“In a larger sense, the biological study of mind is more than a scientific inquiry of great promise; it is also an important humanistic endeavor. The biology of mind bridges the sciences—concerned with the natural world—and the humanities—concerned with the meaning of human experience. Insights that come from this new synthesis will not only improve our understanding of psychiatric and neurological disorders, but will also lead to a deeper understanding of ourselves” (1).
Exciting progress has been made in psychiatric research over the last 20 years, including the development of new brain imaging techniques, clues into the genetic basis of mental illness, and acceptance of the randomized clinical trial as the means for establishing evidence of treatment efficacy. With these advances, neuroscience has rapidly supplanted psychoanalysis as the dominant force in American psychiatry and has changed the focus of scientific inquiry from mind to brain. This paradigm change is discernible in the papers presented at annual meetings of APA and the brain-image-dominated covers of journals such as The American Journal of Psychiatry.
In contrast, the field of internal medicine has followed a different course. More distant 18th and 19th-century technological advances moved medical practice away from the bedside to the laboratory and reading room (2). The growth of government-sponsored research after World War II resulted in a dramatic increase in medical specialization and mechanization (3). The rise of evidence-based medicine restricted what counted as informative data to population-based numbers, and doctors were perceived to lose their focus on the individual (4). By the 1980s, technological optimism was tempered by concern that doctors seemed unable to recognize the meaning of patients’ experience of illness (5). A schism formed between the biomedical sciences and the humanities disciplines such as philosophy, anthropology, and literature, which fed perceptions of a harmful medical reductionism (6). Patients alienated by this reductionistic model of scientific medicine often opted for nonevidence based treatments which could actually be harmful (7).
In response to this growing crisis, some internal medicine physicians turned to the humanities in an attempt to refocus attention on patients’ experience of illness and doctors’ experience of caring for them (8). A new discipline, the medical humanities, emerged to bring perspectives of disciplines such as philosophy, art, literature, film, and anthropology to bear on understanding health, illness, and medicine (9). Importantly, the goal of this movement was not to restore a pretechnological harmony between doctor and patient, but rather to add a humanistic view to a scientifically competent medicine, thereby giving the objective facts of health and illness meaning for individual patients (10).
Many benefits have been adduced by proponents of the medical humanities (11). For example, exposure to the medical humanities may facilitate engagement with patients, giving physicians the skills to empathize with patients facing the tragedy of illness and patients the courage to be open with their doctors (12, 13, 14). Medical humanities have been increasingly important in physician education as a way of enhancing attunement to individual patients’ concerns, the meanings of illness and health, ethical care, and cultural issues (15, 16). Studying art and literature may also rebuild medical idealism and offer a window into suffering and injustice (17). Medical humanities are now a significant part of the curriculum in many medical schools (18, 19) and even form the basis for graduate programs (20).
The aim of this study was to investigate the prevalence of medical humanities articles over time in selected psychiatric and internal medicine journals. The primary hypothesis was that the percentage of overall journal text devoted to the medical humanities would increase in medicine and decrease in psychiatry for the time period studied. In addition, the breadth and diversity of the medical humanities was expected to be greater in the medical journals compared to the psychiatric journals.
Feasibility issues prevented review of all psychiatry and medicine journals. Instead, the three leading clinical journals of each field were identified by referencing the Institute for Scientific Information’s (ISI) 2004 Impact Factor rankings of psychiatry and medicine journals (21). The three highest-ranking journals that were (a) published in the United States and (b) aimed at a clinical audience were selected for review. These journals were The New England Journal of Medicine, Journal of the American Medical Association, Annals of Internal Medicine, The American Journal of Psychiatry, The Archives of General Psychiatry, and The Journal of Clinical Psychiatry.
Two definitions of the medical humanities, one broad and one narrow, were used for the purposes of the study. An article met criteria for the broad construct, termed “Medical Humanities,” if both a significant portion of its content (i.e., greater than or equal to 50% of the written text) and a major focus of the article was a topic in a humanities field. Humanities fields included literature, art, philosophy, ethics, history, religion, anthropology, cultural and gender studies, sociology, and education. Psychology per se was considered a social sciences category rather than a humanities category for the purposes of the study. The “content” and “main point” criteria existed to ensure articles mentioning humanities topics briefly, or only for purposes ancillary to the point of the article, were not included.
The narrower construct, termed the “Arts,” comprised descriptions of first person experience, fiction, poetry, commentary on the doctor-patient relationship, painting, and writing about the arts. While the “Medical Humanities” construct was defined so as to overestimate the humanities present, the “Arts” was defined such that no one could argue whether one of its members constituted humanities writing. Though this construct was defined to quantify humanities writing with which no one could disagree, the humanities literature also suggested such writings may be among the most fertile areas for potential integration with biomedicine.
The selected journals were sampled at six time points: 1950, 1960, 1970, 1980, 1990, and 2000. One author (BRR) hand-searched every issue of all six journals in the 6 years sampled. Each page of text was reviewed to determine whether it met the above criteria. The first page of each article was photocopied to permit examination by a second judge should categorization be unclear. Disputes between judges were resolved by discussion and reference to the inclusion and exclusion criteria. Pages in each predefined category (“Medical Humanities” and the “Arts”) for each year were summed to generate the data set. Partial pages of text were rounded up or down based on whether they occupied more or less than 50% of a page, with one page being the minimum number of pages possible for each piece.
The number of pages categorized as “Medical Humanities” and “Arts” for each journal in each of the 6 years studied was divided by the total number of pages published by the journal that year. This yielded a percentage of total yearly text devoted to humanities for each journal, which made between-journal differences in publishing schedules and total quantity irrelevant. For instance, in the year 1960 The American Journal of Psychiatry published 1121 pages, of which 272 (23.6%) met criteria for “medical humanities” and 13 (1.1%) for the “Arts.”
The percentage data for each journal for each year sampled were submitted to the CURVEFIT regression procedure in SPSS. This analysis was used to quantify the direction and nature of the trajectory of change over time in the percentage of text devoted to medical humanities.
Presence of the Medical Humanities Over Time
As shown in Figure 1
, percentage of pages meeting criteria for “Medical Humanities” in psychiatry journals initially increased from 8.3% in 1950 to 16.9% in 1970. A precipitous decrease was found after 1970 in the psychiatry journals, with the average percentage of pages declining to 2.4% by the year 2000. The three journals had very similar curves, with the exception that pages of “Medical Humanities” in the Journal of Clinical Psychiatry failed to rise in 1960 and 1970, as was observed in the other two journals.
The opposite pattern was noted in the internal medicine journals studied (see Figure 1
). The average percentage of pages devoted to “Medical Humanities” more than doubled from 1950 to 2000, from 4.9% to 11.1%. All three journals had similar curves, though the amount of humanities found in the Annals of Internal Medicine rose more sharply than the others from 1990 to 2000.
The average percentage of pages devoted to “Medical Humanities” over time was computed for the psychiatry and internal medicine journals, and the results were plotted against each other (see Figure 2
). The “Medical Humanities” data for internal medicine were best fit using a linear regression model, (R2=0.697, p=0.039), with an increasing trend over time. Data for the field of psychiatry were best fit using a cubic model (R2=0.99, p=0.001), with a decreasing trend over time. These best fit lines have oppositely signed slopes, confirming that the fields are diverging in the amount of “Medical Humanities” found in their journals.
Presence of the Arts Subset of Medical Humanities Over Time
The average percentage of pages devoted to the predefined subset of humanities termed the “Arts” was also computed for the psychiatry and internal medicine journals, and the results were plotted against each other (see Figure 2
). In the psychiatry journals, text meeting criteria for the “Arts” fluctuated between 0% and 1% from 1950 to 2000. A linear regression model with negative slope fit the data only at the trend level (R2=0.58, p=0.078). In contrast, the “Arts” in internal medicine journals rose from 0.4% to 3.5% of the total text. These data showed significant fit with both linear (R2=0.664, p=0.048, increasing over time) and exponential (R2=0.729, p=0.03, increasing over time) models.
Cross-Sectional Humanities Content
The types of humanities articles contained in the journals examined for the years 1950 and 2000 appear in Table 1
. In 1950, humanities content focused on professional conduct, history of the field, and education of new trainees. The variety of formats and types of content increased in psychiatry and medicine from 1950 to 2000, but the increase in the medical journals was much greater. In 2000, internal medicine journals contained narratives of illness written in the patient’s own words, creative writing by medical students, residents, and graduate physicians, as well as a broad sampling of artistic endeavors. In contrast, much of the psychiatry humanities came from recurring features such as book reviews and historical notes. Apart from these brief features in every issue, there were almost no medical humanities in the rest of the journals’ content.
On a measure of their humanities content, three leading psychiatry journals declined from a high of nearly 17% in 1960 and 1970 to a low of 2% in 2000. An apparent increase in humanities publications within psychiatry journals from 1950 to 1960 may be an artifact introduced by the Archives of Psychiatry and Neurology, which in 1959 divided into two journals, the Archives of General Psychiatry and the Archives of Neurology. Humanities publications in Archives of Psychiatry and Neurology were negligible in 1950, decreasing the summed percentage for that time point. Removing Archives of Psychiatry and Neurology for 1950 results in a new figure of 14.8%, which is similar to the results found in 1960 and 1970.
Not only has the quantity of medical humanities in major psychiatry journals declined, but the diversity appears to have diminished as well. Cross-sectional analysis of the humanities in psychiatry journals in the year 2000 reveals the bulk of the remaining texts to be brief, recurring pieces. This contrasts with a more varied group of texts found in current internal medicine journals. Humanities publications in medical journals increased dramatically over the period studied, more than doubling from 1950 to 2000.
Decline of the Humanities in Psychiatry
It is notable that the humanities in psychiatry have declined so rapidly given their historical importance in the field. Ethics, literature, history, and other humanities disciplines have been intimately entwined with psychiatry from its inception (22). The first psychiatric phenomenology involved close studies of individuals and their courses of illness, described in vivid literary narratives (23). Psychodynamic psychiatry, and, above all, psychoanalysis, have engaged in productive dialogue with literature, anthropology, and sociology to investigate unconscious processes, social systems, sexuality, gender definitions, and innumerable other issues (24). In addition, psychiatrists have traditionally come from humanities-related undergraduate backgrounds, are more interpersonally oriented, and more verbally skilled compared to residents choosing to train in specialties such as surgery or internal medicine (25, 26).
The observed decline in medical humanities occurred after 1970, a period during which biological psychiatry supplanted psychoanalysis as the dominant force in American psychiatry. The humanities disciplines share aspects of psychoanalytic approaches, in that both rely heavily upon narrative and anecdote, usually lack control or comparison groups, and are generally not amenable to statistical or quantitative analyses. While many would justifiably object that psychoanalysis is not necessarily humanistic, the two fields may have been conflated, and both were discarded as the field of psychiatry emphasized replicable, quantifiable observations. The predominantly psychoanalytic focus of humanities texts that remain in psychiatric journals (such as “Introspections” in The American Journal of Psychiatry) lends support to this hypothesis (27).
Should Psychiatry be “Rehumanized”?
No: Against Rehumanization
Given the foregoing, the question arises of whether the medical humanities may still have a role in psychiatry—whether psychiatry should be “rehumanized.” Many reasonable psychiatrists would say no. Generations of psychoanalysts have explored the humanities, and perhaps no further benefit can be obtained. Besides, psychiatrists practice differently from busy internists or surgeons, who may need reminding to listen to their patients. Psychiatry is already humanistic by its very nature, with its focus on the patient’s subjective experience and on patient-clinician interactions. The humanities only offer vague illustrations of clinical concepts about which modern psychiatrists have become very sophisticated, such as projection or transference.
Others might go further. While agreeing that psychiatry does not need to be rehumanized, they might also argue that psychiatry is much better off without the humanities. Given their dubious history, the humanities may be actively harmful to psychiatry by purporting to offer competing, nonscientific explanations of pathology or treatment recommendations. A focus on humanities might also divert attention and resources from proper psychiatric education, treatment, and research.
These positions are understandable given the history of psychiatry, and the purpose of this discussion is not to refute them. Currently, they cannot be refuted, because of an absence of data quantifying the value of medical humanities to psychiatry. What is striking, however, is that the field of psychiatry has proceeded as if one of the above hypotheses were correct, while alternative hypotheses are equally valid at present.
Others have argued that it is “self-evident that whatever benefits the medical humanities may have for the rest of medicine, they are equally relevant for psychiatry” (28). While the standardization of psychiatric nosological nomenclature has been tremendously valuable from a research perspective, commentators both within and outside psychiatry have criticized psychiatrists for reducing patients’ stories to DSM categories or checklists. Humanities readings and writing could enhance clinicians’ ability to describe phenomenology in a more complex and diverse way, just as they have assisted internal medicine physicians in appreciating the human stories underlying diagnoses and chief complaints. Psychiatric humanities could potentially enhance psychiatric care, as there is growing evidence that some psychiatric patients do not feel listened to by their doctors. A recent survey of outpatient clients revealed significant dissatisfaction with the 5 minute psychopharmacologic check-in visit (29).
Many internal medicine physicians have found engagement with the humanities increases physicians’ capacity for self-awareness and self-reflection. Given the strong emotional responses patients with some psychiatric disorders generate in the clinician, the humanities may prove similarly valuable to psychiatrists in metabolizing intense countertransference feelings and using them to better understand the patient. Additionally, by leading to more gratifying patient-doctor interactions, studying the humanities may improve psychiatrists’ work satisfaction, just as it has been reported to do in internal medicine.
There also may be unique potential benefits for psychiatry in reengaging with the medical humanities. Today’s psychiatrists have the opportunity to synergistically integrate diverse disciplines. For instance, discoveries such as mirror neurons provide a brain basis of empathy and intersubjective experiences. Individuality of personality and behavioral patterns are no longer seen solely as “mental” phenomena but may be related to epigenetic DNA changes or alterations in brain connectivity or anatomy resulting from individual experiences. The humanities may provide an integrative, cross-disciplinary perspective as well as a common language for understanding and exploring phenomena that range from intrapsychic to organ-level to cellular to molecular.
Additionally, the stigma of mental illness remains significant, and nonclinicians often find mental illness frightening and difficult to understand. Writing about the experience of having or treating mental illness may increase public understanding of entities like depression and ECT (30, 31). The stigma of mental illness extends to psychiatrists as well, as psychiatrists are typically viewed more negatively than other physicians by patients and colleagues alike (32). Psychiatrists are often frustrated in their efforts to obtain funding for mental health research by the inability of government officials to understand the scope and nature of mental health problems. In conveying this message, narrative descriptions of doctor and patient experiences of mental illness may be an effective supplement to epidemiological data.
It should be noted that proponents of integrating the humanities and biological psychiatry do not advocate a return to the 1960s psychiatry, when the humanities were sometimes held out as a competing explanation or scientific paradigm. Rather than being confronted by the age-old debate between scientific biological psychiatry and nonscientific (or antiscientific) humanistic traditions, today’s neuroscience-informed psychiatrists are offered opportunities for integrating different perspectives (33). In this setting the humanities may produce a different type of evidence, one that can illuminate (and be illuminated by) scientific approaches.
Examples of such integration include an interesting new feature in The American Journal of Psychiatry entitled “Patient Perspectives,” which describes the experiences of individual study participants in randomized controlled trials (34). Others are Kay Redfield Jamison’s An Unquiet Mind (35) and Touched with Fire (36), which provide insight about a clinician-researcher’s own experiences of mania and depression and illustrate why patients are often reluctant to give up their productive “highs” by taking mood stabilizers. Numerous other examples can be found, including the work of neuroscientist Robert Sapolsky (37) and neurologist Oliver Sacks (38).
Our findings should be interpreted with several limitations in mind. First, the most challenging part of the study was defining “Medical Humanities.” The authors felt like Supreme Court Justice Potter Stewart, who famously defined pornography by saying “I know it when I see it.” However, this could not be avoided given the amorphous, multidisciplinary nature of the medical humanities field. Notably, the data revealed similar increases in the internal medicine literature and decreases in the psychiatry literature with respect to both the broad construct of “Medical Humanities” and the narrow construct of the “Arts.”
A second limitation of the study is that journals were reviewed in an unblinded fashion. To minimize the risk of evaluator bias and the potential “halo effect” that perception of emerging patterns may have on reviewers, journals were not reviewed sequentially with respect to year or field, and no data were analyzed until all journals had been reviewed.
Third, the journals studied differ in a number of significant ways, including publication schedule (e.g., weekly, monthly), quantity of pages published, and editorial leadership. Using the percentage of total yearly text for each journal controlled for publication differences, but it may still be objected that a journal’s content is a function of a particular editor’s decision making rather than philosophical changes in the field as a whole. While this is a cogent objection to the study’s methods, we were not able to discern a better proxy for the prioritization of the humanities than their presence in the field’s leading journals. We might also argue that the selection of editors having particular viewpoints may reflect and reinforce larger trends in the field.
This is the first study to our knowledge that quantifies the decrease over time of humanities material in psychiatry journals. While many psychiatrists would likely find these results consistent with their anecdotal impressions, this subject has not been studied to date. It bears emphasizing that this article does not provide data on whether or not the decline of medical humanities in psychiatric journals is good or bad for the field as a whole. In fact, we are critical of strong positions held for or against the humanities’ utility to psychiatry formed in the absence of such data. Our point is to highlight the divergent trends in the fields of psychiatry and internal medicine and call for further discussion and empirical study. Among the most helpful future studies would be determination of whether exposure to the medical humanities in fact improves psychiatric education, patient outcomes, and the physicians’ experience.
FIGURE 2. Comparison of Average Percent of Publication Pages Devoted To “Medical Humanities" and “Arts” in Selected Psychiatry and Internal Medicine Journals from 1950 to 2000
aData for internal medicine were best fit using a linear regression model, (R2=0.697, p=0.039), with an increasing trend over time. Data for psychiatry were best fit using a cubic model (R2=0.99, p=0.001), with a decreasing trend over time.
bData for internal medicine fit linear (R2=0.664, p=0.048, increasing over time) and exponential (R2=0.729, p=0.03, increasing over time) models. Data for psychiatry fit a linear regression model with negative slope only at the trend level (R2=0.58, p=0.078).
A preliminary version of this paper was presented at Columbia University’s Narrative Medicine Rounds on October 5, 2006. A portion of the data reported here were also presented in a workshop at the American Psychiatric Association, 160th Annual Meeting in May 2007. The authors wish to thank Sarai Batchelder, Ph.D., for reviewing this manuscript and providing assistance with the statistical analyses.