Summary

Objective: To investigate how depression, anxiety, and stress interact with severity of medically unexplained physical symptoms (MUPS) among patients in Hong Kong primary care settings.

Design: A cross-sectional study.

Subjects: A total of 168 Chinese participants had been classified into three MUPS groups based on self-report as well as assessment by attending physicians: (a) none, (b) one to three, and (c) more than three MUPS. Main outcome measures: Somatic subscale of SCL-90R and DASS-21 (Depression, Anxiety, and Stress Scales).

Results: There were significant and high positive univariate correlations between the number of MUPS and depression, anxiety, and stress, with larger numbers of MUPS associated with higher levels of dysphoric feelings and stress: Depression, r (167) = 0.52; Anxiety, r (167) = 0.67; and Stress, r (167) = 0.55; ps < 0.05. The vector of mood and stress symptoms explained 44.3% of variance in MUPS. It was found that the Anxiety subscale of DASS had the strongest association with MUPS when other variables were controlled for. (β = 0.569, p < 0.001)

Conclusion: Attending physicians should be aware of the psychosocial factors of MUPS (particularly anxiety) among patients with these symptoms. Further studies should address other risk factors including those without notable psychological disturbances as treatment focus may need to vary as a function of this and other associated features.

Keywords: primary care, medically unexplained physical symptoms, depression, anxiety, stress.

摘要

目的：研究香港基層醫療患者中抑鬱、焦慮和壓力與原因不明身體症狀（MUPS）的嚴重程度之間的相互作用。

設計：橫斷面研究。

研究對象：根據患者自我報告和主診醫生的評估結果，將共計168名華人患者分為三個MUPS組：(a)無；(b)1-3；(c)超過 3 個MUPS。

主要測量內容：SCL-90R和DASS-21的軀體症狀分量表（抑鬱、焦慮和壓力量表）。

結果：MUPS的數量和抑鬱、焦慮及壓力之間存在顯著的單變數正相關關係，MUPS的數量越多，煩躁不安和壓力的水準就越高：抑鬱，r(167)=0.52；焦慮，r(167)=0.67；應激，r ( 1 6 7 )=0 . 5 5；p s<0 . 0 5。情緒和壓力症狀可解釋44.3%的MUPS變異。研究發現，當控制了其他變數時，DAS S的焦慮分量表與MUP S的關聯最強 ( β = 0 . 5 6 9 , p < 0.001)。

結論：主診醫生應瞭解有此類醫學原因不明身體症狀症狀MUPS（尤其是焦慮）患者的社會心理因素。應針對患者(包括無明顯社會心理障礙患者)的其他高危因素做進一步研究，因為治療重點可能要隨上述及其他有關因素而調整。

主要詞彙：基層醫療，醫學原因不明身體症狀，抑鬱，焦慮，壓力。

---

Introduction

In clinical practice, many symptoms cannot be explained or understood within a biological framework. About 10 to 50% of patients seen by medical practitioners have medically unexplained physical symptoms (MUPS).^1-4 In a Hong Kong cross-sectional survey conducted in 30 general practice clinics, the prevalence of functional disorders was 16.9% with most of these patients being female.⁵ As Kirmayer et al (2004, p.664)6 have observed, there is no consensus on how people with ambiguous bodily reactions should be diagnosed based on accepted standards of medical nosology. The above authors noted that "an unexplained physical symptom names a predicament, not a disorder". The common features of patients under the broad spectrum of medically unexplained physical symptoms (MUPS) include (a) having one or more physical symptoms and (b) even after appropriate clinical examination and investigations, classification or aetiology cannot be fully explained by a known general medical condition.

Given the lack of an adequate aetiological understanding, treatment outcome is often unsatisfactory. Epidemiological studies have revealed that people with MUPS were often "frequent attenders" requiring more protracted investigations and/or invasive procedures than would otherwise be expected.^6,7 Not surprisingly, the average health care costs were much higher in this group of patients as compared with those having medically explained symptoms. 

The literature regarding the relationship between affective functioning and the experience of physical symptoms remains mixed, and mostly done in Western countries. As there have not been any previous studies done in Hong Kong's primary care settings investigating psychosocial factors associated with MUPS, the present research was designed in order to increase knowledge in this locality, with an aim to improve treatment of MUPS among our populations.

Method

Subjects

Physicians: These were recruited from fifteen primary care practices located in diverse Hong Kong districts. Fourteen of the participating physicians took responsibility for patient recruitment and data collection. In one setting, these responsibilities were undertaken by a research assistant.

Patients: When patients arrived at the clinic, they were invited to participate in a study about medically unexplained physical symptoms. The specific number of potential patients asked to participate in this research was not quantified as recruitment was done in a general and non-intrusive manner; but this number estimated should be in the hundreds. If a patient indicated willingness to complete questionnaires for the study, they signed the consent form and then completed all assessment instruments in the waiting room. 

All the participants were Chinese, and were categorized into three groups based on the number of MUPS they had: (a) none, (b) 1 to 3 and (c) more than 3 symptoms. Patients were excluded from the study if they were suffering from acute symptoms, had significant language/communication problems, or were receiving psychiatric or psychological treatments.

Measurements

Questionnaire. Apart from demographic data, a questionnaire included the somatic subscale of the Symptom Checklist-90 Revised (SCL-90R), and 21 questions from the Depression, Anxiety, and Stress Scales (DASS-21). 

Somatic Subscale of SCL-90R. The participants were first asked to indicate any symptom or symptoms that they had experienced in the past week and indicated a severity rating using a Likert scale of 1 = not at all to 5 = extremely severe.

The Depression, Anxiety, and Stress Scales (DASS-21). This instrument⁸ comprises three scales, Depression, Anxiety, and Stress, each of which consists of 7 items. Frequency/severity ratings are made on a series of 4-point (0 to 3) scales (0 = did not apply to me at all, 3 = applied to me very much, or most of the time). As it is primarily a dimensional rather than a categorical measure, continuous DASS-21 values were used in the present study rather than converting scores into diagnostic categories. The reliability of the DASS is excellent (e.g., alpha coefficients have been the range of 0.73-0.81) and the measure possesses adequate convergent and discriminated validity.^9-11 This assessment instrument had already been translated into Chinese and validated in Asian samples.¹² 

MUPS scores: MUPS was defined as physical symptoms that, in the opinion of the attending physician, were medically unexplained and having been experienced for a duration of more than one month. In addition to a raw score (that could range from 0 to 12), a total MUPS score was obtained by summing the severity ratings given by patients for all MUPS.

Procedure

Once approval was received from the sponsoring university’s Institutional Research Board (IRB) and the University of Hong Kong/Hospital Authority Hong Kong West Cluster IRB (UW 10-125), a pilot study was performed in one clinic to evaluate the feasibility and acceptability of the patients in responding to SCL-90R questions. No difficulties in understanding and responding to the study questionnaire were found.

In the main study, each patient self-completed all the study instruments including the SCL-90R before the doctor consultation. The primary care physician reviewed to endorse that the symptoms reported were not related to current or known chronic illnesses. Training/briefing was provided to train at least one individual from each participating clinic to follow the procedures.

Data Analyses

All statistical analyses were conducted using SPSS (Version 19). For all the variables, descriptive statistics were computed. Associations between the number of MUPS, stress and psychological symptoms were obtained using Pearson Product-Moment and multiple regression methods. Relationships between categorical demographic variables and group membership were evaluated using chi-square tests. Between-group differences on the continuous variables were evaluated through the use of MANOVA procedures.

Results

Characteristics of the Sample

The total number of patients recruited was 168. The sample consisted of 113 (67.7%) women and 54 (32.3%) men. The mean age was 41.²² years range (18-76) (Table 1). No significant difference among the three MUPS groups was found with respect to age, gender or education level.

Correlations among MUPS, Depression, Anxiety, and Stress

As shown in Table 2, there were significant univariate correlations between the number of MUPS and psychosocial problems: Depression, r = 0.52; Anxiety, r = 0.67; Stress, r = 0.55.

The correlations were somewhat greater when symptom scores were employed, Depression, r = 0.58; Anxiety, r = 0.72; Stress, r = 0.60.

Multiple Regression Analysis between DASS and MUPS

Using a full regression model entering all variables in one step, it was found that the three DASS-21 predictors of depression, anxiety, and stress symptoms explained 44.3% of variance in MUPS ( adjusted R2 = 0.443 ) . Individual scale findings showed that the Anxiety subscale of DASS had a highly significant association with MUPS when all other variables were statistically controlled (p < 0.001) while the contributions made by stress and depression symptoms did not reach significance (see Table 3). As univariate correlations between demographic variables (e.g., age, gender, and education) were not significant, they were not entered into the multiple regressions for reasons of parsimony.

Differences between MUPS Groups on DASS Variables

A MANOVA was performed to investigate the differences between the DASS means (given in Table 4) of three MUPS groups. There were significant differences among the three groups ( p < 0.001) . Follow- up analyses of variance (ANOVAs) revealed significant between-groups differences on each dependent variable. Effect size was large for anxiety and moderate for depression and stress. With respect to multiple comparisons between the means of each group. Tukey tests were performed and it was found that those with more than 3 MUPS had greater mood and stress problems than the participants with fewer or no symptoms, p < 0.001.

Discussion

Similar to other reports,^12-15 stress, anxiety, and depression symptoms (as measured by DASS) were associated with the number of medically unexplained symptoms in our primary care sample. However, the present study is unique in providing quantitative estimates of the associations between depression, anxiety, and stress with MUPS that lasted for only one month. This finding shows the importance of a thorough going assessment of these (and related) psychosocial factors in patients suffering from MUPS in primary care. Particularly suggested by the present finding is the concomitant treatment of anxiety when multiple MUPS are present early in the medical encounter as it was shown that MUPS had temporal relationship with depression and anxiety.¹⁴

That only anxiety was found to be a significant predictor of MUPS in the multiple regression analysis was not surprising as it is well established that anxiety is highly correlated with depression and stress (e.g., r = 0.717 and 0.774 respectively in this study). As a result, when correlated variables are partialled, only the factor with the highest correlation with the dependent variable enters the regression equation as significant.¹⁶ This should not, however, be taken to indicate that depression and stress issues can be ignored by physicians treating patients with MUPS as they may be the cause of the anxiety for many patients. Other studies in the literature showed that the association of MUPS with psychiatric morbidity ranged from 16% to over 50% depending on the inclusion criteria of MUPS used in the studies.^17,18 In the primary care setting, most patients seek medical advice with symptoms of shorter duration and usually with psychological symptoms that may not fulfill psychiatric diagnostic criteria. Therefore this study used less stringent criteria as compared with other studies¹⁸ as it is a more common presentation in our primary care setting. It would be prudent to treat these patients early before the development of full blown psychiatric morbidities. 

The fact that the vectors of DASS variables for the three MUPS groups were highly significant is yet another indication of the importance these psychosocial variables play in the lives of people with multiple MUPS. It is not difficult to speculate on the reasons that as the number of MUPS increases, anxiety and depression symptoms would increase. Obviously, as patients have physical symptoms, they could easily become frustrated and anxious especially when the symptoms are not satisfactorily explained or managed. If the symptoms persisted despite various means were tried to get rid of them, people would feel helpless or even hopeless. A recent study¹⁹ had shown that physical symptoms whether explained or not predicted psychopathology. On the other hand, people with depressive and/or anxiety symptoms could have more physical symptoms. From a physiological point of view, anxiety would stimulate the autonomic nervous system and increase in noradrenergic functioning resulting in a lot of physical symptoms.

With respect to the controlling potential confounding variables; there were no significant differences between Groups A, B, and C with reference to gender, age, and education level. Thus the findings obtained from group contrasts would not be biased by these factors. With reference to gender, the result of this study is contrary to findings of previous studies, including a study on Chinese Americans, that women reported more MUPS than men.^20,21 With this preconception, some studies even used different inclusion criteria for men and women in the study of somatisation disorder.¹⁸

Limitations

The recruitment of participants in this study was by convenience sampling and, therefore, results may be biased since data could not be obtained from those who refused entry into the study. Interestingly, however, the largest group in the present investigation had more than 3 MUPS (58%, n = 98). This finding suggests that every medical practice is likely to include large numbers of people with multiple MUPS. One might even expect similar distributions from random sampling procedures. On the other hand, patients with larger number of MUPS may be more eager to participate in the hope of gaining better understanding of their own problems.

Finally, as this was a cross sectional study, it was not possible to determine if the factors under scrutiny displayed longitudinal changes. As with questionnaire-based research, some people may respond to items haphazardly or with less than total candor. Further, the present study did not focus on any particular symptom or symptoms and it has been reported that different symptoms are associated with different ways of coping.

Conclusions

Our findings demonstrate that the presence of medically unexplained physical symptoms of short duration warrant screening for variables including stress, anxiety, and depression especially for those with more than three MUPS. It would be appropriate for primary care physicians to take into consideration these people’s psychological status early in the

consultation encounters, and the origin of their anxiety such as health anxiety, developmental issues, existential anxiety to decide on the most appropriate form of treatment. While it is prudent to look for affective symptoms in these patients, they can only explain a proportion of MUPS variance and other factors are clearly involved in the experience and response to MUPS.

Even though it is suggested by the DSM-V work group that the name of this group of patients would likely be renamed using different diagnostic criteria (www.dsm5.org), it is most important to understand these patients in a different perspective, noting the interplay amongst biological, psychological, social and even spiritual dimensions. Apart from medications,

other forms of therapies such as cognitive behavior therapy and acceptance and commitment therapy may be useful to help these patients to lead a more satisfying and meaningful life instead of wasting efforts in trying to get rid of all symptoms. Future research should focus on patients who have multiple somatic symptoms in the absence of identifiable physical or psychiatric diseases in order to provide the most appropriate form of treatment irrespective of the name we are going to ‘label’ them with.

Acknowledgments

I would like to thank the following who had dedicated a lot of their efforts to share with me their ideas and/or recruit patients to answer the questionnaires: Drs. Amy Chan, Mark Chan, Chan Suk Yin, Florence Cheung, Anthony Chong, Fung Yuk Kwan, Kong Yim Fai, Kenneth Kwok, :Letical order) and doctors at the Family Medicine Unit, The University of Hong Kong. 

---

---

Reference