Medical Disorders Among Patients Admitted to a Public-Sector Psychiatric Inpatient Unit

Active and significant physical disorders are common among public-sector psychiatric patients and frequently go undetected (1,2). We attempted to improve disease detection in a county mental health center's psychiatric hospital by introducing a brief medical questionnaire and additional laboratory tests and conducting further medical evaluations when indicated by the screening results.

Methods

All adult patients who were admitted from the mental health center's emergency psychiatric service to its contiguous psychiatric hospital during a six-month period (September 1990 through February 1991) were invited to participate in the study. After the study was fully explained to each participant, he or she signed an informed consent document that had been approved by the mental health center's human subjects committee.

Patients were excluded if they were admitted through the medical emergency department (68 patients); had a language barrier (eight patients), deafness, mental retardation, or organic brain syndrome (one each); or refused to participate (34 patients). Seven patients were excluded for miscellaneous reasons. In addition, data were unavailable for 177 consenting patients because they were discharged early, their specimens were lost, or laboratory tests were not ordered.

On admission, patients received a routine physical examination by the admitting psychiatrist. The psychiatrist referred patients with medical problems to the department of medicine for evaluation and treatment recommendations. He routinely ordered a complete blood count, a serologic test for syphilis, urinalysis, and the additional laboratory tests used in the study: a fasting chemistry-20 panel (electrolytes, glucose, blood urea nitrogen, creatinine, uric acid, magnesium, calcium, phosphorous, total protein, albumin, total bilirubin, alkaline phosphates, serum glutamic-pyruvic transaminase, serum glutamic-oxaloacetic transaminase, lactate dehydrogenase, creatine kinase, iron, and cholesterol), free thyroxine, and vitamin B₁₂ and folate levels. A combined blood draw was scheduled for the first morning after admission.

On the first working day after admission, a research nurse administered a medical symptom questionnaire consisting of ten items with high predictive value for active and important physical disorders (2). Patients who were too agitated or psychotic to respond were approached at a later time. If the research nurse concluded that the results of the psychiatrist's examination, any laboratory test, or the questionnaire suggested an active and important physical disorder, she referred the patient to the department of medicine.

To establish a standard for physical disorders and to determine whether they were newly diagnosed or already known to the mental health center, two internists independently reviewed the patient's integrated county hospital medical records—both the psychiatric and the medical record—laboratory test results, responses to the medical questionnaire, and any results of an evaluation by the referring physician or nurse. Disagreements were resolved by discussion.

Active and important physical disorders were those that required treatment or continuing medical surveillance, could threaten life or cause or exacerbate a psychiatric disorder, or had potentially significant consequences for health or functioning (2). Physical disorders were regarded as new if they were not documented in the patient's integrated medical record. The second author reviewed all the information to decide whether the physical disorder was causing, exacerbating, or simply coexisting with the patient's mental disorder.

Results

We enrolled 289 patients, 157 men (54 percent) and 125 women (43 percent). For seven patients (2.4 percent), gender was not recorded. The study sample represented 56 percent of the eligible patients. The patients ranged in age from 18 to 76 years (mean±SD=35.4±12.2 years). Eighteen patients (6 percent) were aged 60 or older. The primary admitting diagnoses were schizophrenia or other psychoses (180 patients, or 62 percent), mood disorders (70 patients, or 24 percent), adjustment disorders (26 patients, or 9 percent), dementia or organic brain syndrome (six patients, or 2 percent), and other disorders (seven patients, or 2 percent).

Eighty-four patients (29 percent) had active and important physical disorders. The admitting diagnoses for the 84 patients were similarly distributed: 51 patients (61 percent) with psychotic disorders, 20 (24 percent) with mood disorders, ten (12 percent) with adjustment disorders, and three (4 percent) with dementia or organic brain syndrome. Of the 119 disorders listed in Table 1, a total of 24 (20 percent) were newly diagnosed for 23 patients (8 percent).

For 49 patients (17 percent), laboratory tests showed no abnormalities. One abnormality was found for 79 patients (27 percent), two for 61 patients (21 percent), three for 48 patients (17 percent), and four to nine for 52 patients (18 percent). The psychiatric unit's routine laboratory tests uncovered new cases of syphilis, anemia, urinary tract infection, and proteinuria. The additional laboratory tests conducted for the study uncovered new cases of hypothyroidism, diabetes, alcoholic liver cirrhosis, hepatitis, and vitamin B₁₂ and folate deficiency. Fourteen of 18 abnormal free thyroxine results were false positives, three led to new diagnoses of hypothyroidism, and one confirmed a known diagnosis of hyperthyroidism. Seven of eight abnormal vitamin B₁₂ levels were elevations, and one led to a diagnosis of vitamin B₁₂ deficiency. Elevated uric acid levels among ten patients were not investigated further.

The ten-item medical questionnaire had a false-negative rate of 26 percent; it missed active physical disorders for 22 of the 84 patients who had such disorders. The questionnaire did not identify 13 of the 23 patients with newly detected disorders, most of which were suggested by laboratory tests.

Only one disorder that was newly diagnosed by the study procedures was judged to be the cause of the patient's psychiatric symptoms. It was determined that the psychotic symptoms of a 25-year-old woman who was admitted with schizophrenia, undifferentiated type, were caused by hypothyroidism. Hypothyroidism was newly diagnosed for two patients with schizophrenia, and it was determined to be exacerbating their psychotic symptoms.

For six patients it was determined that seven previously recognized physical disorders were causing their psychiatric symptoms: drug withdrawal syndrome, lack of normal expected physiological development, alcoholic dementia, epileptic psychosis (two patients), and postconcussion syndrome. Eight patients had previously recognized physical disorders that exacerbated their psychiatric symptoms: lack of normal expected physiological development (three patients), hyperthyroidism plus hypoparathyroidism, myocardial infarction, quadriplegia plus anemia, senile dementia comorbid with schizophrenia, and other cerebral degeneration comorbid with major depression.

A wide range of coexisting physical disorders were newly diagnosed among 20 patients (7 percent): anemia (six patients), alcoholic liver cirrhosis (three patients), syphilis (two patients), and thalassemia, diabetes, vitamin B₁₂ deficiency, hepatitis, chronic obstructive pulmonary disease, urinary tract infection, proteinuria, cellulitis, and cachexia (one patient each).

Discussion

When patients admitted to a public-sector psychiatric hospital received enhanced screening, 29 percent were found to have important physical disorders. For 10 percent of the patients, the physical disorders had not been previously detected. Physical conditions that were causing or exacerbating mental symptoms were not uncommon, and their detection was critical to providing appropriate care.

Our methods may have underestimated the frequency of physical disorders, because some patients were discharged before abnormal findings could be followed up. These patients were advised by mail to seek medical evaluation.

Rates of physical disease among psychiatric inpatients have been found to vary from 17 percent to 77 percent (1). However, only two studies have reported rates of newly detected physical diseases, 17 percent (2) and 27 percent (3); these rates are higher that the 8 percent rate we observed. The differences undoubtedly stem from case-finding methods, patients' characteristics, and definitions of important disease.

Although the detection of physical disease among psychiatric inpatients is acknowledged to be essential, no consensus exists about the most appropriate or cost-effective methods. A careful medical history and physical examination are the most sensitive and specific methods (4). Cursory evaluations produce high rates of misdiagnosis (4,5). Certain patient characteristics are associated with a high risk of having a comorbid physical condition (6). Unfortunately, the high false-negative rate of our medical questionnaire precludes its use as a stand-alone screening method.

Although the laboratory tests introduced by the study uncovered unsuspected disease, many test abnormalities were clinically unimportant. The use of laboratory tests in screening psychiatric patients is thoroughly reviewed elsewhere (7,8). Our results and those of others (1) indicate that thorough medical screening is essential to caring adequately for psychiatric patients. A Medical Evaluation Field Manual derived from a large-scale study conducted in California's public mental health system is available on the Web (9). The manual describes how, when, where, and whom to screen. An apparently cost-effective screening algorithm derived from that study has not yet been independently tested (10).

Conclusions

Active and important physical disorders are common among patients admitted to psychiatric inpatient units. Some patients' mental symptoms are caused or exacerbated by undiagnosed medical conditions. Additional research is needed to define cost-effective medical evaluation methods for patients in this setting and to devise ways to convince program administrators and staff to implement them.

Acknowledgment

The authors thank Lynda Wolfe, R.N., for her assistance in evaluating patients.

Dr. Koran is affiliated with the department of psychiatry and behavioral science at Stanford University Medical Center. Dr. Sheline, Dr. Freedland, Ms. Mathews, and Ms. Moore are with the department of psychiatry at Washington University Medical Center in St. Louis, Missouri. When this study was done, Dr. Sheline was medical director of the inpatient psychiatry service at Santa Clara Valley Medical Center in San Jose, California. Dr. Imai and Dr. Kelsey are with the department of medicine at Santa Clara Valley Health and Hospital System in San Jose. Send correspondence to Dr. Koran at the OCD Clinic, Room 2363, 401 Quarry Road, Stanford, California 94305 (e-mail, [email protected]).

Enlarge table