Physical Illness Among All Discharged Psychiatric lnpatients in a National Case Register Jonathan Rabinowitz, D.S.W. Mordechai Mark, M.D. Miriam Popper, M.A. Dina Feldman, M.A. Abstract Previous studies have found that although psychiatric patients tend to have more physical illness than the rest of the population, itfrequently goes unrecognized and untreated in psychiatric settings. This study investigated rates of reported physical illness among hospitalized psychiatric patients in preparation for national reform in mental health services. Datafrom the Israeli National Psychiatric Case Registry were analyzed on reported physical illness among all 38, 714 psychiatric discharges during 1989-1991. Physical illness was reported for 10.62% of patients under age 25, 14.04% of patients 25 to 44, 34.27% of patients 45 to 65, and 61.26% of patients 65 and older. Rates differed among hospitals. Reported physical illness was considerably lower than expected as compared with other studies, Underdiagnosis is suggested as a possible explanation. Study results were used to add differential payment for physical comorbidity under the new National Health Insurance Law. Other corrective measures are discussed. Psychiatric inpatients and outpatients are likely to have coexisting physical illness. 19 Studies have repeatedly found that about half of psychiatric patients have physical diseases requiring medical treatment or medical surveillance. Often these diseases cause or exacerbate a mental disorder. 4'6'1~ Yet almost half of these physical problems may remain undiscovered in outpatient treatment TM and inpatient treatment. 4"~~ Patients with psychiatric disorders use medical services more than do persons without psychiatric disorders. 1216Physical and psychiatric symptoms tend to cluster in some individuals who dominate use of medical and psychiatric services, ~2'~7'18In two register investigations in Denmark, ~7,~s patterns of general medical hospitalization among psychiatric patients were studied. The results showed that psychiatric patients used proportionately more general medical hospitalizations than did nonpsychiatric patients. In view of the lack of attention given to physical illness in psychiatric settings, and the tendency of psychiatric patients to also have physical illness needing treatment, the current study attempts to explore the extent to which physical illness is diagnosed in Israeli psychiatric hospitals. The impetus for this study was in preparation for major reform in health and mental health care in Israel. One of the goals of the reform is to integrate the mental health and the general health care systems as a Address correspondenceto Dr. JonathanRabinowitz,D.S.W.,SeniorLecturer, Schoolof Social Work, Bar Ilan University, Ramat Gan, Israel; e-mail: rabinowz@shekel.ec.biu.ac.il. Mordechai Mark, M.D., is the director of Mental Health Services for the State of Israel Ministry of Health. Miriam Popper, M.A., is the director of the Department of Information & Evaluation, Mental Health Services. State of Israel Ministry of Health. DinaFeldman, M.A., is the admirtislrativedirector of Mental Health Services for the State of Israel Ministry of Health. 82 The Journal of Mental Health Administration 24:1 Winter 1997 means of providing more comprehensive care as has been suggested in the United States. 19 The reformed mental health system will begin operating during 1997 when the National Health Insurance Act begins providing coverage for mental health services. E~ The current study is one of a series of studies done in preparation for the reform of mental health care. Using data from the Israel National Psychiatric Case Registry, it investigates rates of reported physical illness among all patients hospitalized in psychiatric wards in Israel. (The psychiatric inpatient care system in Israel includes 6 psychiatric wards of general hospitals, 15 public and governmental, and 9 private psychiatric hospitals.) Differences in reporting were explored between types of hospitals (psychiatric, psychiatric wards of general hospital, and private hospitals providing long-term care to chronically mentally ill), age groups, and first admission versus patients with at least one prior hospitalization. Method Data from the National Psychiatric Case Registry of the Ministry of Health on reported physical illness at discharge were analyzed by age, gender, and number of admissions (first in life versus veteran patients, i.e., those with more than one admission). The study included the entire population of 38,714 psychiatric discharges during 1989-1991 (n = 12,887, n = 12,939, and n = 12,888, respectively). Because the entire population of patients was included, differences are not attributable to sampling error. Approximately 25% of the discharges were of first admissions and the rest of veteran patients. About 20% of the veteran patients had more than one discharge in a calendar year and were thus included more than once in the results. Because data are stored by discharge and not by patient, we were not readily able to remove duplicate records. As a result, veteran patients with more than one discharge in a calendar year are overrepresented in the study. However, we were able to compare the first admission discharges (which by definition are unduplicated) to veteran patients. We had no a priori reason to assume that duplicated counts of veteran patients would skew the results. The source of the data is the discharge report of the treating psychiatrist, who is required by law to complete a discharge form that is entered into the case registry for any admission and discharge to a psychiatric bed in Israel. The form includes physical illness divided into nine groups (see Table 1 for list), and other clinical and demographic variables. The treating psychiatrist checks no more than two of the illness groups, or the box for "none of the above illnesses." The form lists most maj or illness categories with the notable exceptions of gynecologic and hematologic illness. Therefore, if the box for "none of the above illnesses" is checked by the psychiatrist, it does not necessarily mean that the person is perfectly healthy because he or she may suffer from some condition that is not coded on the form. Incomplete forms are routinely returned for completion to the hospital by the Ministry of Health. The response rate was 98%. Results The distribution of illnesses was almost identical for the three years studied. Table 1 presents the results for 1989. Differences between 1989 and the other two years are reported in the text. The year 1989 was chosen as the base year because it was the last year before the mass migration from the former Soviet Union to Israel. Immigrants from the former Soviet Union now comprise 10% of the national population, and they have been hospitalized at higher rates than the rest of the population.21 Despite this change in the composition of the national population, the results showed only small differences in reported physical illness in the three years. Table 1 shows that the most prevalent disease group is cardiovascular disease, which was reported for 10.13% of the patients. This was also the group with the largest reporting difference between the years. In 1990 reporting dipped to 9.04% and in 1991 it increased to 11.04%. For about 5% of the patients two illness groups were noted. These patients appear twice in Table 1. Physical Illness RABINOWITZ ET AL 83 oo +~ 9 ',0 c~ II ~o O0 "~- II cq II . ,...~ ~ ~ L"-- (..) v II C) p.., ~0 o,o c~ c~ o o ..o ~J o c~ c~ oo ~o II ~,...~ o c~ o p.., o~ ..c= ~5 84 Illness by Age Group Table 1 also shows that there is a monotonic increase of total physical findings by age group. There is a similar pattern within each illness group. The exceptions are small differences in cardiovascular illness between the two youngest groups (0.30%) and between the two oldest groups for metabolic illness-diabetes (0.06%). Illness by First and Repeat Admission Reported physical illness is higher for veteran patients than for first admissions, with an increase over the three study years (1989 = 0.78%; 1990 = 0.94%; 1991 = 4.46%). As presented in Table 1, the illness categories with at least 1% difference were (1) metabolic illness-diabetes, which was higher for veteran patients; and (2) neurological disorders, which was lower among veteran patients. Differences between illness groups prevailed for 1990 and 1991. Differences Between Hospitals There were differences between the hospitals on rates of reported physical illness. (To allow for comparisons among hospitals, 532 discharges from day hospitals and atypical institutions were removed from the balance of the analysis.) These differences were considerably larger for older patients. For patients under 25, reporting ranged from 0.00% for the hospital reporting the least physical illness to 33.00% for the hospital reporting the most; for patients 25 to 44, it ranged from 3 % to 65%; for patients 45 to 64, it ranged from 4.0% to 76%; and for patients 65 and older, it ranged from 3% to 94%. There were some differences associated with different types of hospitals. The private hospitals (n = 560 patients), which basically serve long-term chronic patients over the age of 44, reported the most physical illness. This is not surprising as it is confounded by age, because many of the discharges are due to death or transfer to a medical facility. The general hospital psychiatric wards and the public and governmental psychiatric hospitals serve patients of all ages. One psychiatric ward of a general hospital (n = 195 patients) stood out with 59% of patients having a reported physical illness (this was 88% of patients over 65, 76% of patients between the ages of 45 and 64, 56% of patients between the ages of 25 and 44, and 33% of patients under 25). The hospital with the lowest rate of reported physical illness had a 3% to 4% prevalence of reported physical illness for all age groups. Figure 1 shows total reported physical illness by hospital type and by patient age. Overall, the psychiatric wards of general hospitals (n = 1,732 discharges) reported almost 9% more physical illness than did the psychiatric hospitals (n = 10,064 discharges; 29.3% vs. 20.5%). The smallest difference (1.88%) is for the group under age 25, and the largest (14.82%) is for the group between the ages of 45 to 64. The group over age 65 breaks the pattern with 6.64% more reported physical illness in the psychiatric hospitals. The largest differences between the general hospital wards and psychiatric hospitals were in reporting of metabolic disorders-diabetes. For the entire cohort, the general hospital psychiatric wards reported 6.13% and the psychiatric hospitals reported 3.38%. This difference is primarily for the patients over age 45. For the patients over 65, the psychiatric hospitals reported neurological problems more than did the wards (10% vs. 6%). There were some differences in the percentage of cases in which no reported illness was recorded among the four largest psychiatric hospitals, which had between 953 and 1,381 discharges. Reported illness ranged from 4% known physical illness to 27%. Discussion The results show that more than three quarters of all patients discharged from psychiatric beds in Israel over the three years studied had no reported physical illness from one of the major illness Physical Illness RABINOWI'IZ ET AL. 85 Figure 1 Reported Physical Illness by Hospital Type by Patient Age P~clnWtCacWardTg'pe :.:.:.:.z.:.:,:.:,:.:, iiiiiiiiii!i!!!ili!i!i ::::::::::::::::::::: "d ii!i?iii?~-[?iii?i! :+:.:.:+;+:,1.: i!i!~iiii!iii!iiiiii ::::::::::::::::::::: .:.:.:.:.:.:.:,:,:,:, ...,............,.... i?iiiii~iiiiii!i!ii!i ..-.........- -- ................... ..................... .>:.:.:,:.:.:.:.:.:. ..................... .................... 9 iiiiiii!i!iiiiiiii i;iiiiii!iiliiiii!iii!ii i! u In I > 2.~ (O=~,061) il;i!i!i!i~i!i!!!~!!~ :::::::;:;::::::::::5 i~iiii!ii!i!i; iiiiiiiiiiiiiii!iigig ::::::::::::::::::::: iii!i:'ii!i!i!!!i!!!i ::::::::::::::::::::: I I A~e groups that hospitals are required to report. There were considerable and expected increases in reported physical illness with increase in patient's age. Psychiatric wards in general hospitals reported more physical illness than did psychiatric hospitals. This is probably due to more medical resources in general hospitals. The current finding of 25% reported physical illness among psychiatric patients is considerably lower than in studies in which careful medical examinations were done of psychiatric patients in which active, important physical illness was diagnosed in about half of the patients. In the most recent such study, Koran et al. 4 investigated rates of medical illness in California's public mental health system by medically evaluating a sample of 529 patients from eight program categories. On the basis of the results, they estimated that 45% of adult psychiatric patients in California had active, important physical diseases. Figure 2 compares the results of the current study, based on physician reporting, with the Koran results, based on medical evaluation. As can be seen, there is considerably more endocrine, neurological, digestive, and respiratory illness among the medically examined patients. Particularly noteworthy is the very large difference in neurological diseases. In the remaining three groups there were small differences in favor of physician reporting. These differences are probably even greater because the Koran et al. sample was somewhat younger than the current study (3% vs. 12.4% for the group over age 65, 69% vs. 48.4% for the group between the ages of 25 and 44, and 7% vs. 16% for the group under age 25). Koran et al. did not report physical illness by age group, so such comparisons were not possible. On the basis of the comparison with the Koran et al. results, which are similar to other studies, reported medical illness in the current study is considerably lower than expected for a psychiatric 86 The Journal of Mental Health Administration 24:1 Winter 1997 Figure 2 Physician Reporting Versus Medical Evaluation 16 15 14 13 12 11 P 9 7 a. 6 5 4 3 2 1 0 i l m iiii l - " ] I ' I Musculoskeletll I Respiratory I Endocdne Neurologl 9 Cardlovlsculsr Digestive Eye & ENT Disease Category l~ Physician Reporting MedicalEvaluation population. A further comparison with the general noninstitutionalized population further illustrates how low reporting is. Because no such data are available in Israel, U.S. data based on the United States National Health Interview Survey 22 are used for this comparison. Despite the differences between the two studies, such as those relating to data collection methods, the comparison offers a rough approximation of a distribution of somatic morbidity in the general population. The reported age groupings are the same as in the current study, with the exception of the 18-year-old cutoff point, which is 24 in the current study. Table 2 presents results of the household survey on digestive, respiratory, and cardiovascular illness by age group. These conditions were the most readily comparable to the current study. A comparison of Table 2 and Table 1 shows that for the selected illness groups, the household survey reported about three times more physical illness on average than did the hospital reporting. For example, whereas almost 80% of household dwellers over 65 reported cardiovascular disease, it was reported for only about one-third of patients. It seems unlikely that such large differences could be attributable to differences in methods. It is also doubtful that the differences are due to definition of illness because the household survey asked about serious chronic conditions that if known to physicians would probably be recorded. The respiratory conditions included were chronic bronchitis, asthma, and emphysema; digestive conditions included ulcer, hernia of abdominal cavity, gastritis or duodenitis, enteritis or colitis, spastic colon, and diverticula of intestines; and the cardiovascular illnesses included were heart disease, cerebrovascular disease, hardening of the arteries, and hypertension. Physical Illness RABINOWITZ ET AL. 87 Table 2 Selected Findings From the United States National Health Interview Survey 22 Age Disease 18-44 45-64 Cardiovascular Respiratory Digestive 10.75 7.83 5.49 43.06 10.77 12.32 65 and over 78.05 14.37 16.02 Note: Numbers represent percentage within age group reporting problem; duplicate counts allowed. The lower rates of reported physical illness in the current study as compared to other psychiatric and general populations may be due to inadequacies of the reporting form, laxity on the part of psychiatrists completing the forms, actual lower rates, or underdiagnosis of physical illness. The consistency of the findings among the three study years, within hospitals and the expected age increases, suggests that the data are reliable. Assuming that the psychiatric patient population in Israel is not considerably healthier than in the Koran study and that reporting is not systematically biased, this suggests that physical illness is underreported because it is underdiagnosed. Such a finding would be similar to those studies 14'1~ that found considerable underdiagnosis of physical illness among psychiatric patients. Implications for Mental Health Service Delivery Previous studies have grappled with ways of improving detection of physical illness among psychiatric patients. The Koran et al. study led to developing a medical algorithm for detecting physical illness in psychiatric patients. 1~They also recommended 4 educating staff, mandating the use of standardized medical history and physical examination forms, encouraging routine dialogue with patients about medical problems, recruiting staff who are knowledgeable and committed in this area, and improving patient access to physical health care services by creating linkages between health and mental health systems. These recommendations, with the exception of the last one, are basically in-hospital solutions. Different from these approaches, systemwide interventions were recommended. Specifically, it was recommended that hospitals should be given additional reimbursement for treating patients with physical illness. This would give hospitals a financial incentive and the wherewithal to diagnose and treat physical illness. This recommendation was included in the reimbursement system under the National Health Insurance Act. Also under National Health Insurance hospitalization is covered by the patient's Health Maintenance Organization (HMO). Previously, hospitalization was covered by the Ministry of Health, primarily through historic allocations to hospitals. Under the new system, HMOs will be scrutinizing the care that their patients receive in the hospital. This should result in closer attention to patients' care including physical illnesses, as well as in sharing of information from the HMO about the patient (e.g., preexisting physical illnesses). Another change that may increase attention to physical illness is that psychiatric beds are being shifted in favor of moving beds from psychiatric hospitals to psychiatric wards of general hospitals where care for medical problems is more abundant. To raise consciousness to this issue, highlights of this study were presented in a Hebrew-language publication that is read by many health care practitioners. A follow-up study is being planned for 1997 to examine changes in reported physical illness. 88 The Journal of Mental Health Administration 24:1 Winter 1997 References 1. Koranyi EK: Morbidity and rate of undiagnosed physical illnesses in a psychiatric clinic population. Archives of General Psychiatry 1979; 36:414-419. 2. Bartsch DA, Shem DL, Feinberg LE, et al.: Screening CMHC outpatients for physical illness. Hospital and Community Psychiatry 1990; 41:786-790. 3. Maricle RA, Hoffman WF, Bloom JW, et al.: The prevalence and significance of medical illness among chronically mentally ill outpatients. CommunityMental Health Journal 1987; 23:81-90. 4. Koran LM, Sox HC, Marton KI, et al.: Medical evaluation of psychiatric patients: Results in a state mental health system. Archives of General Psychiatry 1989; 46:733-740. 5. Davies EW: Physical illness in psychiatric out-patients. British Journal of Psychiatry 1965; 111:27-33. 6. Koranyi EK: Somatic illness in psychiatric patients. Psychosomatics 1980; 21:887-889. 7. Hall RCW, Popkin MK, Devaul RA, et al.: Physical illness presenting as psychiatric disease. Archives of General Psychiatry 1978; 35:1315-1320. 8. Hall RCW, Beresford TP, Gardner ER, et al.: The medical care of psychiatric patients. Hospital and CommunityPsychiatry 1982; 33:25-34. 9. Honig A, Pop P, Tan ES, et al.: Physical illness in chronic psychiatric patients from a community psychiatric unit. The implications for daily practice. British Journal of Psychiatry 1989; 155:58-64. 10. Sox HCJ, Koran LM, Sox CH, et al.: A medical algorithm for detecting physical disease in psychiatric patients. Hospital and Community Psychiatry 1989; 40:1270-1276. 11. Knutsen E, DuRand C: Previously unrecognized physical illnesses in psychiatric patients. Hospital and Community Psychiatry 1991; 42:182-186. 12. Hinkle LE, Wolff HG: The nature of man's adaptation to his total environment and the relation of this to illness. Archives of Internal Medicine 1957; 99:442-460. 13. Eastwood MR, Trevelyan MH: Relationship between physical and psychiatric disorder. Psychological Medicine 1972; 2:363-372. 14. Lipowski ZJ: PsychosomaticMedicine and Liaison Psychiatry. Selected papers. New York: Plenum, 1985. 15. Lipowski ZJ: The interface of psychiatry and medicine: Towards integrated health care. Canadian Journal of Psychiatry 1987; 32:743-748. 16. Hankin JR, Steinwachs DM, Regier DA, et al.: Use of general medical care services by persons with mental disorders. Archives of General Psychiatry 1982; 39:225-231. 17. Fink P: Mental illness and admission to general hospitals: A register investigation. Acta PsychiatricaScandinavica 1990; 82:458-462. 18. Fink P: Physical disorders associated with mental illness: A register investigation. Psychological Medicine 1990; 20:829-834. 19. Mechanic D: Integrating mental health into a general health care system. Hospital and Community Psychiatry 1994; 45:893-897. 20. Mark M, Rabinowitz J, Feldman D, et al.: Reform in mental health services in Israel: The changing role of government, HMO's and hospitals. Administration and Policy in Mental Health 1995; 23(3):253-259. 21. Popper M, Horowitz R: Psychiatric Hospitalization oflmmigrants 1990-1991. Statistical Report No. 7 (In Hebrew) ed. Jerusalem: Israel Ministry of Health, Department of Information and Evaluation, 1992. 22. U.S. Department of Health and Human Services: Vitaland Health Statistics, CurrentEstimatesform the National Health Interview Survey. Series 10, No. 173 ed. Washington, DC: U.S. Department of Health and Human Services, 1989. Physical Illness RABINOWITZ ET AL. 89