30th August 2013, Volume 126 Number 1381

Simon van Rij, Tony Dowell, John Nacey

Prostate cancer is a significant burden to men’s health. Prostate cancer is the most common non-cutaneous malignancy diagnosed in New Zealand and is the third most common cause of cancer death in men.1

The role of prostate-specific antigen (PSA) screening for prostate cancer remains controversial and an ongoing topic of sometimes heated debate. There is the concern that the PSA test causes over diagnosis and overtreatment of indolent prostate cancer.2,3 Conversely there is the ability to recognise cancers at an earlier stage to potentially mitigate the morbidity and mortality that prostate cancer causes.4

Internationally the incidence rates of prostate cancer vary markedly between populations, with New Zealand having the second highest rate behind the United States.5

Certain population groups are known to be at more risk of prostate cancer, however it is believed the incidence of prostate cancer is heavily influenced by the rate of PSA screening within the population. The more people that are screened the more new cancers will be found.6

PSA itself is a kallekrein serine protease that is secreted from the prostate and small amounts circulate in the blood.7 An elevated PSA is not diagnostic of prostate cancer but identifies those more at risk and who may require a prostate biopsy to establish the diagnosis.

The level of PSA may be raised for reasons other than prostate cancer, such as infection, and there is no absolute threshold that can determine a diagnosis of prostate cancer with certainty. In New Zealand the Parliamentary Select Committee investigating the early detection and treatment of prostate cancer reported on 27 July 2011:9

...While a national prostate screening programme was not recommended, the Health Committee did recommend establishing an equity-focused Quality Improvement Programme. This would ensure that men receive evidence-based information about prostate cancer testing and treatment, which they could use to make informed decisions, and timely access to high-quality care along the entire treatment pathway.

The Ministry of Health subsequently formed a Prostate Cancer Taskforce that has now released a consultation document which includes specific recommendations about PSA testing and screening men for prostate cancer.10

Within this climate we have investigated the current practices of GPs towards PSA screening in asymptomatic men.

The first aim was to determine the current rate of PSA screening of the New Zealand male population. The second aim was to survey general practitioners’ (GPs’) utility of PSA and their attitudes towards PSA screening.

Method

After obtaining institutional ethical approval for the study two methods of data collection were performed. A multiple-choice questionnaire targeted for GPs was developed with a working party including urologists, GPs and research staff.
The questionnaire included demographic data of gender, years in practice and decile rating of the respondents practice. Case vignettes of PSA screening scenarios for the respondent to answer management questions based on their usual practice.
Specific questions pertaining to age ranges at which the respondent performs screening tests at, intervals of screening, attitudes/beliefs behind their method of PSA screening, and information sources for PSA education.
A pilot trial was undertaken with a group of GPs and the questionnaire modified from the feedback given. A power calculation was performed which identified a sample size of 341 to reflect within a 5% confidence interval the actual response of all GPs in New Zealand with 95% confidence level.
Using a database of registered GPs in New Zealand, 1000 GPs were selected through a computer-generated random process. This list was generated from an external source and the study researchers were blinded from this and had no access to this selection process. To maintain anonymity no tracking of questionnaires was performed however this did not allow for any confirmation or chase up to be performed.
The results were collected and analysed using Microsoft Excel software. Correlation was performed using logistic regression.
The second method of data collection utilised a non-identifiable prospective audit of all registered New Zealand GPs and their laboratory testing patterns.
All PSA tests performed during 2011 were available for analysis. The actual results of the PSA samples were not recorded but encrypted unique identifiers allowed individual patients to be linked to each test to classify the age of the patient, geographic location and the frequency at which each individual was tested during the year 2011.
Neither the indication for testing nor the status of the patient was known. Therefore, a small percentage of tests may have been to follow up on an already diagnosed prostate cancer.
Patients with four or more PSA tests in a year were excluded from analysis (2470) as these were unlikely to represent typical screening patterns and more likely to represent those with prostate cancer or under close surveillance.
Total population data were gained from Statistics New Zealand and age specific ranges through the use of district health board registration data. Incidence rates of prostate cancer were taken from the New Zealand cancer registry reports and this was available in age specific ranges.
Attempts were made to classify the current prevalence rates of prostate cancer in New Zealand. Unfortunately these data are not held by the cancer registry and cannot be easily extrapolated from incidence data along with death rates from prostate cancer. This is due to the high number of older men with a previous diagnosis of prostate cancer who died from other causes.

Results

During 2011, 334,100 PSA tests were performed by New Zealand GPs. Of the 931,923 males older than 40 years enrolled in general practices 267,037 had a PSA performed (28.3% of that total population).

Figure 1 breaks down this percentage into age bands highlighting the lower percentage of the population tested in the younger age groups and a peak in the 65–75 age groups.

Figure 1. Percentage of New Zealand male population with a PSA test in 2011 by age group

rij

Using a model of the incidence rates of prostate cancer4 and dividing this by the number of men having a PSA test in the calendar year we can get an approximate number of individuals tested each year per new case of prostate cancer. Table 1 shows this for different age groups. Highlighting the large number of men who are tested in the 40–49 age group per each new case of prostate cancer.

To increase the accuracy of this measurement we attempted to exclude those who were tested because of a previous diagnosis of prostate cancer. A model was created to give an estimate of prevalence of prostate cancer utilising incidence and death rates of prostate cancer for the last 10 years. It could only be applied to the 40 to 70 age range, due to the higher number of men outside this age range who died from other causes. The adjusted rates are included in Table 1 below.

Table 1. 2009 incidence rates of prostate cancer in New Zealand men, by age group

rij2

The GP questionnaire was completed by 263 of the 1000 sent. 63% (164) respondents were male, and 80% of all respondents classified their practice as urban. There was an equal distribution of respondents from each of the different decile ratings.

The case vignettes showed GPs were more likely to initiate PSA testing in a man under the age of 50 with a family history of prostate cancer, 92% (95%CI: 89–95%) compared to a man without 50% (95%CI: 44–56%) The cases also showed that 70% (95%CI: 65–75%) of GPs would not initiate discussion of PSA testing in an asymptomatic 79-year-old male.

In the cases that looked at referral patterns for PSA tests already performed 95.5% (243) of the respondents would not refer a 53-year-old man with two recent PSA tests of 3.9 to a urology service, instead 75% would repeat PSA tests at 6–12 month intervals before making further decisions.

Figure 2 shows the age at which the respondents would initiate discussion of PSA screening with 57% (149 respondents; 95%CI: 51.4–62.6%) considering it below the age of 50 years; 21% (95%CI: 16.3–25.7%) of respondents would not initiate PSA screening discussion at any age.

Figure 2. Earliest age at which GPs would initiate discussion of PSA screening

rij3

The average time interval between repeated screening tests for a 55-year-old male, with a previous PSA within the reference range, was annually for 39% of respondents and more than every 2 years for 34% of respondents.

Respondents were asked to describe their approach to PSA screening (Table 2).

Table 2. GPs’ approaches to PSA screening

rij4

The most common influence on the respondents PSA screening habits was the responsibility of providing informed consent for their patients (70%, 186). Many were influenced by the conflicting ideas between overtreatment and causing potential harm 67% (176) versus the risk of missing a treatable cancer 59% (154).

Less than 5% of respondents were influenced by evidence of benefits from screening and 24% felt screening would bring improvement in their patients’ quality of life.

The primary source of information that the respondents used for PSA screening information came from Ministry of Health resources such as the New Zealand Guidelines Group. Seventy-two (27%) respondents used their previous patients’ experiences as a source of influencing their PSA screening habits.

The final question addressed whether or not the respondents had noted a difference in their habits with PSA screening in the last 2 years. Sixty-five percent (165) felt there had been no change in their PSA habits while 12% (30) were testing less.

To determine specific characteristics of GPs that influence testing a regression analysis was performed, after dividing the respondents into two groups based on whether they would initiate discussion of PSA testing in an asymptomatic man. There were no significant correlations between any demographic characteristic and whether or not a General Practitioner would initiate PSA screening.

Discussion

PSA testing as a screening tool for prostate cancer is common in New Zealand. Over 267,000 men over the age of 40 were tested during 2011. This is not an extrapolation of local data to a national level or self-reported survey data used to estimate population total, this is the actual number.

The percentage of the total population being tested shows almost three out of every 10 men over the age of 40 are having a PSA test. Comparing between age ranges the percentage of the total male population having PSA testing peaks in the 65–69 year age group and is a higher proportion for those aged 70–74 than those 55–59.

If PSA is going to be used as a screening tool evidence has shown it to be of most benefit in those with a longer life expectancy due to the slow progression of prostate cancer.10

In contrast, as a man ages his chance of dying from another cause rather than prostate cancer increases, therefore early detection is likely to infer less benefit. An appropriate discussion needs to be made with men at all ages about PSA testing but tailored to the patient’s health status and stage of life. This will lead many in the older patient groups deciding not to commence or continue PSA testing.

Decreasing this high percentage group of the older population having PSA screening may provide more efficient use of health resources. The Prostate Cancer Taskforce as part of their discussion paper have released suggested age adjusted PSA criteria at which GPs could refer patients of all ages to a urologist. With the goal to remove some ambiguity around the subject and provide more standardised care across New Zealand.11

Williams et al in the United Kingdom retrospectively reviewed PSA testing performed for the total male population of 83 randomly selected general practices. In 2007 of the 126,000 eligible males aged 40-89y 6% underwent testing, this had not changed since a similar study performed five years earlier. These rates of testing in the United Kingdom were far less than in other developed countries.12

The rate of testing in the United States is of interest since it has the highest incidence of prostate cancer. In 2001, 49,000 US men completed a self-reported health questionnaire that showed 57% of men over the age of 50 had a PSA test performed within the last year.13

Tuppin et al performed a study for PSA testing in France and from laboratory data of 10 million men found that in a 3-year period from 2008-2010 30% of men over the age of 40 had a PSA test.14

Data from Australia, which has a similar incidence of prostate cancer to New Zealand, show that the percentage of the population being screened has dramatically increased over time. Medicare data from New South Wales men aged 50–64 years showed an increase in PSA testing from 26% of the population in 1996 to 35% of the population in 2006.15 These results show a similarity between New Zealand’s rates of PSA screening and other developed countries, apart from the low rates in the United Kingdom.

From this current study the internationally comparable high rate of PSA testing performed in New Zealand correlates with New Zealand’s previously published high incidence rate of prostate cancer.16 The high uptake of PSA testing and the ability to track this testing data, places New Zealand in an ideal situation to be at the forefront of future research into prostate cancer screening.

These data did not provide clinical details for each patient undergoing PSA testing. Within the total number tested there will be men who were having PSA testing for other reasons such as ongoing testing of an already diagnosed cancer or follow up after a previous negative prostate biopsy.

Attempts were made to minimise this by removing the patients who had four or more tests in a calendar year and also by using a model to predict the prevalence rates of prostate cancer and removing this number of patients from analysis. The numbers removed were small in comparison to the total population tested and did not significantly alter the percentage calculations.

Data providing the number of tests performed for each new case of prostate cancer do not directly infer the rate of prostate cancer diagnosis. Neither the actual result of each test was known nor the decision as to whether or not to proceed to prostate biopsy, the technique used for this or the results. However, it is useful to assess the economic cost of PSA screening in New Zealand in relation to the burden of prostate cancer.

The current study shows there are a large number of tests being performed in younger men for each new case of prostate cancer. Future research could view the number of prostate biopsies performed to assess the percentage of men who have a PSA test going on to biopsy.

From the questionnaire results 79% of GPs would initiate discussion of PSA screening. Durham et al in 2003 surveyed 381 New Zealand GPs and found 97.5% performed some form of PSA screening.17

Our results show an increasing number of GPs who are not performing this test compared to this previous research. Since this first questionnaire was published almost 10 years ago the evidence around prostate cancer and PSA screening has changed dramatically. Firstly there have been two large randomised controlled trials of PSA screening showing conflicting results.4,18

The European Randomised Screening for Prostate Cancer (ERSPC) trial followed 182,000 men for 11 years and showed a reduction in prostate cancer mortality with PSA screening.

However in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial that followed 76,600 US men for 10 years showed a conflicting result of no benefit from organized screening. Also during the last 10 years the management of early “low risk” stage prostate cancer has changed dramatically from radical intervention to active surveillance.

Reports show up to 40% of newly diagnosed men with prostate cancer could enter this treatment pathway.19 This strategy decreases the risks of overtreatment while maintaining similar long term outcomes in appropriately selected patients.20 All of these results will have likely changed the way GPs approach PSA screening.

On regression analysis there were no clear associations between demographics of the GPs and their decision to screen with PSA or not. Previous research by Gormley et al identified factors that increased a GP’s likelihood of using PSA screening. They found male GPs, those in practice for more than 20 years, those running a men’s “health check” service and those who had attended a local urology education session were factors influencing increased testing.20

This is not new research and a number of other authors have attempted to identify these factors and have found similar results.21 What is less known is the effect education and national guidelines have on different groups of GPs’ testing habits.

Current research shows a GPs’ decision whether to initiate testing is influenced by factors outside of evidence based medicine such as previous patient experiences. Government guidelines and other health professional groups are a key source of information for GPs.

With the evidence for PSA testing and prostate cancer changing so rapidly it is important that these resources remain up to date to provide GPs with the latest information. Future research to measure response to these information sources would be of interest and guide internationally this ongoing question of why there is such variability in PSA screening.

Information about the risk factors for prostate cancer are well known by GPs particularly family history. The timing of PSA testing and the use of supplemental testing remains more varied.

Reducing the rates of repeated PSA testing at short intervals after two confirmatory tests could potentially reduce health spending. There is no evidence to show that this practice alters long-term outcomes, particularly in light of the slow progression time of prostate cancer.

GPs are aware of the need to provide patients with appropriate information to make an informed decision. It reflects an understanding of the current guidelines stance on the patients making their own informed decision. The questionnaire also shows that this decision is likely to be influenced by the GP and their views on the topic of PSA screening as not all will initiate this discussion with their patients.

Limitations of the current survey included a low responder rate of 26%. The total number of respondents did not meet the original power calculation of 341 lowering the precision of our results as applying to all GPs and limiting regression analysis between respondents.

Recently the United States Preventative Services Task Force released a statement that evidence is insufficient to assess the risks and benefits of prostate cancer screening in men younger than 75 years.22 This large organisation that influences health policy in the United States sparked controversy with this recommendation.

With such widespread integration of PSA screening in society, as shown by this current study, it makes it difficult to know what impact the Task Force’s recommendation will have on the rates of PSA testing and ultimately morbidity and mortality of prostate cancer.

This current study is in a unique position to help answer those questions in the future as it provides clear current data on habits and rates of testing.

Conclusion

The incidence of yearly PSA testing amongst the New Zealand male population over the age of 40 is 28%. This is comparable to other developed countries around the World.

GPs’ provide appropriate information for men to make an informed decision about PSA screening. This is provided in a variety of ways based on doctor and patient factors. There is an increasing population of GPs who will not initiate any discussion of PSA testing in their male patients.

Summary

Prostate cancer is the second most common cancer in males in New Zealand. A blood test PSA (prostate specific antigen) can be used as a screening tool to identify those at more risk of prostate cancer. This test remains controversial due to the risks of investigation and treatment versus the benefit from early detection of cancer. Using total New Zealand laboratory data we showed 28% of the male population over the age of 40 had this test over a one year time frame. This is the first time that the actual number has been published for New Zealand from accurate data. We also surveyed General Practitioners and found an increasing group who would not initiate discussion of prostate cancer screening with their male patients.

Abstract

Aim

Prostate cancer is the second most common cancer among men in New Zealand. Prostate-specific antigen (PSA) as a screening tool for prostate cancer remains controversial. The aim was to determine the rate of PSA screening in New Zealand and to survey general practitioners’ utility of PSA and their attitudes towards PSA screening.

Method

A questionnaire was sent to 1000 general practitioners (GPs). In addition, a non-identifiable prospective audit of all registered New Zealand GPs’ laboratory PSA tests was accessed for 2011.

Results

Of the 931,923 males older than 40 years, 267,037 had a PSA test performed (28.3%). This percentage peaked in the 65–75 age group (45%). 263 GP questionnaires were completed. 79% of all GPs would initiate discussion of PSA testing. The most common method of testing was at a time of another health need or check-up.

Conclusion

The incidence of yearly PSA testing in the New Zealand male population over the age of 40 is 28%. GPs provide appropriate information for men to make an informed decision about PSA screening. There is an increasing population of GPs who will not initiate any discussion of PSA testing in their male patients.

Author Information

Simon van Rij, Urology Registrar, Urology Department, Wellington Hospital, Wellington; Tony Dowell, Professor General Practice, Department of Primary Health Care & General Practice, University of Otago, Wellington; John Nacey, Professor Surgery, Department of Surgery, University of Otago, Wellington

Correspondence

Simon van Rij, Urology Department, Wellington Hospital, Private Bag 7902, Wellington South, New Zealand. Fax: +64 (0)4 3855856;

Correspondence Email

simonvanrij@gmail.com

Competing Interests

None identified.

References

  1. Ministry of Health. Age-specific rates per 100,000 for the five most common causes of cancer death 2008. Wellington: Ministry of Health, 2011.
  2. Pashayan N, Powles J, Brown C, Duffy SW. Excess cases of prostate cancer and estimated overdiagnosis associated with PSA testing in East Anglia. Br J Cancer. 2006;95(3):401–5.
  3. Draisma G, Boer R, Otto SJ, et al. Lead times and overdetection due to prostate-specific antigen screening: estimates from the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst 2003;95(12):868–78
  4. Schröder FH, et al. Prostate-cancer mortality at 11 years of follow-up, N Eng J Med, 2012;366: 981–990.
  5. Center MM, et al. International variation in prostate cancer incidence and mortality rates. Eur Urol. 2012;61:1079–1092.
  6. Welch HG, Albertsen PC. Prostate cancer diagnosis and treatment after the introduction of prostate-specific antigen screening: 1986–2005. J Natl Cancer Inst. 2009;101:1325–1329.
  7. Balk PB, Ko YJ, Bubley GJ. Biology of prostate-specific antigen, JCO. 2003;21(2): 383–391.
  8. “Inquiry into early detection and treatment of prostate cancer” Report of the Health Committee Forty-ninth Parliament (Dr Paul Hutchison, Chairperson). Presented to the House of Representatives. July 2011.
  9. Ministry of Health. Diagnosis and management of prostate cancer in New Zealand men. Recommendations from the Prostate Cancer Taskforce, 2012.
  10. Ministry of Health. Cancer: new registrations and deaths 2009. Wellington: Ministry of Health, 2012.
  11. Williams N et al. Prostate-specific antigen testing rates remain low in UK general practice: a cross-sectional study in six English cities. BJU Int. 2011;108(9):1402–1408.
  12. Sirovich BE, Schwartz LM, Woloshin S. Screening men for prostate and colorectal cancer in the United States: does practice reflect the evidence? JAMA. 2003;289(11):1414-1420.
  13. Tuppin P, et al. Prostate-specific antigen use among men without prostate cancer in France (2008-2010). Bulletin Du Cancer. 2012;99(5):521–7.
  14. Smith DP, Supramaniam R, Marshall VR, et al. Prostate cancer and prostate-specific antigen testing in New South Wales. Med J Aus. 2008;189:315–8.
  15. Center MM, et al. International variation in prostate cancer incidence and mortality rates. European Urology. 2012;61:1079–1092.
  16. Durham J, Low M, McLeod D. Screening for prostate cancer: a survey of New Zealand general practitioners. N Z Med J. 2003;116(1176): U476.
  17. Andriole GL, et al. Prostate cancer screening in the Randomized Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial: mortality results after 13 years of follow-up. J Natl Cancer Inst. 2012;104(2):125–132.
  18. National Comprehensive Cancer Network (NCCN) guidelines. Available at: www.nccn.org (Accessed on May 15, 2012).
  19. Wilt TJ, Brawer MK, Jones KM, et al. Radical prostatectomy versus observation for localized prostate cancer. N Engl J Med. 2012 Jul 19;367(3):203-13.
  20. Gormley GJ, et al. Prostate-specific antigen testing: uncovering primary care influences. BJU Int. 2006;98(5):996–1000.
  21. Gattellari M, Young JM, Ward JE. GP and patient predictors of PSA screening in Australian general practice. Family Practice. 2003;20(3):294–303.
  22. Moyer VA, U.S. Preventive Services Task Force. Screening for prostate cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med. 2012 Jul 17;157(2):I-44.