Design of study
Prostate cancer is a major cause of morbidity and mortality in men today. It is the second most common male cancer, superseded only by lung cancer, and in 2000 it accounted for 12% of all male cancer deaths in England and Wales 1.
In recent years there has been increasing pressure from the public to introduce a national screening programme for prostate cancer 2. However, there is at present insufficient evidence to support such a screening programme. Current practice measures the amount of Prostate Specific Antigen (PSA) in the serum as a predictor for prostate cancer, however, the test is non-specific and there is much debate about the optimum cut-off value to demarcate prostate cancer and thus the level at which further investigations are required 3,4,5.
Furthermore, there is no universally accepted treatment for prostate cancer. Trials are currently underway to evaluate the effectiveness of the three main treatments available; active monitoring (also known as watchful waiting), radiotherapy, and surgery (total prostatectomy), to demonstrate whether they actually do improve life expectancy or quality of life 5,6. Finally, it is well recognised that more men will die with prostate cancer than from it, and a PSA test will not distinguish between a slow-growing tumour and a highly aggressive metastatic carcinoma 5.
In September 2002 the Department of Health sent all General Practitioners (GPs) in the UK the NHS Prostate Cancer Risk Management Programme (PCRMP) guidelines on 'advising and counselling asymptomatic men who are worried about prostate cancer' with the aim to standardise and validate the information provided to men across the country 7. Previous studies, based in the US and New Zealand, have assessed how GPs manage testing for prostate cancer in asymptomatic men, but there has been no equivalent research regarding practice in the UK 8,9,10. Our study aimed to assess GP awareness of the PCRMP and whether their knowledge and current practice, with regards to prostate cancer screening in asymptomatic men, correlated with the recommendations of the PCRMP. It also aimed to assess whether GP knowledge and practice is affected by awareness of the PCRMP or by GP characteristics; gender, year of qualification, practice type and practice area.
In January 2003 a single questionnaire was posted to all 232 GPs from one urban and one suburban PCT listed on NHS internet sites as practising in the West Midlands. The questionnaire asked whether GPs were aware of the PCRMP, whether they had received the PCRMP information pack and assessed their knowledge of the PCRMP guidelines. GP knowledge was tested by a set of 7 multiple choice questions based on a hypothetical case study as well as specific questions about prostate cancer (See Appendix One).
Characteristics of each GP sent the questionnaire were simultaneously obtained from the Medical Directory 2002, the PCT lists, and telephone calls to each General Practice. This included their genders, years of qualification and surgery type, (whether multi-partner, (two) partner or single-handed practices).
Before distribution the questionnaire was independently reviewed by 4 GPs and a consultant urologist, and any ambiguous questions were amended. To encourage response the questionnaire was designed to take no more than 5 minutes to complete, and was accompanied by a pre-paid reply envelope. Each questionnaire was labelled with a unique identification number to allow a comparison between characteristics of non-responders and responders.
Answers to each of the 7 multiple-choice-questions were marked as either correct or incorrect, taking the information in the PCRMP guidelines as the gold standard. Each GP was then given a total score out of 7 by summing the number of correct responses to the questions asked. Using SPSS for Windows Version 11.0, GPs were grouped according to whether they were aware of the PCRMP or not, whether they had received the PCRMP guidelines or not and also for each GP characteristic. The mean scores achieved by each group were then compared using the t-test, thus enabling an assessment of whether awareness of the PCRMP guidelines or any GP characteristic was associated with knowledge and practice regarding management of prostate cancer. (GP responses to the individual questions are illustrated in Appendix Two and the model answers are indicated by an asterisk.)
The overall response rate was 47% (109/232). Of responders 72% (78/109) were aware of the PCRMP and 58% (63/109) of GPs recalled receiving the PCRMP information pack. The mean score of all responders was 3.45, but no GP got all 7 questions correct.
GPs who said they had received the PCRMP guidelines achieved a mean score of 3.75, and those who had not received the PCRMP guidelines achieved a mean score of 3.00 (t = 2.414, p = 0.018). GPs who qualified before 1979 achieved a mean score of 2.79 compared to a mean score of 4.05 for GPs who qualified in or after 1979 (t = 3.905, p <0.001). The average score achieved by GPs in the suburban PCT was 3.92, compared to 2.75, scored by GPs in the urban PCT (t = 3.895, p <0.001). The mean score achieved by GPs in multipartner practices was 4.01, those in partner practices was 2.17, and those in single-handed practices was 2.14 (p <0.001). There were no significant differences between the scores for each sex, or between those who were and were not aware of the PCRMP. (See Table One)
Comparison of responders to non-responders showed that 49% (34/70) of females and 46% (73/158) of males responded ( 2 = 0.109, p = 0.741). 41% (38/92) of GPs that qualified before 1979 and 56% (58/104) of GPs that qualified in or after 1979 responded ( 2 = 4.087, p = 0.03). 61% (65/107) of GPs from the suburban PCT and 35% (44/125) of GPs from the urban PCT responded ( 2 = 15.106, p <0.001). Finally, 46% (22/48) of single-handed GPs, 20% (11/54) or partners and 59% (76/130) of GPs in multi-partner practices responded ( 2 = 22.256, p <0.001). (See Table Two)
Our results found that 3 out of 10 GPs are completely unaware of the PCRMP, and 4 out 10 did not recall receiving the PCRMP information pack, despite the fact that it was sent to every GP in the UK. It is assumed that GPs simply did not remember being sent the guidelines, or the information pack had been misplaced amongst a host of other mail. As one GP pointed out "who knows [if I have received it]! You should see how much junk we get every day".
The results show some discrepancies between GP prostate cancer management and the recommendations of the PCRMP; no GP got all 7 questions correct and the mean score achieved by all GPs was less then half the possible total. It was found that those GPs who acknowledged receipt of the guidelines achieved a significantly higher score than those who did not. It may be that those GPs who were aware of receiving the PCRMP guidelines had also devoted time to reading them, and thus had knowledge of the latest recommendations and subsequently scored higher. Alternatively it may be that those GPs who were aware of the fact that they had been sent the PCRMP guidelines had a particular interest in prostate cancer and therefore were more likely to achieve higher scores. Without assessing whether GPs had read the guidelines or not, this distinction is not possible.
It was also found that GPs who qualified in or after 1979 achieved a significantly higher score than those who qualified before 1979, suggesting that younger GPs are more aware of current guidelines perhaps because of a more recent education. Likewise, GPs in the suburban PCT or multipartner practices scored higher than those in the urban PCT or single / partner practices. However, the majority of practices within the suburban PCT studied are multipartner and therefore it cannot be confirmed whether working in a suburban practice or being part of a multipartner practice had the greater influence. The suburban PCT is in a more affluent area, therefore GPs working there may be more aware of the issues surrounding prostate cancer screening because they are more likely to be faced with a knowledgeable man requesting a PSA test. Conversely the urban PCT covers a more deprived area where doctors have to treat a larger, less healthy population but have less time and fewer resources available 11. Alternatively, GPs in multipartner practices may be able to exchange new information more easily with their peers. In contrast GPs working in single-handed practices may be disadvantaged due to working in clinical isolation without the close support of colleagues.
The response rate achieved by our study was only 47%; due to restricted time and resources we were unable to send out a 6-week reminder to non-responders as originally intended. The study was also affected by responder bias; responders were more likely to be young GPs working in multipartner practices in the suburban PCT (See Table Two). As a consequence the knowledge and practice of the responders may be different from that of non-responders. Previous studies of general practice postal surveys have suggested that non-responders either view postal questionnaires as unimportant, have limited time to read them, or are not aware of receiving them in the first place 12. Responders in this study, by comparison, may give a higher priority to issues surrounding prostate cancer and have more time to read medical literature. Assuming this to be true the results may reflect fewer discrepancies between GP management of prostate cancer and the recommendations of the PCRMP than actually exist.
Amongst our findings we have shown differences in GP management of prostate cancer, worse in older GPs. This is echoed by previous studies of Primary Care Physicians based in the US and New Zealand. McKnight et al., Hyung et al., and Morris et al., all demonstrate varied practice regarding screening for prostate cancer and moreover, Morris et al., found that that younger GPs were managing patients according to evidence based practice, whilst the practices of older GPs were influenced by past clinical experiences. Unfortunately, there is no equivalent research in the UK to compare our results with 8-10.
Despite the NHS issuing the PCRMP guidelines in September 2002, to 'clarify the confusion about prostate cancer management' in General Practice, there remains significant variation in GP management of prostate cancer screening in asymptomatic men. Hence, our findings suggest that further promotion of the PCRMP is necessary. This could be achieved by advertisement of the PCRMP in peer-reviewed journals before and during the publication of the guidelines, as well as keeping GPs informed by email and thus allowing them the opportunity to obtain a replacement copy if required.
It is evident that greater support should be offered to GPs when to attempting to establish new national guidelines. Research has shown that interactive workshops have the greatest benefit when changing professional practice13. Thus, educational meetings could be one possible solution to this problem, especially if offered to older GPs working in urban, single-handed and partner practices, who according to our study are the least well informed regarding prostate cancer and PSA testing.
To conclude, our study demonstrates variance in GP management of prostate cancer, however, the results are representative of only two PCTs in the West Midlands. Thus, to fully assess the impact of PCRMP a much larger multi-centered study, encompassing PCTs across the UK, is required.
|Mean Score on MCQ||p-value **|
|Aware of the PCRMP * ?||
p = 0.900
|Received the PCRMP guidelines?||
p = 0.018
p = 0.233
|Year of qualification||
|Primary Care Trust||
|p < 0.001|
|p < 0.001|
* Prostate Cancer Risk Management
|Responders to the questionnaire||p-value*|
|p = 0.741|
|Year of qualification||
|p = 0.043|
|Primary Care Trust||
|p = < 0.001|
|p = < 0.001|
* Chi-squared test
Appendix One: Multiple Choice Questions
A fit 69-year old Caucasian asymptomatic man with no family history of prostate cancer presents at your clinic asking for a PSA test.
demonstrates a lack of knowledge with regards to prostate cancer and the PSA
test. What would you do?
Strongly advise him to have a PSA test
Counsel him recommending PSA testing
Counsel him but ask him to come back if he wants the test
Counsel him discouraging PSA testing
Strongly advise him not to have a PSA test
2. A PSA
test is done and he has an elevated PSA. At what level would you recommend further
2.0 normal 3.0 normal 4.0 normal
5.0 normal 6.0 normal
PSA result was 4.5 ng/mL. A Transrectal Ultrasound Guided (TRUS) prostate biopsy
was later performed on him. What is the probability that the biopsy results
show cancerous cells?
20% 21 - 50% 51 - 80% 81%
4. He asks about the treatment options. Which treatment best reduces overall mortality in men with prostate cancer?
Surgery (radical prostatectomy)
None of the above
many untreated men with prostate cancer will die from the disease?
Most or all will
About half will
Most will not
6. What is the median age of diagnosis of prostate cancer?
45yrs 55yrs 65yrs 75yrs 85yrs
is the most significant risk factor for developing prostate cancer?
Smoking Ethnicity Family History Age Diet
Appendix Two: GP responses to Multiple Choice Questions with the model answers indicated by ( * ) .
This project was carried out as part of our public health course. We would like to thank our supervisors Dr Jonathan Mant and Tim Marshall.
Contributors: Both NS and SRT conceived and designed the study and wrote the paper.
Funding: The University of Birmingham Public Health Project
Competing interests: None declared
1. Quinn M, Babb P. Patterns and trends in prostate cancer incidence, survival, prevalence and mortality. Part II: individual countries. BJU Int, 2002; 90:174-184
2. Chapple A et al. Why men with prostate cancer want wider access to prostate specific antigen testing: qualitative study. BMJ, 2002; 325: 737-739
3. Lee F, Patal HRH. Prostate cancer: management and controversies. Hospital Medicine, 2002; 63 (8): 465-470
4. Barry MJ. Prostate-specific-antigen testing for early diagnosis of prostate cancer. N Engl J Med, 2001; 344 (18): 1373-1377
5. Neal DE, Donovan JL. Prostate cancer: to screen or not to screen? Lancet Oncol, 2000; 1:17-24
6. Cancer Research UK; Detatils of trial. [WWW] http://www.cancerhelp.org.uk/trials/trials/trail.asp (1st November 2003)
7. Watson E et al. The PSA test and prostate cancer: information for primary care, 2002. Sheffield, NHS Cancer Screening Programmes
8. MacKnight JT et al. Screening for prostate cancer: a comparison of urologists and primary care physicians. Southern Medical Association Journal, 1996; 89 (9): 885-888
9. Hyung KL et al. Practice trends in the management of prostate disease by family practice physicians and general internists: an internet-based survey. Adult Urology, 2002; 59: 266-271
10. Morris J et al. Screening for prostate cancer: what do GPS think? New Zealand Medical Journal, 1997; 110:178-182
11. Watt G. The inverse care law today. Lancet, 2002; 360: 252-254
12. McAvoy BR, Kaner EFS. General practice: General practice postal surveys: A questionnaire too far? BMJ, 1996; 313: 732-733
13. Thomson O'Brien
MA, Freemantle N, Oxman AD, Wolf F, Davis DA, Herrin J. Continuing education
meetings and workshops: effects on professional practice and health care outcomes
(Cochrane Review). In: The Cochrane Library Issue 2, 2003. Oxford: Update Software.
|Click on these buttons to visit our journals|
All pages copyright ©Priory Lodge Education Ltd 1994-2004.
First Published June 2004