Relationships between spirituality and mental stress in people living with HIV in China: A cross-sectional study

Abstract

There are limited reports on the relationship between spirituality and mental stress in PLWH in China, who may be subject to anti-religious pressures from the government. In this study, we aimed to understand whether spirituality influences Chinese PLWH’s mental stress and, if so, at what level. We recruited 200 PLWHs from Beijing’s Ditan Hospital to complete a cross-sectional survey inquiring about their practice of spirituality as well as their level of mental stress. The study found that PLWH who presented with a mid-level of spirituality have the highest mental stress when compared to those who have a low level of spiritual beliefs or a high level of spiritual beliefs. This study points to the utility of healthcare providers taking PLWH’s potential spirituality into consideration, perhaps in particular for those with a moderate level of spirituality, in order to provide the most comprehensive care possible.

Introduction

Despite recent advances in pharmacotherapy and public health interventions, HIV/AIDS continues to be both a major health and economic burden in China. As of September 2021, it is estimated that about 1.25 million people in China are living with HIV (National Center for AIDS/STD Control and Prevention, China CDC, 2018). As HIV/AIDS has become a chronic, manageable condition with the advent of antiretroviral therapy (ART; Lima et al., 2015), mental stress is an increasing concern for people living with HIV (PLWH) worldwide, with depression and anxiety being their most prevalent expression of acute stress (Jiang et al., 2019).

Globally, the prevalence of depression and anxiety in PLWH ranges from 7.2% to 71.9% and 4.5% to 82.3%, respectively (Chaudhury, et al., 2016). The incidence of depression and anxiety also increases with progression of HIV disease. In China, more than 60% of PLWH present with depression and 40% with anxiety, and PLWH have a higher prevalence of mental health morbidity than individuals in the general population (Aunon et al., 2020; Niu et al., 2019). Furthermore, depression and anxiety among PLWH have been shown to negatively impact their adherence to ART (Jiang et al., 2019), cognition (Batchelder et al., 2017), sleep (Huang et al., 2018), and immune system (Taniguchi et al., 2014) and have been linked to daily functional impairment and decreased quality of life (Niu et al., 2019). Thus, understanding the potential ameliorating factors to depression and anxiety is especially important. One such factor may be spirituality (Arrey et al., 2016; Coleman & Holzemer, 1999).

Spirituality is “an inherent component of being human and involve[s] a personal quest for meaning in life, sense of purpose, connectedness and wholeness, love and commitment, as well as a sense of the Holy amongst us” (Dyson et al., 1997). Spirituality has been shown to play an important role in the mental health of PLWH from various national backgrounds, ranging from South Africans (Arrey et al., 2016), to Americans (Kudel et al., 2011), to Iranians (Martin, 2009), but few studies discuss spirituality among Chinese PLWH and its impact on their mental health.

Spirituality is a substantial component of Chinese culture. Notably, 94% of the 1.3 billion residents have, at some point in their lives, participated in spiritual activities (Lu 2014). Over 40% of Chinese report that Buddhism, Confucianism, Taoism, or Christianity have influenced their spirituality, religious traditions, and practices (Tang & Chen, 2018). Despite the official policy of atheism by the Chinese government (Congressional-Executive Commission on China, 2018, P. 7), it does recognize religions (Albert, 2018). However, many local administrations have promoted atheism and to this end, some have even destroyed churches and temples (Goldman, 2018), leading to some contradictions in public life. Although not widely viewed as a religion, the value and vision of Confucianism has shaped the life of parents, educators, and students.

This study sought to explore if spirituality can buffer the negative effects of depression and anxiety. To date, the evidence has been mixed and inconsistent (Braam & Koenig, 2019; Wang et al., 2019). Traditionally, a negative linear relationship between spirituality and mental health has been presumed (see e.g., Carrico et al., 2006; McFarland, 2010), with most research suggesting that the higher the degree to which spirituality is a part of a person’s life, the less negative mental stress is present (e.g., Coleman & Holzemer, 1999; Lassiter & Parsons, 2016). However, other studies suggest that spirituality can have a deleterious impact on mental stress (Lucchetti et al., 2021). For example, spirituality may create unrealistic expectations or be used as a way of avoiding or denying the reality of stressful or traumatic experiences, which may ultimately lead to more negative mental health outcomes.

Recently, the often-overlooked nonlinear effects that spirituality has on mental health have been reported and have begun replacing previous findings of the linear effects (Lassiter & Parsons, 2016). Currently, there are three theoretically justified nonlinear-effect models: the threshold effect, the exponential effect, and the quadratic relationship models (Braam & Koenig, 2019; McFarland, 2010). The threshold effects model posits that individuals with high levels of spirituality will experience lower depressive symptoms, whereas those with low or moderate levels of spirituality may have similar levels of depressive symptoms. The exponential effects model suggests that there is a positive linear relationship between spirituality and mental health symptoms, but the magnitude of the increase will grow exponentially (Braam & Koenig, 2019). The quadratic relationship model is either a U-shaped or an inverted U-shaped relationship between spirituality and mental health (Vittengl, 2018). Overall, research is inconclusive regarding relationships between spirituality and mental stress and there are limited studies in this area with Chinese PLWH.

As the religious and spiritual coping theory of Gall and Guirguis-Younger (2013) and the stress and coping theory of Folkman and Lazarus (1984) indicate, when individuals are faced with stressful events (e.g., HIV disease or treatment), religion and spirituality can emerge at the points of cognitive appraisal and coping; that is, PLWH may mobilize a variety of religious and spiritual coping strategies, such as activities that are public (e.g., religious service attendance), private (e.g., prayer), or non-traditional in nature (e.g., meditation). These strategies have the potential to change health outcomes, including mental health, in those practice them. Studies have also shown that PLWH who engage in spiritual activities may obtain social support (emotional and tangible support) from their religious peers or family members who accept their HIV serostatus (Cotton et al., 2009; Szaflarski, 2013). These supports can encourage PLWH to seek care and treatment, remind them to take their medication, and provide emotional support during difficult times (Huang et al., 2021). Additionally, family and peers’ support may help to normalize HIV care and treatment and reduce the stigma and fear that can prevent PLWH from seeking care in the first place (Chen et al., 2018).

Thus, when exploring the relationships between spirituality and mental stress, the confounding effect of social support needs to be considered. In this project, we measured the level of spirituality in lieu of individual religious beliefs. This was done in order to avoid the potential self-disclosure of participant’s religious beliefs, as this information may be accidentally shared with the hospital and/or local officials. This paper examines the relationships among spirituality, depression, and anxiety in Chinese PLWH.

Methods

Participants, Settings and Procedures

A cross-sectional quantitative study design was employed. Between June and July 2013, a convenience sample of 200 PLWH was recruited at Beijing’s Ditan Hospital outpatient clinic, one of the premier treatment centers for infectious diseases in China. Various religions are practiced in different regions and communities in China, e.g., Buddhism, Tibetan Buddhism, Christianity, Daoism, Hinduism, and Islam, but recent national policies have tightened regulations on them (Liang, & Xiao, 2022). According to news sources, because these religions can profoundly impact people in China and the doctrines of religions might contradict official policies, individuals and foreign establishments are prohibited from sharing any religious content online in China (Wang, 2021). Therefore, even though the data was collected in 2013, it is still relevant today because current religious practice in China is even more restricted than when the study sample was taken.

The inclusion criteria for the participants were (a) a confirmed HIV-positive diagnosis, (b) able to fill out the survey, and (c) at least 18 years of age. Cognitively impaired or actively psychotic individuals were excluded. After written informed consent was secured by the researchers, participants were asked to complete a 40-minute paper-and-pencil survey and were reimbursed RMB100 (~USD $15) for their time. The institutional review boards of all involved institutions granted approval for the study.

Measures

Demographics

Data was collected on age, gender, ethnicity, sexual preference, education level, work status, economic status (i.e., where they stood in relation to the poverty line of Beijing in 2013, which was 500 RMB per person per month; Pan et al., 2020), marital status, HIV disclosure to family status, and whether they lived alone or not.

Spirituality

We estimated participants’ spirituality using seven items from two scales. The first five items came from the Duke University RELigion Index (DUREL; Koenig et al., 1997). A sample item reads, “In my life, I experience the presence of the Divine.” The Likert scale responses include: 1 - Definitely not true; 2 - Tends not to be true; 3 - Unsure; 4 - Tends to be true; 5 - Definitely true of me. The other item includes, “How often do you attend church or other religious meetings?” Responses range from 1 - Never; 2 - Once a year or less; 3 - A few times a year; 4 - A few times a month; 5 - Once a week; to 6 - More than once/week. The DUREL scale has an overall score range from 5 to 27 with higher scores indicating more faithful religious practices. The remaining two items came from the brief COPE (Coping Orientation to Problems Experienced) inventory which focuses on practicing spirituality (Us et al., 2015). A sample item reads, “I’ve been praying or meditating.” Responses used a Likert scale that ranged from 0 = I have not been doing this at all to 3 = I have been doing this a lot with higher scores indicating more spiritual practices. The score of these two items ranged from 0 to 6. For all of these seven items, the score ranged from 5 to 33.

Even though a number of DUREL items are religious in nature, many of our participants did not claim any specific religion, but they did have a spiritual practice; therefore, we used the term “spirituality” in this study. People in China may not want to openly disclose their religions but we found that many of them engage in spiritual practices. The brief COPE scale contains two questions that focus on practicing spirituality that were merged with the DUREL items; together, their psychometric properties were tested using factor analysis.

Although the original purpose of these two scales were different and the two scales have different response ranges, after we conducted a series of psychometric analyses, we observed that the seven items load onto a single factor. The Bartlett test of sphericity ${\text{x}}_{\left(21\right)}^2=952.27$, p < .000, rejects the null hypothesis that the items’ variables are not intercorrelated; overall, the Kaiser-Meyer-Olkin measure = 0.86 and the first factor determinacy coefficient = 0.93. Taken together, these results suggested that one latent variable may be sufficient to explain the correlation structure among the seven items, despite coming from different scales. To accommodate the fact that these items had different scaling ranges, we fitted a measurement model using confirmatory factor analysis (CFA) using a first-order, one-factor model. The results (${\text{x}}_{\left(9\right)}^2=12.62$, p=0.18, RMSEA=0.045, CFI=0.996, TLI=0.991) showed our CFA model can nicely reconstruct the covariance matrix among the seven items and fit well to the data. Using these results, we further computed the factor scores with a range of −1.53 to 0.92, to represent the underlying latent variable. The Cronbach’s α was 0.91 for these seven items. Finally, we conducted sensitivity analyses in which we used the raw scores from the original scales and achieved similar conclusions.

Depressed mood

The 21-item revised Beck Depression Inventory (BDI-II; Beck et al., 1996) is a brief clinical depression assessment designed to evaluate the psychological and somatic manifestations of depressive symptoms. Each item is rated from 0 to 3, indicating least to most depressed mood. A total score is calculated by adding the scores of each item and ranges from 0 to 63, where higher scores indicate a greater severity of depressed mood. In the present study, the Chinese version of the BDI-II (Wang & Gorenstein, 2013) demonstrated good internal consistency, with a Cronbach’s α coefficient of 0.92.

Anxiety

The 20-item Self-Rating Anxiety Scale (SAS) was used to assess the frequency of anxiety symptoms (Zung, 1971). Participants reported how often they have anxiety-related somatic symptoms on a 4-point Likert scale (1 = none or a little of the time to 4 = most or all of the time), where higher scores indicated a higher level of anxiety. The Chinese version of SAS used in the present study had satisfactory internal consistency (Cronbach’s α coefficient was 0.82).

Social support

The Medical Outcomes Study–Social Support Survey (MOS-SSS; Yu et al., 2015) is a 19-item instrument that asks respondents to choose how often two kinds of support are available to them, social-emotional support and tangible support. Each item is scored on a 5-point Likert scale from 0 = none of the time to 4 = all of the time. A higher item score reflects a higher level of social support. In this study, Cronbach’s α was 0.96.

Data analysis

The R version 3.2.5 software package was used for data analysis. Univariate analyses were used to obtain an overall picture of the study sample and distribution of the selected variables. To test relations between spirituality and depressive and anxiety symptoms, ordinal linear regression was used. We included the following controlling variables to adjust for estimated relations between spirituality and such symptoms: demographic factors, relationship status, and social support. Given prior empirical findings that targeted relationships will be nonlinear, in Models 1 and 3 we put social support and spirituality into regression analysis (Cotton et al., 2009; Szaflarski, 2013). In Models 2 and 4, we added the squared term of spirituality in the analysis, which was used to test for non-linear relationships between spirituality and mental health variables. By using the squared term, accuracy of estimate predictions may be improved by accounting for this non-linearity (Kutner et al., 2005).

To compare the models with and without squared spirituality and to help model selection, we calculated the R² statistics to indicate how much variation in mental health outcomes can be explained by the models. We formally tested fit to the data between models using Likelihood-ratio tests because the model without squared spirituality was considered nested within the model with squared spirituality. The null hypothesis was that the two models did not differ regarding their fit to the data. Rejection of the null hypothesis means that the model with squared spirituality fits the data better than the model without squared spirituality.

Lastly, to aid in interpretation, we visualized the relationships between spirituality and mental health outcomes. Using the selected models, we predicted mental health outcomes across the response range of spirituality, while holding all variables at their mean values. To accommodate uncertainty inherent in the statistical modelling process, we also computed the 95% confidence intervals for the predicted mental health outcomes.

Results

Among the study sample of 200 PLWH, 81% were male (n=162). The mean age of the sample was 36.9 years (SD=9.1; range = 19–78). The mean score of social support was 2.2 (SD=0.9). The average levels of depressive and anxiety symptoms were 16.3 (SD=11.5) and 37 (SD=8.2), respectively. Participants’ mean spirituality score was −0.02 (SD=0.997). Descriptive data are presented in Table 1.

Table 1:

Sample Characteristics.

	Mean or n	SD or %
Age (Years old)	36.9	9.1
Gender
Female	38	19.0
Male	162	81.0
Ethnicity
Han	179	89.5
Non-Han	21	10.5
Sexual Preference
Same-sex	84	42.0
Opposite-sex	67	33.5
Both-sexes	11	5.5
Refused to Answer	38	19.0
Education
≤High School	35	17.8
High School	64	32.5
> High School	98	49.8
Work Status
Not Working	78	39.8
Work Part-time	22	11.2
Work Full-time	96	49.0
Economic Level
Impoverished	70	35.0
1–200% Poverty Line	67	33.5
> 200% Poverty Line	49	24.5
Missing	14	7.0
Marital Status
Married	69	34.5
Single, Married Before	47	23.5
Never Married	84	42.0
HIV Disclosure to Family
Nobody Knows	69	34.7
Some of Them Know	97	48.7
All of Them Know	33	16.6
Live Alone
No	153	76.5
Yes	47	23.5
Social Support	2.2	0.9
Depressive Symptoms	16.3	11.5
Anxiety Symptoms	37.0	8.2
Spiritual Practice	11.10	6.25

Nonlinear Relationships Between Spirituality and Mental Stress

Depressive Symptoms

Model 1 in Table 2 shows that the linear spirituality term was related to depressive symptoms (adjusted β = 3.14, p <0.01, CI: 0.91–5.38), after adjusting for all the controlling factors (e.g., education and social economic status). In Model 2, both the linear spirituality term (adjusted β = 6.11, p< 0.01, 95% CI: 2.73 – 9.50) and the squared spirituality term (adjusted β = −2.22, p< 0.01, 95% CI: −4.35 - −0.08) were significant, yet in different directions, suggesting that the depressive symptoms rose rapidly when spirituality increased initially; however, the increase slowed down, or even began to decrease, when spirituality passed a certain threshold. The R² value was higher for Model 2 (0.31) than for Model 1 (0.29). The Likelihood-ratio test was also significant $\left({\chi }_{\left(1\right)}^2=5.07,p=0.02\right)$, suggesting that Model 2 fitted significantly better than Model 1 to the data. Using Model 2, we plotted the relationship between spirituality and depressive symptoms. As shown in Figure 1 and Figure 2, this relationship was an inverted U-shaped curve, suggesting that those with low and high levels of spirituality reported fewer depressive symptoms than those with values in between, consistent with the hypothesis that “average is bad, extremes are good.” Depressive levels gradually increased and reached the peak value when spirituality was about 1.5, and when between 1.5 and 3, the depressive levels decreased.

Table 2:

Results of Multivariate Multiple Regression with Depressive and Anxiety Symptoms as the Outcomes.

	Depressive Symptoms1				Anxiety Symptoms1
	Model 1		Model 2		Model 3		Model 4
	β	(95% CI)	β	(95% CI)	β	(95% CI)	β	(95% CI)
Social Support	−5.04**	(−6.82, −3.26)	−5.12**	(−6.88, −3.38)	−3.70*	(−5.06, −2.34)	−3.76**	(−5.11, 2.42)
Spirituality	3.14**	(0.91, 5.38)	6.11**	(2.73, 9.50)	1.49†	(−0.18, 3.16)	3.73**	(1.18, 6.29)
Spirituality squared			−2.22*	(−4.35, −0.08)			−1.67*	(−3.20, 0.14)
R-Square	0.29		0.31		0.24		0.26
Likelihood Ratio Test	$x_{\left(1\right)}^2=5.07$, p = 0.02				$x_{\left(1\right)}^2=5.36$, p = 0.02

^1.All the estimations were further adjusted for all the demographic factors;

^†p < 0.1;

^*p < 0.05;

^**p < 0.01.

Figure 1:

Model-adjusted average scores of depressive symptoms along with factor scores of spirituality.

The thick line represents the mean values of depressive and anxiety symptoms, and the broken lines represent the 95% confidence intervals for the estimated depressive and anxiety symptoms.

Figure 2:

Model-adjusted average scores of anxiety symptoms along with factor scores of spirituality.

The thick line represents the mean values of depressive and anxiety symptoms, and the broken lines represent the 95% confidence intervals for the anxiety symptoms.

Anxiety Symptoms

After adjusting for all the controlling factors, Model 3 in Table 2 showed that the linear spirituality term was only marginally related to anxiety symptoms (adjusted β = 1.49, p <0.10, CI: −0.18–3.16). In Model 4, where a squared spirituality was added to the model, both the linear spirituality term (adjusted β =3.73, p< 0.01, 95% CI: 1.18 – 6.29) and the squared spirituality term (adjusted β = −1.67, p< 0.05, 95% CI: −3.20 – 0.14) were significant, yet in different directions, suggesting that anxiety symptoms rose rapidly when spirituality increased initially. However, the increase slowed down, or even began to decrease, when spirituality passed a certain threshold. The R² value was higher for Model 2 (0.26) than for Model 1 (0.24). The Likelihood-ratio test was also significant $\left({\chi }_{\left(1\right)}^2=5.36,\text{p}=0.02\right)$, suggesting that Model 2 fitted significantly better than Model 1 to the data. Using Model 2, we plotted the relationship between spirituality and anxiety symptoms. As shown in Figure 1 and Figure 2, this relationship was an inverted U-shaped curve, similar to depressive symptoms, suggesting that those with low and high levels of spirituality reported fewer anxiety symptoms than those with values in between, again consistent with the hypothesis that “average is bad, extremes are good.” The depressive levels gradually increased and reached peak value when spirituality was about 1.5, and when between 1.5 and 3, anxiety levels decreased.

Discussion

Spirituality continues to be a strong facet of PLWH’s mental health and coping skill set, especially in the face of all the difficulties an HIV infection entails (Dalmida, 2006). This study suggests that PLWH who report mid-level amounts of spirituality experienced more depressive and anxiety symptoms than those who reported no spirituality at all or the highest levels of spirituality. This inverted U-shaped model suggests that while moderate spirituality is associated with high mental distress, weak or strong spirituality is associated with decreased depressed mood and anxiety. This implies that participation in spiritual activities may help PLWH enhance their sense of meaning and life purpose, and consequently, buffer their mental stress levels.

Of note is that in our study, 89.5% of the participants were of Han ethnicity. In China, the vast majority of the Han have no religious affiliation (Wang et al., 2019). Despite China’s atheistic social environment and the Chinese government’s policy of official atheism (Congressional-Executive Commission on China, 2018, P. 7), similar to the findings among Chinese students (Zhang et al., 2017), this study shows that some individuals do have spiritual beliefs and certain levels of spirituality might have a salutary effect on the mental health among Chinese PLWH.

The findings are consistent with previous studies on spirituality and mental stress (Braam & Koenig, 2019; Vittengl, 2018; Wang et al., 2019). The existing evidence is fragmentary and inconsistent with regards to the connection between spirituality and mental health. For example, the curvilinear effect of religious involvement on depressed mood was represented by a U-shaped effect (Schnittker, 2001; Wang et al., 2019), whereas McFarland (2010) reported an inverted U-shaped relationship between organizational religiosity and death anxiety among older adults (McFarland, 2010). Interestingly, in the same study, the relationship of depressive symptoms and religiosity were presented as an exponential model (McFarland, 2010). Similar to McFarland’s findings, this study presented a significant inverted U-shape relationship between spirituality and depressed mood and anxiety among PLWH in China. These studies highlight that the nonlinear relationship is not universal but culturally specific, and that there is a need to examine the relationship between variations in mental health and the extent of people’s spirituality, rather than simply whether someone is spiritual or not.

As described above, spirituality can provide a sense of meaning and personal connectedness to a higher power or truth and it may enhance physical and mental health through biological, psychological, and social mechanisms (Coleman & Holzemer, 1999; Oji et al., 2017). First, we found that Chinese PLWH who report a low level of spirituality have both less anxiety and depressed mood. This echoes a study by Lockenhoff et al., 2009, that found that PLWH who had a low level of spirituality had better physical health and better self- and family-management of their HIV, as well as having more confidence, and consequently, less anxiety and depression (Lockenhoff et al., 2009). Culturally, if people still looked strong and were able to function, family members usually would not pressure them to seek physical or psychological help. Participants in this study also presented with similar results. When Chinese PLWH present with less depressed mood and anxiety, they do not see the need to practice spirituality. However, when a depressive mood is heightened, there appears a need for Chinese PLWH to seek peace through spiritual practices, resulting in the observed peak of depressive mood for those with a moderate level of spiritual practice. Culturally, for members showing certain levels of anxiety or depression, their family would usually encourage them to seek spiritual help to bring them inner peace, and later may encourage them to seek professional help. When family and friends encourage PLWH to pursue spirituality, such external attention may also increase the PLWH’s levels of anxiety and depressed mood, and may also be one of the reasons why the Chinese PLWH in this study who practiced a moderate level of spirituality presented with higher mental stress. However, those with high levels of spiritual practices presented diminished levels of mental distress. This result is similar to other studies and suggest that PLWH with greater spirituality were dealing with difficult HIV-related situations more peacefully and with a greater sense of coherence between the themselves and their environment, affording them protection against depression and anxiety (e.g., Oji et al., 2017; Steglitzet al., 2012).

In this study, those with a moderate spirituality may have experienced some ambivalence within their value system. For example, other studies have suggested that participants with a moderate level of spirituality may see HIV as a form of punishment from God, exacerbating their guilt and other mental stress, rather than helping them cope with HIV and sustain self-management care tasks (Oji et al., 2017). For those Chinese PLWH devoid of spiritual practice, it may be helpful to introduce them to non-judgmental spiritual practices, if they are open to it. This study’s results suggest that, with some support, Chinese PLWH with spiritual beliefs may have less mental stress.

Addressing mental health needs in HIV care is a top priority (Vittengl, 2018). However, spirituality has not been traditionally incorporated into mental health assessments and practice (Carrico et al., 2006; Dalmida, 2006), in particular among PLWH in China. This study presents several clinical implications for improving the mental health of Chinese PLWH through spirituality. First, under the current climate in China, using secular spirituality (Confucianism, mindfulness meditation, etc) with the PLWH population is acceptable. Also, it is important to assess spirituality levels and pay special attention to those who have moderate levels of spirituality and who could benefit from additional psychological assistance. Second, when developing potential prevention and intervention efforts for depressed mood or anxiety among Chinese PLWH, healthcare providers may consider addressing spiritual needs (Ironson et al., 2017). Addressing spiritual needs may facilitate a sense of meaning, purpose, and hope for Chinese PLWH. Last, people living in a country that is officially atheistic still have spiritual needs; therefore, spirituality should be assessed and considered as part of a holistic program to provide interventions that will enhance the quality of life of PLWH.

Study Limitations

There are several limitations in this study. First, this study shows the associations between spirituality and mental health rather than making a causal inference. Although applications of statistical techniques, such as propensity score matching, may assist in estimating robust causal relations, our smaller sample size and the cross-sectional nature of our study design prevented us from adopting such approaches. As such, determining the causal relationships between mental health and spirituality among this vulnerable population should be of high priority in future studies. Second, the convenience sample and data collection periods raise issues regarding external validity; for example, there is a relatively high proportion of men who have sex with men in this study, which is representative of PLWH in Beijing but not in other areas in China. Third, the study data was collected a decade ago, however, the religious or spiritual practices are still limited within China. Therefore, this paper can still reflect on certain spiritual practices in people living with HIV in China. Last, this study included social support as a confounding factor, but other confounding factors (e.g., personality traits) may also influence the relationship between spirituality and mental health. Therefore, future longitudinal or experimental studies should be conducted to further test these relationships.

Conclusions

Our study results suggest that Chinese PLWH who have low and high spirituality presented with better mental health, specifically, with lower depressed mood and anxiety levels, as compared to those with moderate levels of spirituality. This finding was significant even after considering demographic variables and social support factors. These results point to the impact that spirituality has on the mental health of PLWH in Chinese culture. Our results indicate that HIV care providers should pay attention to patients’ spirituality, with perhaps increased attention to those with a moderate level of spirituality, in order to provide the best care possible for those living with HIV.

Data Accessibility statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.