An Open Letter to Acting Administrator of the Office of Information and Regulatory Affairs Sharon Block
Ms. Sharon Block, Acting Administrator
Office of Information and Regulatory Affairs
Office of Management and Budget, White House
Dear Ms. Block,
We are writing to you concerning the function of your office to oversee and maximize "the quality, objectivity, utility, and integrity of information (including statistical information) disseminated by Federal agencies,” as specified by Congress.i For almost four years the National Institute of Mental Health (NIMH) has disseminated incomplete and inaccurate information on its website regarding the prevalence of schizophrenia.
Schizophrenia is the most serious and disabling of psychiatric disorders and is a major contributor to the number of mentally ill individuals who are homeless or incarcerated in jails and prisons. A 2013 study estimated the annual economic burden of schizophrenia in the United States to be $155.7 billion.ii Official statistics on diseases such as schizophrenia are used by both state and federal agencies to access priorities and plan psychiatric services. For example, ascertaining the true prevalence of schizophrenia is important to verify a recent report that claims that the increasing use and potency of marijuana is associated with an increase in schizophrenia.iii
The following timeline details NIMH`s dissemination of false information:
1985: Based on the results of the Epidemiologic Catchment Area (ECA) study, the last thorough mental illness prevalence study carried out in United States, NIMH reported that the one-year prevalence of schizophrenia among adults (18 and older) was 1.1%.iv Based on the 2010 census data, 1.1% would be the equivalent of 2.6 million individuals with schizophrenia (unless otherwise noted, 2010 census data will also be used for other population estimates in this document to ensure comparability.) The 1.1% of adults number was used by NIMH as the official estimate for schizophrenia for the next 32 years, until 2017.
1993: Congress requested an updated ECA prevalence number for schizophrenia. NIMH reported it to be 1.5% of adults, or 3.5 million individuals.v
2006: In an effort to improve transparency regarding the expenditure of NIH research dollars, Congress required NIH to create a public database which would allow the public to ascertain how much each NIH Institute is spending for research for each disease. The result is the Research, Condition and Disease Categorization (RCDC) database which has provided annual data since 2008. As noted on the RCDC website: "The American people want to know how the NIH spends their tax dollars."vi
2016: In a further effort to improve transparency regarding the expenditure of NIH research dollars, Congress required NIH to post prevalence numbers for each disease. Combined with the RCDC database, this allows the public to calculate the number of dollars per patient being spent by NIH for each disease. Advocacy groups can then compare how much NIH is spending on their disease compared to other diseases.
2017: As detailed elsewhere, NIMH was at this time in the process of sharply reducing its support for research on schizophrenia.vii It is highly likely that leadership at NIMH identified the potential public outcry from this drastic abandonment of schizophrenia research should the new transparency requirements call attention to it. The National Institute of Neurological Disorders and Stroke (NINDS) was spending almost twice as much per patient for research on Alzheimer's disease and Parkinson's disease compared to what NIMH was spending for research on schizophrenia (https://report.nih.gov/funding/categorical-spending#/).viii Facing public criticism and backlash, there were two possible solutions; one was to spend more money on and reprioritize schizophrenia research and the second was to re-interpret the data in a way that would reduce the number of people said to have schizophrenia so the decline in investment was safe from public or congressional scrutiny.
In November, 2017, with no public notice or public discussion, NIMH abruptly edited its website and claimed that the official prevalence of schizophrenia was 0.3% of adults, or 704,000 individuals. It claimed that the ECA data that it had been using for 32 years - 1.1% of adults or 2.6 million individuals - "is no longer scientifically defensible."ix NIMH justified its 0.3% figure by citing a household survey of 9,282 individuals carried out in 2000- 2003.x The household survey included only people living at home who acknowledged that they had schizophrenia. It did not include anyone with schizophrenia in hospitals, nursing homes, group homes, jails, prisons, homeless shelters, or those living on the streets. It also did not include the 29% of people living at home who refused to participate in the survey, presumably including some with paranoid schizophrenia. In addition, the screening instrument used in the self-report survey had been found in another NIMH-funded study to be particularly poor at diagnosing schizophrenia.xi
2018: The drastic change in NIMH`s official estimate of the prevalence of schizophrenia, from 2.6 million to 704,000, without any credible supporting data, was publicly ridiculed. The Psychiatric Times entitled "Hocus-Pocus: How the NIMH Made 2 Million People with Schizophrenia Disappear"xii and an op-ed in the Wall Street Journal asked: "Where Did the Schizophrenics Go?"xiii In response, Dr. Joshua Gordon, the director of NIMH, acknowledged that the 0.3% figure did not "reflect the full spectrum of knowledge available in the relevant literature."xiv
2019: In an effort to avoid giving a specific prevalence figure for schizophrenia, NIMH leadership decided to combine schizophrenia with "related psychotic disorders." Rather than providing a specific prevalence, as it does for every other mental disorder on its website, it proposed a range of prevalence "between 0.25% and 0.64%", apparently of the total population. This would be the equivalent of between 772,000 and 2.0 million individuals with schizophrenia and related psychotic disorders. NIMH offered two new references to justify its new range of prevalence estimates.
The first of the new references was a study carried out between 2005 and 2008 of people living at home. Based on their findings, the authors concluded that there were “3.03 million noninstitutionalized US residents” with schizophrenia.xv This study also did not include individuals in hospitals, nursing homes, group homes, public shelters, prisons or jails; the total number would be much higher if these had been included. The second of the new studies cited by NIMH to support its position was a 2006 study using data from Medicaid, Medicare, private insurance companies and the Veterans Administration.xvi This study estimated the number of people with treated schizophrenia in the United States to be 0.53% of the total population, or approximately 1.6 million individuals. However, this number included only those individuals who were receiving treatment and NIMH has estimated that 45% of individuals with schizophrenia in the United States are not being treated.xvii If the 45% of individuals with schizophrenia who are not being treated, about 1.4 million people, are combined with those who are being treated, the total number of people with schizophrenia would be 3.1 million.
2021: Using data from Medicaid and Medicare, a new study from the Johns Hopkins School of
Public Health reported that the one-year prevalence of schizophrenia among adults in the United States is 1.62%; this is the equivalent of 3.8 million individuals.xviii This study was one of the most complete studies done to date, including everyone except for those in jails and prisons.
In summary, for almost four years NIMH has attempted to minimize the number of people who have schizophrenia. It has done so by posting data on its website that NIMH leadership knows to be incomplete and inaccurate – and has defended the practice while also resisting efforts to conduct an objective and scientifically sound prevalence study under its authority to do so. Indeed, its claims for a lower prevalence are contradicted by the very studies it cites in support of its claims. We therefore make the following recommendations:
1. That NIMH post a true prevalence for schizophrenia on its website, based on the best data currently available. This number should probably be in the range of 3.5 million, the revised prevalence from the ECA study, to 3.8 million from the recent Johns Hopkins study.
2. Prevalence is an absolute threshold requirement for both researchers and policymakers. Given the fact that the last thorough prevalence study of serious mental illnesses was undertaken 40 years ago and there seems to be little appetite to obtain this critical data by those in a position to do so, Congress should mandate period prevalence studies at regular intervals, paying special attention to obtaining accurate information on the numbers of such individuals in jails and prisons and among the homeless. By doing so, Congress can exercise its critical role in ensuring that research undertaken on behalf of U.S. citizens is conducted in a manner designed to serve the greater good, not the short-term ends of bureaucrats.
We are sending copies of this letter to selected members of Congress who have a special interest in NIMH research. By reporting inaccurate prevalence data NIMH is, whether intentionally or unintentionally, subverting the intent of Congress to allow the public to ascertain how much the institute is spending on research on one disease compared to other diseases. We are calling for a return to the basic principle of scientific study that the pursuit of knowledge be guided by an objective interest in finding the correct answer to an identified question, not the answer that will best justify decisions already made.
Thank you for your attention to this matter.
Sincerely,
E. Fuller Torrey, M.D.
Associate Director for Research, Stanley Medical Research Institute
Founder, The Treatment Advocacy Center
Lisa Dailey, J.D., M. St.
Executive Director, The Treatment Advocacy Center
Elizabeth Sinclair Hancq, M.P.H.
Director of Research, The Treatment Advocacy Center
References:
i Pub L. 106-554, 114stat. 2763a-154. Office of Management and Budget. 2002. Guidelines for ensuring and maximizing the quality, objectivity, utility, and integrity of information disseminated by federal agencies; Notice; Re-publication. Federal Register, 67 (36), 8452-8460. ii Cloutier M, et al. The Economic Burden of Schizophrenia in the United States in 2013. J Clin Psychiatry. 2016 Jun;77(6):764-71. doi: 10.4088/JCP.15m10278. PMID: 27135986.
iii Hjorthøj C, et al. Development Over Time of the Population-Attributable Risk Fraction for Cannabis Use Disorder in Schizophrenia in Denmark. JAMA Psychiatry. 2021. Online ahead of print. doi:10.1001/jamapsychiatry.2021.1471.
iv Regier DA, et al. The de facto US mental and addictive disorders service system.
Epidemiologic catchment area prospective 1-year prevalence rates of disorders and services. Arch Gen Psychiatry. 1993 ;50(2):85-94. doi:0.1001/archpsyc.1993.01820140007001.
v Health care reform for Americans with severe mental illnesses: report of the National Advisory Mental Health Council. Am J Psychiatry. 1993 Oct; 150(10):1447-65. doi: 10.1176/ajp.150.10.1447.
vi Torrey EF, et al. Using the NIH Research, Condition and Disease Categorization Database for research advocacy: Schizophrenia research at NIMH as an example. PLoS One. 2020 Nov 19;15(11):e0241062. doi: 10.1371/journal.pone.0241062. PMID: 33211693; PMCID: PMC7676683.
vii https://www.nimh.nih.gov/about/strategic-planning-reports/
viii Torrey EF and Simmons W. Where did all the schizophrenics go? Wall Street Journal. March 26, 2019.
ix Gordon JA. On the prevalence of schizophrenia and transforming care through research. Psychiatric Times. 2018; 35 (3):11.
x Kessler RC, et. The prevalence and correlates of nonaffective psychosis in the National Comorbidity Survey Replication (NCS-R). Biol Psychiatry. 2005 Oct 15;58(8):668-76. doi: 10.1016/j.biopsych.2005.04.034.
xi Eaton WW, et al, Case identification in psychiatric epidemiology: a review. Int Rev Psychiatry. 2007 Oct;19(5):497-507. doi: 10.1080/09540260701564906.
xii Torrey EF and Sinclair E. Hocus Pocus: How the NIMH made 2 million people with schizophrenia disappear. Psychiatric Times, March 14, 2018.
xiii Torrey EF and Simmons W. Where did all the schizophrenics go? Wall Street Journal. March 26, 2019.
xiv Gordon JA. On the prevalence of schizophrenia and transforming care through research. Psychiatric Times. 2018; 35 (3):11.
xv Desai PR, et al., Identifying patient characteristics associated with high schizophrenia-related direct medical costs in community-dwelling patients. J Manag Care Pharm. 2013 JulAug;19(6):468-77. doi: 10.18553/jmcp.2013.19.6.468.
xvi Wu EQ, Shi L, Birnbaum H, Hudson T, Kessler R. Annual prevalence of diagnosed schizophrenia in the USA: a claims data analysis approach. Psychol Med. 2006 Nov; 36(11):1535-40. doi: 10.1017/S0033291706008191.
xvii National Institute of Mental Health. (2016). Bipolar disorder among adults. Retrieved from: https://www.nimh.nih.gov/health/statistics/prevalence/bipolar-disorder-among-adults.shtml
xviii Mojtabai R. Estimating the prevalence of schizophrenia in the United States using the multiplier method. Schizophr Res. 2021 Mar 2;230:48-49. doi: 10.1016/j.schres.2021.02.010.