For some, it starts with stealing candy. For others, it’s a reckless car crash, or a sudden penchant to urinate in public. The type of incident varies, but according to a study published January 5 in JAMA Neurology, more than a third of people with frontotemporal dementia (FTD) act out criminal behaviors. In some cases, the odd conduct is the first clear signal to their loved ones that something is way off. Led by Bruce Miller at the University of California, San Francisco, the study also reported bad behavior in people with Alzheimer’s and Huntington’s diseases; however, those patients’ misconduct tended to surface later in the disease and to a lesser degree than it did in people with FTD. The study raises questions about how the criminal justice system should handle people with FTD and, even more pressingly, concerns about the plight of undiagnosed patients who may be languishing in prisons or on the streets.

FTD Behind Bars?

More than a third of people with FTD act out criminal behavior. How many of them wind up in prison is an open question, particularly for the undiagnosed. [Image courtesy of Auntie P., Flickr Creative Commons.]

Dealing with wayward behavior is nothing new to Miller and other seasoned clinicians who work with FTD patients. “We have seen many examples of these cases over the years, so we thought it was time to put them into the literature,” Miller told Alzforum. “We want people to understand that you have to think very seriously about the brain whenever someone commits a crime, particularly when they commit a crime for the first time after age 50.”

Frontotemporal lobar degeneration (FTLD) chips away at regions of the brain that rein in impulsivity and support inhibition and empathy, thus creating a prime environment for behavior that defies social norms. These problems often emerge while cognition remains intact, so family, friends, and co-workers tend to see the person as rude, odd, or worse, rather than as someone suffering from a neurodegenerative disorder, said co-first author Madeleine Liljegren of Lund University in Sweden. “They get judged as a weird person or sent to a psychiatrist, where they might be misdiagnosed with bipolar disorder or depression,” she said. “If society in general were more aware of these disorders, people could receive proper care from the start and avoid getting into trouble.”

Smaller studies have reported a high incidence of criminal behavior in people with FTD. A collection of case studies reported people with FTD stealing, exposing themselves to children, sexually harassing others, and leaving the scene of an accident they caused (see Mendez, 2010). In 2013, a German study of 83 patients reported criminal behavior in more than half of people with behavioral variant FTD or semantic dementia, but only 12 percent of AD patients (see Diehl-Schmid et al., 2013).

In the present study, first authors Georges Naasan from the University of California, San Francisco, and Liljegren teamed up to measure criminal activity in a much larger cohort. They scanned medical records from more than 2,000 patients with different neurodegenerative diseases at the UCSF Memory and Aging Center for words relating to criminal behavior, such as “theft,” “hit and run,” “shoplifting,” and “police.” They found 204 records containing at least one of these buzzwords. Of these, 37.4 percent of behavioral variant FTD patients had records, followed by 27.4 percent of patients with the semantic variant of primary progressive aphasia (svPPA, a type of FTLD), 20 percent of HD patients, and just 7.7 percent of AD patients. Crimes included trespassing, public urination, theft, sexual advances, violence toward others, and driving infractions such as speeding, hit-and-run accidents, and driving under the influence.

On average, people with bvFTD and svPPA were about a decade younger than AD patients when their moral or legal lapses were first noted. Criminal behavior was the first presentation of disease in 14 percent of people with bvFTD, compared with nearly 8 percent of svPPA and just 2 percent of AD patients.

Without access to police records the researchers could not determine whether patients were charged or convicted, but notations in their medical records indicated that nearly 20 percent of bvFTD patients, 14.6 percent of svPPA patients, and 5 percent of those with AD were reported to the police. These numbers are likely an underestimate, the authors speculate, as some caregivers could have held back such information.

Janine Diehl-Schmid of the Technical University of Munich in Germany, who was not involved in the study, agreed. Her previous study interviewed caregivers to look for criminal behaviors in FTD patients, and found a higher incidence than Liljegren’s study did. Nevertheless, Diehl-Schmid said the large number of patients included in Liljegren’s study was impressive and resonated with her findings (see full comment below). Liljegren told Alzforum that she is currently performing the same medical records search of FTD patients at Lund University.

While retrospectively examining medical records did allow the researchers to sift through a large number of patients, prospective studies will be crucial to understand why the patients behaved the way they did, and whether they grasped the moral or legal implications of their actions, wrote Brad Dickerson, who runs a large FTD clinic at Massachusetts General Hospital (see full comment below).

The researchers described some cases in each disease category. A woman with bvFTD was arrested for trespassing after visiting the same property every day despite being told she would be prosecuted. A man with bvFTD became obsessed with asking random people for their phone number, and told police he had no remorse about doing so. Compelled by her sudden attraction to candy and shiny objects, a woman with svPPA started shoplifting and stealing money from strangers’ purses.

Although these were harmless crimes, Miller recalled a patient who narrowly evaded serious trouble for committing a minor act. The man became obsessed with urinating, and did so repeatedly in a public park in front of children and their parents. He was arrested and charged with exposing himself to children—and if convicted would have been labeled a sex offender. After it was determined that the man also urinated obsessively in other situations, and Miller wrote a letter to the court describing his condition, the charges were dropped and the man returned to his family’s supervision.

“Once it is recognized that these people have a degenerative disease, the courts generally are lenient,” Miller said. “The worrisome part is those who aren’t diagnosed. Those cases will fare badly in the courts.”

Unfortunately, Miller said, many people with FTD who commit crimes are likely undiagnosed and thus do not have a physician’s support. Not only is FTD a relatively unknown disease, many people may not have access to a neurologist. The patients in Miller’s study thus represent the exception rather than the rule. Indeed, the bvFTD patients with criminal behavior in this study had an average of 19 years of education—well into the realm of graduate school.

People who go undiagnosed run the risk of feeling the law's full weight for their transgressions, particularly if they appear to be mentally competent. Legal systems in many countries use some form of the M’Naghten rule to decide whether a person should be held fully accountable for their actions. The rule—established following an 1843 court case—holds that a person can plead guilty by reason of insanity only if he or she did not understand the nature of the crime or whether it was wrong. Mario Mendez of the University of California, Los Angeles, has previously noted that people with FTD may appear cognitively sharp and acknowledge that their behavior was inappropriate, but lack remorse for their actions. In essence, they knew what they did was wrong but did not care and did it anyway. Without a diagnosis, this combination of behaviors could take an insanity plea off the table and lead to a conviction. “Those two things are so basic to the M’Naghten rule, and the U.S. legal system is just not up to the task of dealing with that,” said Mendez. “These patients are hard to defend.”

Mendez added that people with FTD who commit crimes are often mistaken for sociopaths. However, unlike sociopaths, people with FTD tend to act on impulse, rather than planning their crimes, weighing pros and cons, and manipulating others to achieve their goals. “That kind of instrumental psychopathy is not what we’re talking about,” he said.

Usually, testifying in court or even a letter suffices to get charges dropped or reduce punishment, Miller said, but in cases where the patient’s actions have harmed another person, the courts may still impose punishment despite the illness.

One such case occurred in 2010, when psychiatrist Joel Stanley Dreyer was convicted of prescribing, selling, and distributing large amounts of addictive painkillers. One person died of an overdose as a result. Despite experts’ opinions that Dreyer had FTD, he was sentenced to 10 years in a prison medical facility. In 2012, a higher court overturned the sentence, holding that the judge should have ordered a special competency hearing during the trial (see United States v. Dreyer, 705 F.3d 951 [9th Cir. 2013]).

Adonis Sfera, a forensic psychiatrist at Patton State Hospital in Los Angeles, California, daily sees the consequences of people with dementia being sent to prison. State hospitals are secure facilities that house people with mental disorders who have committed crimes. Most of the patients at Patton were transferred there from prisons after their mental illness prevented them from following institutional rules, Sfera said. One such case was a teacher who, at age 59, suddenly bought a gun and shot her neighbor in the shoulder. She was convicted of attempted murder and sent to prison, where her mental illness eventually became obvious and she was transferred to Patton. There, Sfera evaluated her and a PET scan revealed frontotemporal atrophy. Based on her diagnosis of likely FTD, the courts recently ruled to release the woman to a lower care setting, Sfera said.

How many FTD patients are behind bars or confined to state hospitals? Sfera is attempting to answer this question for state hospitals in California by searching their medical records. He said that currently there is no systematic screening program in place to check for the disease at the state hospital or prison level, let alone at the level of court trials. He advocates screening all first-time offenders over the age of 55 with neuropsychological testing and neuroimaging (see Sfera et al., 2014). Miller and Liljegren agree that some form of testing is the right thing to do. “Our legal system doesn’t have clear guidelines of how to deal with frontal lobe dysfunction in the setting of the courts, and this is badly needed,” Miller said.

According to the Bureau of Justice Statistics, in 2012 more than 28,000 men and women older than 55 were admitted to federal or state prisons in the United States with sentences of at least one year (see U.S. Department of Justice, Prisoners in 2012: Trends in Admissions and Releases, 1991-2012). Data on how many of these prisoners were first-time offenders is unavailable, according to Elizabeth Ann Carson, the program manager for the bureau. Miller speculated that many older people in prisons or state hospitals may suffer from some form of frontal lobe dysfunction, even if not FTLD per se. “Whether it’s infarct, or subtle frontal lobe dysfunction related to AD, or something else, I think this is common, not rare,” he said.

One problem is that many forensic psychiatrists are unfamiliar with FTD or lack experience diagnosing the disease—an issue Sfera is attempting to rectify by delivering seminars about FTD at forensic conferences. Liljegren added that neurologists and psychiatrists need to work together more often. “After all, we’re studying the same organ,” she said. To that end, both Mendez and Miller have spread the word about FTD at psychiatric conferences. Importantly, Mendez noted that often neurologists and other physicians can miss an FTD diagnosis as well, because FTD is a relatively new kid in town. “There was almost nothing known about FTD up until 1993,” Mendez said. “Since then, knowledge has been rapidly picking up speed, but it needs to be disseminated.”

Besides physicians, the legal profession also needs education about FTD, Miller said. “It’s often going to be a lawyer or judge who will need to pick up on this,” he said.

Where should courts send FTD patients who commit crimes? Sfera said they do not belong in state hospitals, where care is based on the premise of rehabilitation and treatment. Rehabilitation is not possible for FTD, a progressive disease, and treatment often amounts to anti-psychotic drugs that may do more harm than good, Sfera said. Instead, he and Miller believe such patients belong in a palliative-care setting or a facility for patients with cognitive disorders. Many dementia care facilities refuse to admit FTD patients because they are highly mobile and, at least early on, do not have profound cognitive deficits.

Patients who pose no danger to others could be returned to their families, but for others, state hospitals like Patton may remain the only option, Sfera said. Until proper facilities exist, Sfera proposed creating specialized wards within state hospitals to deal with the needs of such patients.—Jessica Shugart


  1. This study is important because there has been little investigation of criminal behaviors in patients with FTD or other neurodegenerative diseases, although clinicians working in this field recognize that many patients come to clinical attention in part due to acquired behaviors that are considered criminal. Strengths of this study include its very large sample size, comparison between patient groups with different neurodegenerative diseases, and analysis of the specific types of criminal behaviors reported. An important weakness, noted by the authors, is that it is a retrospective study based on key word searches in medical records. Prospective research on this topic is needed.

    Furthermore, additional investigation of the behaviors contributing to criminal acts would be of interest, particularly an investigation of whether the individual has insight into the criminal nature of his or her actions and the potential consequences. Based on my own experience with such individuals, I would predict that a subset of persons with FTD are aware of the criminal nature of their actions and potential consequences, but lack inhibitory control and restraint or simply may not be concerned about the aversive value of the consequences. Other patients may lack conceptual understanding of the violation of normative societal behavior. Additional research on this topic is not only of scientific interest with regard to investigations of moral decision-making and the like, but of course of practical importance when it comes to determining how to treat such individuals in the legal system.

    View all comments by Brad Dickerson
  2. This is a great study. Given that this is a single-center study, the number of patients is extremely impressive. The results are similar to the findings of previous studies, but the large numbers give the results more significance. The results emphasize the importance of clinicians asking for criminal behavior when obtaining the patient’s history. I am somewhat puzzled about the low number of speeding and traffic violations in the bvFTD/SD subgroup. In a previous study, we found higher numbers, i.e. one third of patients with bvFTD/ SD (Ernst et al., 2010). It’s unclear if traffic violations were only identified as such in the Liljegren study when patients got a ticket, or whether a caregiver’s report sufficed.

    We previously reported higher incidences of criminal behavior in FTD (Diehl-Schmid et al., 2013). That might be in part because we conducted caregiver interviews exclusively focused on criminal behavior; these might have detected more cases than a retrospective record review. It can be awkward for caregivers to mention this kind of thing during an overall clinical interview, or they may deny it out of worry about future legal problems. In an interview study focused on criminal behavior, much time can be taken to explain the objectives of such a study. Furthermore, we used only one interviewer, while in the Liljegren study it’s probable that numerous persons had obtained the patient history and had found certain kinds of criminal behavior more or less noteworthy. On the other hand, it must be taken into account that our study overestimated the frequency of criminal behavior. As our memory clinic belongs to the Department of Psychiatry, we get more referrals of patients with unusual behavioral disturbances than a neurology department.

    In Germany, a person with FTD is not protected from prosecution, but their dementia is considered in setting the penalty. When a person with dementia and therefore diminished capability poses a danger to others, he or she is generally placed in forensic psychiatry units or closed wards of nursing homes, not prison.

    I see many FTD patients who were misdiagnosed as having depression, but in most cases only because someone doubted that diagnosis and referred them to us for further evaluation.

    Awareness of FTD has increased in the last decade. If patients go undiagnosed or misdiagnosed, that is mostly due to atypical presentations. Thanks to the Internet, some caregivers even diagnose the patients themselves, if not always correctly, as bvFTD. In 2011 the German FTLD Consortium was initiated, a multicenter research network of more than 10 clinical centers throughout Germany. All these centers have expertise in FTLD and raise awareness among the doctors in their respective regional areas.

    In order to protect people with neurodegenerative diseases from criminal prosecution, all first-time offenders over age 50 could be screened. I would also like to educate judges about the fact that criminal behavior might be a symptom of dementia, so that, when in doubt, they call expert witnesses.

    I find it interesting that no “major” crimes have been committed. Why is this so? More research should explore why people with FTD commit crimes, what they think about it, how they judge criminal behavior compared to the general population, and what would they say if they were the victim of criminal behavior. It would be extremely interesting to conduct a large study with prison inmates and persons in forensic psychiatry to screen them for potential neurodegenerative disorders.


    . [Car driving ability of patients with frontotemporal lobar degeneration and Alzheimer's disease]. Nervenarzt. 2010 Jan;81(1):79-85. PubMed.

    . Guilty by suspicion? Criminal behavior in frontotemporal lobar degeneration. Cogn Behav Neurol. 2013 Jun;26(2):73-7. PubMed.

    View all comments by Janine Diehl-Schmid

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . The unique predisposition to criminal violations in frontotemporal dementia. J Am Acad Psychiatry Law. 2010;38(3):318-23. PubMed.
  2. . Guilty by suspicion? Criminal behavior in frontotemporal lobar degeneration. Cogn Behav Neurol. 2013 Jun;26(2):73-7. PubMed.
  3. . Neurodegeneration behind bars: from molecules to jurisprudence. Front Psychiatry. 2014;5:115. Epub 2014 Aug 27 PubMed.

External Citations

  1. United States v. Dreyer
  2. Prisoners in 2012: Trends in Admissions and Releases, 1991-2012

Further Reading


  1. . Neurodegenerative disorder masquerading as psychosis in a forensic psychiatry setting. BMJ Case Rep. 2014 Jun 13;2014 PubMed.

Primary Papers

  1. . Criminal behavior in frontotemporal dementia and Alzheimer disease. JAMA Neurol. 2015 Mar;72(3):295-300. PubMed.