Mental Morbidities and Chronic Fatigue in Severe Acute Respiratory Syndrome Survivors:  Long-term Follow-up FREE

The world was struck by a severe acute respiratory syndrome (SARS) pandemic in 2003, resulting in considerable morbidities and mortalities. A substantial proportion of SARS survivors experienced SARS-related physical complications, including avascular necrosis (AVN),¹ pulmonary fibrosis,² and hypocortisolism.³ The physical conditions of most of the patients continuously improved in the first year of follow-up, but their mental conditions showed limited signs of improvement.⁴ The 1-year follow-up study after the outbreak found that some SARS survivors had persistent mental problems.^4- 5 Many studies found that the status of being a health care worker at the time of SARS infection,^5- 10 the severity of the infection, and the dosage of steroid usage were related to the short-term mental outcomes. However, these studies were limited by the relatively modest sample sizes and the reliance on self-reported questionnaires. In addition, prominent fatigue was commonly reported among SARS survivors, but its magnitude was seldom investigated. In this regard, there was a dearth of long-term follow-up data regarding the mental sequelae and the impact of fatigue among these patients. We investigated the long-term aftermath of SARS survivors, focusing on psychiatric morbidities, chronic fatigue, and associated factors.

When SARS affected the whole territory of Hong Kong, 1755 individuals were infected, and the hospitals of our local region (New Territories East Cluster) treated about one-quarter of the SARS patients. All of these Chinese survivors with confirmed SARS infection were invited to participate in the study. Those with known neurological or pituitary diseases were excluded. The data were collected from December 2005 through July 2007. Personalized letters were sent to all eligible individuals and included an introduction explaining the research aim and method. Subsequent telephone contacts were made for invitation to participate in the study and arrangement of the interviews.

Patients who agreed to participate in the study were asked to complete a set of questionnaires and to undergo a semistructured clinical interview by trained clinicians. The questionnaires included detailed demographics, background history, and psychometric scales, including the Chinese versions of the Hospital Anxiety and Depression Scale (HADS)¹¹ and Impact of Event Scale–Revised (CIES-R).¹² The HADS cutoff was set at 16, for suggesting potential anxiety and depressive problems as suggested by our data at the early and recovery phases.¹³ In addition, a set of self-constructed questions was also included to document the perceived impairment in various aspects, including the perception of stigmatization and medicolegal issues. The degree of social stigmatization was assessed by asking if the participants perceived any stigmatization because of their SARS infection, with a grading of 1 (no stigmatization) to 4 (always perceived stigmatization). Psychiatric diagnosis was determined by a validated Chinese version of the Semistructured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) (SCID)¹⁴ by trained interviewers. They were also assessed for the presence of chronic fatigue by using (1) the modified criteria for chronic fatigue syndrome (CFS) according to the Centers for Disease Control and Prevention (CDC) (1994 version),¹⁵ in which the exclusion criteria for psychiatric disorders were omitted, and (2) a self- reported Chinese version of the Chalder Fatigue Questionnaire¹⁶ (the questionnaire was translated from English to Chinese then back from Chinese to English). Because there were a number of batteries and assessments to be tested, the logistics of the study were divided into 3 parts: (1) general questionnaires that comprised most of the questionnaires described herein, (2) clinical interviews for the SCID, and (3) chronic fatigue assessment. These 3 parts were completed within a month if possible once the individual consented to participate in the study. This study was approved by the local ethics review committee.

The patients were classified into 2 groups: survivors with psychiatric illnesses (hereinafter, psychiatric group) and survivors without psychiatric illnesses (hereinafter, nonpsychiatric group). The groups' comparisons on continuous variables were analyzed by t test while categorical variables were analyzed by χ² test. Fisher exact test was used when the expected count of categorical variables was smaller than 5. To ascertain the associated factors for psychiatric morbidities and chronic fatigue, respectively, the odds ratio (OR) was determined by adjusting the duration of follow-up with logistic regression analysis of backward stepwise procedure in separate models. The cutoff point of the selected variables for logistic regression was fixed at P < .10. The level of statistical significance was set at P  = .05. Statistical analysis was performed by using SPSS software (version 17.0; SPSS Inc, Chicago, Illinois).

There were 410 SARS survivors who had received treatment in the local hospitals of our regional cluster. Compared with the whole population of patients with SARS in Hong Kong, our study participants had no significant difference in terms of sex ratio and median age (whole population vs our cluster of participants, respectively: male sex, 44.3% vs 42.1% [P  = .43]; median age, 40 years vs 41 years [P  = .35]). Eight persons were excluded owing to subsequent confirmation that they had not been infected by SARS. Twenty-eight people had died, and 5 were excluded because they had emigrated (n = 3), were non- Chinese (n = 1), or had been diagnosed as having pituitary disease (n = 1). In total, 369 SARS survivors were eligible for the study. Among them, 74 SARS survivors had been lost to follow-up, and 62 individuals refused to participate (Figure). Eventually, 233 survivors were successfully recruited to the study, giving an overall response rate of 63.1%. Among these participants, 181 (77.7%) participated in the clinical interview, 170 (73.0%) completed the questionnaires, and 146 (62.7%) completed chronic fatigue assessment. A total of 124 (53.2%) completed all 3 parts of the study, including the clinical interview, questionnaires, and chronic fatigue assessment.

The mean (SD) age of the respondents was 43.3 (13.7) years, and 70.4% of the individuals were female. Most of them were married (65.3%), and 40.0% had at least a tertiary education; 68.4% of them were working at the time of the study, while unemployment, retirement, and being a housewife accounted for 8.4%, 8.4%, and 14.9%, respectively. Compared with the nonrespondents, respondents were more likely to be female, but the 2 groups did not differ by age and SARS-related acute treatment (including steroid treatment and intensive care unit [ICU] admission) (Table 1).

The mean duration of post-SARS follow-up was 41.3 months (range, 31-51 months) after the SARS epidemic. Comparing the psychiatric and nonpsychiatric groups, the nonpsychiatric group had a slightly longer duration of follow-up study (psychiatric group: mean duration of follow-up, 39.6 [6.8] months; nonpsychiatric group: 43.0 [5.9] months; P < .001). Those with chronic fatigue symptoms responded to the study earlier than those without (chronic fatigue group: mean duration of follow-up, 37.8 [5.4] months; non–chronic fatigue group, 42.1 [5.9]; P < .001).

Among the 181 individuals who participated in clinical interviews at follow-up, 6 (3.3%) had a history of psychiatric disorders before contracting SARS. At the time of follow-up, a total of 77 (42.5%) had experienced at least 1 active psychiatric illness as determined by the SCID. The most common diagnoses were posttraumatic stress disorder (42 of 77 survivors [54.5%]), depression (30 of 77 [39.0%]), somatoform pain disorder (28 of 77 [36.4%]), panic disorder (25 of 77 [32.5%]), and obsessive compulsive disorder (12 of 77 [15.6%]).

Comparing survivors with psychiatric morbidities vs those without, there were no significant differences in sex, socioeconomic factors, severity of the illness during acute infection, and physical comorbidity or complications like AVN (Table 2). There was impairment of the ability to work among those in the psychiatric group, as suggested by the higher percentage of individuals having no gainful employment at follow-up. The work impairment was also found among health care workers. Health care workers who had been ill with SARS and had psychiatric morbidities were more likely to quit their medical-related work after SARS. Overall, the psychiatric group had higher percentages of survivors involved in SARS-related litigation and application to the SARS survivors' trust fund, which was set up by the local government to compensate for the physical and mental dysfunctions of patients with SARS. Those with psychiatric morbidities perceived more social stigmatization. Logistic regression was performed, controlling for the duration of follow-up and a history of psychiatric disorders. Results suggested that being a health care worker during the SARS epidemic (odds ratio [OR], 3.24; 95% confidence interval [CI], 1.12-9.39; P = .03), being unemployed (including being a housewife or retired) at follow-up (OR, 4.71; 95% CI, 1.50-14.78; P = .008), having the perception of social stigmatization (OR, 3.03; 95% CI, 1.20-7.60; P = .02), and having applied to the SARS survivors' fund (OR, 2.92; 95% CI, 1.18-7.22; P = .02) were associated with an increased risk of psychiatric morbidities at follow-up.

The mean (SD) CIES-R scores of the study cohort were as follows: intrusion domain, 1.66 (1.00); avoidance domain, 1.27 (0.97); and hyperarousal domain, 1.76 (1.04). Sixty-five participants (27.9%) met the cutoff criteria for the intrusion domain; 41 (17.6%) met the cutoff criteria for the avoidance domain; 78 (33.5%) met the cutoff criteria for the hyperarousal domain. The mean (SD) scores for the HADS domains were 7.4 (4.5) for the depression domain and 7.6 (4.4) for the anxiety domain, and 83 participants (35.6%) surpassed the cutoff criteria for HADS.

Chronic fatigue was found to be common among both psychiatric and nonpsychiatric groups. The prevalence rate according to the Chalder fatigue questionnaires (chronic fatigue score ≥4 and symptoms lasting for >6 months) and the modified CDC 1994 criteria¹⁵ for CFS were 40.3% and 27.1%, respectively. Those with fatigue symptoms were more likely to have comorbid active psychiatric disorders (Table 3).

Most sociodemographic and medical variables (including the use of steroid replacement therapy, severity of SARS, and medical complications like AVN) were not associated with the development of fatigue (Table 3). Logistic regression showed that a shorter duration of follow-up study (OR, 0.89; 95% CI, 0.83-0.96; P = .003) and application to the SARS survivors' fund (OR, 2.64; 95% CI, 1.07-6.51; P = .04) were associated with a chronic fatigue problem in SARS survivors.

To our knowledge, this study is one of the largest long-term studies of the mental comorbidities of survivors in the aftermath of the SARS epidemic. The results demonstrated that the rates of psychiatric morbidities, chronic fatigue, and resultant functional disabilities were persistently high and clinically significant. The persistence of psychiatric morbidities among the SARS survivors who participated in our study was alarming. Reviewing the results from previous post-SARS cohorts, the psychiatric morbidities measured by standardized questionnaires ranged from 10% to 35% in the acute phase of the infection (acute phase to 1 month after SARS)^9,17 to 64% at the 1-year follow-up.⁵ Although the prevalence rate of psychiatric disorders in these studies varied with the stringency of the applied psychiatric tools, it showed a trend of progressive rise after the event, and our data further suggest that the trend continued as time passed.

The strong association with occupation as a health care worker at the time of infection¹⁷ was replicated in our study. We also found that those hospital workers having psychiatric morbidities were more likely to give up their employment as medical care workers after infection with SARS. It suggested that the occupational risk among health care workers who face acute, life-threatening disaster should not be neglected, and the associated psychological and functional impairment can be considerable. Implementation of a health care policy that supports service for the emotional needs of ill health care workers is essential. Regarding long-term impairment factors, a lack of gainful employment, perceived social stigmatization, and application to the SARS survivors' fund were found to be closely associated with the persistent psychiatric morbidities in our studies. Although the causal relationships of these factors with psychiatric morbidities could be interactive and bidirectional, our findings suggested that poor functional rehabilitation and adaptation after SARS were major issues among SARS survivors. Perceived social stigmatization among SARS survivors could be present in both the acute and recovery phases. At the time of acute infection, the dreadful nature of the illness, its initial obscure pathologic characteristics, and the associated quarantine procedures could have resulted in disproportionate and undesirable labeling of the patients, even when their infection had cleared.¹⁸ After recovery, the residual physical symptoms of pain and fatigue are often regarded as vague and unsubstantiated. In addition, the functional decline and lengthy compensation and litigation processes associated with SARS infection might further increase others' skepticism about the survivors' illness and possibly the rejection of them. Alternatively or synergistically, the negative cognition related to psychiatric disorders may sensitize SARS survivors' interpersonal difficulties and hence amplify their perception of stigmatization.

Apart from psychiatric disorders, complaint of prominent chronic fatigue was common among the SARS survivors. The high prevalence of chronic fatigue problems in SARS survivors with both psychiatric and nonpsychiatric disorders suggested that psychiatric disorders per se did not fully account for the chronic fatigue problems. Our study found that application to the SARS survivors fund was associated with chronic fatigue, even after controlling for psychiatric morbidities. Instead of being a causative factor, the relationships could be bidirectional. On the one hand, application to the SARS survivors' fund partially reflects the degree of disability resulting from chronic fatigue symptoms, while on the other hand, involvement in lengthy SARS-related litigation might perpetuate the chronic fatigue symptoms.

Given the commonness of chronic fatigue symptoms among SARS survivors and its occurrence independently of the psychiatric morbidities, the etiology and pathophysiologic mechanisms should be investigated. In this regard, there has been postulation about the possible involvement of cortisol status and chronic fatigue.¹⁹ Among SARS survivors, most were treated with high doses of corticosteroids during the acute infection phase of the disease, and some of them continued to be treated with steroid replacement therapy during the recovery phase. Hypocortisolism was reported in almost 40% of SARS survivors 1 year after the infection.³ However, our study did not find any association of a chronic fatigue problem with the use of steroid treatment during the acute phase or at follow-up. Nonetheless, additional measurement of the subtle adrenal status and future workup, including dynamic hormonal challenge testing of these individuals, would be needed for the clarification of the role of cortisol status in the development of chronic fatigue among SARS survivors. Apart from the cortisol involvement, the role of disaster in mediating fatigue via immunological dysfunctions, including cytokine disturbances, has been suggested.^20- 21 During acute SARS infection, cytokine storm had been demonstrated,²⁰ and cytokine disturbance was closely associated with somatic and painful conditions.²² More research is needed to delineate the long-term association of endocrine status and cytokine dysfunctions with psychiatric disorders and chronic fatigue in the cohort of SARS survivors.

Some limitations of this study need to be addressed. First, the overall response rate was about 63%, and some participants did not complete all parts of the study. Nonetheless, the response rate of the current study was already higher than most of those in previous reports, in which the response rates were relatively low, mostly below 50%, for other SARS cohorts.^9,17,23 In addition, there might be some potential response bias as reflected by a higher percentage of female respondents. However, we attempted to minimize this bias by tracing the nonresponders, and the comparison of the demographics of the respondents and nonrespondents suggested a similar age and severity of SARS infection between the 2 groups. Also, there were some slight differences (a few months) in the duration of the follow-up study between the respondents and nonrespondents. We tried to control for this factor of differential duration of follow-up study in the regression analysis. Furthermore, we tried to limit the time frame for completion of the various parts of the study to 1 month for each individual. The other limitation of this study is that the information on medical complications was retrieved from medical records and reports from study participants. Additional physical workup, especially on hormonal profiles, would provide more information and directions for future studies.

The mental impact of SARS persisted and remained clinically significant among the survivors for up to 4 years of follow-up. Chronic fatigue symptoms and syndrome, not accounted for by the psychiatric morbidities, were also prevalent among them. Additional research work is needed on the delineation and clarification of the endocrine status and fatigue problems among SARS survivors. Optimization of the treatment of mental health morbidities by a multidisciplinary approach, especially targeting symptoms of fatigue, functional and occupational rehabilitation, and intensive psychiatric interventions, is also needed.

Because new infectious diseases are emerging at an unprecedented rate and pose a global threat for pandemics, there should be better preparation in public health strategies for dealing with both the acute phase of a disease and the long-term potential mental health complications related to the disastrous outbreaks of new infectious pathogens. Various channels to mental health services should be available to patients, health care workers, and the general public, not only during the acute phase of a disease but also the aftermath of an infectious outbreak.

Correspondence: Yun-Kwok Wing, FRCPsych, Department of Psychiatry, Shatin Hospital, Chinese University of Hong Kong, 33 Ah Kong Kok St, Shatin, New Territories, Hong Kong SAR, China (ykwing@cuhk.edu.hk).

Accepted for Publication: August 7, 2009.

Author Contributions: Dr Wing accepts full responsibility for the conduct of the study, had access to the data, and controlled the decision to publish. All of the authors contributed to, revised, and approved the final manuscript. Study concept and design: Wing, Leung, Ma, and Fong. Acquisition of data: Wing, Yu, Leung, Ma, Kong, Yee, Fong, and Lam. Analysis and interpretation of data: M. H.-B. Lam, Wing, Yu, Ma, Kong, and S.-P. Lam. Drafting of the manuscript: M. H.-B. Lam, Wing, Yu, Leung, and S.-P. Lam. Critical revision of the manuscript for important intellectual content: Wing, Yu, Ma, Kong, Yee, Fong, and S.-P. Lam. Statistical analysis: Wing and Yu. Obtained funding: Wing and Leung. Administrative, technical, and material support: Wing, Ma, Kong, Yee, Fong, and S.-P. Lam. Study supervision: Wing and Ma.

Financial Disclosure: None reported.

Funding/Support: This study was supported by Research Fund for the Control of Infectious Diseases (No. 03040062) from Hong Kong Research Fund Secretariat.