radiograph analysis in the present study was excel- lent. Despite the small sample size, the excellent correlation coefficient (rs5 2.96) between the fingertip-to-floor test and the radiographic measurements suggests that this simple clin- ical measure is very closely related to x-ray measures (criterion validity).
The reliability of the fingertip-to-floor test in LBP pa- tients, assessed by the ICC and the Bland and Altman9 method, is excellent. The present results are inconsistent with those of previous studies in healthy volunteers, which reported coefficients of variation of 76%25 and 14%26 for intraobserver reliability, and 83%25 for interobserver reli- ability. The insufficient reliability was attributed to the multifactorial variables inherent in the full motion: hip flexion, high number of vertebra, and effect of flexion in the fingers, wrist, elbow, and shoulder. The discrepancy be- tween the results of the previous studies and our data can be explained by some differences in method, such as the num- ber of patients, the clinical experience of the investigators, and the relevance of the statistical tests.
Another difference is that the present study concerned pa- tients, not healthy subjects. However, the reliability in healthy subjects should not be lower than that established for patients.
In addition, in 50 healthy volunteers, we found excellent inter- observer reliability of the fingertip-to-floor test (ICC 5 .97) (unpublished data).
The present study is the first to evaluate responsiveness of the fingertip-to-floor test. After a functional restoration pro- gram including many daily exercises for flexibility, the effect size (.87) and SRM (.97) values showed that the fingertip-to- floor test has very good sensitivity to change, higher than that of the Scho¨ber test (effect size 5 .75; SRM 5 .69). These results confirm that the fingertip-to-floor test is more sensitive
Fig 1. Scatter plot of the fingertip-to-floor test versus radiographic measure of trunk mobility. Correlation coefficient was2.96 (Spear- man’s rank correlation).
Fig 2. Bland and Altman graphic representation of the interobserver reliability of the fingertip-to-floor test. Mean of the difference was 20.6 6 2.8cm.
to change than the Scho¨ber test and therefore a more appro- priate way to assess the changes in total trunk flexibility after physical treatments.
Although spine flexibility of persons with LBP does not correlate with disability in some cross-sectional studies,27-29 reduced spinal mobility is a negative predictor for successful rehabilitation.30Furthermore, after chronic back pain rehabili- tation programs, such as functional restoration programs, the increase in trunk mobility is constant31,32 and is probably a potent psychologic stimulation for the patients. Finally, the assessment of the respective participation of the spine (assessed by the Scho¨ber or modified Scho¨ber tests) and the pelvic mobility (deduced from the fingertip-to-floor test and the Scho¨ber test) during trunk flexion could allow therapists to direct exercises toward either spine or pelvic mobility. In ankylosing spondylitis, the relationship between disability and spinal mobility has been poorly studied, but several range-of- motion (ROM) restrictions, and among them trunk mobility (assessed by the fingertip-to-floor test) and lumber mobility (assessed by the Scho¨ber test), correlated with radiologic changes.33 A rehabilitation program can increase trunk and spine mobilities34and participation in a rehabilitation program appears to protect against disability.35As for LBP, the assess- ment of the respective participation of the lumbar segments and pelvis in ROM restriction helps to determine the target joints for exercises.
CONCLUSION
The results of the present study revealed that the finger- tip-to-floor test has excellent metric properties for LBP patients. This very simple clinical test can routinely provide relevant information on trunk mobility. The measure of trunk flexion by this test complements the measure of lum- bar flexion by the Scho¨ber or modified Scho¨ber test and gives information on the respective participation of the spine and pelvic mobility when a patient bends forward in stand- ing position.
References
1. Macrae IF, Wright V. Measurement of back movement. Ann Rheum Dis 1969;28:584-9.
2. Moll JM, Wright V. Normal range of spinal mobility. An objec- tive clinical study. Ann Rheum Dis 1971;30:381-6.
3. Rainville J, Sobel JB, Hartigan C. Comparison of total lumbosa- cral flexion and true lumbar flexion measured by a dual inclinom- eter technique. Spine 1994;19:2698-701.
4. McCombe PF, Fairbank JCT, Cockersole BC, Pynsent PB. 1989 Volvo award in clinical sciences: reproducibility of physical signs in low back pain. Spine 1989;14:908-18.
5. Waddell G, Main CJ, Morris EW, Venner RM, Rae PS, Sharmy SH, et al. Normality and reliability in the clinical assessment of backache. Br Med J 1982;284:1519-23.
6. Mayer TG, Kondraske G, Beals SB, Gatchel RJ. Spinal range of motion. Accuracy and sources of error with inclinometric mea- surement. Spine 1997;22:1976-84.
7. American Psychological Association. Standards for educational and psychological testing. Washington (DC): APA; 1995.
8. Shrout PE, Fleiss JL. The effects of measurement errors on some multivariable procedures. Am J Public Health 1977;67:
1188-91.
9. Bland JM, Altman DG. Statistical methods for assessing agree- ment between two methods of clinical measurement. Lancet 1986;
1:307-10.
10. Atkinson G, Nevill A. Comment on the use of concordance correlation to assess the agreement between two variables. Bio- metrics 1997;52:775-7.
11. I-Kuei Lin L, Chinchilli V. Rejoinder to the letter to the editor from Atkinson and Nevill [letter]. Biometrics 1997;52:
777-8.
12. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care 1989;27:S178-89.
13. Liang MH, Fossel AH, Larson MG. Comparisons of five health status instruments for orthopedic evaluation. Med Care 1990;28:
632-42.
14. Cohen J. Statistical power analysis for the behavioral sciences.
New York: Academic Pr; 1977.
15. Allbrook D. Movements of the lumbar spinal column. J Bone Joint Surg Br 1957;39:339-45.
16. Clayson SJ, Newman IM, Debevec DF, Anger RW, Skowlund HV, Kottke F. Evaluation of mobility of hip and lumbar ver- tebrae of normal young women. Arch Phys Med Rehabil 1962;
43:1-8.
17. Dvorak J, Panjabi MM, Chang DG, Theiler R, Grob D. Functional radiographic diagnosis of the lumbar spine. Flexion-extension and lateral bending. Spine 1991;16:562-71.
18. Froning EC, Frohman B. Motion of the lumbosacral spine after laminectomy and spine fusion. Correlation of motion with results.
J Bone Joint Surg Am 1968;50:897-918.
19. Hayes MA, Howard TC, Gruel CR, Kopta JA. Roentgenographic evaluation of lumbar spine flexion-extension in asymptomatic individuals. Spine 1989;14:327-31.
20. Pearcy M, Portek I, Shepherd J. Three-dimensional x-ray anal- ysis of normal movement in the lumbar spine. Spine 1984;9:
294-7.
21. Pearcy M, Tibrewal SB. Axial rotation and lateral bending in the normal lumbar spine measured by three-dimensional radiography.
Spine 1984;9:582-7.
22. Tanz SS. Motion of the lumbar spine; roentgenologic study. AJR Am J Roentgenol 1953;69:399-412.
23. White AA, Panjabi MM. The basic kinematics of the human spine.
A review of past and current knowledge. Spine 1978;3:12-20.
24. Yamamoto I, Panjabi MM, Crisco T, Oxland T. Three-dimen- sional movements of the whole lumbar spine and lumbosacral joint. Spine 1989;14:1256-60.
25. Merritt JL, McLean TJ, Erickson RP, Offord KP. Measurement of trunk flexibility in normal subjects: reproducibility of three clin- ical methods. Mayo Clin Proc 1986;61:192-7.
26. Gill K, Krag MH, Johnson GB, Haugh LD, Pope MH. Repeat- ability of four clinical methods for assessment of lumbar spinal motion. Spine 1988;13:50-3.
27. Sullivan MS, Shoaf LD, Riddle DL. The relationship of lumbar flexion to disability in patients with low back pain. Phys Ther 2000;80:240-50.
28. Gronblad M, Hurri H, Kouri JP. Relationship between spinal mobility, physical performance tests, pain intensity and disability assessments in chronic low back pain patients. Scand J Rehabil Med 1997;29:17-24.
29. Nattrass CL, Nitschke JE, Disler PB, Chou MJ, Ooi KT. Lumbar spine range of motion as a measure of physical and functional impairment: an investigation of validity. Clin Rehabil 1999;13:
211-9.
30. Pope MH, Rosen JD, Wilder DG, Frymoyer JW. The relation between biomechanical and psychological factors in patients with low back pain. Spine 1980;5:173-8.
Table 1: Responsiveness of the Fingertip-to-Floor Test and the Scho¨ber Test
First Visit Last Visit Difference p SRM Effect Size
Fingertip-to-floor test 27615.4 13.6613.2 213.45613.8 ,.0001 .97 .87
Scho¨ber test 3.561.2 4.461.2 0.961.3 ,.0001 .69 .75
31. Mayer T, Tabor J, Bovasso E, Gatchel RJ. Physical progress and residual impairment quantification after functional restoration.
Spine 1994;19:389-94.
32. Hazard RG, Fenwick JW, Kalisch SM, Redmond J, Reeves V, Reid S, et al. Functional restoration with behavioral support: a one-year prospective study of patients with chronic low back pain.
Spine 1989;14:157-61.
33. Viitanen JV, Kokko ML, Lehtinen K, Suni J, Kautiainen H.
Correlation between mobility restrictions and radiologic changes in ankylosing spondylitis. Spine 1995;20:492-6.
34. Viitanen JV, Suni J, Kautiainen H, Liimatainen M, Takala H.
Effect of physiotherapy on spinal mobility in ankylosing spondy- litis. Scand J Rheumatol 1992;21:38-41.
35. Guillemin F, Briancon S, Pourel J, Gaucher A. Long-term disabil- ity and prolonged sick leave as outcome measurements in anky- losing spondylitis. Possible predictive factors. Arthritis Rheum 1990;33:1001-6.
Supplier
a. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.