‘‘Analysis of Paired Data in Physical Therapy Research: Time to Stop Double-Dipping?'' was a guest editorial that appeared in the August 2005 issue of the JOSPT. This editorial, written by Dr Hylton Menz, raised an important statistical issue about how to properly manage study designs that involve paired data. "Double-dipping'' refers to using an inappropriate statistical analysis for paired data. For example, if a study used 12 subjects and measured both feet of each subject, it would be inappropriate to calculate a standard error that is based on an n of 24 (2 × 12) independent feet. Double-dipping is a problem because the paired limbs within a subject are not likely to be statistically independent. This practice can result in a standard error that is smaller than it should be. Based on a spuriously small standard error and an inflated n, a statistical test may appear to be more ‘‘significant'' than it actually is. Dr Menz concluded his editorial by recommending several strategies to select only 1 of a paired limb per patient for measurement to maintain the statistical independence of the data.2 For example, we could randomly select and measure only 1 foot per subject, and then use an n of 12 (1 × 12) independent measurements in the statistical analysis. However, would we lose valuable information in doing so?

In his efforts to persuade readers to avoid the error of double-dipping, Dr Menz may have inadvertently discouraged the use of valid study designs and appropriate analysis methods for paired data. My purpose in writing this guest editorial is to discuss these designs and encourage investigators to consider all of the options available to them in planning their studies. This editorial discusses 3 situations where an appropriate design and analysis for paired data may be advantageous. I will use feet and ankles in my illustrations of paired data, but these comments also extend to other pairs of limbs and organs that may be the focus of research in physical therapy.

J Orthop Sports Phys Ther. 2006; 36(2):42-44. doi:10.2519/jospt.2006.0102 

Key Words: paired data, statistics

Study Design: Cross-sectional, secondary analysis. Objectives: To examine whether there were differences in the numbers and types of impairments on examination between 2 groups of people with low back pain (LBP), those who participated in symmetric leisure activities and those who participated in asymmetric leisure activities. Background: It has been proposed that people who repeatedly perform an activity that involves trunk movements and alignments in the same direction will develop strategies that are generalized to many activities. The repeated use of these strategies is proposed to contribute to impairments identifiable on examination and to LBP. Methods and Measures: Forty males and 40 females (mean ± SD age, 41.4 ± 13.9 years) with LBP who reported participation in either a symmetric or an asymmetric leisure activity participated in a standardized examination. Responses from 10 trunk-rotation-related impairment tests were analyzed using the Mann-Whitney U and chi-square statistics. Results: Thirty people participated in asymmetric leisure activities and 50 people participated in symmetric leisure activities. The total number of rotation-related impairments was different for the 2 groups (U = 1112, P<.01). The asymmetric group displayed more total rotation-related impairments (median, 4.0; range, 7) than the symmetric group (median, 2.0; range, 6). A greater percentage of the asymmetric group displayed more impairments on 5 out of 10 individual tests, as compared to the symmetric group (P≤.05 for all comparisons). Conclusions: Our results provide preliminary data to suggest that trunk-rotation related impairments, identified on examination, may be related to the general type of movements and alignments used repeatedly by patients with LBP.

 J Orthop Phys Ther. 2006;36(2):58-71. doi:10.2519/jospt.2006.2161

Key Words: examination, lumbar spine, spinal disorders, sports

Study Design: Resident’s case problem. Background: Identifying stress fractures of the hip can be a challenging differential diagnosis. Pain presentation is not always predictable and radiographs may not show the fracture, especially during its early stages. Hip stress fractures left untreated can displace and necessitate open reduction internal fixation or total hip arthroplasty. Diagnosis: A 70-year-old woman presented to the physical therapy clinic with complaints of right hip pain. A physician had evaluated her, and radiographs of the hip, which revealed some arthritic changes, were otherwise normal. Upon examination, the physical therapist observed an antalgic gait, a noncapsular pattern of limitation of hip motion, an empty painful end feel at the end range of motion (ROM) for hip abduction, external rotation, and flexion, and extreme tenderness to palpation over the anterior hip region. The therapist suspected a more pernicious problem than osteoarthritis and discussed his suspicion with the physician. The physician subsequently requested an MRI that revealed a femoral neck and head stress fracture that was later confirmed with a bone scan. The patient was provided with a walker for ambulation with a non-weight-bearing status for 6 weeks, after which she returned to physical therapy for progressive weight bearing and strengthening. She was discharged with a relatively pain-free hip and was ambulating with a cane. A 2-month follow-up examination revealed a pain-free hip and a return to all premorbid activities, including ambulation without an assistive device. Discussion: The presence of a normal radiograph of the hip should not be considered conclusive in ruling out a stress fracture in the hip region. The current case demonstrates how careful evaluation can reveal occult pathologies and prevent potentially catastrophic morbidity.

J Orthop Sports Phys Ther. 2006;36(2):80-88. doi:10.2519/jospt.2006.2065

Key Words: bone scan, imaging, MRI, physical therapy examination

Study Design: Descriptive study. Objectives: To determine the presence of median and ulnar neuropathies in both upper extremities of university guitarists. Background: Peripheral nerve entrapment syndromes of the upper extremities are well documented in musicians. Guitarists and plucked-string musicians are at risk for entrapment neuropathies in the upper extremities and are prone to mild neurologic deficits. Methods and Measures: Twenty-four volunteer male and female guitarists (age range, 18-26 years) were recruited from the Belmont University School of Music and the Vanderbilt University Blair School of Music. Individuals were excluded if they were pregnant or had a history of recent upper extremity or neck injury. Subjects completed a history form, were interviewed, and underwent a physical examination. Nerve conduction status of the median and ulnar nerves of both upper extremities was obtained by performing motor, sensory, and F-wave (central) nerve conduction studies. Descriptive statistics of the nerve conduction study variables were computed using Microsoft Excel. Results: Six subjects had positive findings on provocative testing of the median and ulnar nerves. Otherwise, these guitarists had normal upper extremity neural and musculoskeletal function based on the history and physical examinations. When comparing the subjects’ nerve conduction study values with a chart of normal nerve conduction studies values, 2 subjects had prolonged distal motor latencies (DMLs) of the left median nerve of 4.3 and 4.7 milliseconds (normal, <4.2 milliseconds). Prolonged DMLs are compatible with median neuropathy at or distal to the wrist. Otherwise, all electrophysiological variables were within normal limits for motor, sensory, and F-wave (central) values. However, comparison studies of median and ulnar motor latencies in the same hand demonstrated prolonged differences of greater than 1.0 milliseconds that affected the median nerve in 2 additional subjects, and identified contralateral limb involvement in a subject with a prolonged distal latency. The other 20 subjects demonstrated normal comparison studies of the median and ulnar nerves in both upper extremities. Conclusions: In this descriptive study of a population of 24 university guitarists, 4 musicians (17%) were found to have electrophysiologic evidence of median neuropathy at or distal to the wrist or carpal tunnel syndrome. Ulnar nerve electrophysiological function was within normal limits for all subjects examined.

J Orthop Sports Phys Ther. 2006;36(2):101-111. doi:10.2519/jospt.2006.2155

Key Words: median nerve, nerve conduction study, ulnar nerve