A community-based study of the relationship between somatic and psychological distress in Hong Kong

Abstract

Although the predominantly somatic presentation of distress has been used to explain low rates of emotional illnesses and health service use in Chinese communities, this concept of somatization has not been examined by concurrently studying the profile of somatically and psychologically distressed Chinese individuals. A random population-based sample of 3014 adults underwent a structured telephone interview that examined their sociodemographic characteristics, somatic distress (Patient Health Questionnaire-15, PHQ-15), non-specific psychological distress (Kessler Scale-6, K6), health service use, and functional impairment. Four groups of individuals identified by PHQ-15 and K6 cut-off scores were compared. Results showed that PHQ-15 and K6 scores were positively correlated. The large majority of respondents (85.9%) reported both somatic and psychological distress. The proportions of Low Distress Group, Somatically Distressed Group, Psychologically Distressed Group, and Mixed Distress Group were 69.2%, 5.0%, 15.8%, and 10.0%, respectively. Specific age range, male gender, greater family income, higher education level, and retirement were associated with decreased odds of somatic and/or psychological distress. Although psychological distress best predicted impairment, somatic distress best predicted health service use. Mixed distress predicted most impairment and health service use. Thus, psychological distress and somatic distress commonly coexist across Chinese sociodemographic groups. This speaks against the conventional notion of somatization and is consistent with recent findings of a higher prevalence of emotional illnesses in Chinese people. That psychologically distressed individuals are more impaired but less inclined to seek help than somatically distressed individuals may partly explain low levels of help-seeking for mental disorders found in epidemiological studies.

Keywords

Chinese help-seeking impairment psychological distress somatic symptoms somatization

Introduction

Somatization is a multi-faceted construct (Kirmayer & Young, 1998) that can be studied across as well as within cultural groups. Recent cross-national studies of group differences in somatic and psychological symptoms reporting indicated that the somatization effect can be understood as a relative difference between cultural groups (Parker, Cheah, & Roy, 2001; Ryder et al., 2008; Yen, Robins, & Lin, 2000). Thus, Ryder and Chentsova-Dutton (2012) noted that Chinese somatization was a matter of symptom emphasis, in that Chinese people were more likely to acknowledge somatic symptoms than their Western counterparts for various cultural reasons. Within the Chinese cultural group, somatization has also been understood as the predominantly or exclusively somatic presentation of distress at the level of individuals (Kirmayer & Robbins, 1991). With respect to this approach, there has been an enduring belief that Chinese people are particularly prone to exhibiting somatic distress and denying psychological distress (Hsu & Folstein, 1997; Kleinman, 1982). This notion of somatization has been routinely used to explain the community and clinical epidemiological findings of low rates of common emotional illnesses, such as depression, in Chinese people (Guo, Tsang, Li, & Lee, 2011; Hsu & Folstein, 1997). However, the empirical support for this explanation is mixed.

Various studies across the decades have shown that somatic symptoms were common among Chinese patients with mental health problems in primary and tertiary care settings in Chinese communities. For example, 70%–88% of psychiatric outpatients initially presented with somatic complaints in the absence of dysphoric affect (Kleinman, 1977; Tseng, 1975). High rates of a reportedly culture-specific somatic syndrome of neurasthenia were also found in clinical (Lee et al., 2000) and community studies (Zhang, 1989). These findings suggest that the somatic expression of distress is dominant in Chinese societies. This somatic emphasis is believed to contribute to very low rates of help-seeking for emotional illnesses (Lee et al., 2010).

Nonetheless, there is evidence against this notion of Chinese somatization. The somatic presentation of distress has repeatedly been found to be common in a variety of both Western and non-Western communities (Kirmayer & Young, 1998; Simon, VonKorff, Piccinelli, Fullerton, & Ormel, 1999; Zhou et al., 2011). Although the use of restrictive samples and dissimilar measurement tools has made it difficult to compare somatization cross-nationally, a recent population-based study (Lee, Ma, & Tsang, 2011) suggested that the prevalence and profile of 15 kinds of commonly experienced somatic distress in the general population of Hong Kong were mostly similar to those of their Western counterparts (Kroenke, Spitzer, & Williams, 2002). Moreover, recent studies have found, by and large, comparable rates of depression and anxiety disorders in Chinese and Western communities (Hu et al., 2009; Lee, Tsang, & Kwok, 2007; Phillips et al., 2009). This suggests that psychological distress may be more common in Chinese groups than previously thought.

One major methodological limitation of the previous studies on Chinese somatization is that they did not concurrently measure somatic and psychological distress at the level of individuals. These studies therefore did not directly demonstrate that the presence of somatic distress was accompanied by a relative absence of psychological distress (Isaac et al., 1995; Zhang, 1989). Likewise, recent studies on psychological distress did not examine how emotionally distressed Chinese people might experience somatic distress (Lee et al., 2007; Phillips et al., 2009). This is undesirable because somatic distress may only be the initial mode of expressing distress among Chinese people. Moreover, psychological distress in Chinese people might be readily reported when facilitative methods of symptom elicitation are used (Guo et al., 2011; Kleinman, 1982; Yeung, Chang, Gresham, Nierenberg, & Fava, 2004). At this stage, the relationship between mode of distress expression and low rates of help-seeking for emotional illnesses in Chinese people remains unclear.

The evidence briefly reviewed above argues against a simple picture of either somatization or psychologization in Chinese people. It suggests a more complex picture in which somatic and psychological distress may lie on a continuum and coexist in different degrees across groups of individuals. Although previous cross-national primary care studies suggested that a mixed presentation of somatic and psychological symptoms was common (Simon, Gater, Kisely, & Piccinelli, 1996; Simon et al., 1999), the extent to which Chinese people in Chinese community settings somatize distress is unknown. Those studies typically oversampled somatically distressed individuals by using outpatients from primary and tertiary psychiatric settings. Such clinical samples were often biased by help-seeking behavior. For example, higher-income, educated and younger people were more likely to seek psychiatric treatment (Wang et al., 2005). There was one community study on somatization in which Mak and Zane (2004) found that the reporting of somatic symptoms by Chinese Americans was comparable to that of Russian immigrants in Israel. Nonetheless, since people's tendency for somatization could be modified by their having lived in a Western community (Parker, Chan, Tully, & Eisenbruch, 2005), the findings of this community-based study of Chinese people living in the United States may not be generalizable to Chinese people living in Chinese communities.

The present study aims to examine both somatic and non-specific psychological distress in a Chinese community sample. We delineated four admittedly arbitrary groups of Chinese individuals with regard to distress expression as follows: 1) somatically distressed, 2) psychologically distressed, 3) both somatically and psychologically distressed, and 4) healthy persons with low somatic and psychological distress. Using tools specific for measuring somatic and psychological distress respectively, we aimed to examine the relative prevalence and correlates of these four hypothesized groups including their relationships with impairment and health service use.

Method

Sampling

This study was approved by the research ethics committee of The Chinese University of Hong Kong. A random telephone survey of the general population aged 15–65 years was conducted between September 2 and 22, 2009. A total of 11,120 calls had successfully established contact with the household, with 1625 calls having no interviewee aged between 15 and 65 years, 4004 calls being hung up immediately by receivers, and 2477 calls being rejected for an interview. Thus, 3014 telephone interviews were successfully completed with informed verbal consent. Of the households that were successfully contacted and had interviewees within the age range, the participation rate was 54.9% (3014/[3014 + 2477] × 100%) in accordance with the recommendation for reporting response rate in telephone surveys (Johnson & Owens, 2003). Informed verbal consent was obtained prior to each interview. The sample was weighted according to the gender distribution of different age groups in Hong Kong as reported by the Census and Statistics Department of the Hong Kong Government. With a 95% confidence level, the maximum sampling error was ±1.76%.

Instruments

An independent survey research organization, the Hong Kong Institute of Asia-Pacific Studies of The Chinese University of Hong Kong, was commissioned to conduct the survey. The interviewers were university students with experience in administering telephone survey interviews, including those on mental health issues. The interview was conducted in Cantonese Chinese, the predominant dialect used in Hong Kong. Besides sociodemographic information, the questionnaire included Chinese versions of the Patient Health Questionnaire-15, the Kessler Scale-6, Sheehan Disability Scale, and items probing the frequency of health service use.

Patient Health Questionnaire-15 (PHQ-15)

This questionnaire (Cantonese Chinese version) was used to measure the somatic distress of respondents. The respondents were asked to rate how much they have been bothered by each of the 15 somatic symptoms in the past month on a “0” (not bothered at all) to “2” (bothered a lot) scale, yielding a total score of 0 to 30 for females and 0 to 28 for males (Kroenke, 2007). While the PHQ-15 is a continuous measure of somatic distress, respondents can also be categorized into four grades of severity: minimal (score = 0–4), mild (score = 5–9), moderate (score = 10–14), and severe (score = 15–30) (Kroenke et al., 2002). Using the same cut-offs, a study that used the same sample of respondents as the present study reported that the level of impairment and health service use increased with each increasing level of somatic distress. That study supported the reliability and validity of the Chinese PHQ-15 in our sample (Lee et al., 2011). The Cronbach's alpha of the PHQ-15 was .79 in this study.

Kessler Scale-6 (K6)

The K6 has been widely used for measuring non-specific psychological distress. It is sensitive to both mood and anxiety disorders as stipulated in the Diagnostic and Statistical Manual of Mental Disorders, Fourth edition, Text Revision (DSM-IV-TR; American Psychiatric Association [APA], 2000) (Kessler et al., 2002). The Cantonese Chinese version (Lee et al., 2012) was used in this study. It consists of six questions that inquire how often the respondents feel (i) nervous, (ii) hopeless, (iii) restless or fidgety, (iv) so depressed that nothing could cheer you up, (v) that everything was an effort, and (vi) worthless in the past month. Response options included the whole 30 days (4), most of the time (3), some of the time (2), a little of the time (1), and none of the time (0), yielding a score range of 0–24. A score of 13 or higher indicates high psychological distress, a score of 8 to 12 indicates moderate psychological distress, and a score of 0 to 7 indicates low psychological distress (Wang et al., 2007). The internal consistency, reliability, and validity of this Cantonese Chinese version of K6 have been established in the same sample of respondents who participated in the present study (Lee et al., 2012). The Cronbach's alpha of the K6 was .76 in this study.

Sheehan Disability Scale (SDS)

Impairment in four domains of life in the previous four weeks was assessed using the SDS. The four domains include household responsibilities (“doing housework, such as cleaning and grocery shopping”), work/school (“the ability to work, such as working, studying, or taking exams”), close relationships (“the ability to form and maintain close relationships with other people, such as romantic partner, family members, or close friends”), and social life (“the ability to form social relationship”). Severity was rated on a scale of 0–10 (none [0], mild [1–3], moderate [4–6], severe [7–9] and very severe [10]) (Leon, Olfson, Portera, Farber, & Sheehan, 1997). The SDS has been widely used in both Western and Chinese community psychiatric surveys (Lee et al., 2011). The Cronbach's alpha of the SDS was .86 in this study.

Frequency of health service use

The question “in the previous year, how often on average did you seek help from health professionals?” was asked to assess frequency of health service use. Respondents were given a choice of seven response categories ranging from “nil in the previous year” to “once a week or more”, with a higher score representing more frequent health service use.

Statistical analyses

The PHQ-15 and K6 were used as dimensional measures of somatic distress and psychological distress respectively. Pearson correlation was computed to examine the linear relationship between somatic distress and psychological distress.

Respondents were divided into four groups according to their scores on PHQ-15 (cut-off: 9/10) (Kroenke et al., 2002) and K6 (cut-off: 7/8) (Wang et al., 2007): 1) Somatically Distressed Group who had high level of somatic distress and low level of psychological distress (PHQ-15 score > = 10 and K6 score < = 7); 2) Psychologically Distressed Group who had low level of somatic distress and high level of psychological distress (PHQ-15 score < = 9 and K6 score > = 8); 3) Mixed Distress Group who were high on both somatic and psychological distress (PHQ-15 score > = 10 and K6 score > = 8); 4) Low Distress Group who were low in both somatic and psychological distress (PHQ-15 score < = 9 and K6 score < = 7). Cross-tabulations and Chi-square tests were conducted to estimate the distributions of sociodemographic variables across the four subgroups. Multi-nominal logistic regression analysis was used to compare demographic differences in the Somatically Distressed Group, Psychologically Distressed Group, and Mixed Distress Group against the Low Distress Group. Additional ordinal logistic regression analysis was conducted to test if rates of somatic symptoms were higher in males.

To examine the differences in impairment and health service use across these four subgroups, Chi-square test, Welch test, Kruskal-Wallis test, Mann-Whitney U Test (as non-parametric post hoc test), and regression analyses were used. For dimensional analyses, Pearson and Spearman correlations were conducted for examining linear relationships between somatic and psychological distress with impairment and health service use. Linear regression and ordinal regression analyses were used to determine the association of somatic and psychological distress with impairment and health service use while adjusting for sociodemographic variables. Receiver operating characteristics (ROC) curve and area under curve (AUC) analyses were also conducted to examine whether severe impairment (dichotomously defined as a score of 7 or higher on at least one domain of SDS) and frequent health service use (dichotomously defined as having sought professional help once a month or more in the previous year) were attributable to somatic distress or psychological distress. 95% confidence intervals were reported. The results were evaluated based on an alpha level of .05 of a two-tailed test.

Results

PHQ-15 and K6 profiles

Of the respondents, 48.7%, 36.8%, 11.1%, and 3.5% were classified as having minimal, mild, moderate, and severe somatic distress, respectively. The mean PHQ-15 score for the whole sample is 5.43 (SD = 4.10). As for the K6, 74.2%, 19.7%, and 6.1% of the respondents had low, moderate, and high psychological distress respectively. The mean K6 score for the whole sample is 5.24 (SD = 12.58). Both PHQ-15 and K6 scores showed a positively skewed distribution. Figure 1 illustrates the number of respondents scoring different levels on the K6 and PHQ-15.

Figure 1.

Number of respondents scoring different levels on the Kessler Scale-6 and Patient Health Questionnaire-15.

Only 3.3% of the respondents did not report any somatic and psychological distress (PHQ-15 score = 0 and K6 score = 0). While 6.3% of the respondents reported any somatic distress (PHQ-15 score > 0) without psychological distress (K6 score = 0), 4.5% of the respondents presented with any psychological distress (K6 score > 0) without somatic distress (PHQ-15 score = 0). Thus, the overwhelming majority of respondents (85.9%) reported a varying mix of somatic and psychological distress. Correlation between PHQ-15 and K6 total scores was positive and high (r = .60, p < .001).

Prevalence and sociodemographic profile

The majority of respondents were classified as Low Distress Group (69.2%), while the proportions of those classified as Somatically Distressed Group, Psychologically Distressed Group, and Mixed Distress Group were 5.0%, 15.8%, 10.0%, respectively. These four groups differed significantly in the distributions of gender, age, marital status, educational attainment, occupational status, and family monthly income (p < .05) (Table 1).

Table 1.

Sociodemographic characteristics of the whole sample and the four subgroups.

	Whole sample (N = 3014, 100%)		Low Distress Group^c (n = 2087, 69.2%)		Somatically Distressed Group^d (n = 150, 5.0%)		Psychologically Distressed Group^e (n = 477, 15.8%)		Mixed Distress Group^f (n = 300, 10.0%)
	n	%	n	%	n	%	n	%	n	%	Chi-square	p-value
Gender
Male	1404	46.6	947	75.2	36	2.9	203	16.1	73	5.8	68.25	<.001
Female	1610	53.4	1140	65.0	114	6.5	274	15.6	227	12.9
Age^a
15–24	608	20.2	400	65.8	27	4.4	124	20.4	57	9.4	33.94	<.001
25–34	405	13.5	258	63.7	18	4.4	85	21.0	44	10.9
35–44	616	20.5	424	68.8	32	5.2	92	14.9	68	11.0
45–54	787	26.2	560	71.2	43	5.5	111	14.1	73	9.3
55–65	592	19.7	442	74.7	29	4.9	65	11.0	56	9.5
Educational Status^a
Primary (Grade 1–6)	373	12.4	244	65.4	23	6.2	45	12.1	61	16.4	52.41	<.001
Secondary (Grade 7–11)	1377	45.9	939	68.2	85	6.2	203	14.7	150	10.9
Post-secondary	361	12.0	261	72.3	13	3.6	62	17.2	25	6.9
University or above	887	29.6	631	71.1	29	3.3	167	18.8	60	6.8
Marital Status^a
Single	1065	35.6	701	65.8	47	4.4	218	20.5	99	9.3	41.01	<.001
Married/Cohabited	1796	60.0	1293	72.0	97	5.4	231	12.9	175	9.7
Divorced/Widowed	134	4.5	78	58.2	6	4.5	27	20.1	23	17.2
Employment Status^a
Employed	1661	55.4	1143	68.8	82	4.9	278	16.7	158	9.5	39.11	<.001
Unemployed	146	4.9	89	61.0	6	4.1	34	23.3	17	11.6
Retired	302	10.1	241	79.8	9	3.0	29	9.6	23	7.6
Fulltime students	441	14.7	298	67.6	21	4.8	82	18.6	40	9.1
Homemakers	449	15.0	306	68.2	32	7.1	54	12.0	57	12.7
Family Monthly Income (HKD)^a,b
<$10,000	525	18.9	327	62.3	24	4.6	86	16.4	88	16.8	55.15	<.001
$10,000 – $29,999	1359	48.8	917	67.5	79	5.8	235	17.3	128	9.4
$30,000 – $59,999	624	22.4	454	72.8	31	5.0	98	15.7	41	6.6
>$60,000	274	9.8	217	79.2	8	2.9	34	12.4	15	5.5

Missing data: age, n = 6; educational status, n = 16; marital status, n = 19; employment status, n = 15; family monthly income, n = 232.

USD$1 = HKD$7.8.

Respondents scored 0–7 on Kessler-6 (K6) and 0-9 on Patient Health Questionnaire-15 (PHQ-15).

Respondents scored 0–7 on K6 and 10 or above on PHQ-15.

Respondents scored 8 or above on K6 and 0–9 on PHQ-15.

Respondents scored 8 or above on K6 and 10 or above on PHQ-15.

When compared with Low Distress Group, males and those with university-level education were less likely to be in the Somatically Distressed Group. Male gender, age 55–65, and family income of HKD30,000 or above were significantly associated with decreased odds of being in the Psychologically Distressed Group. Finally, individuals with male gender, aged 55–65, retired from work, with educational attainment higher than primary education, with family income of HKD10,000 or above were less likely to be in the Mixed Distress Group (Table 2). Additional analysis revealed that females were more likely than males to report somatic distress after controlling for demographic variables and psychological distress (OR = 2.49, Wald = 117.52, p < .001).

Table 2.

Multivariate analysis for demographic factors associated with inclusion in Somatically Distressed, Psychologically Distressed, and Mixed Distress Groups^a

	Somatically Distressed Group^d (n = 150, 5.0%)		Psychologically Distressed Group^e (n = 477, 15.8%)		Mixed Distress Group^f (n = 300, 10.0%)
	Odds ratio	95% CI	Odds ratio	95% CI	Odds ratio	95% CI
Gender
Female	+		+		+
Male	0.39***	0.25, 0.58	0.80^#	0.64, 1.00	0.37***	0.27, 0.50
Age^b
15–24	+		+		+
25–34	0.95	0.35. 2.54	0.99	0.64, 1.54	1.10	0.57, 2.12
35–44	0.98	0.37, 2.63	0.71	0.44, 1.15	0.89	0.45, 1.76
45–54	0.91	0.34, 2.46	0.65	0.40, 1.06	0.50	0.25, 1.01
55–65	0.90	0.31, 2.62	0.53*	0.30, 0.94	0.46*	0.21, 0.99
Educational Status^b
Primary (Grade 1–6)	+		+		+
Secondary (Grade 7–11)	0.90	0.51, 1.56	0.97	0.65, 1.45	0.52**	0.35, 0.78
Post-secondary	0.52	0.24, 1.15	0.97	0.59, 1.60	0.36**	0.20, 0.66
University or above	0.46*	0.22, 0.94	1.12	0.71, 1.77	0.38**	0.22, 0.65
Marital Status^b
Single	+		+		+
Married/Cohabited	1.06	0.60, 1.87	0.76	0.55, 1.04	1.09	0.72, 1.67
Divorced/Widowed	0.85	0.30, 2.45	1.53	0.89, 2.63	1.76	0.92, 3.36
Employment Status^b
Employed	+		+		+
Unemployed	0.88	0.34, 2.29	1.31	0.84, 2.05	1.01	0.54, 1.88
Retired	0.50	0.22, 1.12	0.61	0.36, 1.03	0.53*	0.29, 0.96
Full-time students	1.01	0.39, 2.60	0.69	0.44, 1.08	0.85	0.44, 1.64
Homemakers	0.84	0.51, 1.38	0.77	0.53, 1.12	0.77	0.53, 1.14
Family Monthly Income (HKD)^b,c
<$10,000	+		+		+
$10,000 – $29,999	1.18	0.72, 1.94	0.87	0.64, 1.16	0.51***	0.37, 0.71
$30,000 – $59,999	1.08	0.59, 1.98	0.66*	0.46, 0.94	0.34***	0.22, 0.54
>$60,000	0.57	0.23, 1.42	0.52**	0.32, 0.84	0.27***	0.14, 0.52

Reference group = Low Distress Group (respondents scored 0–7 on Kessler-6 (K6) and 0–9 on Patient Health Questionnaire-15 (PHQ-15), n = 2087 (69.2%).

Missing data: age, n = 6; educational status, n = 16; marital status, n = 19; employment status, n = 15; family monthly income, n = 232.

USD$1 = HKD$7.8.

Respondents scored 0–7 on K6 and 10 or above on PHQ-15.

Respondents scored 8 or above on K6 and 0–9 on PHQ-15.

Respondents scored 8 or above on K6 and 10 or above on PHQ-15.

reference category.

p = .05.

p < .05.

p < .01.

***

p < .001.

Impairment

Of the respondents, 17.2% were found to be severely impaired in at least one of the four SDS domains; 8.2% of the Low Distress Group, 16.7% of the Somatically Distressed Group, 32.3% of the Psychologically Distressed Group, and 55.7% of the Mixed Distress Group reported severe impairment in at least one SDS domain (Chi-square = 507.90, p < .001).

The Mixed Distress Group had the highest mean total and subscale SDS scores, followed by the Psychologically Distressed Group, the Somatically Distressed Group, and the Low Distress Group (Table 3). The differences between groups were significant (p < .05). After controlling for demographic variables, Mixed Distress Group (OR = 14.90, 95% CI = 10.95–20.28, p < .001), Psychologically Distressed Group (OR = 5.62, 95% CI = 4.31–7.33, p < .001), and Somatically Distressed Group (OR = 2.42, 95% CI = 1.50–3.91, p < .001) exhibited elevated risks for severe impairment when compared to the Low Distress Group.

Table 3.

Distress and help-seeking profile among the four subgroups.

	Whole sample (N = 3014, 100%)		Low Distress Group^a (n = 2087, 69.2%)		Somatically Distressed Group^b (n = 150, 5.0%)		Psychologically Distressed Group^c (n = 477, 15.8%)		Mixed Distress Group^d (n = 300, 10.0%)
	mean	SD	mean	SD	mean	SD	mean	SD	mean	SD	Welch^e	p-value
Patient Health Questionnaire-15 score	5.429	4.10	3.747	2.54	11.613	2.04	5.824	2.25	13.410	3.35	1329.773	<.001
Kessler-6 score	5.241	4.14	3.115	2.12	4.900	1.74	10.451	2.59	11.920	3.03	1730.622	<.001
Sheehan Disability Scale score
Household responsibilities (0–10)	2.012	2.37	1.426	1.97	2.353	2.24	2.937	2.47	4.493	2.73	158.629	<.001
Work/study (0–10)	2.631	2.66	1.863	2.20	2.980	2.56	4.015	2.60	5.600	2.72	242.554	<.001
Close relationships (0–10)	2.150	2.45	1.476	2.01	2.387	2.26	3.497	2.57	4.623	2.66	199.082	<.001
Social life (0–10)	2.030	2.46	1.353	2.00	2.400	2.39	3.359	2.55	4.453	2.76	188.558	<.001
Total (0–40)	8.833	8.37	6.118	6.65	10.120	7.33	13.807	7.98	19.170	8.41	319.093	<.001
12-Month Professional Help^f	3.140	1.47	2.900	1.37	3.910	1.50	3.290	1.42	4.110	1.63	147.043^g	<.001

Respondents scored 0–7 on Kessler-6 (K6) and 0-9 on Patient Health Questionnaire-15 (PHQ-15).

Respondents scored 0–7 on K6 and 10 or above on PHQ-15.

Respondents scored 8 or above on K6 and 0–9 on PHQ-15.

Respondents scored 8 or above on K6 and 10 or above on PHQ-15.

Welch's variance-weighted statistic was used because the variances were not equal.

Respondents were given a choice of seven response categories: 1 = “nil in the previous year”; 2 = “once every 6–12 months”; 3 = “once every 4–5 months”; 4 = “once every 2-3 months”; 5 = “once a month”; 6 = “2-3 times every month”; 7 = “once a week or more”, with a higher score representing more frequent health service use.

Chi-square value generated by Kruskal-Wallis test.

Health service use

Of the respondents, 14.6% used a health service once a month or more in the previous year. For the remaining respondents, 33.3% reported health service use once every 2–3 months, while 36.4% used a health service once every 6–12 months.

The Mixed Distress Group exhibited the highest frequency of health service use in the previous year, followed by the Somatically Distressed Group, the Psychologically Distressed Group, and the Low Distress Group (Table 3). Further analyses revealed significant differences between all the groups except for the Mixed Distress Group and the Somatically Distressed Group (results available upon request). After controlling for demographic variables, Somatically Distressed Group (OR = 3.72, Wald = 64.27, p < .001), Psychologically Distressed Group (OR = 1.79, Wald = 35.43, p < .001), and Mixed Distress Group (OR = 4.74, Wald = 154.47, p < .001) reported significantly more frequent health service use than the Low Distress Group.

Prediction of impairment and health service use

Impairment had a significant positive relationship with somatic distress (r = .49, p < .001) and psychological distress (r = .60, p < .001). After adjusting for demographic variables and psychological distress, somatic distress significantly predicted impairment (β = .45, t = 11.60, p < .001). Likewise, psychological distress predicted impairment after controlling for demographic variables and somatic distress (β = .93, t = 24.81, p < .001).

Health service use was significantly and positively correlated with somatic distress (Spearman's rho = .34, p < .001) and psychological distress (Spearman's rho = .22, p < .001). Higher somatic distress was associated with more frequent health service use after adjusting for demographic variables and psychological distress (OR = 1.16, Wald = 189.91, p < .001). Likewise, higher psychological distress was associated with increased health service use after controlling for demographic variables and somatic distress (OR = 1.03, Wald = 5.73, p < .05).

The K6 total score (AUC = 0.78, 95% CI = 0.76–0.81) was superior to the PHQ-15 total score (AUC = 0.73, 95% CI = 0.71–0.76) in predicting severe impairment. On the other hand, PHQ-15 total score (AUC = 0.71, 95% CI = 0.68–0.74) was more superior to K6 total score (AUC = 0.64, 95% CI = 0.61–0.67) in predicting frequent health service use.

Discussion

This is the first study to examine both somatic and psychological distress among Chinese individuals in a representative Chinese community sample. It confirmed both previous clinical findings and the long-held belief that somatic distress is common in Chinese community. Moreover, such distress is dimensionally distributed and common at both clinically meaningful and milder levels. Nonetheless, it is worth noting that psychological distress is just as common among Chinese people in Hong Kong. Such “psychological distress” applies not only to depressive symptoms as suggested in the extant literature because the K6 measures non-specific psychological distress that occurs in the spectrum of DSM-IV anxiety and mood disorders (Kessler et al., 2002). Although the arbitrary PHQ-15 and K6 cut-offs precluded firm conclusions from being drawn, the findings tentatively suggested that, with reference to the definitions we adopted, there might arguably be significantly more individuals who were psychologically rather than somatically distressed in the community. This was especially so when those who experienced psychological distress in the context of mixed distress were considered.

Consistent with Western studies and a seminal Chinese clinical study on neurasthenia and depression (Beirens & Fontaine, 2011; Kleinman, 1982), our findings indicated that differences in somatic or psychological distress across Chinese people do not result from a trade-off between somatization and psychologization. Rather, somatic distress and psychological distress were moderately and positively correlated. Moreover, the percentage of respondents who had only/predominantly one type of distress was very small. A varying mix of somatic and psychological distress is therefore the most common mode of experiencing distress among Chinese people in Hong Kong.

Prevalence and sociodemographic profile

Apart from showing that both somatic distress and psychological distress are common, our findings indicated that several patterns of distress, as suggested by the proportions of respondents captured in the Somatically Distressed Group, Psychologically Distressed Group, and Mixed Distress Group, could exist among Chinese people in Hong Kong. Although some researchers have argued that Chinese people preferentially experience distress in somatic rather than psychological terms (Chang, 2007), we found that the Psychologically Distressed Group may be more prevalent than the Somatically Distressed Group. Moreover, the Somatically Distressed Group is less common than the Mixed Distress Group. The finding of a relatively high percentage of respondents in the Mixed Distress Group is consistent with studies across nations showing a positive correlation between somatic and psychological distress (Escobar et al., 2010; Simon et al., 1996) and high rates of comorbidity between somatic syndromes and anxiety/depressive disorders in both clinical and community epidemiological studies (Barsky, Orav, & Bates, 2005; De Waal, Arnold, Eekhof, & Van Hemert, 2004; Löwe et al., 2008).

Using a community sample that allows the investigation of sociodemographic profiles associated with different modes of distress expression, the present study confirmed that Chinese people are far from being obligatory somatizers. It suggested diversity rather than homogeneity in the expression of distress. Our results also indicated possible gender differences in the expression of distress. Kirmayer and Robbins (1991) revealed that gender ratios were different for patients with three different forms of somatization in primary care. Their study indicated that while female patients had more somatoform somatization, there was no statistically significant difference in the sex ratios between non-somatizers and both hypochondriacal and presenting somatizers (i.e., those who presented with exclusively somatic complaints that could be attributed to a psychiatric disorder). Our study used a non-clinical sample and did not measure specific mental disorders. Consequently, we would refrain from comparing our finding of males being less likely to be in the Somatically Distressed Group with the finding of Kirmayer and Robbins (1991) regarding gender and presenting somatization. It is of interest to note that the rates of somatic symptoms were lower in males even after controlling for psychological distress. This was in line with previous clinical and community studies that female gender was associated with higher somatic scores (Gureje, Simon, Ustun, & Goldberg, 1997; Lee et al., 2011). Moreover, that females were more likely to report psychological distress was consistent with the finding of female gender being a risk factor for depression in both Chinese and Western societies (Lee, Tsang, Huang, et al., 2009; Lee, Tsang, & Kwok, 2007; Lucht et al., 2003). The reporting of lower somatic and psychological distress in males does not support the possibility that Chinese males expressed distress by using somatic symptoms to replace psychological symptoms. In addition, it might help explain the generally lower level of help-seeking for emotional illnesses in males than females (Shek, 1992).

Impairment and health service use

It is not surprising that mixed distress was associated with the highest levels of impairment and health service use. Empirical evidence indicated that both psychological distress (Hansen, Fink, Frydenberg, & Oxhøj, 2002; Kessler, DuPont, Berglund, & Wittchen, 1999) and somatic symptoms (Escobar et al., 2010) predicted impairment and health service use. The Mixed Distressed Group can be viewed as not only categorically different but also being on the farther end of a continuum of severity.

Although different kinds of barriers to health service use exist for psychiatrically distressed individuals, the question of how psychological distress may relate to service use differently from somatic distress has rarely been studied. Somatoform disorders have been shown to be associated with high rates of health service use (Barsky et al., 2005), but direct comparison with other common mental disorders regarding service utilization is lacking.

The present study showed that psychological distress predicted impairment and health care use differently from somatic distress. Specifically, the Psychologically Distressed Group had a higher level of impairment in all four domains of functional impairment but less frequent health care use than the Somatically Distressed Group. This is in line with the recent finding of a community study showing that Chinese Americans with somatoform disorders were more likely to seek professional help than those with anxiety and depressive disorders, albeit the latter group of disorders were more impairing (Kung & Lu, 2008).

Our findings may partly be explained by the help-seeking behavior of distressed Chinese people. It has been observed that Chinese people tend not to view depressive and other emotional illnesses as a disease and are less likely to seek help for such “psychological” problems (Parker, Chan, & Tully, 2006) even if the latter are quite impairing. Many of them may not consider such distress an illness (Hwu, Chang, Yeh, Chang, & Yeh, 1996) and thus manage it by using self-help methods instead (Lee et al., 2007). For example, a community study showed that the lifetime prevalence of major depressive disorder was much lower in Taiwan than in the United States. Nonetheless, depression in Taiwan was accompanied by twice the number of lost workdays compared to the same disorder in the U.S. sample. The authors argued that “stoicism” in the Chinese culture made people endure their psychological suffering and refrain from seeking professional help (Liao et al., 2012). That psychological distress was more impairing but less predictive of health service use than somatic distress in the present study is partially consistent with this view. Moreover, our findings suggest that the assessment of health service use in distressed Chinese people should take account of both somatic and psychological presentations of distress. This is because epidemiological research instruments such as the widely used Composite International Diagnostic Interview (CIDI) typically assess service use by asking about people's help-seeking for psychological or emotional problems (e.g., Bonnewyn et al., 2009). This might underestimate health service use, which was repeatedly found to be very low in Chinese communities (Phillips et al., 2009; Wang et al., 2005).

Limitations

The present study has several limitations that call for further research. In order to report on the relative prevalence of somatic and psychological distress expressions in a Chinese community, we used a categorical approach to analysis even though a dimensional approach could better fit some of the data. The choice of cut-offs for differentiating high versus low somatic/psychological distress is arbitrary. Thus, the prevalence and correlates of the four hypothesized groups will change when the cut-offs are changed to more lenient or stringent ones. We therefore refrain from concluding that psychological distress must be more common than somatic distress in the Hong Kong Chinese general population. Additionally, the PHQ-15 and K6 are brief measures. This would limit their comprehensiveness in screening for somatic and psychological distress. While this study provides preliminary estimates for the relative prevalence of somatic and psychological distress among Chinese people, more detailed scales that include a wider range of forms of distress are needed in future studies. This applies also to the examination of local idioms of distress that are not captured by scales originally developed in Western populations (Cheng, 1989; Kleinman, 1982). Another limitation is that the data on impairment relies on the self-report of respondents. It is possible that those prone to reporting more psychological distress may report more impairment when the latter is assessed via self-report rather than detailed objective rating by interviewers. In addition, although the PHQ-15 and K6 allow people to report a number of common somatic and psychological symptoms, our instruments do not generate clinical diagnosis. It remains unclear whether the Psychologically Distressed Group is equivalent to those individuals with anxiety, depression and other common mental disorders. Similarly, we do not know if the Somatically Distressed Group includes a large number of individuals with somatoform or somatic symptom disorders which have not been adequately studied in population epidemiological studies. Moreover, somatization may only occur among those who have substantial distress (e.g., people with major depressive disorder). Although a community sample can shed light on somatization in a Chinese general population, the somatization effect in the small number of highly distressed people may have been counteracted by the lack of somatization effect in the large number of non-distressed people in the community. Our findings do not exclude the possibility that somatization may still occur in some of those with greater symptom severity and under the complex influence of clinical settings. In addition, there is only one question on health service use. To understand better the relationship between distress, impairment and health service use, it is important to investigate what health service the four groups used and the problems they encountered. Finally, the present study was based on the Hong Kong Chinese community sample. The findings might not be generalizable to non-Chinese and other Chinese populations.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Biographies

Sing Lee, MBBS, FRCPsych, is a Professor in the Department of Psychiatry and Director of the Hong Kong Mood Disorders Center and Hong Kong Eating Disorders Center, Faculty of Medicine, The Chinese University of Hong Kong. He is a clinician, teaches medical students, and supervises psychiatric residents. His principal research interest is in social psychiatry, including the relationship between culture and psychopathology, psychiatric epidemiology, and other aspects of mental health and social change in Chinese society.

Candi Man Chi Leung, MPhil, is a Research Assistant at the Hong Kong Mood Disorders Center, Faculty of Medicine, The Chinese University of Hong Kong. She has keen research interest in mood disorders and suicide. She has coauthored a number of papers on the epidemiological studies of health anxiety and generalized anxiety disorder, unemployment and suicide rates during economic recovery, and temporal patterns of charcoal burning suicides.

Kathleen Pik San Kwok, MSSc, is a Clinical Psychologist at the Hong Kong Eating Disorders Center and Hong Kong Mood Disorders Center of the Faculty of Medicine, The Chinese University of Hong Kong. She is a full-time clinician. For over a decade, she has used cognitive behavioral therapy and other psychotherapies to treat Chinese patients with emotional and eating disorders. Her research interests include anxiety, depressive and eating disorders.

King Lam Ng, MPH, was a Senior Research Assistant at the Department of Social Work and Social Administration, Faculty of Social Sciences, The University of Hong Kong. She has coauthored several papers on the epidemiology of mood and anxiety disorders, the personal recovery of bipolar disorder and schizophrenia, the psychometric properties of various instruments, and the clinical profile of eating disorders.

References

American Psychiatric Association (APA) (2000) Diagnostic and statistical manual of mental disorders, (4th ed., Text Revision. Washington, DC: American Psychiatric Publishing.

Barsky

A. J.

Orav

E. J.

Bates

D. W.

(2005) Somatization increases medical utilization and costs independent of psychiatric and medical comorbidity. Archives of General Psychiatry 62: 903–910.

Beirens

Fontaine

J. R. J.

(2011) Somatic complaint differences between Turkish immigrants and Belgians: Do all roads lead to Rome? Ethnicity & Health 16: 73–88.

Bonnewyn

Katona

Bruffaerts

Haro

J. M.

De Graaf

Alonso

Demyttenaere

(2009) Pain and depression in older people: Comorbidity and patterns of help seeking. Journal of Affective Disorders 117: 193–196.

Chang

(2007) Depressive symptom manifestation and help-seeking among Chinese college students in Taiwan. International Journal of Psychology 42: 200–206.

Cheng

T. A.

(1989) Symptomatology of minor psychiatric morbidity: A crosscultural comparison. Psychological Medicine 19: 697–708.

De Waal

M. W. M.

Arnold

I. A.

Eekhof

J. A. H.

Van Hemert

A. M.

(2004) Somatoform disorders in general practice: Prevalence, functional impairment and comorbidity with anxiety and depressive disorders. British Journal of Psychiatry 184: 470–476.

Escobar

J. I.

Cook

Chen

C. N.

Gara

M. A.

Alegría

Interian

Diaz

(2010) Whether medically unexplained or not, three or more concurrent somatic symptoms predict psychopathology and service use in community populations. Journal of Psychosomatic Research 69: 1–8.

Guo

W. J.

Tsang

Lee

(2011) Psychiatric epidemiological surveys in China 1960–2010: How real is the increase of mental disorders? Current Opinion in Psychiatry 24: 324–330.

10.

Gureje

Simon

Ustun

T. B.

Goldberg

D. P.

(1997) Somatization in cross-cultural perspective: A World Health Organization study in primary care. American Journal of Psychiatry 154: 989–995.

11.

Hansen

M. S.

Fink

Frydenberg

Oxhøj

(2002) Use of health services, mental illness, and self-rated disability and health in medical inpatients. Psychosomatic Medicine 64: 668–675.

12.

Hsu

L. K.

Folstein

M. F.

(1997) Somatoform disorders in Caucasian and Chinese Americans. Journal of Nervous and Mental Disease 185: 382–387.

13.

J. Z.

C. Y.

Duan

W. D.

Gao

Zhang

Tang

Z. R.

Shen

Q. J.

(2009) Survey on mental disorders among registered residents and non-registered residents in Shenzhen [in Chinese]. Zhonghua Liu Xing Bing Xue Za Zhi 30: 543–548.

14.

Hwu

H. G.

Chang

I. H.

Yeh

E. K.

Chang

C. J.

Yeh

L. L.

(1996) Major depressive disorder in Taiwan defined by the Chinese Diagnostic Interview Schedule. Journal of Nervous and Mental Disease 184: 497–502.

15.

Isaac

Janca

Burke

K. C.

Costa e Silva

J. A.

Acuda

S. W.

Altamura

A. C.

Tacchini

(1995) Medically unexplained somatic symptoms in different cultures: A preliminary report from phase I of the World Health Organization International Study of Somatoform Disorders. Psychotherapy and Psychosomatics 64: 88–93.

16.

Johnson, T. P., & Owens, L. (2003, May). Survey response rate reporting in the professional literature. Paper presented at the 58th Annual Meeting of the American Association for Public Opinion Research, Nashville, TN. Retrieved from http://www.srl.uic.edu/publist/Conference/rr_reporting.pdf.

17.

Kessler

R. C.

Andrews

Colpe

L. J.

Hiripi

Mroczek

D. K.

Normand

S. L.

Zaslavsky

A. M.

(2002) Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychological Medicine 32: 959–976.

18.

Kessler

R. C.

DuPont

R. L.

Berglund

Wittchen

H. U.

(1999) Impairment in pure and comorbid generalized anxiety disorder and major depression at 12 months in two national surveys. American Journal of Psychiatry 156: 1915–1923.

19.

Kirmayer

L. J.

Robbins

J. M.

(1991) Three forms of somatization in primary care: Prevalence, co-occurrence, and sociodemographic characteristics. The Journal of Nervous and Mental Disease 179: 647–655.

20.

Kirmayer

L. J.

Young

(1998) Culture and somatization: Clinical, epidemiological, and ethnographic perspectives. Psychosomatic Medicine 60: 420–430.

21.

Kleinman

(1977) Depression, somatization, and the “new cross-cultural psychiatry”. Social Science & Medicine 11: 3–10.

22.

Kleinman

(1982) Neurasthenia and depression: A study of somatization and culture in China. Culture, Medicine and Psychiatry 6: 117–190.

23.

Kroenke

(2007) Somatoform disorders and recent diagnostic controversies. Psychiatric Clinics of North America 30: 593–619.

24.

Kroenke

Spitzer

R. L.

Williams

J. B.

(2002) The PHQ-15: Validity of a new measure for evaluating the severity of somatic symptoms. Psychosomatic Medicine 64: 258–266.

25.

Kung

W. W.

P. C.

(2008) How symptom manifestations affect help seeking for mental health problems among Chinese Americans. Journal of Nervous and Mental Disease 196: 46–54.

26.

Lee, S., Guo, W. J., Tsang, A., He, Y. L., Huang, Y. Q., Liu, Z. R., … Kessler, R., C. (2010). Perceived barriers to mental health treatment in metropolitan China. Psychiatric Services, 61, 1260--1262.

27.

Lee

Y. L.

Tsang

(2011) Psychometric properties of the Chinese 15-item Patient Health Questionnaire in the general population of Hong Kong. Journal of Psychosomatic Research 71: 69–73.

28.

Lee

Tsang

Huang

Y. Q.

Y. L.

Liu

Z. R.

Zhang

M. Y.

Kessler

R. C.

(2009) The epidemiology of depression in Metropolitan China. Psychological Medicine 39: 735–747.

29.

Lee

Tsang

Kwok

(2007) Twelve-month prevalence, correlates, and treatment preference of adults with DSM-IV major depressive episode in Hong Kong. Journal of Affective Disorders 98: 129–136.

30.

Lee

Tsang

K. L.

Y. L.

Guo

Mak

Kwok

(2012) Performance of the 6-item Kessler scale for measuring serious mental illness in Hong Kong. Comprehensive Psychiatry 53: 584–592.

31.

Lee

Wing

Chan

Lee

A. M.

Lee

D. T.

Weiss

M. G.

(2000) Psychiatric morbidity and illness experience of primary care patients with chronic fatigue in Hong Kong. American Journal of Psychiatry 157: 380–384.

32.

Leon

A. C.

Olfson

Portera

Farber

Sheehan

D. V.

(1997) Assessing psychiatric impairment in primary care with the Sheehan Disability Scale. International Journal of Methods in Psychiatric Research 27: 93–105.

33.

Liao

S. C.

Chen

W. J.

Lee

M. B.

Lung

F. W.

Lai

T. J.

Liu

C. Y.

Chen

C. C.

(2012) Low prevalence of major depressive disorder in Taiwanese adults: Possible explanations and implications. Psychological Medicine 42: 1227–1237.

34.

Löwe

Spitzer

R. L.

Williams

J. B.

Mussell

Schellberg

Kroenke

(2008) Depression, anxiety and somatization in primary care: Syndrome overlap and functional impairment. General Hospital Psychiatry 30: 191–199.

35.

Lucht

Schaub

R. T.

Meyer

Hapke

Rumpf

H. J.

Bartels

John

(2003) Gender differences in unipolar depression: A general population survey of adults between age 18 to 64 of German nationality. Journal of Affective Disorder 77: 201–211.

36.

Mak

W. W. S.

Zane

N. W. S.

(2004) The phenomenon of somatization among community Chinese Americans. Social Psychiatry and Psychiatric Epidemiology 39: 967–974.

37.

Parker

Chan

Tully

(2006) Depression and help-seeking in a western sample “highly acculturated” Chinese and controls. Journal of Affective Disorders 94: 239–242.

38.

Parker

Chan

Tully

Eisenbruch

(2005) Depression in the Chinese: The impact of acculturation. Psychological Medicine 35: 1475–1483.

39.

Parker

Cheah

Y. C.

Roy

(2001) Do the Chinese somatize depression? A cross-cultural study. Social Psychiatry and Psychiatric Epidemiology 36: 287–293.

40.

Phillips

M. R.

Zhang

Shi

Song

Ding

Pang

Wang

(2009) Prevalence, treatment, and associated disability of mental disorders in four provinces in China during 2001–05: An epidemiological survey. Lancet 373: 2041–2053.

41.

Ryder

A. G.

Chentsova-Dutton

Y. E.

(2012) Depression in cultural context: “Chinese somatization”, revisited. Psychiatric Clinics of North America 35: 15–36.

42.

Ryder

A. G.

Yang

Zhu

Yao

Heine

S. J.

Bagby

R. M.

(2008) The cultural shaping of depression: Somatic symptoms in China, psychological symptoms in North America? Journal of Abnormal Psychology 117: 300–313.

43.

Shek

D. T.

(1992) Reliance on self or seeking help from others: Gender differences in the locus of coping in Chinese working parents. Journal of Psychology: Interdisciplinary and Applied 126: 671–678.

44.

Simon

Gater

Kisely

Piccinelli

(1996) Somatic symptoms of distress: An international primary care study. Psychosomatic Medicine 58: 481–488.

45.

Simon

G. E.

VonKorff

Piccinelli

Fullerton

Ormel

(1999) An international study of the relation between somatic symptoms and depression. New England Journal of Medicine 341: 1321–1335.

46.

Tseng

W. S.

(1975) The nature of somatic complaints among psychiatric patients: The Chinese case. Comprehensive Psychiatry 16: 237–245.

47.

Wang

P. S.

Gruber

M. J.

Powers

R. E.

Schoenbaum

Speier

A. H.

Wells

K. B.

Kessler

R. C.

(2007) Mental health service use among Hurricane Katrina survivors in the eight months after the disaster. Psychiatric Services 58: 1403–1411.

48.

Wang

P. S.

Lane

Olfson

Pincus

H. A.

Wells

K. B.

Kessler

R. C.

(2005) Twelve-month use of mental health services in the United States. Archives of General Psychiatry 62: 629–640.

49.

Yeung

Chang

Gresham

R. L. J.

Nierenberg

A. A.

Fava

(2004) Illness beliefs of depressed Chinese-Americans in primary care. Journal of Nervous and Mental Disease 192: 324–327.

50.

Yen

Robins

C. J.

Lin

(2000) A cross-cultural comparison of depressive symptom manifestation: China and the United States. Journal of Consulting and Clinical Psychology 68: 993–999.

51.

Zhang

M. Y.

(1989) The diagnosis and phenomenology of neurasthenia: A Shanghai study. Culture, Medicine and Psychiatry 13: 147–161.

52.

Zhou

Dere

Zhu

Yao

Chentsova-Dutton

Y. E.

Ryder

A. G.

(2011) Anxiety symptom presentations in Han Chinese and Euro-Canadian outpatients: Is distress always somatized in China? Journal of Affective Disorders 135: 111–114.