Reliability of clinical measurements of forward bending using the modified fingertip-to-floor method.

Physical therapists who examine and treat patients with low back pain (LBP) usually assess the patients' ability to flex the lumbar spine. Many therapists believe it is important to assess a patient's lumbar flexion. Most therapists, however, use observational method, and do not actually measure the amount of lumbar flexion. Measurements of lumbar flexion are often used to help determine whether a patient has a functional loss attributable to LBP.(1) The selection of treatments for patients with LBP may also be based partially on the assessment of lumbar flexion.(2) Measurements of lumbar flexion are also used to assess patient progress.(3) A variety of methods for measuring lumbar flexion have been described in the literature. Angular measurements of lumbar flexion have been made using spondylometers,(4) inclinometers,(1,5) and standard goniometers.(6) Burton(7) used a flexible rule to measure lumbar flexion. A tape measure has been used to take measurements over the lumbar spine, and these measurements have been used to represent the amount of lumbar flexion.(8,9) All of these methods are similar in that the therapist palpates and marks the patient's spinous processes and places the measurement instrument directly over the lumbosacral area. Some clinicians have used the fingertip-to-floor (FTF) method for assessing lumbar flexion.(10) The FTF method differs from the methods previously discussed because FTF measurements are not taken directly over the lumbosacral area. Fingertip-to-floor measurements are obtained by having the patient forward bend in an attempt to touch the floor with the fingertips. The distance between the tip of the middle finger and the floor is usually estimated or measured with a tape measure(11) or yardstick.(12) Some clinicians have taken serial FTF measurements on patients to assess the efficacy of treatments for LBP.(13-16) Therapists who use the FTF method to measure changes in lumbar flexion of patients with LBP are making assumptions. They are assuming that any change in the FTF distance is a result of a change in lumbar flexion because they are also assuming that motion in the hips, thoracic spine, and upper extremities remains unchanged between measurements. One problem with the FTF method is that measurements cannot be taken on patients who are able to touch the floor with their fingertips. Some authors have modified the FTF method by having the subjects stand on a stool so that measurements could be taken when subjects were able to touch the floor with the fingertips.(11,12,17) Measurements are taken with the modified fingertip-to-floor (MFTF) method by having the subject forward bend while standing on a stool in an attempt to touch the floor with the fingertips and then measuring the distance between the tip of the middle finger and the top of the stool. The MFTF method allows the therapist to take measurements on patients who are able to touch the floor or reach beyond the level of the floor. Physical therapists who use measurement methods, such as the MFTF method, to monitor patient progress, attribute changes in the measurement to changes in the variable being measured. However, changes in measurements may also be partially due to measurement error. The amount of change attributable to measurement (8)Dritz Corp, Spartanburg, SC 29304. error is dependent on the reliability of the measurement. Frost et al,(11) Broer and Galles,(12) and Buxton(17) concluded that measurements made with the MFTF method were reliable on healthy subjects. The reliability of MFTF measurements taken on patients with LBP, however, has not been studied. Frost et al(11) studied the reliability of measurements made with the MFTF method because they were interested in using the MFTF method to measure patients' trunk motion. In selecting their sample, however, they excluded subjects with a history of back surgery and subjects who had visited a physician because of LBP within the 5-year period prior to their study. There are characteristics unique to a population that can affect reliability. Therefore, a sample from the population of interest should be selected when studying the reliability of measurements. Patients with LBP, for example, might experience increased pain when asked to forward bend and touch the floor with their fingertips. Any change in a patient's pain with repeated tests might result in inconsistent measurements. Healthy subjects typically have no pain during forward bending. Therefore, the reliability of measurements made with the MFTF method on patients with LBP might be different than the reliability of measurements taken on healthy subjects. The purpose of this study was to assess the intratherapist and intertherapist reliability of measurements of forward bending taken with the MFTF method on patients with LBP. Method Subjects Subjects for this study were 73 patients (47 male, 26 female) referred for treatment of LBP to the Department of Physical Therapy, Malcolm Grow Medical Center, Andrews Air Force Base, Maryland. The subjects' ages ranged from 18 to 73 years X 43.5, s = 13.9). Patients were included in the study if their therapist was confident they could tolerate repeated forward bending and considered assessment of the amount of lumbar flexion to be an appropriate part of the patients' examination. A total of 38 subjects reported having had pain for greater than I month prior to data collection. The remaining 35 subjects reported having had episodes of pain for less than 1 month before data collection. A total of 14 of the 73 patients participating in the study reported having no history of LBP. All patients read and signed an informed consent form prior to inclusion in the study. Testers Six physical therapists on the staff of the Department of Physical Therapy, Malcolm Grow Medical Center, participated in this study. The therapists' experience ranged from 6 months to 30 years. All therapists were given a description of the MFTF method prior to the beginning of the study by one of the investigators (MGG). The description was limited to instruction in the use of the tape measure and the stool that was used in the study. Therapists were not given specific instructions on how to position patients or on how to instruct patients to bend forward. None of the therapists reported they routinely took FTF or MFTF measurements. At the beginning of the study, each therapist was assigned a number. These numbers and a table of random numbers were used to make a random list for each therapist. These lists were used to randomly choose a second therapist for each patient measured. instrumentation A cloth tape measure(*) marked in 0.10cm increments was used to obtain measurements with the MFTF method. The reverse side of the tape measure was covered with silver duct tape to prevent the therapists from seeing the numbers on the tape measure and to prevent examiner bias. To ensure that the length of the tape measure did not change with repeated use, one of the investigators (MGG) checked the accuracy of the tape measure twice a day by comparing the centimeter scale on the tape with a metal meter stick. The length of the tape measure did not change during the study. Procedure An investigator (MGG) was notified whenever a therapist (called the referring therapist) identified a patient who met the criteria for inclusion in the study. The investigator described the study to the patient and asked the patient to read and sign the consent form. After the consent form was signed, the referring therapist used his random-number list to identify another therapist (called the retest therapist), who took a second set of measurements on the patient. The referring therapist then instructed the patient to stand on a 32.4-cm-high stool and bend forward. Using the covered side of the tape measure, the therapist measured the distance between the tip of the patient's right long finger and the top of the stool. The therapist marked the distance on the tape measure with his thumbnail (Fig. 1). The therapist then held the tape measure so that the investigator (MGG) could see the ruled side of the tape measure. The investigator used his thumbnail to mark the point on the centimeter scale that corresponded to the referring therapist's thumbnail and then took the tape from the referring therapist Fig. 2). The referring therapist was not allowed to see the ruled side of the tape measure and was not informed of the measurement. The MFTF distance was recorded to the nearest 0.10 cm. Measurements of the distances above the stool were positive; those below the stool were negative. A zero was recorded if the patient reached the top of the stool. After the first measurement was taken, the patient was allowed to step off the stool and move around the measurement area for 15 to 30 seconds. The patient then stepped back onto the stool, and the referring therapist took a second measurement. After the referring therapist took two measurements, the retest therapist was notified and took two measurements using the same method as the referring therapist. The retest therapist was not informed of the measurements obtained by the referring therapist. Data Analysis Intraclass correlation coefficients (ICC[1,1](18) were calculated to assess intratherapist and intertherapist reliability. The ICC(1,1) was chosen because it is the appropriate form of ICC to use when assessing the reliability of single measurements taken by random pairs of therapists. The ICC(1,1) is also the most conservative form of ICC. The ICC for intratherapist reliability was calculated by comparing the first and second measurements taken by all therapists. Therefore, a total of 146 paired measurements were used to determine the intratherapist reliability of MFTF measurements. The ICC for intertherapist reliability was calculated by comparing the first measurement taken by the referring therapist with the first measurement taken by the retest therapist. A total of 73 paired measurements were used to determine the intertherapist reliability of MFTF measurements. Results The descriptive statistics for the measurements obtained in this study are reported in Table 1. The ICC for intratherapist reliability of all paired measurements was .98. The ICC for intertherapist reliability was .95 (Table 2). Discussion The results of this study appear similar to those of Frost et al,(11) who reported high intratester and intertester reliability for MFTF measurements taken on healthy subjects. We feel, however, that the reliability values reported by Frost and colleagues are of limited value to clinicians. They used forms of the ICC that were less conservative than the ICC used in this study. A less conservative form of the ICC would result in inflated ICCs compared with those reported in this study. in addition, there are a number of methodological differences between the two studies. The factors that potentially affect MFTF measurements obtained on healthy subjects are different from those factors that may affect MFTF measurements obtained on patients. For example, patients with LBP commonly have an increase in pain when forward bending. Because Frost et al examined healthy subjects, they were unable to examine the effect of pain on the reliability of MFTF measurements. in our study, 22 of the 73 patients stated that they had an increase in pain during forward bending. To determine whether an increase in pain had any effect on the intertherapist reliability of measurements, a posteriori analyses were performed. Table 2. intraclass Correlation Coefficients for Reliability

                                    N              ICC
Intratherapist reliability         146           .98[Alpha]
Intertherapist reliability          73           .95[Beta]

[Alpha]ICC Calculated by comparing first and second measurements of referring and retest therapists.

[Beta]ICC calculated by comparing first measurement taken by referring therapist with first measurement taken by retest therapist. Separate ICCs were calculated for the first measurements taken by paired therapists on patients who had an increase in pain with forward bending and patients who had no change in pain with forward bending. The ICC for patients who had an increase in pain was .97, and the ICC for patients who had no change in pain was .94. The ICCs for intertherapist reliability were essentially the same for patients who had no change in pain during forward bending and for patients who had an increase in pain. An increase in pain during forward bending apparently did not influence reliability, but it may have affected the patient's ability to forward bend. The results of an a posteriori t test revealed that the mean of the first MFTF measurements (20.6 [+ or -] 14.5 cm) for patients who had an increase in pain when forward bending was significantly greater (p [is not greater than] .05) than the mean of the first MFTF measurements (12.6 [+ or -] 12.9 cm) for patients who had no increase in pain when forward bending. The results of the t test suggest that patients who report an increase in pain when forward bending may experience a decrease in the ability to forward bend when compared with patients who do not report an increase in pain during the test. Another difference between our study and the study by Frost et al(11) was in the instructions the therapists gave to the subjects. In the study by Frost et al, the therapists who took the measurements used a specific set of instructions to position the patients and to take the measurements. In our study, the therapists were allowed to use their own instructions. No pair of therapists gave the same instructions to the patients, and the instructions varied greatly. For example, one therapist gave very brief instructions, such as "bend forward as far as you can." Another therapist gave very specific instructions, telling the patient to "bend forward by rolling the spine, bending at the hips and keeping the knees straight." The therapists in our study allowed the patient to stand in the position that was most comfortable. Most patients stood with their feet approximately shoulder-width apart. Instructions to the patients and patient positioning were not standardized in order to more closely replicate clinical procedure. The results of this study reveal that standardized instructions and patient positioning are not required for MFTF measurements to be highly reliable. Data collected in this study also indicate that the MFTF method may be preferable to the FTF method for therapists who take serial measurements on patients. Out of the 73 patients measured in this study, 20 (27%) were able to reach the top of the stool or beyond the top of the stool. A therapist would need to use the MFTF method to take measurements on these 20 patients. Also, some of the remaining 53 patients might be able to reach to the top of the stool or beyond as their LBP diminished. Having patients stand on a stool did not appear to cause them any inconvenience or discomfort. None of the 73 patients reported any fear of falling off the stool when bending forward. Because MFTF measurements were shown to be highly reliable, this study is a first step in determining the usefulness of the MFTF technique. The usefulness of MFTF measurements, however, depends on the construct validity and, ultimately, the criterion-related validity of measurements obtained with the MFTF method. The construct validity, or theoretical basis for the use of FTF and MFTF methods, can be questioned because the FTF and the MFTF distance is not just a measurement of flexion of the lumbar spine. Motion in die joints of the remainder of the spine, the hips, and the upper extremities can affect the measurements. Some authors(19,20) argue that the FTF method should not be used to measure lumbar flexion because the FTF distance is dependent on motion occurring at a number of joints. Biering-Sorenson(21) compared single FTF measurements with measurements of lumbar flexion obtained with the skin-distraction method. The skin-distraction method has been shown to yield valid measurements of lumbar flexion.(8) Measurements were taken on 479 female subjects and 449 male subjects. The author reported a correlation coefficient of .22 for female subjects and -.35 for male subjects. BieringSorenson's results reveal that a strong linear relationship might not exist between the FTF distance and the amount of lumbar flexion. However, because Biering-Sorenson used the FTF method and not the MFTF method, he was not able to measure subjects who were able to reach the floor or beyond. He merely assigned these subjects a value of zero. if Biering-Sorenson had used the MFTF method, the results might have been different. We believe that the construct validity of one FTF measurement for assessing lumbar flexion must be questioned. Individual differences in hip, spinal, and upper extremity range of motion make the use of one FTF measurement as a measurement of lumbar flexion questionable. most therapists who assess forward bending of patients with LBP, however, probably do not use a single FTF measurement as a measurement of lumbar flexion but take serial measurements over time. Any change in the FTF distance is then attributed to a change in lumbar flexion. if a patient with LBP gains lumbar flexion as he or she improves, there should be some change in the FTF distance, because the ability of a patient to reach toward the floor is partially dependent on motion occurring in the lumbar spine. The question of whether a change in serial FTF measurements reflects a change in the amount of lumbar flexion could be answered with a criterion-related validity study. The criterion-related validity of FTF measurements could be assessed by comparing changes in serial FTF measurements with measurements of changes in lumbar flexion obtained radiographically. The degree of correlation between changes in the FTF or MFTF distance and changes in lumbar flexion has not been determined. Because some patients can reach beyond the level of the floor, the MFTF method would be preferable to the FTF method in a validity study. The MFTF method, as described in this study, seems to be reliable and therefore appropriate for any future study addressing the criterion-related validity of MFTF measurements. Conclusions The MFTF method, as used in this study, appears to be a very reliable method for measuring forward bending of patients with LBP. Measurements are easily and quickly obtained with the MFTF method, and the reliability of MFTF measurements is apparently unaffected by an increase in pain during forward bending. The results of this study, however, do not provide evidence as to whether the MFTF method can be used to assess changes in a patient's ability to flex the lumbar spine. Further study is needed to determine the criterion-related validity of serial MFTF measurements for predicting the change in the amount of lumbar flexion. Acknowledgment We thank the therapists at Malcolm Grow Medical Center for their assistance in data collection. References 1 Mayer TG, Tencer AF, Kristoferson S, et al. Use of noninvasive techniques for quantification of spinal range of motion in normal subjects and chronic low back dysfunction patients. Spine. 1984;9:589-595. 2 McKenzie RA. The Lumbar Spine: Mechanical Diagnosis and Therapy. Waikanae, New Zealand: Spinal Publications Ltd; 1981. 3 Maitland GD. Vertebral Manipulation. 5th ed. London, England: Butterworth & Co (Publishers) Ltd; 1986. 4 Twomey L, Taylor. A description of two new instruments for measuring the ranges of sagittal and horizontal plane motion in the lumbar region. Australian Journal of Physiotherapy. 1979;25:201-203. 5 Loebl WY. Measurements of spinal posture and range of spinal movement. Annals of Physical Medicine, 1967;9:103-110. 6 Cox JM. Lou, Back Pain: Mechanism, Diagnosis and Treatment. 4th ed. Baltimore, Md: Williams & Wilkins, 1985. 7 Burton K. Regional lumbar sagittal mobility. Clinical Biomecbanics. 1986;1:20-26. 8 Macrae IF, Wright V. Measurement of back movement. Ann Rheum Dis. 1969;28:584-589. 9 Schober P. The lumbar vertebral column and backache. Muenchener Medizinesch Worchenschrift. 1937;84:336-338. 10 Hoppenfeld S. Physical Examination of the Spine and the Extremities New York, NY: Appleton-Century-Crofts; 1976. 11 Frost M, Stuckey S, Smalley LA, et al. Reliability of measuring trunk motions in centimeters. Phys Ther. 1982;62:1431-1437. 12 Broer MR, Galles NRG. Importance of relationships between various body measurements in performance of the toe-touch test. Research Quarterly. 1958;29:253-263. 13 Ponte DJ, Jensen GH, Kent BE. A preliminary report on the use of the McKenzie protocol versus the Williams protocol. Journal of Orthopaedic and Sports Physical Therapy. 1984;6:130-139. 14 Kendall PH, Jenkins JM. Exercises for backache: a double-blinded controlled trial. Physiotherapy 1968;54:154-157. 15 Lidstrom A, Zachrisson M. Physical therapy on low back pain and sciatica: an attempt at evaluation. Scand J Rehabil Med. 1970;2:37-42. 16 Haldeman S, Gilles J, Haldeman J, et al. Low back pain: a study of 50 patients on a group exercise program. Physiotherapy Canada. 1975;27:71-77. 17 Buxton D. Extension of the Kraus-Weber test. Research Quarterly 1957;28:210-217. 18 Shrout PE, Fleiss JL: Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-428. 19 Rae PS, Waddell G, Venner RM. A simple technique for measuring spinal flexion: its use in orthopedic practice. J R Coll Surg Edinb. 1984;29:281-284. 20 Moll JMH, Wright V. Measurement of spinal movement. In: Jayson M, ed. The Lumbar Spine and Back Pain. 3rd ed. New York, NY: Churchill Livingstone Inc; 1987. 21 Biering-Sorenson F. Physical measurements as risk indicators for low back pain. Spine. 1984;9:106-119.

[TABULAR DATA OMITTED]

M Gauvin, MS, PT, is Chief, Physical Therapy, USAF Hospital, Tyndall Air Force Base, FL 32404. This research was conducted while Capt Gauvin was a physical therapy master's degree candidate at the Medical College of Virginia, Virginia Commonwealth University, Richmond, Va. Address all correspondence to Capt Gauvin at 7509 Linda Ln, Panama City, FL 32404. D Riddle, MS, PT, is Assistant Professor, Department of Physical Therapy, School of Allied Health Professions, Medical College of Virginia, Virginia Commonwealth University, PO Box 224, MCV Station, Richmond, VA 23298. J Rothstein, PhD, PT, is Associate Professor, Department of Physical Therapy, School of Allied Health Professions, Medical College of Virginia, Virginia Commonwealth University. This study was approved by the Research Committee of Malcolm Grow Medical Center at Andrews Air Force Base, Maryland. The opinions expressed herein are solely those of the authors and may not be construed as an official position of the US Air Force Medical Service or the US Department of Defense. This article was submitted April 28, 1989, and was accepted February 14, 1990