|Year : 2015 | Volume
| Issue : 2 | Page : 61-67
Accuracy of clinical diagnosis of prostate cancer: Implications for patients' management
Igwebuike Victor Onyiaorah1, Cornelius Ozobia Ukah1, Daniel Chukwuemeka Anyiam1, Michael Emeka Onwukamuche1, Chilota Chibuife Efobi2
1 Department of Histopathology, Nnamdi Azikiwe University, Awka, Anambra, Nigeria
2 University of Port Harcourt, Rivers State, Nigeria
|Date of Web Publication||9-Jun-2015|
Dr. Michael Emeka Onwukamuche
Department of Histopathology, Faculty of Medicine, Nnamdi Azikiwe University, Awka, Nnewi, Anambra
Background: Cancer of the prostate is the commonest cancer and the most frequent cause of cancer deaths in Nigerian males. In Nigeria, prostate cancer patients frequently present in the late stage of the disease with consequent poor prognosis. The outcome of the disease in our setting is also worsened by diagnostic errors; this fact has been underappreciated. In this study, we have assessed the accuracy of clinical diagnosis of prostate tumors by clinicians working in Anambra State, Nigeria. Materials and Methods: Retrospective data were retrieved for all histopathology requests for prostatectomy specimens submitted to three pathology laboratories in Anambra State, Nigeria. The working clinical diagnoses were compared with the corresponding histopathology diagnoses, and the diagnostic accuracy of the medical doctors was calculated statistically. Results: Our results showed that the clinicians studied have a low overall sensitivity (37,500 per 100,000 patients) in diagnosing prostate cancer. Furthermore, doctors working in public hospitals have a slightly lower sensitivity but higher specificity than doctors serving in private hospitals when it comes to diagnosing prostrate tumors. Conclusion: This study revealed a high specificity but low sensitivity in clinicians' ability to accurately diagnose prostatic cancer. Reliance on PSA needs to be reviewed. More so the serving laboratories in the index State may need to apply internal and external quality assurance systems to their PSA analysis. It will be instructive to search for other diagnostic tools with appreciably higher predictive value. Generally, more education, research and development is needed for improvement.
Keywords: Accuracy, diagnosis, false rates, prostate cancer, sensitivity, specificity
|How to cite this article:|
Onyiaorah IV, Ukah CO, Anyiam DC, Onwukamuche ME, Efobi CC. Accuracy of clinical diagnosis of prostate cancer: Implications for patients' management. Trop J Med Res 2015;18:61-7
|How to cite this URL:|
Onyiaorah IV, Ukah CO, Anyiam DC, Onwukamuche ME, Efobi CC. Accuracy of clinical diagnosis of prostate cancer: Implications for patients' management. Trop J Med Res [serial online] 2015 [cited 2019 May 24];18:61-7. Available from: http://www.tjmrjournal.org/text.asp?2015/18/2/61/158396
| Introduction|| |
Prostate cancer is the commonest cancer in men and the second leading cause of cancer deaths among males in the United States.  Similarly, cancer of the prostate is the commonest malignancy as well as the commonest cause of cancer deaths in Nigerian males constituting 29.2% of all cancers in males. ,, An appreciable proportion of Nigerian patients with prostate cancer are discovered late in the advanced stages of the disease with a high Gleason score. , The unavailability of effective treatment for advanced prostate cancer has necessitated intensive efforts aimed at detecting low-stage, curable cancers with the goal of helping to improve prognosis and survival of patients with prostate cancer. 
A medical diagnosis is a clinical tool that assists in categorizing patients into disease groups that tend to share a similar pathogenesis, prognosis, and response to a particular treatment.  From the foregoing, medical diagnosis is a reasoned inference derived from the use of logic often considering aggregated information. The aim of making medical diagnoses is to appropriately define medical problems or conditions with the goal of planning and executing prompt and targeted solutions. The patient's management and wellness particularly hinge on making prompt and accurate diagnosis. Delays or errors in medical diagnosis, therefore, portends dismal outcome in patient care. Although diagnostic errors are common in research and are critical, it remains an underappreciated and less researched subject in patients' management.  A diagnostic error "is a diagnosis that is missed, wrong or delayed, as detected by some subsequent definitive test or finding. The ensuing harm results from the delay or failure to treat a condition present when the working diagnosis was wrong or unknown or from treatment provided for a condition not actually present." 
The primary purpose of analyzing diagnostic errors is not to apportion blame or accuse physicians of wrongdoing. On the other hand, it is an acceptable form of medical auditing to find and correct errors particularly in patients' management and the medical system in general.  However, in recent times, lawyers and patient safety advocates are increasingly interested in diagnostic errors. Truly, misdiagnosis accounts for more malpractice claims than medication errors, or any other medical management flaw. ,,, Studies have shown that diagnostic errors vary between 10% and 15% in frequency.  Researchers say that at least one in four health conditions are misdiagnosed. , Unfortunately, more serious health conditions are more frequently misdiagnosed than less serious diseases.  Cancers may not be the most frequently misdiagnosed condition; however, because of the frequent dismal outcome often associated with the nondetection or a delay in detection, it has become the foremost in frequency of diagnostic error lawsuits. 
Traditionally, only digital rectal examination (DRE) was employed by surgeons in diagnosing and differentiating prostate tumors prior to the introduction of the tumor marker, prostate-specific antigen (PSA), and transrectal ultrasound (TRUS).  Currently, most hospitals and medical doctors in our environment use DRE in combination with PSA estimation for the clinical diagnosis of prostate tumors and follow-up assessment of patients with prostate cancer. These evaluations also inform their decision for biopsy and request for histopathologic investigation. PSA is a chemical marker, the serum level of which is found to correlate well with the clinical stage in patients known to have prostate cancer; it was subsequently found to be able to detect prostate cancer in men not initially known to have the disease, which justifies its use in the screening for prostate cancer. ,
Findings on DRE are considered abnormal if the prostate gland is enlarged, asymmetric, nodular, or tender. A firm nodule, generalized nodularity, and asymmetry are considered more serious; meanwhile, symmetric enlargement is common in aging men. When the rectal mucosa overlying the prostate is attached and not independently mobile from the prostate, it is suggestive of locally invasive cancer. The test interpretation of DRE findings is ultimately based on the physician's impression, but DRE has poor inter-rater reliability.  SchrÖder et al. 1998 found that the overall sensitivity and specificity of DRE in detecting prostate cancer were 31% and 91%, respectively. They further stated that the sensitivity and positive predictive value of DRE in detecting prostate cancer decrease with decreasing serum PSA and vice versa while the specificity of DRE in detecting prostate cancer remains 83% throughout the range of serum PSA. 
It is noteworthy that increase in PSA values can be encountered in conditions other than prostate cancer. Such conditions include but are not limited to ejaculation, perianal trauma, prostate biopsy, and prostatitis. These, therefore, may confound screening for prostate cancer by serum PSA estimation; this validates the reason why increased serum PSA is not always pathognomonic of prostate cancer.  Although false negatives are encountered, the true rate is unknown due to infrequent biopsies in men with normal screening test results. It has been documented that 5-alpha-reductase inhibitors (finasteride [Proscar] and dutasteride [Avodart]) cause a decrease in serum PSA. These agents can reduce the PSA value by 50%, and this should be considered when commencing therapy or interpreting PSA results.  Given that prostatectomy may be accompanied by some untoward effects like impotency, urinary incontinence, other postsurgery complications, and even death, it will be instructive to ensure the diagnostic accuracy of prostate tumors before embarking on such surgical therapy. ,, Also, unnecessary prostatectomy is associated with avoidable pains, hospitalization, and financial burden. Such patients have been found to report a poor quality of life and psychosocial well-being. 
Evidence suggests that while years of experience and strong diagnostic skills reduce the risk of diagnostic mistakes, these do not definitely prevent all the factors that may lead to diagnostic errors. Indeed, overconfidence by well-experienced physicians has been found to contribute to diagnostic mistakes. , It will seem that the clinician's preconception plays a role in misdiagnosis, but in real situations misdiagnosis is scarcely attributed to a physician's cognitive error alone. Most studies have revealed the involvement of summation of individual and system factors.  Hence, the need for medical scientists and physicians to employ good logic in fashioning a better diagnostic algorithm that will have a high sensitivity and specificity in the diagnosis of prostate cancer.
This study aims to evaluate the accuracy of doctors working in Anambra State in South East Nigeria in diagnosing prostate tumors.
| Materials and Methods|| |
This is a retrospective study and the data for this study are drawn from information and results inputted into SPSS software (IBM, Chicago). The data are census of all the results of total prostatectomy specimens from patients with clinical diagnoses presenting with symptoms and those operated after screening interventions (of PSA) and noted to have abnormality on DRE from 2005 to 2012 submitted to the histopathology departments of Nnamdi Azikiwe University Teaching Hospital (NAUTH), Nnewi and two other private histopathology laboratories, namely, Pathocon Specialist Clinics and Research Institute, Nnewi and Nkeoma Specialist Hospital, Onitsha. These laboratories are the only three histopathology laboratories in Anambra State. The NAUTH laboratory is accredited and regulated by the Medical and Dental Council of Nigeria, while the other two laboratories are accredited and regulated by Anambra State Department of Hospital Services. The pathologists working in these laboratories are certified by the National Postgraduate Medical College of Nigeria. The choice of results from total prostatectomy specimens and the exclusion of results from needle biopsy specimens are informed by the fact that needle biopsy in some cases may miss tumors due to its limited access and low sensitivity.  Adenocarcinoma of the prostate is multifocal as such needle biopsy may miss some tumors; therefore, needle core biopsies are not "gold standard" in the diagnosis of prostate adenocarcinoma.
From these data, provisional diagnoses, as written by the clinicians, are compared with the definitive diagnoses derived histologically. When a provisional diagnosis is not written, we captured it as a "no-cancer diagnosis." In the end, the accuracy of clinical diagnoses by the doctors serving in the public and private hospitals were calculated in statistical terms that included their sensitivity, specificity, and false rates. Also, the accuracy of public hospitals in the clinical diagnosis of prostate cancer is compared with that of private hospitals. The differential accuracy in diagnosis of prostate cancer was calculated using the formulae developed by Oyeka et al., 2012. 
Results from all properly labeled total prostatectomy specimens submitted by registered physicians to the three histopathology laboratories were included in the study.
- Results from prostate specimens other than that acquired by total prostatectomy were excluded
- Prostate specimens considered inadequate for histopathologic evaluation were excluded from the study
- Autolyzed prostate specimens were excluded
- Mislabeled specimens were also excluded. 4.
| Results|| |
Excerpts from [Table 1] showed that out of 28 total prostatectomy specimen that are histologically diagnosed as prostate cancer, 9 were wrongly considered to be clinically cancer free by their doctors in all hospital combined. These doctors however, diagnosed correctly 196 of the 228 patients that had prostatectomy and were histologically negative for prostate cancer. Half of the prostatectomy cases that had histologic diagnoses of prostate cancer seen in public hospital were correctly diagnosed, and as much as 76 out of 83 (91.6%) of the cancer free prostatectomy patient were correctly segregated as prostate cancer free [Table 2]. More total prostatectomy cases (167) were handled by the private hospital in the period under review [Table 3]. Fifteen 68.2% of the 22 prostate cancer patients who had prostatectomy were diagnosed accurately, while 120 out of 135 (88.9%) were rightly diagnosed as free from prostate cancer [Table 3].
The data in [Table 4] are derived from [Table 1], [Table 2] and [Table 3] and it shows that the sensitivity and specificity of the clinical diagnosis are respectively 36.4% and 97.4%, for public hospitals and 37.5% and 94.5%, respectively, in private hospitals. Thus, the diagnostic screening test is less sensitive but more specific in public hospitals than in private hospitals. The specimens were received from both private and public hospitals in all the aforementioned facilities.
|Table 4: Statistical analyses of diagnostic accuracy of medical doctors |
Click here to view
When both hospital groups are combined, the false positive rate (F +ve) is estimated as 0.98817. This aggregate value of false positive rate comprised 0.98236 in public hospitals and 0.99136 in private hospitals. Thus, for every 100,000 patients screened in public hospitals and found to have been diagnosed with prostate cancer, 98,236 are actually free of the disease compared with 99,136 in private hospitals, a relative difference of 900 per 100,000 in favor of public hospitals' diagnostic accuracy. The true positive rates in public hospitals and private hospitals are 0.01764 and 0.00846, respectively, altogether giving a common true positive rate 0.01083. In other words, for every 100,000 patients clinically diagnosed as having prostate cancer, only 1,764 and 864 actually have malignancy in public and private hospitals, respectively, giving an overall true positive rate of 1,083 per 100,000 patients.
Similarly, the estimated false negative rate for all hospitals is 0.00084. This rate is estimated to be 0.00083 and 0.00086 in public hospitals and private hospitals, respectively. Thus, for every 100,000 patients diagnosed and found to be free of prostate cancer in public hospitals and private hospitals, 83 patients and 86 patients in these two groups of hospitals, respectively, actually have prostate cancer, giving an overall misdiagnosis rate of about 84 per 100,000 patients who should have been informed that they have prostate cancer. The overall true negative rate for all hospitals is, therefore, 99,916 per 100,000 cases.
The odds of a positive response is 0.01095 for all hospitals combined while it is 0.01796 and 0.00872 in public hospitals and private hospitals, respectively.
In other words among those testing positive, these estimated rates show that for every 100,000 patients who are diagnosed as having prostate cancer 1095, 1796, and 872 patients are confirmed to actually have the disease by clinical diagnoses in all the hospitals combined, public hospitals, and private hospitals, respectively. Or more specifically, for every 100,000 patients who are actually free of prostate cancer among those screened and found to have the disease, 1095, 1796, and 872 actually have the disease if screened by all hospitals combined, public hospitals, and private hospitals, respectively. Thus, there is a relative difference of 924 per 100,000 in better diagnostic accuracy, which points in the favor of public hospitals than private hospitals.
It should be noted that the estimated odds of negative response indicates that for every 100,000 cases that actually have prostate cancer among those tested and found not to have the disease, about 84, 83, and 86, respectively, actually have prostate cancer if screened by all hospitals combined, public and private hospitals, respectively.
Finally, the resulting odds ratios are found to be 13.03511, 21.63855, and 10.38095 for all hospitals combined, public hospitals, and private hospitals, respectively. This means that for every one patient incorrectly diagnosed as not having prostate cancer there are about 22 prostate cancer patients who are correctly diagnosed as having the disease if diagnoses are done by public hospitals and about 10 prostate cancer patients correctly diagnosed if diagnoses are done by private hospitals, giving an overall rate of about 13 prostate cancer patients correctly diagnosed as having the disease for every cancer patients incorrectly diagnosed in all hospitals.
Although there is significant positive association between clinical diagnoses and histologic diagnoses of prostatic cancer (the calculated Chi-square values are at least 17.517, P = 0.0000) in all the screening centers, there is clearly a stronger association in public than in private hospitals, in view of the relative sizes of the estimated odds ratios. Thus the relative sizes of the corresponding estimated odds ratios indicate that public hospitals are probably more efficient than private hospitals in their ability to screen and isolate prostate cancer cases more accurately.
| Discussion|| |
It is to be noted that the estimated odds of a negative response indicate that for every 100,000 cases actually having prostate cancer among those tested and found not to have the disease, about 84, 83, and 86 actually have prostate cancer if screened by all the hospitals combined, public hospitals, and private hospitals, respectively. The overall sensitivity in prostate cancer diagnoses for all hospitals combined is low, just as it is low for each of the hospital groups. This may probably be due to over-reliance of the doctors on absolute serum PSA value. Though absolute serum PSA has better positive predictive value than any other marker in predicting the presence of prostate cancer since its introduction, , research has shown that absolute serum PSA may rise in other conditions other than prostate cancer.  SchrÖder et al. 1998 found the sensitivity and specificity of DRE in detecting prostate cancer to be 31% and 91%, respectively.  However, since our study evaluated the combined accuracy of both DRE and PSA, it may be difficult to calculate the contribution of DRE alone to the diagnostic accuracy or inaccuracy in our study.
Medical doctors serving in the private hospitals have a slightly better sensitivity in diagnosing prostate tumors than those serving in the public hospitals. This may be explained by the tendency of medical doctors serving in the public hospitals to be more cautious, often requiring biopsy to be able to exclude prostate malignancy with reasonable certainty. Furthermore, doctors in the private hospitals tend to practice cost-effective medicine more than doctors in the public hospitals. Since the biopsy may be outsourced to other more qualified physicians such as the urologists, efforts are more likely to be made on the part of private practitioners to ensure that such referrals and biopsies are requested only when they are absolutely necessary. However, our study was not designed to capture the level of training or experience of the medical doctors who made the diagnoses under review. Nevertheless, experience has been shown to enable better sensitivity in prostate cancer diagnosis. 
Specificity in prostate cancer diagnosis is appreciably high for all hospitals combined. Specificity in prostate tumor diagnosis is 0.97436 and 0.94488 for doctors serving in public hospitals and private hospitals, respectively. This shows that doctors in public hospitals are more thorough in their scrutiny of the prostate and are probably not in a hurry to certify a patient who has presented with symptom (s) as healthy. More so, doctors in public hospital often undergo retraining and development, function as a synergic team, and conscientiously abide by the unit's protocol that have often been tested and found to yield positive results.  In addition, doctors serving in private hospitals often know their patients more closely, and are hence more likely to dismiss symptoms and uncertain clinical signs on the assumption that they would be able to readily reach their patients if a situation arises. Furthermore, the psychosocial relationship between doctors in private hospitals and their patients may adversely affect and incline the diagnosis toward the patient's wish.
The false positive rates are 0.98917, 0.98236, and 0.99136 for all the hospitals combined, public hospitals, and private hospitals, respectively. These high false positive rates may have been caused by the same factors that resulted in the very low sensitivity noted for prostate cancer diagnoses. The inference from our result is that in all hospitals combined, 98.9% of the patients who actually do not have prostate tumor may have been diagnosed with it, thereby exposing them to unnecessary biopsies and/or surgery with the attendant costs and adverse effects. ,,, Also, the false negative rates showed that 0.084% of the patients actually having prostate tumor may have been dismissed as not having it, resulting in the worsening condition and eventual poor prognosis of patients. The false negative rate obtained for all hospitals combined is remarkably low. However, continuous efforts are to be encouraged so as to improve the specificity, particularly for patients with prostate cancer where misdiagnosis or delay in diagnosis portend immeasurable bad consequences. 
With the reality of globalization, enhanced information technology, and efficiency in news dissemination, Nigerian patients are becoming increasingly aware of quality medical practices like their counterparts in the industrialized countries; hence, the number and activities of patient safety advocates will also increase like in the industrialized nations.  Therefore, there are chances that in the coming years litigations for wrong diagnoses will remarkably increase unless doctors and hospitals do more to improve their diagnostic accuracy.
Studies have recorded gaps in the accuracy of DRE and the more respected PSA. ,, For this reason, efforts are continuously being made by medical scientists and medical statisticians to develop better tests and workable algorithms to further improve the accuracy of diagnosis of prostate cancer. Researchers have suggested that application of logistic regression will enable better positive predictive value in the diagnosis of prostate cancer.  Yet another study has indicated the merits of classification and regression tree (CART) analysis in improving the accuracy of clinical diagnosis of prostate cancer and further espoused that CART analysis has some advantages over logistic regression in detecting prostate cancer.  Currently, research in the field of molecular biology has brought about potential prostate cancer markers, which may overcome the weaknesses of PSA. ,, In addition, good experience from cumulative exposure and training are crucial to improving diagnostic accuracy. 
It cannot be gainsaid that to successfully improve the accuracy of prostate tumor diagnosis, medical education must be encouraged and supported, while research in medical science must be enhanced and positive awareness created.
| Conclusion|| |
Our results showed that the clinicians studied have low sensitivity in diagnosing prostate cancer. Furthermore, doctors working in public hospitals have slightly lower sensitivity than those serving in private hospitals. The general diagnostic specificity is high. Reliance on PSA needs to be reviewed. ,, More so the serving laboratories in the index State may need to apply internal and external quality assurance systems to their PSA analysis.  From the foregoing, it will be beneficial to create more robust instrument that will have better accuracy and positive predictive value in prostate cancer diagnosis in order to save patients from the adversity and cost of unnecessary biopsies and afford those that actually have cancer an early detection at stages when the disease is more manageable.
| References|| |
Parker SL, Tong T, Bolden S, Wingo PA. Cancer Statistics, 1996. CA Cancer J Clin 1996;46:5-27.
Ogunbiyi JO, Shittu OB. Increased incidence of prostate cancer in Nigerians. J Natl Med Assoc 1999;91:158-64.
Mohammed AZ, Edino ST, Ochicha O, Gwarzo AK, Samaila AA. Cancer in Nigeria: A 10 year analysis of Kano cancer registry. Niger J Med 2008;17:280-4.
Jedy-Agba EE, Curado MP, Oga E, Samaila MO, Ezeome ER, Obiorah C, et al
. The role of hospital-based cancer registries in low and middle income countries-The Nigerian case study. Cancer Epidemiol 2012;36:430-5.
Obiorah CC, Nwosu SO. A Histopathological study of carcinoma of the prostate in Port-Harcourt, Nigeria. Niger J Clin Pract 2011;14:363-7.
Polascik TJ, Oesterling JE, Partin AW. Prostate specific antigen: A decade of discovery--what we have learned and where we are going. J Urol 1999;162:293-306.
Elder DE, Elenitas R, Johnson BL, Murphy GF, Xu X. Introduction to dermatopathologic diagnosis. In: Elder DE, Elenitas R, Johnson BL, Murphy GF, Xu X, editors. Lever′s Histopathology of the Skin. 9 th
ed. Philadelphia: Lippincott Williams and Wilkins; 2004. p. 1-8.
Schiff GD, Kim S, Abrams R, Lambert B, Elstein AS, Hasler S, et al
. Diagnosing diagnostic errors: Lessons from a multi-institutional collaborative project. In: Henriksen K, Battles JB, Marks ES, Lewin DI, editors. Advances in Patient Safety: From Research to Implementation. Vol. 2. Rockville, MD: Agency for Healthcare Research and Quality (AHRQ); 2005. p. 255-78.
Newman-Toker DE, Pronovost PJ. Diagnostic errors: The next frontier for patient safety. JAMA 2009;301:1060-2.
Phillips RL Jr, Bartholomew L, Dovey SM, Fryer GE Jr, Miyoshi TJ, Green LA. Learning from malpractice claims about negligent, adverse events in primary care in the United States. Qual Saf Health Care 2004;13:121-6.
Chandra A, Nundy S, Seabury SA. The growth of physician medical malpractice payments: Evidence from the National Practitioner Data Bank. Health Aff (Millwood) 2005;W5 (Suppl Web Exclusives):W5-240-9.
Berner ES, Graber ML. Overconfidence as a cause of diagnostic error in medicine. Am J Med 2008;121 (Suppl):S2-23.
Garzotto M, Beer TM, Hudson RG, Peters L, Hsieh YC, Barrera E, et al
. Improved detection of prostate cancer using classification and regression tree analysis. J Clin Oncol 2005; 23:4322-9.
McDonald C, Hernandez M, Gofman Y, Suchecki S, Schreier W. The five most common misdiagnoses: A meta-analysis of autopsy and malpractice data. Int J Fam Pract 2009;7:2.
Schröder FH, van der Maas P, Beemsterboer P, Kruger AB, Hoedemaeker R, Rietbergen J, et al
. Evaluation of the digital rectal examination as a screening test for prostate cancer. Rotterdam section of the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst 1998;90:1817-23.
Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS, Redwine E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 1987;317:909-16.
Catalona WJ, Smith DS, Ratliff TL, Dodds KM, Coplen DE, Yuan JJ, et al
. Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med 1991;324:1156-61.
Smith DS, Catalona WJ. Interexaminer variability of digital rectal examination in detecting prostate cancer. Urology 1995;45:70-4.
Wilbur J. Prostate cancer screening: The continuing controversy. Am Fam Physician 2008;78:1377-84.
Roehrborn CG, Marks LS, Fenter T, Freedman S, Tuttle J, Gittleman M, et al
. Efficacy and safety of dutasteride in the four-year treatment of men with benign prostatic hyperplasia. Urology 2004;63:709-15.
Ellison LM, Heaney JA, Birkmeyer JD. The effect of hospital volume on mortality and resource use after radical prostatectomy. J Urol 2000;163:867-9.
Stanford JL, Feng Z, Hamilton AS, Gilliland FD, Stephenson RA, Eley JW, et al
. Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer: The Prostate Cancer Outcomes Study. JAMA 2000;283:354-60.
Steineck G, Helgesen F, Adolfsson J, Dickman PW, Johansson JE, Norlén BJ, et al
.; Scandinavian Prostatic Cancer Group Study Number 4. Quality of life after radical prostatectomy or watchful waiting. N Engl J Med 2002;347:790-6.
Friedman CP, Gatti GG, Franz TM, Murphy GC, Wolf FM, Heckerling PS, et al
. Do physicians know when their diagnoses are correct? Implications for decision support and error reduction. J Gen Intern Med 2005;20:334-9.
Berner ES. Diagnostic error in medicine: Introduction. Adv Health Sci Educ Theory Pract 2009;14 (Suppl 1):1-5.
Ghandi TK, Kachalia A, Thomas EJ, Puopolo AL, Yoon C, Brennan TA, et al
. Missed and delayed diagnoses in the ambulatory setting: A study of closed malpractice claims. Ann Intern Med 2006;145:488-96.
Wefer AE, Hricak H, Vigneron DB, Coakley FV, Lu Y, Wefer J, et al
. Sextant localization of prostate cancer: Comparison of sextant biopsy, magnetic resonance imaging and magnetic resonance spectroscopic imaging with step section histology. J Urol 2000;164:400-4.
Oyeka ICA, Onyiaorah IV, Onyiaorah A, Chilota Efobi, Okeh Umeh. Use of sensitivity and specificity as measure of association in screening tests. Am J Res Biol 2012;1:7-11.
Chodak G. Prostate cancer: Epidemiology, screening and biomarkers. Rev Urol 2006;8 (Suppl 2):S3-8.
Onyiaorah IV, Ukah CO, Anyiam DC, Onyiaorah AA, Efobi CC. Statistical comparison of clinical and histologic diagnoses of breast tumours in public and private hospitals. Am J Epidemiol Prev 2013;1:113-8.
Virtanen A, Gomari M, Kranse R, Stenman UH. Estimation of prostate cancer probability by logistic regression: Free and total prostate-specific-antigen, digital rectal examination, and heredity are significant variables. Clin Chem 1999;45:987-94.
[Table 1], [Table 2], [Table 3], [Table 4]