Related Experiment Video
Updated: Jul 27, 2026
The Multiple Sclerosis Performance Test (MSPT): An iPad-Based Disability Assessment Tool
Published on: June 30, 2014
Comparison of commonly used orthopaedic outcome measures using palm-top computers and paper surveys
Khaled J Saleh1, David M Radosevich, Rida A Kassim
1Department of Orthopaedic Surgery, School of Medicine, University of Minnesota, 492-420 Delaware Street SE, P.O. Box 492, Minneapolis, MN 55455, USA. saleh002@tc.umn.edu
Introduction:
Measuring patient-perceived outcomes following orthopaedic procedures have become an important component of clinical research and patient care. General and disease-specific outcomes measures have been developed and applied in orthopaedics to assess the patients' perceived health status. Unfortunately, paper-based, self-administered instruments remain inefficient for collecting data because of: (a) missing data (b) respondent error, and (c) the costs to administer and enter data.
Objective:
To study the comparability of palm-top computer devices and paper-pencil self-administered questionnaires in the collection of health-related quality of life (HRQL) information from patients.
Methods:
The comparability of administering HRQL questionnaires using palm-top computer and traditional paper-based forms was tested in a sample of 96 patients with complaints of hip and/or knee pain. Each patient completed mailed versions of the Medical Outcomes Study (MOS), 36-item Health Survey (SF-36), and Western Ontario and McMasters University Arthritis Index (WOMAC) three weeks prior to presenting to clinic. At the clinic they were asked to complete the same outcomes measures using the palm-top computer or a paper-and-pencil version.
Analysis:
In the analysis, scale distributions, floor and ceiling effects, internal consistency and retest reliability of scales were compared across the two data collection methods. Because the baseline characteristics of the groups were not strictly comparable according to age, the data were analyzed for the entire sample and stratified according to age.
Results:
Few statistically significant differences were found for the means, variances and intra-class correlation coefficients between the methods of administration. While the scale distribution between the two methods was comparable, the internal consistency of the scales was dissimilar.
Conclusions:
Administration of HRQL questionnaires using portable palm-top computer devices has the potential advantage of decreased cost and convenience. These data lend some support for the comparability of palm-top computers and paper surveys for outcomes measures widely used in the field of orthopaedic surgery. The present study identified the lack of reliability across modes of administration that requires further study in a randomized comparability trial. These mode effects are important for orthopaedic surgeons to appreciate before implementing innovative data-capture technologies in their practices.
