J ClinEpidemtoi Vol. 45, No. 10, pp. 1101-l 109, 1992 Printed in Great Britain. All rightsreserved

0895-4356/92 S5.00 + 0.00 Copyright 0 1992 Pcrgamon Press Ltd

THE EDINBURGH CLAUDICATION QUESTIONNAIRE: AN IMPROVED VERSION OF THE WHO/ROSE QUESTIONNAIRE FOR USE IN EPIDEMIOLOGICAL SURVEYS G. C. LENG and F. G. R. FOMXJJS Wolfson Unit for the Prevention of Peripheral Vascular Diseases, Department of Public Health Sciences, University of Edinburgh, Teviot Place, Edinburgh EHS 9AG, Scotland (Received in revisedform 8 April 1992)

WHO/Rose Questionnaire on intermittent claudication was developed in 1962 for use in epidemiological surveys, and has been widely used. Several population studies have shown, however, that it is only moderately sensitive (60-68%) although highly specific (90-100%). In this study, reasons for the poor sensitivity and good specificity were determined following its application to 586 claudicants and to 61 subjects with other causes of leg pain. The results showed two important findings: firstly, that over half of the false negatives were produced by one question alone; and secondly that only three questions were required to maintain the specificity of the questionnaire. This knowledge, in conjunction with the pre-testing of additional questions, enabled a new questionnaire to be constructed: the “Edinburgh Claudication Questionnaire”. This questionnaire was tested on 300 subjects aged over 55 years attending their general practitioner, and found to be 91.3% (95% CI 8&l-94.5%) sensitive and 99.3% (95% CI 98.9-100%) specific in comparison to the diagnosis of intermittent claudication made by a physician. The repeatability of the questionnaire after 6 months was excellent (kappa = 0.76, p < 0.001). These results suggest that this revised version of the WHO/Rose Questionnaire should be adopted for use in future epidemiological surveys of peripheral vascular disease. Abstract-The

Epidemiology Vascular

Questionnaire

Claudication

INTRODUCMON

disease of the lower limb arteries is a major cause of morbidity in many Western countries. Indeed, the majority of middle-aged subjects are thought to have some degree of underlying atheroma, although only a small proportion of these have suthcient disease to produce symptoms of ischaemia [I]. There are two symptoms of lower limb ischaemia, intermittent claudication and rest pain. Rest pain is a severe continuous pain which usually results in early specialist referral and treatment; intermittent claudication is much Atherosclerotic

Peripheral

Atherosclerosis

milder and is less likely to require hospital referral, hence its prevalence can be accurately determined only in a population survey. It describes the characteristic history of pain in a muscle, typically the calf, present only on exertion and never at rest. These characteristics distinguish claudication from other causes of leg pain, such as arthritis, venous disease and neurological disorders, and make it a good marker of vascular disease morbidity in the general population. Prineas et al. [2] suggested that all epidemiological surveys of peripheral vascular disease should include a questionnaire to determine the prevalence of intermittent

1101

G. C. LENGand F. G. R. FOWKES

1102

based on a small group of hospital patients in which the majority had intermittent claudication. A subsequent study on a larger sample of the general population also showed a high specificity (100%) compared to physicians’ diagnoses, but produced a much lower sensitivity (68%) [5]. In using the questionnaire in large epidemiological surveys, several authors have commented on the inadequacies of the questionnaire, particularly the low sensitivity [6, 71; indeed in a recent paper on the Whitehall study [8], Professor Rose and colleagues attempted to improve the sensitivity of the questionnaire by including an additional category, “possible claudication”. The aim of this study was to develop a new claudication questionnaire with greater sensitivity than the existing WHO/Rose questionnaire, whilst maintaining a high specificity.

claudication, and at least one non-invasive test to identify the large number of subjects with asymptomatic disease. In 1962, Geoffrey Rose at the London School of Hygiene designed a cardiovascular questionnaire to be used in epidemiological surveys to identify subjects with intermittent claudication, angina and myocardial infarction [3] (Appendix A). It was subsequently adopted by the World Health Organization, and has been applied in over 20 countries to estimate the prevalence of intermittent claudication. In 1977, Rose et al. adapted the questionnaire for self-administration [4] to eliminate observer bias and to enable postal surveys to be carried out. Initial validation of the questionnaire showed it to be 92% sensitive and 100% specific in confirming a diagnosis of intermittent claudication as made by a physician [3], but this study was

DEVELOPMENT

IDENTIFICATION THE WHO/ROSE

PROBLEMS WITH QUESTIONNAIRE

OF

586 subjects with intermittent claudication 61 subjects with leg pain due to other causes

PRE-TESTING

ALTERNATIVE

QUESTIONS

80 subjects with intermittent claudication 100 subjects with leg pain due to other causes

Y THE EDINBURGH

CLAUDICATION

QUESTIONNAIRE

EVALUATION

~1~1

Fig.1.Studydesign.

1103

The Edinburgh Claudication Questionnaire METHODS Study design

This study consists of two parts, the development of the Edinburgh Claudication Questionnaire and its evaluation (Fig. 1). Development of the Edinburgh Claudication Questionnaire IdentiJication of problems with the WHO}Rose Questionnaire The patients selected for this part of the study were chosen in order that strengths and weaknesses of the WHO/Rose Questionnaire could be identified, on the basis that accuracy in the general population had already been established. The questionnaire was applied only to subjects either with known claudication or other causes of leg pain. The self-administered version of the WHO/ Rose Questionnaire was sent by post to 586 claudicants (a group participating in a follow-up study), and to 61 subjects with leg pain due to other causes (identified from a computerized diagnostic register). A positive questionnaire diagnosis of claudication was made only if the “correct” answer was given to all questions. All subjects had previously attended the Peripheral Vascular Clinic at the Royal Infirmary of Edinburgh, and had a definite diagnosis of intermittent claudication, or not, made by a consultant. The ankle : brachial pressure ratio (ABPI) was also measured in all subjects, and a treadmill exercise test performed when the ABPI result was borderline. All claudicants had a positive history of intermittent claudication, plus a low ABPI ( 20% drop in ankle systolic pressure following exercise. The 61 subjects with other causes of leg pain did not have intermittent claudication, or any underlying vascular abnormalities, and included diagnoses of: “referred pain from the back” (21%), “unknown cause” or “idiopathic leg pain” (23%), “neurogenic leg pain” (16%), “arthritis” (12%), “venous” and “orthopaedic” (28%). Pre-testing alternative questions

Six possible alternative questions were developed following discussion with several vascular specialists. These questions (listed in Table 2) concerned the onset, course and quality of pain, plus a diagram of the lower torso to enable the subject to mark the exact site of pain. The

questions were tested on 50 consecutive patients attending the Peripheral Vascular Clinic with intermittent claudication, and was also sent by post to 100 subjects with leg pain on exercise due to other causes (identified from a computerized outpatient diagnostic register). One followup letter was sent after 4 weeks, and a total of 87 questionnaires were returned of which 21 were unanswered because the subjects were no longer experiencing pain. All subjects were defined primarily on the basis of history. The claudicants also had a low ABPI or drop in post-exercise ankle pressure, whereas those with “other leg pain” did not. Comparisons were made between the new and old questions, and any potential changes in sensitivity and specificity identified. Evaluation of the Edinburgh Claudication Questionnaire The Edinburgh Claudication Questionnaire (Appendix B) was developed in the light of the above results, and was then tested for validity and repeatability. These need to be determined in subjects similar to those at whom the questionnaire will ultimately be directed [9]. Therefore the questionnaire was applied to a reasonably representative sample of the general population. It was also applied to a clinic sample with a high prevalence of claudicants (58%) in order to supplement the low numbers of claudicants found in the general population. of Validity Questionnaire

the

Edinburgh

Claudication

The accuracy of the new questionnaire in diagnosing intermittent claudication was determined by comparison to the diagnosis made by a clinician. Again, a positive questionnaire diagnosis of claudication was made only if the “correct” answer was given to all questions. Vascular clinic patients. Fifty new patients attending the Peripheral Vascular Clinic with “leg pain”, aged over 55 years, were selected consecutively. Suitable patients were identified by a clerk on the basis of information provided by the referring doctor. Each subject was given a questionnaire to complete on arrival, and was then seen by two clinicians, independently, and each “blind” to the responses to the questionnaire and the findings of the other clinician. Each clinician was instructed to ask a variety of questions concerning leg pain, phrased in a number of different ways depending on the understanding and ability of the patient; this

1104

G. C. LENG and F. G. R. Fowm

interview was much more extensive than the questionnaire and employed “open” questions throughout. Each subject was classified as having “definite”, “possible” or “no” claudication, on the basis of history alone. A clinical examination was also performed after the initial diagnosis had been made, including measurement of the ABPI, plus a treadmill exercise test where indicated. There were only two differences of opinion between the two clinicians, and in each case one of the diagnoses was of “possible” claudication; the final decision was made in the light of the test results. Community patients. Three hundred patients attending their General Practitioner with any complaint, aged over 55 years, were selected consecutively. Patients were seen at three different practices within the Edinburgh area, chosen to represent different social and geographical regions. As above, each subject initially completed a questionnaire, and was then seen “blind” by one of the above clinicians and a diagnosis of “definite”, “possible” or “no” intermittent claudication made. A check was made of the medical records in all cases where any type of leg pain was reported, and the case discussed with the General Practitioner if the diagnosis remained in doubt. Repeatability of the Edinburgh Questionnaire

Claudication

The repeatability of responses to the Edinburgh Claudication Questionnaire was determined by asking 150 subjects to complete a second questionnaire after an interval of approximately 6 months. The questionnaire was sent by post to the 50 patients seen in the Peripheral Vascular Clinic, and to the first 100 patients seen in general practice. RESULTS

However, if the definition of the questionnaire was extended to include “possible claudicants” as defined by Criqui et al. in 1985 [6], (those with “exercise calf pain not present at rest, but otherwise not fully concordant with the Rose criteria”), then there was a dramatic increase in sensitivity to 91% (95% CI 89-93%), but a correspondingly large fall in specificity to 72% (95% CI 61-83%). To identify the questions which caused a loss of sensitivity and those which were important in maintaining a high specificity, the responses to individual questions were analysed. In the group with known intermittent claudication, question (f) is significantly less sensitive than the other questions, and is thus the main cause of false negative results (Table 1). If(f) is simply omitted, then sensitivity increases from 60% (95% CI 56-64%) to 79% (95% CI 68-82%) with no loss of specificity. Question (a) was found to be highly sensitive, but despite this, several subjects misinterpreted “walking” as meaning “going out for a walk”, rather than just going shopping or out visiting; this question also provides no alternative to “yes” or “no” for those who are unable to walk. In the “other leg pain” group, questions (b), (c), (f) and (h) had the highest specificities (Table 1). However, the majority of these subjects answered more than one question in the negative, and in only 11 questionnaires was a diagnosis of intermittent claudication excluded on the basis of one question alone. In these 11 questionnaires the 3 important questions were (b), (c) and (h), but not question (f) which was always found in association with Table 1. Evaluation of individual questions in the WHO/ Rose Questionnaire

Question (abbreviated)

Development of the Edinburgh Claudication Questionnaire

(a) Pain on walking?

Ident#cation of problems with the WHO/Rose Questionnaire

(c) Pain in the calf?

Analysis of the questionnaires completed by 586 known claudicants (mean age 67.9 years) showed the sensitivity of the questionnaire in detecting intermittent claudication as diagnosed by a physician to be 60% (95% CI 56-64%). The questionnaires from the 61 subjects with leg pain due to other causes (mean age 62.6 years) showed a specificity of 91% (95% CI 85-99%).

(b) Pain standing/sitting?

(d) Pain walking uphill? (f) Pain disappears on walking? (g) Action when in pain (h) Pain gone in 10 minutes of stopping?

Claudicants (n = 586)

Other leg pain (n = 61)

Sensitivity (95% CI)

Specificity (95% CI)

99.3% (99.6100%) 99.3% (99.6100%) 94.9% (93.1-96.7%) 98.8% (97.9-99.7%) 78.3% (75&81.6%) 96.6% (95.1-98.1%) 90.6% (86.8-94.4%)

13.1% (4.621.6%) 80.3% (70.3-90.3%) 24.6% (13.8-35.4%) 13.1% (4x&21.6%) 23.0% (12.433.6%) 9.8% (2.3-17.3%) 63.9% (51.8-76.0%)

Note: Question (e) (“Do you get it when you walk at an ordinary pace on the level?“) is not included because it is involved only in grading the severity of disease.

The Edinburgh Claudication Questionnaire

1105

Table 2. Accuracy of alternative claudication questions

Question (Resvonse indicatinx claudication)

(0 (ii) (iii) (iv) (v) (vi)

What happens to the pain if you try to carry on walking at the same pace? (Slays the same or gets worse) F 3”” have this pain in your leg(s) before you start walking? Dzyou ever get this pain when you are resting at home? (No) When you get this pain does it start: (Graduafly) Mark where you get this pain on the diagram: (Cu.!!) Description of the pain: Dull and aching Cramp Heaviness Pins and needles Sharp Stabbing Burning

the other three. Further analysis showed that, in this study, a questionnaire constructed of only these 3 questions would still have produced a specificity of 91%. The least specific question was (g), which also had only moderate sensitivity. Pre-testing alternative questions The sensitivity and specificity of each of the alternative questions are given in Table 2. The first 4 questions were either less accurate, or no better than the original WHO/Rose questions; the subjects’ assessment of the quality of pain was also unhelpful due to the great variability of responses given. The most useful Table 3. Summary of problems associated with the WHO/ Rose Questionnaire on intermittent claudication WHO/Rose question

(4 Do you get a pain in either leg on walking7

(b) Does this pain ever begin (d

when you are standing still or sitting7 Do you get this pain in your calf (or calves)?

(4 Do you get it when you

Conclusions -“On walking” rather ambiguous. -No alternative to “yes” or “no”. -Important for specificity. -Important for specificity. -Greater sensitivity if substituted with a diagram. -No problems.

walk uphill or hurry?

(e) Do you get it when you walk at an ordinary pace on the level? Does the pain ever disappear while you are still walking? What do you do if you get it when you are walking? What happens to it if you stand still?

-Primarily for grading the degree of symptoms. -Main cause of an overall low sensitivity. -Low specificity-better omitted for simplicity. -Important for

Claudicants (?I = 50)

Other leg pain (n=66)

Sensitivity (95% CI)

Specificity (95% CI)

88.0% (79.0-97.0%)

6.1% (0.3-I 1.9%)

90.0% (81.7-98.3%)

65.2% (53.7-76.7%)

84.0% (73.8-94.2%) 88.0% (79.0-97.0%) 98.0% (94.1-102%)

66.7% (55.3-78.0%) 31.8% (20.743.0%) 30.3% (19.2-41.4%)

50.0% 40.0% 20.0% 4.0% 8.0% 10.0% 8.0%

45.5% 68.2% 71.2% 80.3% 78.8% 86.4% 71.2%

(36.1-25.1%) (26.453.6%) (8-e31.1%) (- l&9.4%) (0.5-15.5%) (1.7-18.3%) (0.5-15.5%)

(33.5-57.5%) (57.&79.4%) (60.2-82.1%) (70.7-89.9%) (68.988.7%) (78.1~94.7%) E&2-82.1%)

question was the diagram to illustrate the site of pain (v). This was slightly more sensitive and specific than the original WHO/Rose question (c) (“Do you get this pain in your calf?‘). The marks placed on the diagram were consistent with the responses to question (c) in all but three cases, where calf pain was denied at question (c) but clearly marked on the diagram. A summary of problems associated with the WHO/Rose Questionnaire, and ways in which it might be improved, are given in Table 3. The Edinburgh Claudication Questionnaire was then developed based on this information: 5 questions in the WHO/Rose Questionnaire were left completely unchanged or modified slightly; 2 were omitted; and a diagram was included to allow subjects to mark the exact site of their pain. The new questionnaire is thus shorter and simpler, which should theoretically improve accuracy by reducing the chances of an inappropriate response. Evaluation of the Edinburgh Claudication Questionnaire Validity In the 300 community patients seen in general practice (mean age 70.2 years), the new questionnaire had a sensitivity of 91.3% (95% CI 88.1-94.5%) and a specificity of 99.3% (95% CI 98.9-100%) in detecting those with intermittent claudication as diagnosed by a physician; in the group of 50 clinic patients with leg pain (mean age 62.4 years), the sensitivity was 82.8% (95% CI 72.3-93.3%) and the specificity 100% (95% CI ,lOO-100%) (Table 4). The positive predictive value was 9 1% and the negative predictive value

G. C. LENGand F. G. R. FOWKES

1106

Table 4. Validity of the Edinburgh Claudication Questionnaire

Prevalence of intermittent claudication (physician’s diagnosis) Sensitivity (95% CI)

Community patients (n = 300)

Clinic patients (n = 50)

7.4% (23/300)

58% (29/50)

91.3% (21/23) 82.8% (24129) (88.1-94.5%) (72.3-93.3%) Specificity (95% CI) 99.3% (275/277) 100% (21/21) (98.9-K@%) (10&100%) Positive predictive value 91% 100% Negative predictive value 99% 81%

was 99% for the community patients, and 100 and 8 1% respectively for the clinic patients. All clinic patients diagnosed as having claudication by a physician were also found to have peripheral vascular disease on non-invasive testing. Five clinic patients were initially classed as “possible” claudicants by the physician; after examination, the diagnosis agreed with the questionnaire in 4 out of 5 cases. Five patients seen in general practice were also classed as “possible” claudicants, but as no investigations were performed, a definitive diagnosis was made following a detailed examination of the patients’ records and consultation with their general practitioner. All cases showed no evidence of peripheral arterial disease, and were negative on the questionnaire. In the clinic population, two cases had claudicating pain in an atypical site (the buttock). Repeatability

Seventy-five completed questionnaires were returned by the patients seen in general practice (75%); two were returned incomplete because one had suffered a stroke and the other died. Forty-two questionnaires were received from the clinic patients (84%). In the community patients, 6 subjects consistently reported intermittent claudication (8%), and 65 consistently reported no claudication (87%). In the clinic patients, 19 consistently reported claudication (45%), and 17 were consistently negative (40%). The correlation coefficient kappa was calculated for each group, and was 0.78 (p < 0.001) for the community patients and 0.71 (p -c 0.001) for the clinic patients. In the community patients there were three cases where the second questionnaire diagnosis differed from the first (4%): in two of these, leg pain had been completely denied in the first questionnaire, and answers typical of claudication given in the second; in the third case, all

responses were the same except that the pain had been described as being present standing and sitting in the first questionnaire, but not in the second. Six questionnaires disagreed over the 6 month interval in the group of clinic patients (14%): 2 cases were again due to a discrepancy in whether the pain was present standing still or sitting; 1 case had apparently developed claudication who had denied any pain on the first occasion; and the remaining 3 reported typical calf claudication on the first questionnaire, but after 6 months also had pain in the foot which was present standing and sitting, possibly indicating the development of rest pain. DISCUSSION

Epidemiological studies of peripheral vascular disease in different countries have shown that the prevalence of intermittent claudication varies from as low as 0.4% in Odense [lo] to as high as 10% in East Finland [l l] in male subjects aged over 55 years. However, the majority of studies show prevalences between 2 and 3% [12]. For a symptom of such low prevalence it is particularly important that any test used to detect it has a high specificity. For example, for a symptom prevalence of 2.5%, a test with 90% sensitivity and 70% specificity would greatly overestimate the prevalence (31.5%) due to the inclusion of many false positives; even if the specificity were increased to 95%, the estimated result would still be too high (7.1% prevalence). Therefore if a single test, or questionnaire, is required to produce an accurate estimation of the prevalence of intermittent claudication, maintaining a high specificity is essential. The sensitivity of the WHO/Rose Questionnaire has been shown to be very low (g-30%) in comparison to non-invasive tests of peripheral vascular disease [6, 131, which is not surprising as the vast majority of peripheral vascular disease is asymptomatic [l]. Indeed, it is unrealistic to expect a questionnaire designed to diagnose a symptom complex to agree closely to anything other than a physician’s diagnosis, also based on the history. Therefore in this study, the physician’s diagnosis was chosen as the “gold standard”, although in most cases the presence of underlying peripheral vascular disease was also confirmed. Ideally, non-invasive tests would have been performed on all patients, but this was not feasible and may therefore represent a possible limitation to the results.

The Edinburgh Claudication Questionnaire

Development of the Edinburgh Claudication Questionnaire The moderate sensitivity (60%) and the relatively high specificity (91%) of the WHO/Rose Questionnaire demonstrated in this study are similar to the figures found by Richard et al. where the questionnaire was also validated against a physician’s diagnosis [5]. The specificity was slightly lower in the Edinburgh sample probably because this was not a population sample, but deliberately included “difficult” cases with other causes of leg pain on exertion to identify questions involved in maintaining a high specificity, and possibly because the questionnaire was self-administered. There are no other published reports of the accuracy of the self-administered version of the Rose Questionnaire, although it was said to show “good agreement” with the original version when it was first tested on a group of over 1800 male civil servants in Whitehall [4]. However, an independent comparison of the two versions showed only 80% concordance [14]. The alternative classification suggested by Criqui et al. in 1985 [6] which included “possible” claudicants would have greatly improved the sensitivity of the WHO/ Rose Questionnaire in this study, but would have reduced specificity to an unacceptably low level. A similar effect was identified by Criqui et al. when they first applied this definition to identify subjects with large vessel peripheral vascular disease. The analysis of the responses to the WHO/ Rose Questionnaire by the group with “other leg pain” clearly pin-pointed three questions which were of prime importance in maintaining the specificity of the questionnaire. The low sensitivity of the questionnaire was mainly produced by question (f) (Does the pain ever disappear while you are still walking?“), which was also identified as being problematical in the recent report from the Whitehall Study by Davey Smith et al. [8], where subjects were included whose pain disappeared whilst walking but who otherwise fulfilled all the WHO/Rose criteria. Using this extended classification, the prevalence of intermittent claudication in the Whitehall study increased from 0.8 to 1.0%. There was little benefit obtained from using most of the questions that were tested as possible replacements for the original WHO/Rose questions, as the greatest improvement could be obtained by simply omitting question (f). The analysis showed, not unexpectedly, that any

1107

assessment of the quality of pain was extremely variable; this confirms the original findings of Rose [3] who assessed the quality of the pain of myocardial ischaemia, but not of claudication. However, the use of a diagram to describe the site of leg pain was both accurate and easy for the subject to complete. The WHO/Rose Questionnaire on chest pain employs a diagram of the upper torso to aid location of cardiac pain, but no equivalent was ever included for the lower limbs. The pain of peripheral vascular disease is not complicated by radiation to different sites, as is the pain of myocardial ischaemia, but confusion can still arise when subjects do not know where the calf is, or who experience pain only in the thigh or buttock. Evaluation of the Edinburgh Claudication Questionnaire Application of the Edinburgh Claudication Questionnaire to the general practice population showed a higher prevalence of disease (7%) than in most epidemiological surveys [l], possibly because the sample was older or because of selection bias. These patients were not a true population sample, but they were more typical of subjects examined in epidemiological surveys than those in the clinic group. It was therefore encouraging to find an extremely high sensitivity in this group, even though the result was based on a relatively small number of claudicants. Unfortunately it was not feasible to test all the community patients for the presence of underlying peripheral vascular disease, but in each case a check of the medical records was made. However, in view of the 100% specificity of the physician’s diagnosis of claudication in detecting underlying peripheral vascular disease in the clinic patients, this omission is probably unimportant. The clinic subjects were a highly selected group of patients with leg pain, 58% of which had intermittent claudication. However, even in this “difficult” group, the questionnaire remained highly specific, and was still over 80% sensitive in detecting claudication. A direct comparison has not been made because of the extremely large sample size that would have been required. There is only one published report of the repeatability of the Rose Questionnaire, where it was retested after an interval of 6 months in a large population survey in Finland [15]. The reliability coefficient kappa was calculated for both men (0.45) and women (0.38), showing

G. C. LENG and F. G. R. FOWKE.Y

1108

reasonable, but not good, agreement. The repeatability of the Edinburgh Claudication Questionnaire after an interval of 6 months was excellent, with a kappa value of at least 0.71 in both the clinic and the community patients. The agreement was slightly less good in the clinic patients, probably because this was a group who all had leg pain due to a variety of causes. In the 9 cases where a discrepancy in the diagnosis of claudication occurred over the 6 month period, 6 cases could probably be explained by a true worsening of symptoms. The only real difficulty was probably due to the reporting of whether the pain was present standing or sitting, which may be due either to difficulty in understanding the question, or to a true change in the nature of the condition. In conclusion, we have developed a new questionnaire, based on the original WHO/Rose Questionnaire, to diagnose intermittent claudication in epidemiological surveys of peripheral vascular disease. It is highly specific, and the results indicate that it has a greater sensitivity than the WHO/Rose Questionnaire. This improvement in accuracy and the excellent repeatability, suggest that the new Edinburgh Claudication Questionnaire should now replace the WHO/Rose Questionnaire in future epidemiological surveys of peripheral vascular disease. Acknowledgements-We thank the following for their contribution to this studv: Dr E. Houslev. Mr P. Donnan. Mr M. R. Whyman, ani the doctors and.ancillary staff ai the Health Centres in West Calder, Bruntsfield and East Trinity, Edinburgh.

2.

3. 4.

5.

6.

I.

8.

9.

10.

11. 12.

13. 14.

REFERENCES 1. Fowkes FGR. Epidemiology of atherosclerotic arterial disease in the lower limbs. Eur J Vast Surg 1988; 2: 283-291.

15.

Prineas RJ, Harland WR, Janzon L et al. Recommendations for use of non-invasive methods to detect atherosclerotic peripheral arterial disease in population studies. Circulation 1982: 65: 1561A-1566A. Rose GA. The diagnosis of ischaemic heart pain and intermittent claudication in field surveys. Bull WHO 1962; 27: 645-658. Rose G, McCartney P, Reid DD. Self-administration of a questionnaire- on chest pain and intermittent claudication. Br J Prev Sot Med 1977: 31: 42-48. Richard JL, Ducimetiere P, Elgrishi I dt al. Depistage par questionaire de l’insuffisance coronarienne et de la claudication intermittente. Rev Epidemiol Med Sot Sante Pub1 1972; 20: 735-755. Criqui MH, Fronek A, Klauber MR et al. The sensitivity, specificity and predictive value of traditional clinical evaluation of peripheral arterial disease: results from non-invasive testing in a defined population. Circulation 1985; 71: 516-522. Fowkes FGR, Housley E, Cawood EHH et al. Edinburgh Artery Study; prevalence of asymptomatic and symptomatic peripheral vascular disease in the general population. Int J Epidemiol 1991; 20: 384392. Davey Smith G, Shipley MJ, Rose G. Intermittent claud&ation, heart d&aHe risk factors and mortality. The Whitehall Studv. Circulation 1990: 82: 19251931. Del Greco, Walop W, McCarthy RH. Questionnaire development: the pretest. Can Med Assac J 1987; 136: 1025-1026. Rose GA, Ahmeteli M, Checgacci L et al. Ischaemic heart disease in middle-aged men. Prevalence comnarisons in Eurone. Bull WHO 1968: 38: 885895.Heliovaara M, Karvonen MJ, Vilhunden R et al. Smoking, carbon monoxide and atherosclerotic diseases. Br Med J 1978; 1: 268-270. Leng GC, Fowkes FGR. Epidemiology of peripheral vascular disease. Curr Med Literature: Thrombosis 1991; 1: 35-43. Isaacson S. Venous occlusion plethysmography in 55 year old men: a population study in Malmo, Sweden. Acta Med Seand 1972; 537 (SUDD~.):l-62. Van Ganse W, Van Hoornd G,-de Backer G et al. L’interview et le questionaire auto-administre de Rose dans une Etude pilot d’athbrosclerosis. Rev Epidemiol Med Sot Sante Pub1 1972; 20: 7-14. Reunanen A, Takkunen H, Aromaa A. Prevalence of intermittent claudication and its effect on mortality. Actn Med Stand 1982; 211: 249-256.

(Appendices A and B opposite)

1109

The Edinburgh Claudication Questionnaire APPENDIX

A

The WHO/Rose Questionnaire on Intermittent Claudication (a) (b) (c) (d) (e) (f) (9)

(h)

Do you get a pain in either leg on walking? 2. 0 No 1. 0 Yes Does this pain ever begin when you are standing still or sitting? 2. lJ No 1. 0 Yes Do you get this pain in your calf (or calves)? 2. 0 No 1. 0 Yes Do you get it when you walk uphill or hurry? 2. ? ?No 1. 0 Yes Do you get it when you walk at an ordinary pace on the level? 2. 0 No 1. 0 Yes Does the pain ever disappear while you are still walking? 2. 0 No 1. 0 Yes What do you do if you get it when you are walking? 1. 0 stop 2. 0 Slow down 3. ? ?Continue at same pace What happens to it if you stand still? 1. 0 Usually continues more than 10 minutes 2. 0 Usually disappears in 10 minutes or less

Definition of positive classification requires all of the following responses: “Yes” to (a), (c) and (d); “No” to (b) and (f); “stop” or “slow down” to (g); and “usually disappears in 10 minutes or less” to (h). Grade 1 = “no” to (e) and Grade 2 = “yes” to (e).

APPENDIX

B

The Edinburgh Claudication Questionnaire

(1) Do you get a pain or discomfort in your leg(s) when you walk?

If you answered “Yes” to question (I)-please

Yes No I am unable to walk

0

answer the followiug questiolls. Otherwise you need not continue.

(2) Does this pain ever begin when you are standing still or sitting?

Yes 0

No Cl

(3)

Do you get it if you walk uphill or hurry?

Yes IJ

No 0

(4)

Do you get it when you walk at an ordinary pace on the level?

Yes 0

No 0

(5)

What happens to it if you stand still? Usually continues more than 10 minutes Usually disappears in 10 minutes or less

(6)

Where do you get this pain or discomfort? Mark the place(s) with “ x ” on the diagram below.

Front

Back

De&iti011 of positive cJasdcati00 requires all of the following responses: “Yes” to (I), “No” to (2), “Yes” to (3), and “usually disappears in 10 minutes or less” to (5); grade 1 = “No” to (4) and grade 2 = “Yes” to (4). If these criteria are fulfilled, a defioite claudicant is one who indicates pain in the calf, regardless of whether pain is also marked in other sites; a diagnosis of atypical claudication is made if pain is indicated in the thigh or buttock, in the absence of any calf pain. Subjects should not be considered to have claudication if pain is indicated in the hamstrings, feet, shins, joints or appears to radiate, in the absence of any pain in the calf.

Rose Questionnaire for use in epidemiological surveys.

The WHO/Rose Questionnaire on intermittent claudication was developed in 1962 for use in epidemiological surveys, and has been widely used. Several po...
953KB Sizes 0 Downloads 0 Views