Ben Stiller and the PSA test at 46: is it a medical error not to order the test for someone without symptoms?

By Ramon Martins Andrade, attorney (OAB/RJ 188.374)

English translation of an article first published in Portuguese: read the original.

On 4 October 2016, Ben Stiller published on Medium a first-person account titled “The Prostate Cancer Test That Saved My Life”. He was diagnosed on 13 June 2014, at 48, and had surgery the same year; the starting point was an internist who ordered a “baseline” PSA test when he was “about 46”, with no symptoms and no family history, and who “discussed it with me”. His PSA rose over a year and a half; then came the MRI, the biopsy and a Gleason 7 (3+4). The line that went around the world on the day he also told the story on Howard Stern’s programme, reported by STAT and by the Associated Press, was that “taking the PSA test saved my life”. Near the end of the text comes a more cautious sentence: “This is a complicated issue, and an evolving one”. The account is his and ends there: what is known about the doctors, the services and the clinical choices along that path is what the actor himself made public, and this article goes no further.

His conviction is not in dispute. The question that matters to the Brazilian reader has two sides: if the doctor did not order a PSA test for a man without symptoms, did the doctor err? And if the doctor ordered it without explaining what would come next, did the doctor get it right? The article examines what the clinical trials showed, what changed in the guidelines, the curious situation in Brazil, where a statute requires the test to be offered and INCA (the Brazilian National Cancer Institute) advises against doing it routinely, what other systems decided in the face of the same uncertainty, and what Brazilian law offers to someone who lived through a decision that was not informed.

Screening and investigating: the distinction that organises everything

INCA separates two situations. Early diagnosis is the investigation of people who have signs or symptoms; screening is the search for the disease in “people without signs or symptoms”. The tests are the same; what changes is the person in front of the doctor. For the man with difficulty urinating, a weak stream, urgency or blood in the urine, INCA is direct: “it is important that they be investigated”. For the man with no complaint, the official position has been the same for more than a decade: “There is no scientific evidence that prostate cancer screening brings more benefits than risks”, and for that reason the institute “does not recommend routine tests for this purpose”; if a man seeks the test on his own initiative, it recommends that men “be informed about the risks involved”. The version for health professionals adds the expression that organises the rest of this article: whoever requests the test should be guided “towards shared decision-making”. The disease is not rare: INCA’s estimate for 2026-2028 is 77,920 new cases a year, the most frequent tumour among Brazilian men apart from non-melanoma skin cancer, with 17,826 deaths in 2024. The risk rises with age and with family history, a “father or brother with prostate cancer before the age of 60”. Stiller had neither.

What the clinical trials showed, and why the debate continues

The controversy has a name and a number. The largest study is the European ERSPC, 182,000 men from eight countries randomised between PSA screening and usual care. At 13 years (Schröder and colleagues, The Lancet, 2014), prostate cancer mortality fell by 21% in the invited group: one death averted for every 781 men invited and for every 27 additional cancers diagnosed. At 16 years (Hugosson and colleagues, European Urology, 2019), the reduction stood at 20%, with 570 men invited and 18 additional diagnoses per death averted. The American PLCO trial, with 76,000 men, found no difference (Andriole and colleagues, Journal of the National Cancer Institute, 2012), with the caveat that about half of the control group had PSA tests on their own initiative. The British CAP trial, with 419,000 men and a single invitation to a PSA test, showed a small effect at 15 years, from 0.78% to 0.69% of deaths from the disease (Martin and colleagues, JAMA, 2024).

The systematic reviews explain why serious agencies diverge. The Cochrane review by Ilic and colleagues (2013) recorded that ERSPC and PLCO “provided contradictory results”, and the same group’s meta-analysis in the BMJ (2018), with 721,000 men, concluded that screening “probably has no effect” on all-cause mortality, but that, taking only the study at lowest risk of bias, there is one fewer death from the disease per thousand men screened over ten years. On the other side of the scale, per thousand men screened, about 1 more hospital admission for sepsis, 3 more cases of incontinence and 25 more of erectile dysfunction, from the treatment of tumours that, in part, would never have caused harm. The ProtecT study made this overdiagnosis concrete: among 1,643 British men with localised cancer detected by PSA, randomised between active monitoring, surgery and radiotherapy, mortality from the disease at 15 years was 2.7%, with no difference between the groups, and one in four men in the monitoring group ended follow-up alive and untreated (Hamdy and colleagues, New England Journal of Medicine, 2023). The United States task force sums up both sides in figures that apply to groups, not to individuals: about 1.3 deaths averted per thousand men screened over 13 years, between 20% and 50% of diagnoses possibly overdiagnosed, 1 in 5 men operated on left with incontinence and 2 in 3 with erectile dysfunction.

There is, finally, the finding that gives meaning to the “baseline PSA” Stiller described. In Malmö, Sweden, 21,000 men followed for 27 years showed that 44% of prostate cancer deaths occurred among the 10% with the highest PSA at ages 45-49, above 1.6 µg/L, while those below the median had a 15-year risk of metastasis of 0.09% (Vickers and colleagues, BMJ, 2013). A single measurement in midlife separates a small group that deserves monitoring from a majority that can wait for years, and it was this reasoning, not the annual campaign, that the guidelines adopted.

What changed in the guidelines: from “don’t” to “decide informed”

When Stiller had his first PSA test, the United States Preventive Services Task Force had just recommended, in 2012, against the test for all men; this is what he was referring to when he wrote that, had that guideline been followed, “I would have never gotten tested at all”. On 8 May 2018 the task force changed course: for men aged 55 to 69, the decision to undergo periodic PSA screening “should be an individual one”, after a conversation about benefits and harms, and “clinicians should not screen men who do not express a preference for screening”; at 70 or older, the recommendation remains against. The recommendation has been under update since 2023, with no new recommendation published as of October 2026.

The urological societies went further. The European Association of Urology guideline, 2026 edition, recommends to “offer an individualised risk-adapted strategy for early detection to well-informed males with a life expectancy of at least 15 years” and, with a strong rating, to offer PSA testing to “well-informed men at elevated risk”: from 50 for all men, from 45 for those with a relative who had the disease before 60 and for men of African descent, and from 40 for carriers of a BRCA2 mutation. The American Urological Association guideline (Wei and colleagues, Journal of Urology, 2023) begins with shared decision-making as a clinical principle, allows a baseline PSA between 45 and 50, directs that testing begin between 40 and 45 for Black men, carriers of germline mutations and men with a strong family history, and allows MRI before the initial biopsy, which in the PRECISION trial spared 28% of men with suspected cancer a biopsy (Kasivisvanathan and colleagues, New England Journal of Medicine, 2018).

In Brazil, the two official voices do not say the same thing, but they converge on this point. INCA concluded in 2013, in the document Rastreamento do câncer de próstata (Prostate cancer screening), that screening “produces more harm than benefit” and that “screening programmes should not be organised”, and it maintains that position on its website, updated in June 2025. The Brazilian Society of Urology (SBU), in a statement of October 2020, advises that “men aged 50 and over, even without symptoms, should seek a specialist”, that the risk group, “Black men or those with first-degree relatives with prostate cancer”, begin at 45, and that screening be done “after a broad discussion of risks and potential benefits” and “in a decision shared with the patient”. The Federal Council of Medicine (CFM), which in a 2018 news item reproduced the SBU’s guidance, has, as far as could be found, no resolution of its own on screening. A 46-year-old Brazilian man with no risk factors falls outside what either body says should be offered routinely, and within what all of them say must be explained, if he asks.

Brazil: the statute requires the test to be offered, INCA requires information, and the queue decides

There is a little-read legal text. Lei 10.289/2001 (the National Prostate Cancer Control Programme Act) established the National Prostate Cancer Control Programme and, in art. 4, II, provided for partnerships with states and municipalities “making tests for the prevention of prostate cancer available to the male population over forty years of age”. Lei 13.045/2014 added art. 4-A: “The units that make up the Unified Health System are required to carry out tests for the early detection of prostate cancer whenever, at the physician’s discretion, such a procedure is considered necessary”. The statute does not impose the population screening that INCA advises against; it makes the test conditional on medical judgement, and it is that judgement, exercised in a consultation and recorded in the medical record, that shared decision-making fills in. Once the hypothesis becomes cancer, Lei 12.732/2012 (the 60-Day Cancer Treatment Act) guarantees first treatment in the SUS (the Unified Health System, Brazil’s public health system) within 60 days of the pathology report (art. 2) and, under § 3 added by Lei 13.896/2019, “in cases where the main diagnostic hypothesis is that of a malignant neoplasm, the tests necessary for clarification must be carried out within a maximum period of 30 (thirty) days, upon a reasoned request from the responsible physician”.

Between the statute and the test lies the regulation of access, and the opinions of the technical support unit to the Judiciary of Rio de Janeiro show the gap. In NATJUS opinion no. 4562/2024, a man from Nova Iguaçu with a PSA “above the reference markers” had his prostate biopsy entered into the regulation system on 11 June 2024 and scheduled for 27 November, five and a half months later. In opinion no. 1895/2025, from Niterói, the MRI already showed a 7 mm nodule classified as PI-RADS 4, and the biopsy had to be done “promptly”, because “excessive delay” could harm the prognosis. In neither case was the PSA request in question; the problem was what comes after it. The instruments, there, are the 30-day period of Lei 12.732 and the right to care “in a timely manner” under art. 8 of the Estatuto dos Direitos do Paciente (the Patients’ Rights Statute), Lei 15.378/2026, triggered by a reasoned medical request with the hypothesis written down.

The social dimension completes the picture. The National Policy for Comprehensive Men’s Health Care (Portaria GM/MS 1.944/2009, an ordinance of the Ministry of Health) started from the finding that “male mortality rates are considerably higher”; ten years later, the 2019 National Health Survey measured that 82.3% of women had seen a doctor in the previous 12 months, against 69.4% of men. The man who does not go to the doctor does not have the conversation the guidelines presuppose, and it is in him that the urinary symptom grows old without investigation. There is also racial inequality, which INCA’s pages on the prostate do not mention and the Brazilian literature documents: a systematic review of 12 studies and 41,000 men found a prevalence of 9.6% among Black men and 5.6% among white men (Romero and colleagues, International Brazilian Journal of Urology, 2012), which the European and American guidelines recognise by bringing the test forward to 40 to 45 in this group.

And under a health plan

In private health care, the question is not whether the test is recommended, but whether it was ordered. RN ANS 465/2021 (the resolution of ANS, the National Supplementary Health Agency, that sets the list of mandatory coverage) guarantees, under an outpatient plan, “medical consultations in unlimited number in basic and specialised clinics” (art. 18, I) and the diagnostic support services requested by the attending physician (art. 18, II). The time limits are those of RN 566/2022: a consultation with a urologist within 14 working days (art. 3, II), a laboratory test within three working days (art. 3, X), MRI and biopsy within ten working days (art. 3, XI). Being covered does not mean being indicated for everyone; but, once ordered by the doctor, the PSA test admits no refusal on the ground of “absence of symptoms”, and a urologist booked two months ahead is a breach of the time limit with a protocol number. Robotic surgery has its own rule in art. 12 of RN 465, discussed in the article on robotic prostatectomy in the ANS list (in Portuguese).

Comparative law: different answers to the same uncertainty

1. United States: shared decision-making as the standard, and the lawsuit that tested it

The most discussed PSA case in the courts is that of a family medicine resident in Virginia. On 19 July 1999, Daniel Merenstein saw a 53-year-old patient, explained the risks and benefits of prostate cancer screening and documented that conversation; the patient did not have the test, was later diagnosed with advanced disease and sued. Merenstein told the story in JAMA in 2004, under the title “Winners and Losers”: the plaintiff’s theory was that he had failed to meet the standard of care in the Commonwealth of Virginia, and four physicians testified that they do not discuss screening with the patient but simply order the test. The jury acquitted the resident and found against the residency programme, which had taught him to practise evidence-based medicine, in the amount of one million dollars. His conclusion, that under the legal system the winners are the physicians who are slow to change, became an obligatory reference on defensive medicine, and the 2018 recommendation gave that conduct the status it lacked in 1999. In today’s United States, neither ordering nor not ordering the PSA test is, in itself, the standard; the documented conversation is.

2. United Kingdom: no population programme, testing on request and screening targeted at BRCA2

The United Kingdom has no population screening programme and lived for years with a compromise rule: the Prostate Cancer Risk Management Programme, of 2016, withdrawn in 2021, stated that the PSA test was available free of charge to any man aged 50 or over who asked for it, once he had carefully weighed the implications. In March 2026, the UK National Screening Committee concluded that “screening is more likely to cause more harm than good in the whole population”, that “for black men, there is ongoing uncertainty”, and recommended for the first time a targeted programme: a PSA test every two years for men aged 45 to 61 with a pathogenic variant in the BRCA2 gene and a family history of breast, ovarian, pancreatic or prostate cancer. The National Health Service keeps the door open: “men and anyone with a prostate can ask a GP about PSA testing”, and the doctor “will discuss the benefits and risks”.

3. Sweden, Lithuania and the European Union: organising what was already happening

Sweden chose an intermediate path. The health authority “continues to advise against a national screening programme”, but, since autumn 2020, regions such as Skåne and Västra Götaland have offered “organised prostate cancer testing”, which the regions’ 2026 document describes as an offer, “not a summons”: the man receives a letter with “neutral information about the possible advantages and disadvantages”, decides whether to have the PSA test, and anyone with a result equal to or above 3 µg/L goes on to MRI before any biopsy. Lithuania adopted, in 2006, a population programme with PSA testing for men aged 50 to 74 (Patasius and colleagues, Journal of Clinical Medicine, 2020). The Council Recommendation of the European Union of 9 December 2022 followed the line of the Swedish model: “countries should consider a stepwise approach, including piloting and further research”, with PSA testing “in combination with additional magnetic resonance imaging (MRI) scanning as a follow-up test”, and benefits and risks presented in such a way that people can “express informed consent”. Seen from this angle, Brazil has Lithuania’s statute, an INCA position similar to the British one and the American practice of 1999: PSA ordered en masse, with little conversation, and an access-regulation system that cannot cope with what a positive test sets in motion.

From omission to liability: three situations the law distinguishes

A doctor’s liability is measured by the conduct expected in the face of the concrete clinical picture and the rule of the time, not by the outcome. The Código de Defesa do Consumidor (the Consumer Protection Code) is express: “the personal liability of self-employed professionals shall be determined upon a finding of fault” (art. 14, § 4), while the hospital, the clinic and the health plan operator are strictly liable for the services that are their own (art. 14, caput). The Código de Ética Médica (the Code of Medical Ethics) forbids “failing to use all available means of health promotion and of prevention, diagnosis and treatment of diseases, scientifically recognised and within the physician’s reach” (art. 32) and “failing to inform the patient of the diagnosis, the prognosis, the risks and the objectives of the treatment” (art. 34). Lei 15.378/2026 placed these rules on the patient’s side: to take part “in the decision about their health care and the therapeutic plan” (art. 11), information “about the risks and benefits of procedures”, “accessible, up to date and sufficient” to decide (art. 12 and § 1), a second opinion and “sufficient time to make decisions” (art. 18), the medical record “without needing to give a justification” (art. 19). Three situations can be distinguished.

The first is Stiller’s case in reverse: the man with no symptoms and no risk factors for whom no one ordered a PSA test, and who years later receives the diagnosis. A doctor who followed the official recommendation acted within the technical standard; cancer discovered later does not turn the omission into an error, because what is required is conduct, not outcome. The search carried out for this article in the database of the São Paulo Court of Justice found no appellate judgment that held a doctor liable for not ordering a PSA test for an asymptomatic man. What the case law sanctions is the second situation, the disregarded symptom. In an appeal before the 3rd Private Law Chamber A of the São Paulo Court (case 9194646-86.2002.8.26.0000, rapporteur Judge Rubens Hideo Arai), the headnote states that “physicians who, notwithstanding the symptoms presented by the patient, fail to order necessary tests act with lack of skill and negligence, incurring in diagnostic error”. Here the measure of damage is the loss of a chance, which the Superior Court of Justice accepted for medical error in REsp 1.254.141/PR (3rd Panel, rapporteur Justice Nancy Andrighi, decided 4 December 2012), a case of inadequate cancer treatment, not of omitted screening, with the limit that applies to all: “The chance, however, can never reach the value of the asset lost. A proportional reduction is necessary”. The firm has developed the argument in late cancer diagnosis and the loss of a chance (in Portuguese) and in what is required of the general practitioner in a short consultation (in Portuguese).

The third situation is that of the test done and not followed up: the abnormal PSA that no one communicated, the biopsy that no one scheduled, the MRI with a nodule that stayed in the report. The duty to communicate the finding is a topic of its own, discussed in who is liable when the report contained the finding and no one said anything (in Portuguese). The same search found the opposite case: in appeal 0016706-37.2009.8.26.0198 (5th Private Law Chamber, rapporteur Judge J.L. Mônaco da Silva, decided 7 August 2018), the PSA pointed to cancer, the biopsy did not confirm it and the court expert “concluded that there was no failure in the provision of the service”. The false positive is part of the test, not a failure of whoever ordered it, and that is why prior information matters. The question of the Merenstein case remains: what of the doctor who ordered a PSA test for a 46-year-old man without discussing what would come next? The answer lies not in the indication for the test, which the statute leaves “at the physician’s discretion”, but in art. 12 of Lei 15.378 and art. 34 of the Code of Medical Ethics: the chain of PSA, biopsy, surgery and sequelae has known risks and benefits, and the patient is entitled to know them before the first link. Recording that conversation in the medical record is, for the doctor, the best protection in both directions and, for the patient, what makes it possible to reconstruct, years later, whether the decision was his.

Before deciding

In my view, Stiller’s account is true and does not answer the question in the title, because it describes the case in which the test worked, not the many in which it leads to an unnecessary biopsy or to surgery for an indolent tumour; the trials exist to count both sides, and that is why this article is neither a campaign for the PSA test nor against it. Three distinctions resolve most doubts. Screening a man without symptoms is not an official recommendation in Brazil, and not ordering the PSA test in that situation is within the technical standard; ordering the test for someone who is informed and wants it is also within it, and Lei 10.289 leaves this “at the physician’s discretion”. A urinary symptom calls for investigation, and unjustified delay is measured as the loss of a chance. An abnormal test requires communication and follow-up within the time limit. In every case, what counts is the written conversation. If there was an ignored symptom, an abnormal PSA without follow-up, a biopsy or consultation outside the time limit, or a chain of procedures started without anyone explaining where it led, get in touch with the firm: the analysis starts from the medical record and the chronology of the tests, and does not replace medical assessment.

About the author

Ramon Martins Andrade (OAB/RJ 188.374) is a lawyer who graduated from UFRJ in 2011, with master’s degrees from Université Sorbonne Nouvelle and from the Naval War College (EGN, Brazilian Navy). He is currently a researcher in Human Rights and Health at ENSP/Fiocruz.

This content is for information and education only. It does not constitute legal advice, advertising of results or any guarantee of outcome, under Rule 205/2021 of the Brazilian Bar Association (OAB).