- Email: [email protected]
An Audit of Hip Radiographs Performed for General Practitioners
Clinical Radiology (2000) 56: 970±972 doi:10.1006/crad.2001.0715, available online at http://www.idealibrary.com on
An Audit of Hip Radiographs Performed for General Practitioners R A J B H AT T, A R U M U G A M R A J E SH , B R UN O MO R G A N, D AVID F IN L AY Leicester Royal In®rmary, Leicester, U.K. Received: 6 September 2000 Revised: 29 November 2000
Accepted: 27 December 2000
AIM: We carried out a prospective study of the GP referrals for hip radiographs to ®nd out the proportion of radiographs falling within the RCR guidelines and to ascertain the reasons for not following the guidelines. MATERIALS AND METHODS: One thousand two hundred and ®fty six consecutive hip radiographs were performed for GPs over a 12-month period. Questionnaires were sent to GPs for 873 referrals during an audit period of 18 months Four hundred and ®fty three questionnaires were returned with adequate information. The proportion of change in management between the patients falling within the guidelines and those outside the guidelines was compared using Fisher's Exact Test (SPSS Package Version 9) and the corresponding 95% con®dence interval (CI) was calculated. RESULTS: Only 194 (43%) of the 453 referrals were within RCR guidelines and 259 (57%) were outside guidelines. In the group where RCR guidelines were followed, 60% were normal, 34% had degenerative changes and 6% other diagnoses. The radiological reports changed management in 44%. In the group where guidelines were not followed 76% were normal, 18% had degenerative changes and 6% other diagnoses. The radiological reports changed management in 37%. There was no evidence of dierence in the change in management between these two groups. Twenty per cent of the radiographs showed abnormalities other than hip changes. Of these, only 17% were abnormalities other than degenerative changes and considered signi®cant. CONCLUSIONS: This audit does not reveal why the majority of the requests fell outside guidelines. This may be due to lack of awareness amongst GPs about guidelines, patient pressure, medico-legal problems and the need for reassurance. However, the audit does not show any evidence that following guidelines would increase eciency of referral. Bhatt R., et al. (2001). Clinical Radiology 56, 970±972. # 2001 The Royal College of Radiologists Key words: general practitioners, Royal College of Radiologists, hip radiographs, audit.
Open access to radiological services for GPs has been shown to help in primary management, reduce specialist referral and improve the quality of the service oered [1±3]. However, these investigations may involve high radiation dose, be clinically inappropriate and thereby cause an unnecessary burden on resources [4]. The Royal College of Radiologists (RCR) has published guidelines concerning indications for imaging investigations. The guidelines state that an X-ray pelvis is not indicated routinely for hip pain with full movement and should be performed only if there is a complex history or if symptoms and signs persist. In patients with hip pain with limited movement, X-ray pelvis is not indicated initially but can be performed if hip replacement might be considered. Hip X-rays are indicated if avascular necrosis is being Author for correspondence and guarantor of study: Raj Bhatt FRCR, Specialist Registrar in Radiology, Leicester Royal In®rmary, Leicester, LE1 5WW, U.K. Tel: 44 (0)116 2541414, Ext. 5468: Fax: 44 (0)116 2585584. 0009-9260/01/120970+03 $35.00/0
considered as the diagnosis [2]. The RCR Working Party predicted that the guidelines would only be eective if their use was supported by monitoring and review of referral practice [5]. This audit studies GP referrals for hip radiographs in the context of RCR guidelines. Our Department of Radiology performs approximately 1300 hip radiographs annually for GPs, using a standard AP projection for the pelvis. Occasionally, a lateral view of the aected hip is performed if deemed necessary.
MATERIALS AND METHODS
One thousand two hundred and ®fty six consecutive hip radiographs were performed for GPs over a 12-month period. Questionnaires with radiological reports were sent to GPs for 873 consecutive referrals during an audit period of 18 months. Five hundred and ®fty three questionnaires were returned with adequate information and were included in # 2000 The Royal College of Radiologists
971
AUDIT OF HIP RADIOGRAPHS PERFORMED FOR GPS
the audit. The remaining questionnaires were not returned or returned with inadequate information. All ®lms were reported by one musculo-skeletal radiologist (DF). The questionnaire concerned the reasons for radiographic examinations, whether patient pressure was a main factor, if there was a speci®c diagnosis in mind, the sign and symptoms involved, whether any treatment was instituted before the result and whether the result had altered management. GPs were asked to complete and return the forms. These and the radiological reports were then analysed. The proportion of change in management between the patients falling within the guidelines and those outside the guidelines was compared using Fisher's Exact Test (SPSS Package Version 9) and the corresponding 95% con®dence interval (CI) was calculated. RESULTS
A questionnaire was sent in 873 cases and returned in 453 cases. The group with returned questionnaires was well matched to the whole group in terms of age, sex and the proportion of the abnormalities detected (Table 1) and forms the audited group. Thirty per cent of the patients Table 1 ± Demographic analysis of dierent groups in the audit
Age Sex Hip X-rays
Whole group (873)
Returned questionnaire (453)
Min: 19 yrs Max: 88 yrs Male 264 Female 609 Normal: 631 Abnormal: 242
Min: 19 yrs Max: 88 yrs Male 137 Female 316 Normal: 315 Abnormal: 138
were male in both groups and there were 69±72% normal radiographs in both groups. Table 2 demonstrates the range of abnormalities in the whole group. Table 3 demonstrates the reasons for referral in the whole group (in some cases this was not clear from the request card alone although reasons for referral were available in all cases from the audited group). Only 194 (43%) of the referrals were within RCR guidelines (Table 4). In the group where RCR guidelines were followed, 60% were normal, 34% had degenerative changes and 6% other diagnoses. The radiological reports contribute to change in a management in 44%. In the group where guidelines were not followed, 76% were normal, 18% had degenerative changes and 6% other diagnoses. The radiological reports contributed to change in a management in 37%. The exact nature of the change in management was not recorded. There is no evidence of a dierence in the frequency of change in management between two groups; dierence in proportions ( patient falling within guidelines±patients falling outside guidelines) 0.07 (7%), 95% CIÿ0.3 to 0.17; p 0.123). Twenty per cent of the radiographs showed abnormalities other than hip changes. Of these, only 17% were signi®cant abnormalities like metastases, sacro-ileitis, fracture and spondylolisthesis. Degenerative disease was not considered a signi®cant abnormality (Table 5). With respect to speci®c diagnosis, three important lesions (malignancy, infection and fractures of hips) were identi®ed from the data. None of the 16 patients referred for suspected malignancy had any evidence of malignancy. Thirteen had normal hips, 2 patients had degenerative changes and one had a hip replacement. In one patient referred for infection, there was no evidence of infection. Fracture was suspected in 20 patients. No fractures were Table 4 ± Results of radiographs related to RCR guidelines (453)
Table 2 ± Results of hip radiographs (873) Radiological report
No. of radiographs
%
Normal Degenerative changes Hip replacement Hip dysplasia Loose prosthesis Chondrocalcinosis Avascular necrosis
631 135 30 6 5 5 2
72 15 3 1 1 1 1
Within guidelines Normal Degenerative changes Others Change in management Outside guidelines Normal Degenerative changes Others Change in management
No. of radiographs
%
194 116 65 13 86 259 198 47 14 96
22 60 34 6 44 30 76 18 6 37
Table 3 ± Reasons for referral for hip radiography (653) Reason for referral
No. of radiographs
%
Patient symptoms and/or signs No speci®c diagnosis in mind To exclude benign treatable disease To exclude malignancy Patient pressure as main reason Failure of treatment Other reasons
266 117 164 46 56 2 12
31 13 19 5 6 1 1
Table 5 ± Results not related to hip joint No abnormality 520 Lower lumbar spine degenerative disease Spondylolisthesis Vertebral fracture Sacro-ileitis Metastases
142 11 8 4 5
972
CLINICAL RADIOLOGY
identi®ed. Nine had normal hip appearances, 6 degenerative changes, 1 hip dysplasia, 3 hip replacement and one had loose prosthesis. In the patients falling within the guidelines (194), an abnormality was expected in (115) 59%. There was an abnormality other than degenerative changes only in 7 (4%). In the patients falling out with the guidelines (259), an abnormality was expected in (109) 42%. There was an abnormality other than degenerative changes only in 9 (3%). No pattern is seen to explain why the majority of referrals fall outside the guidelines.
due to lack of awareness amongst GPs about guidelines, patient pressure, medico-legal problems and reassurance. Questionnaires not returned and inadequately completed questionnaires were inherent problems in this audit. Barton et al. concluded in their study that open access to GPs reduces hospital referrals but has little eect on the proportion being actively treated [3]. Our study suggests that most hip radiographs are normal and there is no signi®cant dierence in the prevalence of normal ®ndings or change in management in the two groups. Therefore, enforcing guidelines more rigidly would have little impact on pattern of referral for hip radiographs.
DISCUSSION
The majority of requests from GPs for hip radiographs fall outside Royal College of Radiologists' guidelines. The majority of radiographs are normal. Although there is a lower rate of abnormality in the `outside guidelines' group, there was no evidence of a statistically signi®cant dierence in the change in management between the two groups. The lack of dierence between the two groups presumably relates to the high percentage of normal radiographs overall. The exact nature of the change in management is not known, as this information was not included in the questionnaire. A previous study showed that GPs regard radiology reports as having a direct eect on patient management in 22±36% of cases, a normal report being as helpful as an abnormal one [7]. It is dicult to assess the exact reasons for the majority of the requests falling outside the guidelines. This may be
APPENDIX 1 QUESTIONNAIRE 1. Why was the X-ray requested? a) Patient symptoms b) Patient signs c) Both d) Others. Please specify 2. Was there a speci®c diagnosis in mind? a) Yes. Please specify, then go to Q4 b) No. Please go to Q3 3. Did you have a main reason for the X-ray? a) Exclude malignancy b) Exclude more benign treatable disease c) Patient pressure d) Others. Please specify 4. What symptoms did the patient have? a) None b) Pain c) Limp d) Others. Please specify
REFERENCES 1 Royal College of General Practitioners and the Royal College of Radiologists. Joint working party reports on radiological services for General Practitioners. J R Coll Gen Pract 1981;31:528±563. 2 Royal College of Radiologists Working Party. Making the best use of a department of clinical radiology: Guidelines for Doctors, 4th edn. London: Royal College of Radiologists, 1998. 3 Barton E, Gallagher S, Flower CDR, et al. In¯uence on patient management of general practitioner direct access to radiological services. Br J Radiol 1987;60:893±896. 4 Royal College of Radiologists' Working Party. In¯uence of Royal College of Radiologists' guidelines on referral from general practice. BMJ 1993;306:110±111. 5 Morgan B, Mullick S, Finlay D, et al. An audit of knee radiographs performed for general practitioners. Br J Radiol 1997;70:256±605. 6 Stoddart PGP, Hall SG. Radiology is valuable to general practitioners: but who pays?. Clin Radiol 1989;40:183±856. 7 Mills KA, Reilly PM. Laboratory and radiological investigations in general practice. BMJ 1983;287:1033±1036.
5. What signs did the patient have? a) None b) Restricted movement c) Pain on movement d) Others. Please specify 6. Did you give any treatment before the X-ray results were available? a) Yes. Please go to Q7 b) No. Please go to Q8 7. What treatment was given? a) Physiotherapy b) Steroids c) Hospital referral d) Simple analgesia e) Others. Please specify 8. Did the X-ray alter your management? a) Yes b) No 9. Were you expecting an abnormality? a) Yes b) No
An Audit of Hip Radiographs Performed for General Practitioners R A J B H AT T, A R U M U G A M R A J E SH , B R UN O MO R G A N, D AVID F IN L AY Leicester Royal In®rmary, Leicester, U.K. Received: 6 September 2000 Revised: 29 November 2000
Accepted: 27 December 2000
AIM: We carried out a prospective study of the GP referrals for hip radiographs to ®nd out the proportion of radiographs falling within the RCR guidelines and to ascertain the reasons for not following the guidelines. MATERIALS AND METHODS: One thousand two hundred and ®fty six consecutive hip radiographs were performed for GPs over a 12-month period. Questionnaires were sent to GPs for 873 referrals during an audit period of 18 months Four hundred and ®fty three questionnaires were returned with adequate information. The proportion of change in management between the patients falling within the guidelines and those outside the guidelines was compared using Fisher's Exact Test (SPSS Package Version 9) and the corresponding 95% con®dence interval (CI) was calculated. RESULTS: Only 194 (43%) of the 453 referrals were within RCR guidelines and 259 (57%) were outside guidelines. In the group where RCR guidelines were followed, 60% were normal, 34% had degenerative changes and 6% other diagnoses. The radiological reports changed management in 44%. In the group where guidelines were not followed 76% were normal, 18% had degenerative changes and 6% other diagnoses. The radiological reports changed management in 37%. There was no evidence of dierence in the change in management between these two groups. Twenty per cent of the radiographs showed abnormalities other than hip changes. Of these, only 17% were abnormalities other than degenerative changes and considered signi®cant. CONCLUSIONS: This audit does not reveal why the majority of the requests fell outside guidelines. This may be due to lack of awareness amongst GPs about guidelines, patient pressure, medico-legal problems and the need for reassurance. However, the audit does not show any evidence that following guidelines would increase eciency of referral. Bhatt R., et al. (2001). Clinical Radiology 56, 970±972. # 2001 The Royal College of Radiologists Key words: general practitioners, Royal College of Radiologists, hip radiographs, audit.
Open access to radiological services for GPs has been shown to help in primary management, reduce specialist referral and improve the quality of the service oered [1±3]. However, these investigations may involve high radiation dose, be clinically inappropriate and thereby cause an unnecessary burden on resources [4]. The Royal College of Radiologists (RCR) has published guidelines concerning indications for imaging investigations. The guidelines state that an X-ray pelvis is not indicated routinely for hip pain with full movement and should be performed only if there is a complex history or if symptoms and signs persist. In patients with hip pain with limited movement, X-ray pelvis is not indicated initially but can be performed if hip replacement might be considered. Hip X-rays are indicated if avascular necrosis is being Author for correspondence and guarantor of study: Raj Bhatt FRCR, Specialist Registrar in Radiology, Leicester Royal In®rmary, Leicester, LE1 5WW, U.K. Tel: 44 (0)116 2541414, Ext. 5468: Fax: 44 (0)116 2585584. 0009-9260/01/120970+03 $35.00/0
considered as the diagnosis [2]. The RCR Working Party predicted that the guidelines would only be eective if their use was supported by monitoring and review of referral practice [5]. This audit studies GP referrals for hip radiographs in the context of RCR guidelines. Our Department of Radiology performs approximately 1300 hip radiographs annually for GPs, using a standard AP projection for the pelvis. Occasionally, a lateral view of the aected hip is performed if deemed necessary.
MATERIALS AND METHODS
One thousand two hundred and ®fty six consecutive hip radiographs were performed for GPs over a 12-month period. Questionnaires with radiological reports were sent to GPs for 873 consecutive referrals during an audit period of 18 months. Five hundred and ®fty three questionnaires were returned with adequate information and were included in # 2000 The Royal College of Radiologists
971
AUDIT OF HIP RADIOGRAPHS PERFORMED FOR GPS
the audit. The remaining questionnaires were not returned or returned with inadequate information. All ®lms were reported by one musculo-skeletal radiologist (DF). The questionnaire concerned the reasons for radiographic examinations, whether patient pressure was a main factor, if there was a speci®c diagnosis in mind, the sign and symptoms involved, whether any treatment was instituted before the result and whether the result had altered management. GPs were asked to complete and return the forms. These and the radiological reports were then analysed. The proportion of change in management between the patients falling within the guidelines and those outside the guidelines was compared using Fisher's Exact Test (SPSS Package Version 9) and the corresponding 95% con®dence interval (CI) was calculated. RESULTS
A questionnaire was sent in 873 cases and returned in 453 cases. The group with returned questionnaires was well matched to the whole group in terms of age, sex and the proportion of the abnormalities detected (Table 1) and forms the audited group. Thirty per cent of the patients Table 1 ± Demographic analysis of dierent groups in the audit
Age Sex Hip X-rays
Whole group (873)
Returned questionnaire (453)
Min: 19 yrs Max: 88 yrs Male 264 Female 609 Normal: 631 Abnormal: 242
Min: 19 yrs Max: 88 yrs Male 137 Female 316 Normal: 315 Abnormal: 138
were male in both groups and there were 69±72% normal radiographs in both groups. Table 2 demonstrates the range of abnormalities in the whole group. Table 3 demonstrates the reasons for referral in the whole group (in some cases this was not clear from the request card alone although reasons for referral were available in all cases from the audited group). Only 194 (43%) of the referrals were within RCR guidelines (Table 4). In the group where RCR guidelines were followed, 60% were normal, 34% had degenerative changes and 6% other diagnoses. The radiological reports contribute to change in a management in 44%. In the group where guidelines were not followed, 76% were normal, 18% had degenerative changes and 6% other diagnoses. The radiological reports contributed to change in a management in 37%. The exact nature of the change in management was not recorded. There is no evidence of a dierence in the frequency of change in management between two groups; dierence in proportions ( patient falling within guidelines±patients falling outside guidelines) 0.07 (7%), 95% CIÿ0.3 to 0.17; p 0.123). Twenty per cent of the radiographs showed abnormalities other than hip changes. Of these, only 17% were signi®cant abnormalities like metastases, sacro-ileitis, fracture and spondylolisthesis. Degenerative disease was not considered a signi®cant abnormality (Table 5). With respect to speci®c diagnosis, three important lesions (malignancy, infection and fractures of hips) were identi®ed from the data. None of the 16 patients referred for suspected malignancy had any evidence of malignancy. Thirteen had normal hips, 2 patients had degenerative changes and one had a hip replacement. In one patient referred for infection, there was no evidence of infection. Fracture was suspected in 20 patients. No fractures were Table 4 ± Results of radiographs related to RCR guidelines (453)
Table 2 ± Results of hip radiographs (873) Radiological report
No. of radiographs
%
Normal Degenerative changes Hip replacement Hip dysplasia Loose prosthesis Chondrocalcinosis Avascular necrosis
631 135 30 6 5 5 2
72 15 3 1 1 1 1
Within guidelines Normal Degenerative changes Others Change in management Outside guidelines Normal Degenerative changes Others Change in management
No. of radiographs
%
194 116 65 13 86 259 198 47 14 96
22 60 34 6 44 30 76 18 6 37
Table 3 ± Reasons for referral for hip radiography (653) Reason for referral
No. of radiographs
%
Patient symptoms and/or signs No speci®c diagnosis in mind To exclude benign treatable disease To exclude malignancy Patient pressure as main reason Failure of treatment Other reasons
266 117 164 46 56 2 12
31 13 19 5 6 1 1
Table 5 ± Results not related to hip joint No abnormality 520 Lower lumbar spine degenerative disease Spondylolisthesis Vertebral fracture Sacro-ileitis Metastases
142 11 8 4 5
972
CLINICAL RADIOLOGY
identi®ed. Nine had normal hip appearances, 6 degenerative changes, 1 hip dysplasia, 3 hip replacement and one had loose prosthesis. In the patients falling within the guidelines (194), an abnormality was expected in (115) 59%. There was an abnormality other than degenerative changes only in 7 (4%). In the patients falling out with the guidelines (259), an abnormality was expected in (109) 42%. There was an abnormality other than degenerative changes only in 9 (3%). No pattern is seen to explain why the majority of referrals fall outside the guidelines.
due to lack of awareness amongst GPs about guidelines, patient pressure, medico-legal problems and reassurance. Questionnaires not returned and inadequately completed questionnaires were inherent problems in this audit. Barton et al. concluded in their study that open access to GPs reduces hospital referrals but has little eect on the proportion being actively treated [3]. Our study suggests that most hip radiographs are normal and there is no signi®cant dierence in the prevalence of normal ®ndings or change in management in the two groups. Therefore, enforcing guidelines more rigidly would have little impact on pattern of referral for hip radiographs.
DISCUSSION
The majority of requests from GPs for hip radiographs fall outside Royal College of Radiologists' guidelines. The majority of radiographs are normal. Although there is a lower rate of abnormality in the `outside guidelines' group, there was no evidence of a statistically signi®cant dierence in the change in management between the two groups. The lack of dierence between the two groups presumably relates to the high percentage of normal radiographs overall. The exact nature of the change in management is not known, as this information was not included in the questionnaire. A previous study showed that GPs regard radiology reports as having a direct eect on patient management in 22±36% of cases, a normal report being as helpful as an abnormal one [7]. It is dicult to assess the exact reasons for the majority of the requests falling outside the guidelines. This may be
APPENDIX 1 QUESTIONNAIRE 1. Why was the X-ray requested? a) Patient symptoms b) Patient signs c) Both d) Others. Please specify 2. Was there a speci®c diagnosis in mind? a) Yes. Please specify, then go to Q4 b) No. Please go to Q3 3. Did you have a main reason for the X-ray? a) Exclude malignancy b) Exclude more benign treatable disease c) Patient pressure d) Others. Please specify 4. What symptoms did the patient have? a) None b) Pain c) Limp d) Others. Please specify
REFERENCES 1 Royal College of General Practitioners and the Royal College of Radiologists. Joint working party reports on radiological services for General Practitioners. J R Coll Gen Pract 1981;31:528±563. 2 Royal College of Radiologists Working Party. Making the best use of a department of clinical radiology: Guidelines for Doctors, 4th edn. London: Royal College of Radiologists, 1998. 3 Barton E, Gallagher S, Flower CDR, et al. In¯uence on patient management of general practitioner direct access to radiological services. Br J Radiol 1987;60:893±896. 4 Royal College of Radiologists' Working Party. In¯uence of Royal College of Radiologists' guidelines on referral from general practice. BMJ 1993;306:110±111. 5 Morgan B, Mullick S, Finlay D, et al. An audit of knee radiographs performed for general practitioners. Br J Radiol 1997;70:256±605. 6 Stoddart PGP, Hall SG. Radiology is valuable to general practitioners: but who pays?. Clin Radiol 1989;40:183±856. 7 Mills KA, Reilly PM. Laboratory and radiological investigations in general practice. BMJ 1983;287:1033±1036.
5. What signs did the patient have? a) None b) Restricted movement c) Pain on movement d) Others. Please specify 6. Did you give any treatment before the X-ray results were available? a) Yes. Please go to Q7 b) No. Please go to Q8 7. What treatment was given? a) Physiotherapy b) Steroids c) Hospital referral d) Simple analgesia e) Others. Please specify 8. Did the X-ray alter your management? a) Yes b) No 9. Were you expecting an abnormality? a) Yes b) No