When patients with schizophrenia initiate conversations, they often talk about food, sex, and religion — usually with some eccentricity. Typically, a psychiatrist's initial response is to assume the patient is simply psychotic. Because the rule of thumb is never to argue with a delusional patient, the most fundamental issues are thus dismissed as mere primitive urges that failed to be inhibited by higher cortical functions. Patients soon realize that their doctors are not very fascinated with the topics of their interest and quickly learn to avoid any further such discussions. We as psychiatrists, then, happily conclude that we have cured them of their most recent psychotic episode and expeditiously discharge them with victorious contentment.
Similarly, our lack of preparedness in addressing these issues has led to the general assumption that patients with schizophrenia are helpless in dietary control even when it involves controlling diabetes, hypertension, dyslipidemia, and obesity. Recent research in metabolic comorbidities has centered on chemical causes, such as medication and satiety and hunger signals. Many seem to have assumed that the paradigm shift in obesity research — from metabolic homeostasis to complex behavioral models — does not apply to patients with schizophrenia.
There is an inclination to presume patients with schizophrenia eat more unhealthy foods and have unhealthy lifestyles overall because they don't know any better. However, with an emerging interest in the metabolic morbidity and mortality in patients with schizophrenia, it would be appropriate to reflect further on their dietary behaviors. This investigation should begin by focusing on disturbances in eating behaviors that could potentially be modified.
The Starved Brain
Eating behaviors in schizophrenia often have been described as voracious gorging due to lack of control or even as a form of environmental automatism.1 Some patients gulp down an entire tray of food and then begin to fight others for more. My first on-call crisis as an intern involved a patient with schizophrenia who had stepped into the visiting room and frightened another patient's family by eating their donuts and coffee. During his hospitalization, he was frequently asking for bananas, oranges, and sandwiches. When gratification was delayed or denied, he threw tantrums on the floor. Many chronically refractory patients, even those with perplexing generosity, zealously protect food— some going so far as to sleep with oranges in their hands or hide food inside their underwear.
After my initial experiences, I began to examine why food is so fundamental to patients with schizophrenia. Perhaps their brains are so constantly starved that what seems to be disinhibited eating is really a physiological compensatory effort. Indeed, there was a time when schizophrenia was treated with megadoses of vitamins and proteins. This stemmed from the belief that malnutrition caused brain and behavioral malfunction because it was already well accepted that starvation could induce psychosis.
The most famous illustration of starvation-induced psychosis is from a Minnesota study conducted a half-century ago.2 Healthy male volunteers underwent 24-week semi-starvation, during which many began to behave similarly to patients with schizophrenia and anorexia nervosa. The participants toyed with food and made distasteful concoctions similar to the bizarre concoctions of patients with schizophrenia that have been explained as a form of pica. The study participants also displayed hoarding behaviors that worsened to the extent of rummaging through rubbish, which is similar to behavior in obsessive-compulsive schizophrenia.
The participants became withdrawn, isolated, and apathetic, often neglecting personal hygiene. They became increasingly preoccupied with food. Their thoughts were not entirely logical, and one man ended up chopping his fingers off in response to stress. Many patients from the starvation study, as with many clinical anorexia nervosa patients, were observed to eventually become disinhibited from their conscious restriction of food and consequently engage in binge eating.
If schizophrenia and starvation-induced psychosis are behaviorally similar and starvation eventually leads to disinhibited binging that resembles schizophrenic gorging and preoccupation with food, then the schizophrenic brain may somehow be a starved brain. Much recent literature correlates psychosis with reductions in essential polyunsaturated fatty acids and subsequent disturbances in the neural phospholipid membrane. In addition, it is well established that certain areas of the schizophrenic brain lack glutamate, an essential amino acid. It is equally well known that NMDA antagonists can produce psychotic symptoms in normal people, which further supports the glutamate deficiency hypothesis in the pathogenesis of schizophrenia.
Admittedly, this theory does not explain why antipsychotics, which indirectly activate the glutamate system, often increase appetite. However, it must be kept in mind that antipsychotics also manipulate other systems, including blockade of D2 receptors, which modulate the hedonic response associated with food intake. Newer antipsychotics additionally may influence satiation and satiety by blocking the anorectic signals such as that mediated by the 5-HT2c receptor. There are also debates over H1 receptors, various adrenergic receptors, ghrelin, leptin, NPY, POMC, and so on, but these signals are for an entirely separate discussion.
In our longitudinal observation study, we observed greater macronutrient preferences for protein and fat in the normal weight as compared to the morbidly obese group of patients with schizophrenia (unpublished data, 2004). Perhaps the patients with normal weights are providing their brains with more of the deficient nutrients through their preferential intakes, while the obese patients have continuous drives for food intake due to insufficient intake of the nutrients of which they are deprived.
In any case, not all patients with schizophrenia have insatiable appetites. At the other extreme of schizophrenic eating described in classic literature is the complete refusal to eat due to anhedonia, which leads to displeasure and even disgust with food.3 Some patients refuse to eat due to persecutory delusions of their foods being poisoned or tampered with. Although it may not make sense intuitively, such diametrically opposed symptoms may still be different manifestations of starvation, which may improve with better nutrition.
During my medicine internship, I was assigned a cachectic patient with schizophrenia who was transferred from the psychiatric inpatient unit due to dehydration resulting from food refusal. She was force-fed via nasogastric tube while receiving no antipsychotic due to a systems error. Her positive and negative symptoms as well as her appetite improved with refeeding, even before she had received antipsychotics.
Likewise, some patients with anorexia nervosa reach a point of complete refusal to eat accompanied by an extreme fear of gaining weight or the actual belief that they are fat despite being severely emaciated. Starvation leads to further distortion of body image and subsequent food refusal, which becomes a self-perpetuating cycle. Refeeding often improves the patients' perceptions and restores their appetites.
In some patients, treatment with atypical antipsychotics alone can restore appetites. It must then be asked whether the return of appetitive drive is associated purely with resolution of psychosis or whether it can be attributed to an independent property of the medication. Patients often report that atypical antipsychotics have changed the hedonic value of food. One patient gained more than 200 pounds after a switch from haloperidone to risperidone. She had been stable on outpatient management for many years with no overt psychotic symptoms. She reported after the change that food “just tastes so good,” and she began eating much more as a result. Perhaps the newer generation antipsychotics are harmful, as they may cause metabolic complications by reinforcing appetitive drive.
However, it appears that patients on typical antipsychotics can gain just as much weight as those on atypicals. Some patients managed on typical antipsychotics — and often lacking the pleasurable sensation associated with food — also gained enormous amounts of weight (up to 145 pounds) in search of something that tastes better. A parallel phenomenon has been reported in patients with impaired olfactory sensation who ate more and grew heavier in search of delicious food.
Many researchers have postulated that compulsive eating may be a result of the reward deficiency syndrome, consistent with observations of decreased striatal D2 receptors availability in morbidly obese patients.4 Although patients with schizophrenia have been known to quickly gain significant weights long before the neuroleptic era, anti-dopaminergic medications may further exacerbate their tendency to weight gain by decreased activation of the reward circuit. Even those patients appearing largely anhedonic often seem to want pleasurable food in its absence, and some patients may even compulsively seek their reward.
On the other hand, some patients have maintained stable weights for decades, even on the most notorious weight-increasing antipsychotics, such as olanzapine. Some of these patients lack both the liking and the wanting for food. They seek pleasure elsewhere; for example, through exercise, comic books, music, religion, sex, or sometimes drugs. As an aside, this opens up the possibility of exploring motivational tradeoffs in our patients. In patients with an impaired ability to derive pleasure from food, it may be the presence or the absence of such wanting that differentiates those who gain weight from those who do not.
Some patients with stable weights score high on measures of cognitive inhibition of food intake and have more physical activities than other patients. Additionally, these patients tend to score quite high on anorexic cognitive scales while viewing body image as an important feature defining them and allowing them to navigate the modern weight-conscious society.
Finally, there are those who have no obvious reason to gain weight but do so anyway, while some others stay thin despite sedentary lifestyles and imprudent dietary habits. Although it seems that there is lack of any discernable trend between schizophrenia, antipsychotics, weight, and eating behaviors, enough connections have been raised to warrant further investigation. It is widely accepted that obesity and metabolic disorders have multifactoral pathogenesis. While the variety of symptoms makes predictions uncertain, it is the recognition of such variability that will allow us to explore possible therapy for individual patients.
As a general trend, however, our observations suggest that characteristic psychopathology and behavioral effects may underlie weight gain in patients with schizophrenia. Hunger sensation appears to be increased in patients with higher body mass index, suggesting differences in the integration of psychoneuroendocrine factors associated with hunger and satiety across weight groups.
Additionally, weight appears to correlate with a variety of psychopathological dimensions. In particular, impulsivity seems to correlate significantly with disinhibited eating behaviors, which occur more often in obese patients with schizophrenia than patients with normal weights.5
Eating Disorders and Schizophrenia
In attempting to tackle the behavioral problems related to eating, weight and metabolism, I noticed many similarities between eating disorders and schizophrenia that raise the question of how they are related. Some patients with schizophrenia have very limited control over what they eat, either with no apparent compensation or with purposeful purging. Others are able to exert self-control over their food intake in ways other than refusing to eat due to paranoid delusions or due to anhedonia. Many of these patients have preoccupations with their body weight and image, perhaps reflective of modern cultural values; some patients additionally have severely distorted body image. Such eating disturbances seem to occur concurrently as well as independently from the severity of the schizophrenic symptoms.6
If patients with schizophrenia can have disinhibited eating as well as restrictive eating while being preoccupied with a distorted body image, perhaps they can have coexistent eating disorders. Conversely, some patients with eating disorders have such severe perceptual distortions with complete absence of insight that they appear quite psychotic. In addition, because of the associated social withdrawal and apathy, some eating disorder patients appear to exist on a schizophrenic spectrum.
In terms of symptom phenomenology, eating disorders share many features with schizophrenic thought, namely the fixity in thinness schema, body image distortion, perseveration over hips, thighs, and cheeks, and the bizarreness of beliefs or rituals related to eating. Do these patients have overvalued obsessive ideas, or are these ideas perceptual distortions severe enough to be somatic delusions? Due to the ego-syntonicity with the preoccupations and rituals related to food and the lack of insight in many patients, an eating disorder could be more of a psychotic phenomenon than a neurotic one in some patients.7
Eating researchers have looked extensively into orexigenic and anorexigenic signals, various neurotransmitters and their topographical variance, and the dysfunctional beliefs and behaviors related to food and weight. They have proposed top-down, bottom-up, neuroendocrine, and motivational models of eating behavior.8 These models have significant overlap with such models in schizophrenia.
Although the most appropriate term currently in use, “comorbidities” is not completely appetizing. As many would argue, the distorted perception of body image often may be difficult to tease apart from the rest of the psychotic phenomena. Moreover, it is not strictly necessary to label patients with multiple categorical diagnoses and present them in a big package. The collective symptoms exist on dynamic points on multidimensional planes.7,9
Labeling does allow for many conveniences, but it often limits our perspectives and may even provide excuses to be neglectful in some aspects of patient care. Furthermore, whether we are able to categorically diagnose patients with schizophrenia with coexisting eating disorders is largely beside the point. Many patients have disturbed eating behaviors and would benefit from management of these problems. With encouraging literature on cognitive-behavior therapy as a therapeutic module in schizophrenia,10 we might do well to focus on the possibility of restructuring the cognitive distortions related to body image, weight, food, and subsequently modifying behavior towards healthier outcome.
Obesity and metabolic disorders in schizophrenia have become topics of great concern. Prudent diets are known to reduce such excess morbidity and consequent mortality in the general population, and there is no reason to suspect otherwise in patients with schizophrenia. However, food may have greater importance to patients with schizophrenia, as their brains may be nutritionally deprived. Thus, it may be comparatively more difficult to modify dietary habits.
It appears that patients with schizophrenia can have disturbances in eating behaviors that comprise an entire spectrum of eating disorders. Eating is a complicated integration of psychoneuroendocrinology. Hunger, satiety, and cognitive perceptions toward food all appear to have individual variance. Some may have a loss of correlation between hunger and fullness. Others may excessively engage in pleasure driven appetitive behaviors. These patients are likely to be amenable to intervention once the core symptoms are accurately identified. Given the many overlapping models between eating disorders and schizophrenia, the therapeutic modules that are effectively employed in eating disorders may render success in modifying the disturbed eating behaviors in patients with schizophrenia.
- Kraepelin E. Dementia Praecox and Paraphrenia. Edinburgh, Scotland: Thoemmes Press; 2002:87.
- Key A. Psychology. In: The Biology of Human Starvation. Minneapolis, MN: University of Minnesota Press; 1950:767–920.
- Myerson A. Anhedonia. Am J Psychiatry. 1922;2:87–103. doi:10.1176/ajp.79.1.87 [CrossRef]
- Wang GJ, Volkow ND, Logan J, et al. Brain dopamine and obesity. Lancet. 2001;357(9253):354–357. doi:10.1016/S0140-6736(00)03643-6 [CrossRef]11210998
- Yum SY, Hwang MY. Demographic, metabolic and psychopathologic profiles in normal weight and overweight/obese patients with schizophrenia. Presented at: American Medical Association Residents & Fellows Section 1st Annual Research Poster Symposium. ; Atlanta, GA. ; December 3, 2004. .
- Yum SY, Hwang MY. Eating disorders in patients with schizophrenia. Presented at: American Medical Association Residents & Fellows Section 1st Annual Research Poster Symposium. ; Atlanta, GA. ; December 3, 2004. .
- Hwang MY, Bermanzohn PC. Schizophrenia and Comorbid Conditions: Diagnosis and Treatment. Washington DC: American Psychiatric Publishing; 2001.
- Halmi KA. Classification, diagnosis and comorbidities of eating disorders: a review. In: Maj M, Halmi KA, Lopez-Ibor JJ, Sartorius N eds. Eating Disorders. Chichester, England: Wiley; 2003:1–33.
- Strauss JS. Hallucinations and delusions as points on continua function. Arch Gen Psychiatry. 1969;21(5):581–586. doi:10.1001/archpsyc.1969.01740230069010 [CrossRef]5823480
- Kingdon DG, Turkington D. Cognitive-Behavioral Therapy of Schizophrenia. New York, NY: Guilford Press; 1994.