Subsequent Risk of Hospitalization for Neuropsychiatric Disorders in Patients With Rheumatic Diseases: Title and subTitle BreakA Nationwide Study From Sweden FREE

Rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and ankylosing spondylitis (AS), also known as Bekhterev disease, are caused by misdirected autoimmune inflammation in the joints and/or spine. Concurrent inflammation in the internal organs is associated with nephritis, myocarditis, colitis, and other diseases.

These 3 rheumatic diseases are relatively common in the general population. The prevalence of RA and AS in industrialized countries is 1% to 2% and 0.5% to 4%, respectively.¹ Both RA and SLE are more common among women, whereas AS is more common among men. However, SLE exhibits the strongest sex difference; the prevalence rates are approximately 4 per 100 000 men and 45 per 100 000 women.^{2 - 3}

In most cases, the medication for these diseases is lifelong, rigorous, and often associated with adverse effects. In addition, these chronic diseases cause long-lasting suffering for many of the affected individuals, which is characterized by chronic pain, a worsened quality of life,^{4 - 7} and neurological and psychiatric disorders.^{4 ,8 - 10} For example, SLE is associated with affective disorders^{8 - 10} and psychosis.^{8 ,10} There is evidence that neuropsychiatric disorders in patients with SLE are caused by immunological mechanisms¹¹ and a decreased regional cerebral blood flow.¹² However, psychological distress and stressful life events are also important factors in the development of maladaptive coping styles, according to a study of patients with RA and SLE from Denver, Colorado.¹³ A study of patients with AS found that disease status scores correlated significantly with anxiety and depression.¹⁴ It is, therefore, likely that these diseases have a negative effect on the mental health of affected individuals and possibly on their risk of hospitalization for psychiatric disorders.

Results from previous studies are relatively consistent. However, a novel approach of the present study is that we included all hospitalizations in Sweden for these rheumatic diseases between 1973 and 2004 as well as hospitalizations for subsequent affective, psychotic, neurotic, and personality disorders. The aim was to analyze the association between RA, SLE, or AS and severe psychiatric disorders leading to hospitalization. In addition, we also analyzed these associations by sex, age, and various follow-up intervals between the first hospitalization for RA, SLE, or AS and for psychiatric disorders and the association between hospitalization for SLE and dementia or delirium.

Patients were followed up from January 1, 1973, until hospital diagnosis, death, emigration, or the end of the study on December 31, 2004. Patients with RA, SLE, or AS were defined as those who, during the study period, were hospitalized for RA, SLE, or AS for the first time before being hospitalized for a neuropsychiatric disorder for the first time.

Data used in this study were retrieved from the MigMed database, Center for Family and Community Medicine, Karolinska Institute. MigMed is a single, comprehensive database that has been constructed using several national Swedish data registers, including, but not limited to, the Total Population Register and the Swedish Hospital Discharge Register.^{15 - 16} Information from the various registers in the database is linked at the individual level via the 10-digit national registration number assigned to each person in Sweden for his or her lifetime. Before inclusion in the MigMed database, national registration numbers were replaced by serial numbers to ensure the anonymity of all individuals, and serial numbers were used for all data linkages. This study was approved by the Ethics Committee of the Karolinska Institute.

Diagnostic codes were retrieved from the Swedish Hospital Discharge Register according to the 8th (1973-1986), 9th (1987-1996), and 10th (1997-2004) editions of the International Classification of Diseases (Table 1).^{17 - 19} The Swedish Hospital Discharge Register covers all hospitals in Sweden. Only first hospitalizations during the study period were included. Hospitalization for the diagnostic subcategories are referred to by the name of the disease.

Age at hospital diagnosis of neuropsychiatric disorders was categorized in 5-year groups, and the groups were merged as necessary. Geographic regions were divided into large cities (cities with a population of >200 000, ie, Stockholm, Gothenburg, and Malmö), Southern Sweden, and Northern Sweden. Geographic region was included as an individual variable to adjust for possible differences with regard to hospital admissions. In addition, time period was included to adjust for possible differences in hospitalization rates over time. The follow-up interval between first hospitalizations for RA, SLE, or AS and for any psychiatric disorder was categorized as less than 1, 1 to 4, 5 to 9, and 10 or more years.

Person-years were calculated from the start of follow-up on January 1, 1973, until hospital diagnosis, death, emigration, or the end of the study on December 31, 2004. Age-standardized incidence ratios (SIRs) were calculated for the entire follow-up period, divided into 3-year periods. The SIRs were calculated as the ratio of the observed to the expected number of cases of subsequent neuropsychiatric disorders²⁰ in patients with rheumatic diseases. The expected number of cases was based on the expected number of cases in the general population, calculated for age (in 5-year groups), sex, time period, and geographic region. Confidence intervals (95% CIs) were calculated assuming a Poisson distribution. Interaction tests were performed to determine whether the association between the 3 rheumatic diseases and psychiatric disorders differed by age. No interaction was found.

Table 2 shows the number of cases and hospitalization rates for rheumatic diseases, hospitalization rates for psychiatric disorders in the general population, and hospitalization rates for psychiatric disorders following rheumatic diseases. Mean age at diagnosis was higher for RA than for AS and SLE. Hospitalization rates for RA, SLE, and AS varied by sex.

We compared age-specific rates of psychiatric disorders in the general population and subsequent psychiatric disorders following rheumatic diseases (Figure). Rates of psychiatric disorders following rheumatic diseases tended to be higher than rates of psychiatric disorders in the general population.

In most age groups, an increased risk was found among men and women with RA, SLE, and AS compared with the reference group (the general population, matched for age, sex, and other factors) (Table 3). Overall, the highest risks were found among men and women with SLE. Their significant SIRs were 2.38 and 2.16, respectively. For RA, the overall significant SIRs were 1.45 and 1.36 for men and women, respectively.

Men with AS had increased risks of affective, neurotic, and personality disorders (Table 4). Women with AS had increased risks of affective and neurotic disorders. Men with SLE had increased risks of affective and personality disorders, whereas women with SLE had an increased risk of psychotic, affective, neurotic, and personality disorders. Men and women with RA had an increased risk of affective, neurotic, and personality disorders. Women with RA also had an increased risk of psychotic disorders. The risk of severe depression was increased among women with SLE and women with RA.

Among men and women, the highest SIRs were observed when the follow-up interval between first hospitalization for RA, SLE, or AS and subsequent psychiatric disorder was less than 1 year (Table 5). This tendency was especially strong among patients with SLE.

For dementia, the SIR was significantly increased only among women with SLE (Table 6). However, the SIRs for delirium were particularly high among men and women with SLE. The significant SIRs were 3.73 and 4.99 for men and women, respectively.

In an additional analysis, we investigated whether multiple (>3) hospitalizations for RA implied an increased risk of psychiatric disorders. We found that the risk was similar compared with that of patients hospitalized only once for RA (data not shown).

This study shows that patients with rheumatic diseases have a higher risk of psychiatric disorders than the general population. Patients with SLE or AS tended to have a higher risk of subsequent psychiatric disorders than patients with RA, after accounting for age. In addition, differences were found in the risk of subsequent psychiatric disorders by subtype of the rheumatic disease, psychiatric disorder, sex, and various follow-up intervals. For example, only women with RA or SLE had an increased risk of psychotic disorders and severe depression. In addition, SLE carried an increased risk of dementia and delirium.

A key strength of the present study is that the study population represented the entire population of Sweden. This allowed us to present risk estimates by sex and subtypes of the 3 rheumatic diseases and the psychiatric disorders. In addition, the data in the Swedish Hospital Discharge Register are nearly 100% complete. In 2001, the main diagnosis was missing in only 0.9% of the hospitalized cases.¹⁵ Because of the national registration number assigned to each individual in Sweden, it was possible to track the records of every person for the whole follow-up period and calculate the exact risk-time. Finally, we analyzed the associations between RA, SLE, or AS and psychiatric disorders by various follow-up intervals. We have not found similar previous studies for comparison. However, a possible explanation for the finding that the highest SIRs were observed when the follow-up interval was less than 1 year is that the individuals go through a psychological crisis when they receive their diagnosis.

This study also has limitations. First, we were unable to draw any conclusions about causal relationships; although an advantage of this study is that it shows that certain subgroups seem to be more vulnerable than others. Second, we lacked information on environmental risk factors for rheumatic diseases and psychiatric disorders. Third, we had no data on psychological distress and stressful life events because this study was based on the entire population. Fourth, data on outpatients are not available in nationwide registers, such as the registers used in this study. This implies that only the most severe cases were included in the present study, ie, those requiring hospitalization. The “true” incidence rates in the population were thus underestimated. Fifth, some associations may not have been detected because of insufficient numbers of cases in certain diagnostic subgroups, especially among men with SLE. Sixth, data on medication were not available to us, which is a limitation because psychosis is sometimes induced by corticosteroids. Finally, we were not able to test for the validity of the diagnoses because our data were based on the entire population. For example, there is an interobserver variability in clinical settings. In addition, diagnostic criteria were changed several times during the follow-up. However, we only used main diagnoses recorded in the hospital registers, which partly increases the probability that the diagnoses are valid. In addition, 61% of RA patients had more than 1 hospitalization for RA.

The results of previous studies are consistent with those of the present study. For example, the purpose of a recent case report was to remind health care providers of the strong association between depression and chronic medical illness, such as RA.²¹ A cohort study observed 537 patients from Brazil and showed that acute psychosis is related to SLE.²² A recent review study from Belgrade in Serbia and Montenegro found that psychiatric abnormalities are common in SLE with a prevalence of 17% to 75%.²³ Other studies have also demonstrated the association between SLE and psychiatric disorders.^{4 ,8 - 10 ,24 - 27} A cross-sectional study of 82 patients with RA found a significantly increased prevalence of anxiety and depression compared with the control group.²⁷ Other studies of patients with RA have confirmed these findings.^{28 - 31} Previous studies of the association between AS and psychiatric disorders are rare, although some studies based on questionnaires and depression scores have shown that the psychological status of patients with AS is affected.^{14 ,32} The findings of an increased risk of dementia and delirium among patients with SLE suggest a primary involvement of the brain.

To sum up, this study adds further knowledge to the literature because of its novel approach; it is the first study of an entire population examining the association between 3 different rheumatic diseases and the most common neuropsychiatric disorders in men and women.

Patients with rheumatic diseases have a higher risk of hospitalization for psychiatric disorders than the general population, and patients with SLE have a higher risk of hospitalization for dementia and delirium. Differences in risk magnitudes were found by subtype of the rheumatic disease and psychiatric disorder, by sex, and by various follow-up intervals. These results show that health care providers who encounter patients with rheumatic diseases should be aware that these patients are more likely to develop severe neuropsychiatric disorders than the general population and that some subgroups seem to be more vulnerable than others.

Correspondence: Kristina Sundquist, MD, PhD, Center for Family and Community Medicine, Karolinska Institute, Alfred Nobels alle 12, SE-14183, Huddinge, Sweden (kristina.sundquist@ki.se).

Submitted for Publication: July 27, 2007; final revision received October 31, 2007; accepted November 29, 2007.

Author Contributions: All authors take responsibility for the integrity of the data and the accuracy of the data analysis. All authors had full access to all the data in the study.

Financial Disclosure: None reported.

Funding/Support: This study was supported by grant K2005-27X-15428-01A from the Swedish Research Council (Dr K. Sundquist), ALF projektmedel, and the Stockholm County Council.

Additional Contributions: We thank Leena Maria Johansson, MD, PhD, at the Center for Family and Community Medicine, Karolinska Institute, and Jehns Christian Martinéus, MD, at the Rheumatologic Clinic, University Hospital, Lund, Sweden.