BMJ No 7123 Volume 315
Paper Saturday 20/27 December Christmas 1997 issue
Reliability of distance estimation by doctors and patients: cross sectional study
Basil Sharrack, Richard A C HughesAbstract
Objective: To assess the reliability and accuracy of distance estimated by doctors and patients.
Design: Comparison between estimated and measured distances of six familiar sites around Guy's Hospital, London.
Subjects: 100 hospital consultants and 100 patients.
Main outcome measures: Median (range) of estimated distances, and mean (SD) of the difference between estimated and measured distances.
Results: Both doctors and patients gave a wide range of estimates of distance. The estimates differed by up to 14.6-fold from the measured distances, and the difference between minimum and maximum estimates was up to 62.5-fold.
Conclusion: Doctors and patients were inaccurate at estimating distances, which implies that estimates of distances walked are not reliable indicators of a person's health.
The assessment of a patient's walking ability is a simple and practical method of evaluating the state of respiratory, cardiovascular, peripheral vascular, and neurological disease.(1,2) Such assessment correlates well with more sophisticated assessments of cardiorespiratory function or muscle strength(3) and is important in assigning scores in many clinical disability rating scales - for example, Kurtzke's expanded disability status scale for multiple sclerosis.(4)
The two most common methods for patient assessment are the maximum distance a patient can walk or the distance they can walk until the onset of symptoms. These distances are infrequently measured in clinical practice. Doctors have traditionally relied on their own or patients' estimates of the distance walked around familiar places. One study assessing the accuracy of trained and untrained artillery observers in estimating target distances ranging from 600m to 1550m showed wide variability.(5) To our knowledge, there are no published studies assessing the accuracy of distance estimates made by doctors and patients.
Subjects and methods
We sent a questionnaire to all 198 consultants in our hospital asking them to estimate (in yards or metres) the dimensions of a hospital ward and the distances between five familiar sites at the hospital. A category for don't know was provided to prevent guessing. One hundred and five (53%) questionnaires were returned, of which 100 were completed. The same questionnaire was given to 100 consecutive adult patients from a general medical and neurological ward and a neurology outpatient clinic. None of the consultants or patients had an overt psychiatric disorder or cognitive dysfunction. All study sites were later measured with an architect's tape measure in metres.
ResultsThe consultants were more familiar with the hospital sites than the patients. The number of consultants giving estimates for the six distances varied between 45 and 97 and the number of patients between 10 and 62 (table). Both consultants and patients inaccurately estimated the distances. Their mean estimates correlated moderately with the measured distances (r=0.73 and 0.56 respectively), and the range of estimates was wide and generally greater for consultants. The estimates for the whole group differed by up to 14.6-fold from the measured distances, and the difference between minimum and maximum estimates was up to 62.5-fold. This variability was partly due to the presence of a few outliers (figure) since the differences between the measured distances and the median estimates of both groups were small. |  | |
| Differences between estimated and measured distances (in metres) for consultants (n=100) and patients (n=100). |
| Estimates of distance (in metres) by consultants and patients | ||||||
| Measured distance (metres) | Consultants (n=100) | Patients (n=100) | ||||
|---|---|---|---|---|---|---|
| No of estimates | Median estimate (range) |
Mean (SD) difference* | No of estimates | Median estimate (range) |
Mean (SD) difference* | |
| 6.6 | 45 | 8.2 (64.9) |
39.3 (10.15) |
31 | 9.1 (87.8) |
12.3 (24.2) |
| 18.8 | 46 | 27.4 (260) |
22.5 (42.2) |
30 | 30.2 (175.6) |
23.3 (36.3) |
| 41.7 | 95 | 36.6 (449.9) |
6.7 (50.8) |
15 | 24.7 (448) |
22.8 (111.2) |
| 116 | 94 | 91.4 (401) |
-4.3 (67.2) |
10 | 64 (234.1) |
-30.8 (76.1) |
| 128 | 97 | 109.7 (478.2) |
-1.3 (83.7) |
62 | 150 (438) |
24.3 (100.2) |
| 319.1 | 94 | 274.3 (2909.6) |
19.9 (343)** |
58 | 402.3 (101.7) |
133.8 (263.2)** |
| * Between estimated and measured distances. **P=0.023. | ||||||
When estimates were expressed as a percentage of the measured distances the estimates for shorter distances were more inaccurate than those of the longer distances. The patients' mean estimate of a ward 6.6 m wide was 17.4 m, an error of 163.9%, while the consultants' mean estimate of the same ward was 10.1 m, an error of 52.8%. On the other hand, the patients' mean estimate of a 319.1 m walk to the local station was 452.6 m, a 41.8% error, while the consultants' mean estimate of the same distance was 339.2 m, a 6.3% error. However the differences between the mean estimates of both groups were not significant, with the exception of the 319.1 m distance (table).
Discussion
This study suggests that people are inaccurate at estimating distances and that medical education is no safeguard. The range of estimates was wide suggesting that any decisions about health based on estimating distance are unreliable. Although estimates were proportional to the distance measured - indicating that consultants and patients were capable of comparing distances and therefore possibly able to estimate changes - the potential value of this observation needs to be evaluated against the test-retest variability of these estimates. The comparable inaccuracy of estimates for both groups suggests that selecting mainly patients with neurological disease (90%) did not bias results or limit their generalisability to other patient groups. Participants are also unlikely to have deliberately provided spurious estimates since the outlier values were from participants who had given more reasonable estimates. As both groups comprised adults the results cannot be generalised to other ages.
Clinical assessments and therapeutic decisions are often based on estimates of distance - for example, the severity of angina, claudication, and chronic respiratory failure - and the effect of treatment on these conditions is usually assessed by the distance a patient can walk before the onset of symptoms. The results of some of the most hailed clinical trials in multiple sclerosis have been based on clinical disability scales related to ambulation, including the expanded disability status scale.(6,7) A 1.0 step change on this 20 grade scale is regarded as an important change, although the difference between grades 5.5, 5.0, 4.5, and 4.0 is the ability to walk 100, 200, 300, or 500 metres respectively.(2) Such distances are usually estimated and infrequently measured in hospital wards and outpatient clinics. The economic implication of distance estimation is considerable as over 1.2 million people in the United Kingdom are in receipt of the higher rate component of the disability living allowance (currently £33.90 ($54.20) a week) (Department of Social Security, personal communication). This allowance is paid to five categories of patients, including people who have difficulty in walking.(8) A person is eligible for this allowance on the basis of a self assessment questionnaire.(9)
| Key Messages | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
We thank the patients and consultants who participated, and
Noor and Sana Sharrack for their help in measuring some of the study
distances.
Funding: None.
Conflict of interest: None.
Department of Neurology,
United
Medical and Dental Schools of Guy's and St Thomas's Hospital,
Guy's
Hospital,
London SE1 9RT
Basil Sharrack, research
fellow
Richard A C Hughes, professor of neurology
Correspondence to: Professor Hughes
email: r.hughes@umds.ac.uk
References 1 Bernstein M L, Despars J A, Singh N P, Avalos K,
Stansbury D W, Light R W. Reanalysis of the 12-minute walk in patients
with chronic obstructive pulmonary disease. Chest
1994;105:163-7.
2 Sharrack B, Hughes R A. Clinical scales for multiple sclerosis.
J Neurol Sci 1996;135:1-9.
3 Sinclair D J, Ingram C G. Controlled trial of supervised
exercise training in chronic bronchitis. BMJ
1980;280:519-21.
4 Kurtzke J F. Rating neurologic impairment in multiple
sclerosis: an expanded disability status scale (EDSS).
Neurology 1983;33:1444-52.
5 Fine B J, Kobrick J L. Individual differences in distance
estimation: comparison of judgments in the field with those from
projected slides of the same scenes. Percept Motor
Skills1983;57:3-14.
6 IFNB Multiple Sclerosis Study Group, University of British
Columbia MS/MRI Analysis Group. Interferon beta-1b in the treatment of
multiple sclerosis: final outcome of the randomized controlled trial.
Neurology 1995;45:1277-85.
7 Jacobs L D, Cookfair D L, Rudick R A, Herndon R M, Richert J R,
Salazar AM, et al. Intramuscular beta-1a for disease progression in
relapsing multiple sclerosis. Ann Neurol 1996;39:285-94.
8 Steadman P. The new disability living allowance. Arch
Dis Child 1993;68:73-4.
9 Steadman P. Applying for disability living allowance.
BMJ 1992;305:
Home | Current issue | Past issues | Classified ads | Career Focus | Feedback
Collections | About this site | About the BMJ | BMA | Medline