Introduction

The prevalence of mental disorders is high and contributes significantly to the burden of disease globally [1]. With the escalation of socioeconomic development and population fluctuation, the prevalence of mental disorders increases annually [2]. In 2016, mental disorders accounted for 32.4% of Years Lived with Disability (YLDs) and 13.0% of disability-adjusted life-years (DALYs) globally [3]. As a matter of fact, mental disorders contribute to the suffering of patients and burden societies and families in general [4]. Mental disorders are also prevalent in China. Approximately 15% of China’s total health expenditures and 1.1% of its gross domestic product were spent on mental disorders [5]. According to the China Mental Health Survey (CMHS) conducted in 2013, the weighted lifetime prevalence of any mental disorder excluding dementia, was 16.6% (95% CI: 13.0-20.2), and the weighted 12-month prevalence was 9.3% (95% CI: 5.4-13.3). Anxiety disorders had the highest 12-month prevalence of all mental disorders at 5.0%, followed by mood disorders at 4.1% and substance use disorders at 1.9%. Therefore, anxiety disorders, mood disorders, and substance use disorders are regarded as common mental disorders in China [6].

Functional disability resulting from mental disorders is on the rise [7]. The percentage of people with mental health disabilities increased from 2.0% in 1997-1999 to 2.7% in 2007-2009 [8]. According to the International Classification of Functioning, Disability and Health Framework (ICF), disability is defined as a personal dysfunction, including impairments, activity limitations, and participation restrictions [9,10,11]. There is an unmet need for effective treatment of mental disorders due to the early age of symptom onset, long tendency to relapse, and long duration [12,13,14]. This may result in a long-term functional disability in patients. For example, a four-year longitudinal study in the Netherlands evaluated the long-term disability of anxiety disorder patients and found that patients with any anxiety disorder had higher levels of long-term disability than healthy controls [15]. Another study of long-term disability in depressive disorder showed that patients reported more severe work disabilities on all measures than healthy controls [16].

However, the reported studies mainly focus on specific categories of mental disorders, leaving gaps in comprehensive studies of the common mental disorders. Moreover, there are few longitudinal follow-up studies in China that discuss the relationship between mental disorders and long-term disability [15, 17]. Furthermore, the current researches have small sample sizes and are slightly less representative [18, 19]. In order to monitor long-term disability, a follow-up study was conducted 5 years after the CMHS completed. Long-term disability in this study was defined as functional disability at 5 years. The purpose of our study was to investigate the extent of long-term disability in common mental disorders and to interpret the relationship between common mental disorders and long-term disability.

Materials & methods

Sample and procedure

Sample

The CMHS is the first nationally representative community survey on mental disorders and mental health services in China. The World Health Organization (WHO) Composite International Diagnostic Interview, Version 3.0 (CIDI-3.0), was used to interview 28,140 community individuals aged 18 and up in the CMHS by trained lay interviewers. Full methodologic details of the CMHS have been published [6, 20, 21]. From April to June 2018, all CMHS participants were subjected to a follow-up study using a computer assisted telephone interview (CATI) system [22, 23]. Individuals diagnosed with mood disorders, anxiety disorders, or substance-use disorders in the CMHS were included in this study, as well those without other mental disorders. Details can be found in Fig. 1.

Fig. 1
figure 1

Flowchart for the participants of CMHS and follow-up study

Full size image

Procedure

This study utilized the CATI system to interview all participants. A total of fifty interviewers attended a two-day training on the use of CATI, the study procedure, telephone interviewing skills, questionnaire content, quality control methods, and passed the final examinations. In addition, informants were interviewed if participants rejected, were unable to, or died during the follow-up period. Please refer to the published article for details on whether alternative answers are allowed [24]. The informant should be at least 18 years old and well-versed in the participant’s situation.

As well as the pre-survey preparation, various aspects of quality control were considered and designed during the formal stage of the questionnaire survey and the post-survey data verification stage to ensure the results were as accurate, usable, and representative as possible. The CATI system made it simpler to get para data during the interviews. Daily data checks were performed on all interviews to detect systematic errors and their distribution patterns, focusing on non-response, stoppage of interviews, and contact attempt times. Audio checks were performed on around 10% of respondents to uncover interviewer problems such as data entry errors, erroneous informant information, inadequate probing, and irregular behaviors while asking a question. Monitoring checks were performed on at least three telephone calls by each interviewer to ensure that the interviewers acted appropriately during the interviews. Further quality control details were published [24].

The study was authorized by the Ethical Committee of Tianjin Anding Hospital and the Sixth Hospital of Peking University. Prior to their involvement in the study, all participants and informants were given oral informed consents, and the consent procedure was recorded to audio files.

Materials

Diagnose of common mental disorders

Past 12-month diagnoses of mental disorders were assessed with the CIDI-3.0, which classifies diagnoses according to the American Psychiatric Association’s Diagnostic and Statistical Manual, Edition IV (DSM-IV) criteria. The CIDI is a fully-structured diagnostic interview conducted by trained lay interviewers [22]. The CIDI has been proven to be valid and is used in China [6] and many other countries for epidemiological surveys [25].

The diagnosis of common mental disorders in this study was assessed according to the criteria of DSM-IV [21], and divided into three categories: (1) anxiety disorders [panic disorder (PD), specific phobia (SP), social phobia (SO), obsessive-compulsive disorder (OCD), generalized anxiety disorder (GAD)]; (2) mood disorders [major depressive disorder (MDD), dysthymic disorder (DD), and bipolar disorder (BD)]; and (3) substance-use disorders (alcohol use disorder (AUD)).

Assessment for long-term disability

At both the baseline and this 5-year follow-up, the functional disability of participant was assessed using the revised version of the WHO Disability Assessment Schedule (WHODAS-2.0), based on the International Classification of Functions [26]. The internal consistency and validity of the World Mental Health Survey WHODAS-2.0 have been demonstrated [27]. The Chinese version of the WHODAS-2.0 has good internal consistency reliability and empirical validity, and the Cronbach’s α was 0.73-0.99 [28]. WHODAS-2.0 includes six domains: 1) cognition – understanding and communicating; 2) mobility – moving and getting around; 3) self-care – attending to one’s hygiene, dressing, eating and staying alone; 4) getting along – interacting with other people; 5) life activities – domestic responsibilities, leisure, work and school; 6) participation – joining in community activities, participating in society.

Based on the distributional features of the instrument, the global disability score was dichotomized at the 90th percentile to reflect the presence or absence of considerable disability, as recommended by Von Korff et al. [27].

Contact information

Baseline and follow-up contact information was collected for tracking purposes, including the address, plans to move to a new home, and contact information for participants and informants.

Covariates

Covariates were used to adjust for possible differences in disability experience, including sex, age, participant/proxy, residence area, marital status, and education. The WHODAS category at baseline was also adjusted as a covariate, since disability at baseline is related to long-term disability.

Statistical analysis

Sampling design weights, non-response adjustment weights, post-stratification adjustment, and trimming of the weights were used when calculating prevalence and the risk ratios (RRs). The methods of weighting were similar as the way to those used in the CMHS, which have been published elsewhere [21]. This study differed from the CMHS in terms of weights primarily because it used a different method for calculating non-response weights, which was also adjusted according to the diagnosis of mental disorder [24].

Frequencies and chi-square (χ2) tests were used for descriptive analysis and comparisons of rates in different categories. The RRs for common mental disorders-related disability were estimated using Poisson regression. Throughout the study, a significance threshold of 0.05 was applied. SAS 9.4 and R 4.0.2 were used for statistical analysis.

Results

General sociodemographic characteristics

During the follow-up study, interviewers attempted to contact participants who had valid telephone numbers for themselves or their informants. The final weighted sample was 6269. Overall, 5435 (86.3%) participants completed the questionnaires themselves, while 864 (13.7%) were informants who completed proxy interviews. A total of 51.01% (n = 3198) of the 6269 participants interviewed were males, while 48.99% (n = 3071) were females. As for education levels, illiteracy or below primary school was the highest (19.94%, n = 1246), whereas university or above was the lowest (9.95%, n = 622). In terms of marital status, 5406 married (86.25%) and 862 single (13.75%) participants were present. Table 1 presents the general sociodemographic characteristics of the sample.

Table 1 General sociodemographic characteristics
Full size table

30-day prevalence of disability of common mental disorders

In this study, 15.41% (86/559) of patients with common mental disorders met the disability criteria (95% CI: 9.67-21.15). In China, mood disorders were the leading cause of disability. 33.27% (58/175) of participants with any mood disorder reported disability (95% CI: 19.94-46.60). The disability rates for MDD and DD were 32.17 and 43.94%, respectively. This rate was 13.74% (35/256, 95% CI: 8.09-19.39) among those with any anxiety disorder, with GAD (29.63%) being the most prevalent. 7.62% (11/139) of those with any substance-use disorder (95% CI: 0.99-1.75). Table 2 displays the specific 30-day prevalence of disability.

Table 2 30-day prevalence of disability
Full size table

Association between disability and common mental disorders

The results indicate that after adjusting for sex, age, participant/proxy, residence area, marital status, education, and the WHODAS category at baseline, the relationship between common mental disorders and long-term disability was: the association between any mood disorder and long-term disability was significant, with participants with any mood disorder 1.94 (95% CI: 1.65-2.27) times more likely to report disability than participants without mood disorders. The relationship between any anxiety disorder and any substance use disorder and long-term disability was not significant. DD, MDD, GAD, OCD, and AUD were statistically significantly associated with long-term disability. The most disabling condition was DD (RR: 2.40; 95% CI:1.87-3.03). Table 3 shows the relationship between disability and common mental disorders.

Table 3 Association of common mental disorders with disability
Full size table

Discussion

This is the first study to explore long-term disability among Chinese people suffering from common mental disorders in a comprehensive way. In the CMHS 5-year follow-up study, the weighted disability rate for patients with mental disorders increased from 9% to 15.41%. The main finding of this study was that any mood disorders and long-term disability were significantly correlated, with DD and MDD being particularly influential. In addition, GAD, OCD, and AUD were all strongly associated with long-term disability. After 5 years, approximately 7.62 to 43.94% of patients with common mental disorders became disabled. The findings of this study may be valuable for policymakers in determining whether the current mental health care system can meet demand, rationalize resource allocation, and improve the long-term prognosis of people with mental disorders.

According to the present study, DD was the most disabling condition. Patients with DD were 2.4 times more likely to be disabled than health controls. This is similar to previous research. Sandhu et al. pointed out that the chronic nature of DD may result in greater functional impairment than acute depression [29]. Bernhard et al. noted that DD often co-occurs with other more severe medical comorbidities, resulting in higher long-term disability than MDD [30, 31]. Hellerstein et al. observed that DD sufferers had difficulty working full-time, which increased their susceptibility to long-term disability in terms of function [32]. In addition, a diagnosis of DD, requires 2 years of symptoms (DSM-IV), whereas a diagnosis of MDD requires only 2 weeks of symptoms. No distinct episodes and long duration are the two defining characteristics of DD [31, 33]. Moreover, the delayed treatment rate for DD is high, some people may seek help decades after the onset of symptoms [34,35,36], but DD is treatable to alleviate symptoms [37]. DD is a recurrent long-term depressive disorder with ill-defined episodes, high rates of delayed treatment, and high rates of co-morbidity, and these factors may further contribute to long-term disability [30, 31, 38, 39]. Similar to the findings of previous research, MDD was also associated with long-term disability [40,41,42]. MDD is associated with substantial disability across many domains of life [43], affecting patients’ ability to fulfill family roles, work, and participate in society at levels similar to or worse than those reported in chronic somatic diseases [44].

In our study, BD did not disclose any significant outcomes [45,46,47,48]. This finding was not in line with previous research. According to Antunes et al., there existed an association between BD and disability, but their study was cross-sectional and cannot describe the long-term disability of patients [45]. Martinez-Aran et al. concluded that BD was associated with functional disability, however, they used a clinical population as opposed to ours [48]. According to the report [49], the number of patients with BD in community was lower than in institution due to the fact that patients with BD are more likely to require hospitalization. As a result of the small number of patients with BD that participated in this study, we may have lacked statistical power to obtain significant effects [6].

In addition to mood disorders, GAD, OCD and AUD showed statistically significant associations with disability. This was similar to previous studies. These disorders are associated with long-term disability more often than not due to the symptomatic features of the disease itself, and they rarely receive formal systemic treatment [15, 50,51,52,53,54,55,56,57]. In this study, people with GAD, OCD, MDD, or AUD were nearly twice as likely to report a disability as people without these disorders. Therefore, while increasing public medical and health investment, the guarantee of long-term medical and health services for patients with mental disorders should be the focus issue.

Limitations

There are several limitations in the research. First, long-term disability was assessed only 30 days prior to follow-up, whereas mental disorders were 12-month based. For episodic conditions, the past month of disability may not include the period of the disorder. A 12-month diagnosis also allowed for the inclusion of remitted disorders that might have residual adverse effects on disability [56]. Second, the number of certain diseases patients followed up in this study was relatively small, which may be biased to some extent in the representativeness of the sample. Nevertheless, for the follow-up study, we adopted complex weighting and non-response weighting, as well as multiple regression analysis to adjust for bias due to the different distributions of participants who were retained and those who were lost-to-follow-up. Third, since the skip strategy used in the follow-up process and the use of informants as responses if participants were unable to respond individually, it is possible that participant disability was underestimated. Even so, considering the difficulties of conducting an interview with CATI design, such tactics might be the most effective way to increase the response rate.

Conclusions

In the Chinese community, patients with common mental disorders were at an increased risk of long-term functional disability. To reduce the risk of long-term disability, it is necessary to pay additional attention to the long-term functional status of patients with DD, MDD, GAD, OCD, and AUD. Furthermore, in the future, mental health resources should be more inclined toward these disorders in order to improve their quality of life and reduce the disability rate.