American Journal of Epidemiology Advance Access originally published online on November 10, 2006
American Journal of Epidemiology 2007 165(3):319-324; doi:10.1093/aje/kwk003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ORIGINAL CONTRIBUTIONS |
Psychiatric Hospitalizations in a Cohort of Danish Polio Patients
1 Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark
2 Department of Infectious Diseases, National University Hospital, Copenhagen, Denmark
3 The Bandim Health Project, Bissau, Guinea-Bissau
Correspondence to Nete Munk Nielsen, Department of Epidemiology Research, Statens Serum Institut, Artillerivej 5, 2300 Copenhagen S, Denmark (e-mail: nmn{at}ssi.dk).
Received for publication December 28, 2004. Accepted for publication June 6, 2006.
| ABSTRACT |
|---|
|
|
|---|
Although previous polio infection remains a considerable cause of long-term morbidity worldwide, few studies have examined the psychiatric consequences of poliomyelitis. The authors followed 4,660 polio patients hospitalized at the primary infectious disease hospital in Copenhagen, Denmark, between 1922 and 1954 as well as 19,017 age- and gender-matched Danes for psychiatric hospitalizations from January 1, 1977, to December 31, 1993. Incidence rates of all psychiatric disorders combined and of separate diagnostic groups of psychiatric diseases in the two cohorts were compared, yielding the incidence rate ratio, a measure of relative risk. Overall, polio patients had a 40% increased risk of being hospitalized for a psychiatric disorder (incidence rate ratio = 1.43, 95% confidence interval: 1.23, 1.66). Apparently, the overall increased risk of psychiatric hospitalizations could not be confined to specific groups of psychiatric disorders but seemed to be explained by slightly increased risks of several different disorders, especially milder psychiatric disorders. Finally, psychiatric morbidity did not differ between paralytic and nonparalytic polio patients. History of hospitalization for polio might be associated with subsequent risk of hospitalization for psychiatric disorders. The underlying mechanism for this association remains uncertain.
Denmark; hospitalization; mental disorders; poliomyelitis
Abbreviations: CI, confidence interval; ICD-8, International Classification of Diseases, Eighth Revision; IRR, incidence rate ratio
| INTRODUCTION |
|---|
|
|
|---|
Chronic and life-threatening diseases are known to be accompanied by increased risk of anxiety, depression, and suicide (17). It has furthermore been suggested that certain medical conditions such as childhood infections of the central nervous system could subsequently increase the risk of psychosis and schizophrenia (810).
Poliomyelitis may cause a very serious central nervous system infection, which may be accompanied by paralysis of the limbs, trunk, or respiratory system (11, 12). Although many of the persons who contracted poliomyelitis during the large epidemics in the 1940s and 1950s recovered completely, thousands were left with lifelong disabilities (13), which makes poliomyelitis an important contributor to the burden of chronic diseases.
Three to four decades after the acute infection, many polio patients experience new signs of neuromuscular dysfunction, the so-called post-polio syndrome defined by new muscle weakness, pain, atrophy, and fatigue (13, 14). What remains uncertain is whether polio survivors are also at risk of psychiatric sequelae. Some studies have shown elevated depression and distress scores among polio patients (1517), whereas other studies have shown normal levels (1820). However, previous studies have been limited by few participants, self-reported symptoms, use of different depression scale systems, and inclusion of selected groups of polio patients, often paralytic polio patients seeking help for other symptoms (1520).
In the present study, we took advantage of historical medical archives on patients hospitalized for poliomyelitis in Copenhagen, Denmark, between 1922 and 1954 and of the nationwide Danish Psychiatric Central Research Register to assess the risk of being hospitalized for a psychiatric disorder among patients with a history of polio.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Patients treated for poliomyelitis between 1922 and 1954 at the Blegdamshospital, the main infectious disease hospital in Copenhagen, were identified as described previously (21). Data for all cases with a discharge diagnosis of paralytic polio, nonparalytic polio, or primary lymphocytic meningitis were included in the analysis. Primary lymphocytic meningitis was generally considered nonparalytic polio after other possible viral infections such as mumps were excluded (22).
A total of 4,660 patients (paralytic: n = 1,903 (including 67 persons admitted for polio sequelae); nonparalytic: n = 2,202; primary lymphocytic meningitis: n = 555) were alive and residing in Denmark on January 1, 1977. For each patient, four persons were identified in the Danish Civil Registration System, matched on sex, age, and geographic residence as of January 1, 1977. Both cohorts have previously been included in a study of the incidence of somatic diseases among polio patients for the period 19771999 (21). By means of the unique personal identification number assigned to all Danish citizens, the exposed (poliomyelitis) and the unexposed cohorts were linked with the Danish Psychiatric Central Research Register, which, since 1970, has registered electronically all admissions to Danish psychiatric hospitals and to psychiatric departments at general hospitals. Day admissions were included in the register in 1974 (23). From 1969 until December 31, 1993, psychiatric diseases were coded according to the International Classification of Diseases, Eighth Revision (ICD-8). Beginning on January 1, 1994, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes were used. Because the ICD-10 and ICD-8 codes are not easily comparable, follow-up ended on December 31, 1993. The study was approved by the Danish Data Protection Agency.
Outcomes
As outcomes, we looked at first hospitalization (date of admission) for any psychiatric disorder and each of the specific psychiatric disorders in the list below, with the further requirement that it should be the worst outcome so far according to the hierarchy. The latter requirement reflects the probability that later psychiatric disorders could be manifestations of earlier, more severe psychiatric disorders and thus not really incidence outcomes. Accordingly, psychiatric diagnoses for each patient were arranged in a hierarchy (24) in the following order modified from Lynge et al. (25):
- Organic disorders: ICD-8 codes 290, 292, 293, 294, 309
- Schizophrenia: ICD-8 code 295
- Manic-depressive psychosis: ICD-8 code 296
- Other psychosis: paranoid states, ICD-8 code 297; reactive psychosis, ICD-8 code 298; unspecified psychosis, ICD-8 code 299
- Neurosis: ICD-8 code 300
- Personality disorder: ICD-8 code 301
- Substance or alcohol abuse: ICD-8 codes 291, 303, 304
- Other nonpsychotic mental disorders: transient maladaptation, ICD-8 code 307; other diagnoses, ICD-8 codes 302, 305, 306, 308
Thus, if a person was diagnosed with a neurosis, he or she would still be at risk of subsequently developing diseases in group 14 but no longer at risk of developing diseases in group 68. The present article deals with primary and auxiliary diagnoses among all inpatients, including day and night patients. We considered only those psychiatric diseases categorized as either psychoses, neurosis, personality disorders, or other nonpsychotic mental disorders (ICD-8 codes 290309).
Members of the cohorts were followed from January 1, 1977, until the date of outcome considered, disappearance, emigration, death, a more severe outcome (in the disease-specific analysis), or December 31, 1993, whichever occurred first. To avoid prevalent cases of psychiatric diseases, patients diagnosed with any psychiatric disorder before January 1, 1977, were excluded from the analyses.
Statistical methods
The incidence rate for each outcome in the cohorts was calculated as the number of persons experiencing the outcome during follow-up, divided by the total follow-up time in person-years for that outcome in the respective cohorts. The ratio of incidence rates in the exposed and unexposed cohorts, the incidence rate ratio (IRR), served as a measure of the relative risk. Ninety-five percent confidence intervals for the IRR were estimated from Wald's test assuming a Poisson distribution of the observed cases. Whenever the exposed cohort was divided, to maintain confounder control, the unexposed cohorts were defined as the individually matched controls for the exposed individuals. All tests of statistical significance were two-sided, likelihood-ratio tests.
| RESULTS |
|---|
|
|
|---|
Persons included in the study were on average age 38.9 years on January 1, 1977 (median, 34.8 years; interquartile range, 29.846.7 years) and were followed for an average of 15.4 years, yielding a total of 365,773 person-years of follow-up. Patients contracted poliomyelitis between 1922 and 1954, the majority (60 percent) between 1950 and 1953, at an average age of 11.3 years (median, 7.4 years; range, 14 days59.2 years).
Overall, history of poliomyelitis was associated with a 40 percent increased risk of being hospitalized for a psychiatric disorder (IRR = 1.43, 95 percent confidence interval (CI): 1.23, 1.66) (table 1). Similar risks were observed among the three groups of polio patients. The risk of psychiatric hospitalization tended to be highest before age 45 years (IRR = 1.82, 95 percent CI: 1.50, 2.20 vs. IRR = 1.02, 95 percent CI: 0.79, 1.30; p < 0.01) (table 1) and was slightly higher among those hospitalized for polio before age 7 years (preschool children) (IRR = 1.71, 95 percent CI: 1.38, 2.09 vs. IRR = 1.19, 95 percent CI: 0.95,1.48; p < 0.05) (table 1).
|
The overall increased risk of psychiatric hospitalizations could not be confined to specific groups of psychiatric disorders. Risk estimates for several of the psychiatric disorders seemed increased, but statistically significantly IRRs were observed for only organic disorders (IRR = 1.49, 95 percent CI: 1.05, 2.08) and milder psychiatric disorders such as personality disorders (IRR = 2.10, 95 percent CI: 1.56, 2.79), substance/alcohol abuse (IRR = 1.65, 95 percent CI: 1.22, 2.22), and other nonpsychotic mental disorders (IRR = 2.39, 95 percent CI: 1.39, 4.02) (table 2). This risk distribution applied equally to all three groups of patients (paralytic polio, nonparalytic polio, and primary lymphocytic meningitis).
|
| DISCUSSION |
|---|
|
|
|---|
In the present cohort study, we observed a modestly increased risk of hospitalization for psychiatric disorders among patients with a history of poliomyelitis, especially those diagnosed with polio before age 7 years. Although the IRRs of grouped psychiatric diseases did not reveal an obvious pattern, the risk of especially milder psychiatric diseases seemed to be increased.
Surprisingly, neither overall nor disease-specific psychiatric morbidity differed between paralytic and nonparalytic polio patients. Even milder polio virus infections seem to affect and damage certain areas of the cerebral cortex and centers of the brainstem, including the Reticular Activating System (26). This system is responsible for the process of attention, and it has been suggested that this damage might be associated with the difficulties in mental functioning observed among polio patients even months after acute infection (26, 27). One might therefore speculate that the central nervous system infection per se has an important influence on the risk of psychiatric disorders among polio survivors. However, a complex interaction between biologic, psychological, and sociologic mechanisms might be a more plausible explanation.
Stressful life events are considered important in the etiology or triggering of psychiatric diseases (28). For those affected, contracting poliomyelitis was a very painful and fearful event. The stringent isolation of hospitalized polio patients meant separation from parents and friends for weeks or even months and total dependence on hospital staff (12, 29). When reentering society, paralytic polio patients not only experienced all the physical problems and limitations associated with being disabled, but some of them were also exposed to social prejudice, isolation, and inappropriate parental rearing because of their disabilities (29). Interestingly, traumatic childhood events such as major illness, hospitalization, separation from parents, and physical handicaps have been suggested to be associated with an increased risk of personality disorders (28, 30). Furthermore, to survive and succeed in a society full of barriers, it is believed that polio patients acquired the highly stressful type A behavior characterized by being hard-driving overachievers and perfectionists (29, 31). Thus, the observed psychiatric morbidity among previous polio patients might be related to the traumatic experience of contracting polio, the following parental and social attitudes, and the struggle to achieve social normalcy.
Nonparalytic polio patients might have been spared many of the psychological and sociologic traumas associated with suddenly being handicapped. However, a considerable proportion of nonparalytic polio patients were left with undiagnosed muscle weakness (32). One might therefore speculate that this group of so-called nonparalytic polio patients did not receive adequate treatment and attention during the acute phase of the disease and that further symptoms might have been ignored by the health authorities or considered not related to their previous polio disease. In addition, marginally disabled people tend to maladjust to a greater extent than severely disabled persons because of the more frequent role conflict (33). Obviously, severely handicapped persons are disabled, whereas those marginally disabled may act or seem to be nondisabled in a wide range of situations (33). Compared with paralytic polio patients, nonparalytic polio patients might therefore experience a greater role ambiguity and presumably a greater role conflict (33). Such circumstances might contribute to the observed increased risk of psychiatric hospitalizations among nonparalytic polio patients.
A supplementary analysis revealed that the risk of specific depressive disorders (ICD-8 codes 296.09, 296.29) was not increased among previous paralytic polio patients (IRR = 0.66, 95 percent CI: 0.22, 1.55; nonhierarchic analysis), a finding in contrast to some (1517) but not all of the previous studies dealing with mental symptoms among paralytic polio patients (1820) and to observations among patients suffering from other chronic neurologic diseases such as Parkinson's disease, epilepsy, and spinal cord injury (7, 34, 35).
The observed association between poliomyelitis and psychiatric disorders might, however, be influenced by different kinds of bias. An overestimation could be explained by the phenomenon that patients already in contact with the health care system are more likely to be diagnosed with another disease. However, such bias would have led to a particularly high risk of psychiatric diseases among paralytic polio patients, contrary to our observations. Underestimation could be the result of "survival" bias because only those polio patients alive on January 1, 1977, could participate in the study (beginning of follow-up). Accordingly, only the healthiest polio patients were included in the study and, furthermore, patients were on average age 39 years at the beginning of follow-up (i.e., psychiatric diseases presenting in childhood and adolescence were not included in the study). Furthermore, only those patients with major psychiatric diseases are admitted to psychiatric hospitals. Accordingly, the risk of minor psychiatric difficulties such as those treated by a general practitioner or a psychiatrist could not be investigated in the present study.
An unknown percentage of polio patients, nonparalytic (32) as well as paralytic, might develop postpolio syndrome (36, 37), which also seems to be accompanied by symptoms of anxiety and depression (36, 38, 39). Accordingly, psychiatric disorders could be misinterpreted as part of the postpolio syndrome and vice versa, leading to diagnostic ambiguity. In addition, mental symptoms associated with postpolio syndrome are probably more prone to be misinterpreted as psychiatrically relevant among nonparalytic polio patients than among paralytic polio patients with obvious signs of the previous polio infection.
Information on exposure (poliomyelitis) and outcome (psychiatric disorders) was collected independently from the historical archives of the Blegdamshospital and the Psychiatric Central Research Register. Coverage of the Psychiatric Central Research Register is high, between 95 and 100 percent (23). However, data in the Psychiatric Central Research Register were originally collected for administrative purposes and not epidemiologic research, which may affect their validity. Still, diagnoses such as schizophrenia and affective psychosis are considered to have a high validity (40). The diagnosis of personality disorders is less reliable, and substance abuse is considerably underreported (41). However, we have no reason to believe that the validity of the data should differ between the two cohorts. Accordingly, a diagnostic misclassification would be nondifferential and render our risk estimates conservative.
Finally, the diagnosis of paralytic polio, nonparalytic polio, and primary lymphocytic meningitis was based on clinical observations, which implies risk of misclassification. This is particularly true for patients without paralysis who might have suffered from infections with other viruses or other conditions. Still, this possibility would have no impact on the results obtained for paralytic patients.
To our knowledge, this is the first population-based assessment of the occurrence of psychiatric disorders among previous polio patients. The observed association merits attention and emphasizes the importance of supportive vigilance toward this group of people.
| ACKNOWLEDGMENTS |
|---|
This study was financed by the Danish Medical Research Council, the Danish Development Research Council, the Danish National Research Foundation, the Wedell-Wedellsborg's Foundation, A.P. Møller's Foundation, and the National Polio Society (PTU).
The authors are grateful to Dr. Povl Munk-Jørgensen, former Chief of the Department of Psychiatric Demography, Institute for Basic Psychiatric Research, Aarhus University Hospital, Denmark, for very helpful comments on the manuscript. They are also grateful to the staff at the Copenhagen City Archives, who helped identify polio patients' records.
Conflict of interest: none declared.
| References |
|---|
|
|
|---|
- Fruehwald S, Loeffler-Stastka H, Eher R, et al. (2001) Depression and quality of life in multiple sclerosis. Acta Neurol Scand 104:25761.[CrossRef][Web of Science][Medline]
- Saikkonen J, Karppi P, Huusko TM, et al. (2004) Life situation of spinal cord-injured persons in Central Finland. Spinal Cord 42:45965.[CrossRef][Web of Science][Medline]
- Kennedy P and Rogers BA. (2000) Anxiety and depression after spinal cord injury: a longitudinal analysis. Arch Phys Med Rehabil 81:9327.[CrossRef][Web of Science][Medline]
- Bankier B, Januzzi JL, Littman AB. (2004) The high prevalence of multiple psychiatric disorders in stable outpatients with coronary heart disease. Psychosom Med 66:64550.
[Abstract/Free Full Text] - Harris EC and Barraclough BM. (1994) Suicide as an outcome for medical disorders. Medicine (Baltimore) 73:28196.[Medline]
- Kishi Y, Robinson RG, Kosier JT. (2001) Suicidal ideation among patients with acute life-threatening physical illness: patients with stroke, traumatic brain injury, myocardial infarction, and spinal cord injury. Psychosomatics 42:38290.
[Abstract/Free Full Text] - Nilsson FM, Kessing LV, Sorensen TM, et al. (2002) Major depressive disorder in Parkinson's disease: a register-based study. Acta Psychiatr Scand 106:20211.[CrossRef][Web of Science][Medline]
- Rantakallio P, Jones P, Moring J, et al. (1997) Association between central nervous system infections during childhood and adult onset schizophrenia and other psychoses: a 28-year follow-up. Int J Epidemiol 26:83743.
[Abstract/Free Full Text] - Gattaz WF, Abrahao AL, Foccacia R. (2004) Childhood meningitis, brain maturation and the risk of psychosis. Eur Arch Psychiatry Clin Neurosci 254:236.[CrossRef][Web of Science][Medline]
- Leask SJ, Done DJ, Crow TJ. (2002) Adult psychosis, common childhood infections and neurological soft signs in a national birth cohort. Br J Psychiatry 181:38792.
[Abstract/Free Full Text] - Christie AB. (1980) Acute poliomyelitis. In: Infectious diseases. Epidemiology and clinical practice. 3rd ed (Churchill Livingstone, New York, NY)572604.
- Mulder DW. (1995) Clinical observations on acute poliomyelitis. Ann N Y Acad Sci 753:110.[CrossRef][Web of Science][Medline]
- Pascuzzi RM. (1992) Poliomyelitis and the postpolio syndrome. Semin Neurol 12:1939.[Web of Science][Medline]
- Halstead LS. (1998) Post-polio syndrome. Sci Am 278:427.[Web of Science][Medline]
- Conrady LJ, Wish JR, Agre JC, et al. (1989) Psychologic characteristics of polio survivors: a preliminary report. Arch Phys Med Rehabil 70:45863.[CrossRef][Web of Science][Medline]
- Berlly MH, Strauser WW, Hall KM. (1991) Fatigue in postpolio syndrome. Arch Phys Med Rehabil 72:11518.[Web of Science][Medline]
- Freidenberg DL, Freeman D, Huber SJ, et al. (1989) Postpoliomyelitis syndrome: assessment of behavioral features. Neuropsychiatry Neuropsychol Behav Neurol 2:27281.
- Schanke AK. (1997) Psychological distress, social support and coping behaviour among polio survivors: a 5-year perspective on 63 polio patients. Disabil Rehabil 19:10816.[Web of Science][Medline]
- Tate DG, Forchheimer M, Kirsch N, et al. (1993) Prevalence and associated features of depression and psychological distress in polio survivors. Arch Phys Med Rehabil 74:105660.[CrossRef][Web of Science][Medline]
- Tate D, Kirsch N, Maynard F, et al. (1994) Coping with the late effects: differences between depressed and nondepressed polio survivors. Am J Phys Med Rehabil 73:2735.[Web of Science][Medline]
- Nielsen NM, Rostgaard K, Askgaard D, et al. (2004) Life-long morbidity among Danes with poliomyelitis. Arch Phys Med Rehabil 85:38591.[CrossRef][Web of Science][Medline]
- Lassen HCA. (1939) Comparative studies of primary lymphocytic meningitis and paralytic polio. (In Danish). Ugeskr Læger 3:7380.
- Munk-Jorgensen P, Mortensen PB. The Danish. (1997) Psychiatric Central Register. Dan Med Bull 44:824.[Web of Science][Medline]
- Kendell RE. (1975) The role of diagnosis in psychiatry. (Blackwell Scientific Publications, Oxford, United Kingdom).
- Lynge I, Mortensen PB, Munk-Jorgensen P. (1999) Mental disorders in the Greenlandic population. A register study. Int J Circumpolar Health 58:18897.[Medline]
- Bruno RL, Cohen JM, Galski T, et al. (1994) The neuroanatomy of post-polio fatigue. Arch Phys Med Rehabil 75:498504.[Web of Science][Medline]
- Meyer E. (1947) Psychological considerations in a group of children with poliomyelitis. J Pediatr 31:3448.[CrossRef][Web of Science]
- Barlow D and Durand V. (1999) Abnormal psychology. 2nd ed. (Brook/Cole Publishing Company, Pacific Grove, CA).
- Bruno RL and Frick NM. (1991) The psychology of polio as prelude to post-polio sequelae: behavior modification and psychotherapy. Orthopedics 14:118593.[Web of Science][Medline]
- Bandelow B, Krause J, Wedekind D, et al. (2005) Early traumatic life events, parental attitudes, family history, and birth risk factors in patients with borderline personality disorder and healthy controls. Psychiatry Res 134:16979.[CrossRef][Web of Science][Medline]
- Bruno RL and Frick NM. (1987) Stress and "type A" behavior as precipitants of post-polio sequelae: the Felician/Columbia Survey. Birth Defects Orig Artic Ser 23:14555.[Medline]
- Bruno RL. (2000) Paralytic vs. "nonparalytic" polio: distinction without a difference? Am J Phys Med Rehabil 79:412.
- Colman AM. (1971) Social rejection, role conflict, and adjustment: psychological consequences of orthopaedic disability. Percept Mot Skills 33:90710.[Web of Science][Medline]
- Nilsson FM, Kessing LV, Sorensen TM, et al. (2003) Affective disorders in neurological diseases: a case register-based study. Acta Psychiatr Scand 108:4150.[Web of Science][Medline]
- Kemp BJ and Krause JS. (1999) Depression and life satisfaction among people ageing with post-polio and spinal cord injury. Disabil Rehabil 21:2419.[CrossRef][Web of Science][Medline]
- Dalakas MC. (1995) The post-polio syndrome as an evolved clinical entity. Definition and clinical description. Ann N Y Acad Sci 753:6880.[CrossRef][Web of Science][Medline]
- Rekand T, Karlsen B, Langeland N, et al. (2002) Long-term follow-up of patients with nonparalytic poliomyelitis. Arch Phys Med Rehabil 83:5337.[CrossRef][Web of Science][Medline]
- Tillett SG and Mozena JD. (1999) The reappearance of polio. Postpolio syndrome. J Am Podiatr Med Assoc 89:1837.[Abstract]
- Hazendonk KM and Crowe SF. (2000) A neuropsychological study of the postpolio syndrome: support for depression without neuropsychological impairment. Neuropsychiatry Neuropsychol Behav Neurol 13:11218.[Web of Science][Medline]
- Kessing LV, Olsen EW, Mortensen PB, et al. (1999) Dementia in affective disorder: a case-register study. Acta Psychiatr Scand 100:17685.[Web of Science][Medline]
- Hansen SS, Munk-Jorgensen P, Guldbaek B, et al. (2000) Psychoactive substance use diagnoses among psychiatric in-patients. Acta Psychiatr Scand 102:4328.[CrossRef][Web of Science][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||