Exploring poor sleep, mental health, and help‐seeking intention in university students

Abstract

Objective

University students experience common mental health problems such as depression, anxiety, and stress along with poor‐sleep quality. This study explores the relationships between these concepts and help‐seeking intention in a general Australian university student population. The primary aim was to examine the moderating effects of sleep quality on help‐seeking intention for common mental health problems. The secondary aim was to examine sex differences in help‐seeking behaviour.

Method

University students, between 18 and 55 years of age (M = 30.18, SD  = 11.37, N = 117) of which 98 were female, completed an on‐line survey assessing help‐seeking intentions, common mental health problems, and sleep quality.

Results

High levels of depression, anxiety, and stress were significantly associated with decreased sleep quality or decreased help‐seeking intention. A multiple regression analysis predicted that students were more likely to report intention to seek help if they had lower scores of depression, but higher scores of stress. Help‐seeking intention levels were lower for males than females. Poor‐sleep quality was not found to be a moderator of help‐seeking intention.

Conclusion

Although the proposed moderation effect of poor‐sleep quality on the relationship between common mental health problems and help‐seeking intention was not supported, the study advanced our knowledge of university students’ low intention to seek help, despite high scores of poor‐sleep quality. Implications for on‐campus interventions and raising awareness among students about these issues are discussed.

Key words:

• anxiety
• depression
• help‐seeking intention
• poor sleep
• stress
• university/college students

What is already known about this topic?

• University students experience common mental health problems such as depression, anxiety, and stress.

• Some students do not seek help as symptoms increase.

• Poor‐sleep quality is often comorbid to mental health problems.

What this topic adds?

• Increased levels of depression and anxiety reflect decreased sleep quality and decreased help‐seeking intention.

• Help‐seeking intention levels were lower for males than females.

• Shows the importance of investing in mental health literacy campaigns on campuses aimed at increasing students’ insight and decreasing negative perceptions of mental health problems.

Mental health problems and poor‐sleep quality are prevalent amongst the general university student population (Zochil, 2013). Although the comorbid relationship of these problems has been established in the literature (Atalay, 2011), people often seek help for poor‐sleep quality only, unaware of potential comorbid problems impacting their treatment and recovery (Malhi et al., 2014). As universities aim to enhance services for students (Vogel, Wester, Larson, & Wade, 2006), it is beneficial to understand the association between poor‐sleep quality, mental health, and university students’ intention to seek help (referred to as help‐seeking intention) in order to inform preventative and treatment services on campus.

A high percentage of university students suffer from mental health problems. Previous research indicates mental health problems in Australian university students are higher than that of the general population: 67.4% (Stallman, 2010) versus 25.5% (Ryan, Shochet, & Stallman, 2010). Research into the mental health problems of university students in Australia, and how sleep quality and help‐seeking behaviour relate to such problems contributes to an emerging research field (e.g., O'Keeffe, 2013; Simpson & Ferguson, 2012; Vivekananda, Telley, & Trethowan, 2011).

Similar to the general population, when university students present to counselling services, they frequently report symptoms consistent with depression, anxiety, and/or associated stress (Andrews & Chung, 2011), collectively referred to as common mental health problems. Depression, anxiety, and stress can present separately, but often occur together, greatly impairing an individual's quality of life (Collins, Westra, Dozois, & Burns, 2004). These common mental health problems are often comorbid with poor‐sleep quality (Savard, Savard, & Morin, 2010) that can further impact students’ functioning (Brown, Buboltz, & Soper, 2002; Cheng et al., 2012).

Poor‐sleep quality is defined as difficulty initiating, maintaining, and/or achieving restful sleep (Atalay, 2011). It has been estimated that between 16% and 23% of young adults have symptoms of poor‐sleep quality (Taylor et al., 2011). This is particularly prevalent amongst university students compared to other young people (Brown et al., 2002; Cheng et al., 2012). Students often experience poor‐sleep quality as they attempt to balance their competing priorities including academic demands, employment, and social life (Andrews & Chung, 2011; Brown et al., 2002). Although some students who suffer from poor‐sleep quality and common mental health problems may recover independently, many students will likely need to seek help in order to successfully overcome these problems (Collins et al., 2004).

An individual's reported future commitment to seek help for a given problem is known as help‐seeking intention (Rickwood, Deane, Wilson, & Ciarrochi, 2005). Generally, young people report a preference for seeking non‐professional (e.g., partner, friends, and/or family members) over professional (e.g., psychologists, counsellors, and/or general practitioners) help for mental health problems and the same appears true when examining their intention to seek help (Wilson, Deane, Ciarrochi, & Rickwood, 2007). Not all students that need help will seek it (Ryan et al., 2010); thus, it is imperative to understand who is likely to seek help and for what problem in order to target those less likely to seek help of their own accord.

However, seeking help relies on an awareness that there is a problem that warrants intervention (Jorm et al., 1997). A potential lack of awareness of mental health problems (Gibbons, Thorsteinsson, & Loi, 2015; Thorsteinsson, Loi, & Moulynox, 2014) can lead to a need for treatment being unmet (Hess & Tracey, 2013). In support of this, Czyz, Horwitz, Eisenberg, Kramer, and King (2013) found that 66% of university students with mental health problems (e.g., depression, alcohol abuse, suicidal ideation, and/or past history of suicide attempt) reported not engaging in help‐seeking as they did not perceive their symptoms warranted intervention. Additionally, the literature suggests that university students’ help‐seeking intention tends to decrease as symptoms increase (Ryan et al., 2010) and so they are at an increased risk for developing common mental health problems (Orsal, Orsal, Unsal, & Ozalp, 2013; Stallman, 2010). Investigating if help‐seeking intention can be predicted using scores of common mental health problems and sleep quality may assist in clarifying how sleep affects help‐seeking intention in university students with common mental health problems.

Common mental health problems and related comorbid conditions, such as poor‐sleep quality, seem to elicit different help‐seeking behaviours (Malhi et al., 2014). Students may seek help for poor‐sleep quality (Reinhold, 1973), but are less likely to seek help for common mental health problems (Ryan et al., 2010), especially those who suffer from comorbid conditions (Malhi et al., 2014). Past literature suggests that individuals experiencing poor‐sleep quality tend to have more severe psychiatric symptoms (e.g., anhedonia) than individuals who do not (Leblanc et al., 2007) and the relationship between sleep and mental health is complicated (Atalay, 2011). Sleep may influence the strength of the relationship between common mental health problems and help‐seeking intention. However, the underlying mechanisms involved in the relationship between common mental health problems and help seeking appears unclear and to the authors knowledge no studies have explored poor sleep, mental health, and help‐seeking intention together. Thus, an exploration of how poor‐sleep quality may interact in the relationship between mental health problems and help‐seeking intention is warranted.

Additionally, there is a need to investigate differences between males and females. According to BeyondBlue (2015), there is a common myth that males should not seek help by sharing their emotions. A systematic review suggests that being young and male are significant barriers to seeking help for mental health problems (Clement et al., 2015). Rickwood et al. (2005) surveyed 2,721 young people in Australia between 14 and 24 years of age, which indicated that males were less likely to report help‐seeking intention for mental health problems as compared with females of the same age. Reavley, McCann, and Jorm (2012) suggest that these sex differences may be caused by higher perceptions of stigma associated with seeking professional help for mental health treatment in males than females.

To the authors’ knowledge, to date, no studies have explored poor sleep, mental health, and help‐seeking intention simultaneously in a university student population; nor has any study investigated if poor‐sleep quality moderates the relationship between mental health and help‐seeking intention. As such, this study will focus on students with mental health ranging from ‘normal’ to ‘extremely severe’ for depression, anxiety, and stress. The study aims to examine the (1) associations between mental health problems (i.e., depression, anxiety, and stress), poor‐sleep quality, and students’ help‐seeking intention; (2) predictors of students’ help‐seeking intention; (3) moderating effects of poor‐sleep quality on the relationship between students’ common mental health problems and help‐seeking intention; and (4) differences between male and female students’ help‐seeking intention.

METHOD

Participants

In total, 143 university students from multiple universities volunteered to participate in the study. However, data from 26 participants was removed due to incomplete data. Therefore, the final sample consisted of 117 students ranging in age from 18 to 55 years (M = 30.18, SD = 11.37), of which 98 were female (one participant identified as ‘other’). Table 1 shows the key demographic characteristics of the participants.

Table 1. Summary of participant demographics (N = 117)

Characteristic	Participants (n)	Percentage (%)
Education
High school certificate (year 12)	50	42.7
Bachelor degree	28	23.9
TAFE certificate or diploma	8	6.9
Postgraduate diploma	19	16.2
Master's degree	9	7.7
Doctoral degree	3	2.6
Current course/degree of enrolment
Undergraduate	83	70.9
Postgraduate	34	29.1
Enrolment status
Full‐time	63	53.8
Part‐time	54	46.2
Student status
Local	106	90.6
International	11	9.4
Previously saw a mental health professional
Yes	84	70.9
No	33	28.1

Materials

Help‐seeking intention

The General Help‐Seeking Questionnaire (GHSQ; Rickwood et al., 2005) was used to measure students’ reported help‐seeking behaviour. The initial 10 items measured help‐seeking intention over the next 4 weeks for personal or emotional problems, if warranted. Students rated how likely they would be to seek help from a variety of supportive individuals (e.g., ‘partner’ or ‘mental health professional’) on a 7‐point‐Likert scale from 1 (extremely unlikely) to 7 (extremely likely). The final four items of the GHSQ were used to assess participants’ previous professional help‐seeking behaviour (How many visits did you have with the mental health professional?) and experience (How helpful was the visit to the mental health professional?; Rickwood et al., 2005). This scale demonstrated high internal reliability (α = .83) with a non‐clinical sample (n = 218) of young people (Wilson et al., 2007). Although there is no agreed upon measure of help seeking in the literature to date (Li, Dorstyn, & Denson, 2014), the GHSQ has been recommended and employed in a number of studies with a student population (Ryan et al., 2010; Smith & Shochet, 2011; Wilson et al., 2007).

The GHSQ was adapted to be more applicable to Australian university students with ‘school’, ‘teacher’, and ‘classroom teacher’ changed to ‘university/college’, ‘lecturer’, and ‘classroom tutor’, respectively. The tense was changed for one question to ask about future intentions replacing ‘please describe who this was’ with ‘please describe who this would be’. Alpha levels for measures in this study are reported in Table 2.

Table 2. Descriptive statistics and correlation matrix for key variables (N = 117)

Variable	1	2	3	4	5
1. Depression		.66**	.63**	.47**	−.30**
2. Anxiety			.72**	.55**	−.27**
3. Stress				.56**	−.12
4. Poor‐sleep quality					−.21*
5. Help‐seeking intention
M	6.61	8.22	14.53	8.51	3.43
SD	10.74	10.11	9.67	3.91	0.99
α	.94	.92	.89	.69	.72
Possible range	0–42	0–42	0–42	0–21	1–7
Actual range	0–40	0–40	0–42	0–17	1.33–5.56

*p < .05. **p < .01 (two‐tailed).

Common mental health problems

The Depression Anxiety Stress Scales 21‐items (DASS21; Psychology Foundation of Australia, 2014), a shortened version of the original DASS42 (Lovibond & Lovibond, 1995), is commonly used to measure university students’ mental health problems (Simpson & Ferguson, 2012; Zochil, 2013). The DASS21 is a set of three scales: depression, anxiety, and stress, with seven items for each scale. Students rated each item (e.g., ‘I found it hard to wind down’) on a 4‐point‐Likert scale of 0 (did not apply to me at all) to 3 (applied to me very much) on how they felt over the past week. Each score is doubled with each scale ranging from 0 to 42 (extremely severe). The DASS21 has high specificity, clinical and research utility, and discriminates well among the three conditions (Dozois & Dobson, 2010). All scales have high internal consistency: depression (α = .88), anxiety (α = .82), and stress (α = .90; Henry & Crawford, 2005).

Poor‐sleep quality

The Pittsburgh Sleep Quality Index (PSQI) is a measure of sleep quality and disturbance over a 1‐month period (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Students rated statements (e.g., ‘Cannot get to sleep within 30 min’) on a 4‐point‐Likert scale from 0 (not during the past month) to 3 (three or more times a week). Nineteen items are scored to provide a global PSQI score, ranging from 0 to 21 with a cut‐off score of 5 or above to indicate poor‐sleep quality (Buysse et al., 1989). The participant's bed partner, if applicable, rated an additional five items about the participant's potential sleep disturbances (e.g., ‘loud snoring’) over the past month. The PSQI is a well‐established measure with demonstrated reliability and validity (see systematic review by Mollayeva et al., 2016) showing α from .70 to .83.

Procedure

The University's Human Research Ethics Committee granted ethics approval (HE15‐030). Potential participants had access to a flyer with a brief explanation of the study and a URL link. The link allowed access to the participant information sheet and a confidential on‐line survey via Qualtrics’ secure site (http://www.qualtrics.com). Three main portals were used to recruit participants. First, potential participants were emailed an invitation to complete the survey and were encouraged to forward the email to other students in their social network. Second, the flyer was made available on social network sites (e.g., http://www.facebook.com) and the University's e‐learning platform. Third, participants were recruited via coordinators of teaching units at other universities who agreed to participate.

Participants gave consent after reading a study information sheet and consent form by clicking a ‘proceed’ button thus starting the on‐line survey. The survey commenced with demographic questions followed by the three previously described questionnaires, which were presented in random order to reduce potential priming and order effects. At the conclusion of the survey, participants were thanked for their participation and could opt to enter a prize draw to win one of three $50 vouchers.

Statistical analysis

All analyses were carried out using SPSS version 23 and PROCESS version 2.13.2, model 1 (Hayes, 2015). Effect size was reported and a significance level criterion of .05 was employed. The interpretation of results focuses on the magnitude of effect sizes (e.g., r and g) while also considering statistical significance. The model for predicting help‐seeking intention included depression, anxiety, stress, and poor‐sleep quality as predictors; no confounders were adjusted for in the model. Moderators were tested using Model 1 in PROCESS using 5,000 bootstrap samples.

RESULTS

Eighteen students (15.4%) in the sample reported good sleep quality and 99 (84.6%) reported poor‐sleep quality. The severity range for depression, anxiety, and stress is shown in Table 3. The findings suggest that 18.0%, 20.5%, and 14.6% of students suffered severe or extremely severe depression, anxiety, or stress, respectively. Table 2 shows the associations between the key outcome measures. Help‐seeking intention was significantly associated with depression (r = −.30) and anxiety (r = −.27). Contrary to expectation, help‐seeking was not positively associated with poor‐sleep quality (r = −.21). Poorer sleep quality was associated with poorer mental health (rs > .46, ps < .01 two‐tailed).

Table 3. Frequency (and percentages) severity range for the Depression, Anxiety, and Stress Scales (21‐items)

Severity range	Depression	Anxiety	Stress
Normal	73 (62.4)	75 (64.1)	72 (61.5)
Mild	15 (12.8)	6 (5.1)	13 (11.1)
Moderate	8 (6.8)	12 (10.3)	15 (12.8)
Severe	7 (6.0)	7 (6.0)	10 (8.6)
Extremely severe	14 (12.0)	17 (14.5)	7 (6.0)

Help‐seeking intention was predicted by mental health problems and poor‐sleep quality with 13% of the variance explained, see Table 4. Examining the beta coefficient depression and stress seem the strongest predictors in the given model.

Table 4. Predicting help‐seeking intention from depression, anxiety, stress, and poor‐sleep quality

Predictor	b	CI95% for b		β	r	sr^2
		Lower	Upper
Depression	−0.03	−0.05	−0.00	−0.27	−.30	.04
Anxiety	−0.02	−0.05	0.00	−0.23	−.27	.02
Stress	0.03	0.00	0.06	0.28	−.12	.03
Poor‐sleep quality	−0.03	−0.08	0.03	−0.11	−.21	.00

Note. Fit for model R² = .13, adjusted R² = .10, F(4, 112) = 4.30, p = .003. CI = Confidence interval.

Three separate moderator analyses were conducted to examine if the relationship between each common mental health problem (i.e., depression, anxiety, and stress) and help‐seeking intention was moderated by poor‐sleep quality. No significant interactions were found (sr² < .02).

Help‐seeking intention was lower for males (M = 2.99, SD = 0.97) than females (M = 3.51, SD = 0.98), t(114) = 2.06, p = .021 (one‐tailed), g = 0.53 [0.02, 1.04].

DISCUSSION

Overall, the results of this study are consistent with the literature. Thus students who reported experiencing more severe symptoms of common mental health problems were less likely to report help‐seeking intention (Ryan et al., 2010). Further, poor‐sleep quality was more prevalent in students with higher scores of depression, anxiety, and/or stress (Taylor et al., 2011). Low levels of depression, but high levels of stress, predicted greater likelihood to report help‐seeking intention, further suggesting that students are more likely to seek help for minor issues, but less likely to seek help as mental health symptoms worsen (Ryan et al., 2010). Furthermore, male university students were less likely than females to report help‐seeking intention (Ciarrochi, Wilson, Deane, & Rickwood, 2003). The difference between males and females for help‐seeking may be due to perceived stigma associated with seeking help. Reavley et al. (2012) found that male students and students with severe mental health symptoms reported experiencing more stigma associated with help seeking for mental health problems compared with female students and/or students with less severe symptoms. Thus stigma may play a significant role as a barrier in preventing students from seeking help (Eisenberg, Golberstein, & Gollust, 2007). One way to address mental health‐related stigma is by investing in mental health literacy campaigns on campuses aimed at increasing university students’ insight and decreasing negative perceptions of mental health problems (Smith & Shochet, 2011).

In this study, students who reported poor‐sleep quality were less likely to report help‐seeking intention (Malhi et al., 2014; Reinhold, 1973). Additionally, poor‐sleep quality was not found to moderate the relationship between students’ common mental health problems and help‐seeking intention. Examining potential moderating models of the mental health effects on help‐seeking intentions indicated only small effects. Future studies need to investigate further these significant aspects of sleep quality, for example, help‐seeking behaviours in relation to sleep disturbance.

Overall, the results of this study suggest that students do not have high intention to seek help despite experiencing poor‐sleep quality or low to moderate mental health symptoms. It is unclear what the specific reasons are for not seeking help (Ryan et al., 2010). Previous research has shown that lack of insight is a barrier to help seeking (Czyz et al., 2013) as a person needs to be aware there is a problem that requires intervention in order to make the decision to seek help (Jorm et al., 1997). Awareness campaigns aimed at improving insight to accurately identify common mental health problems (Yap, Reavley, & Jorm, 2014) as well as poor‐sleep quality (Brown et al., 2002) are recommended. Such campaigns promote effective therapeutic strategies on campus, improve awareness amongst students, and time to receive treatment (Barney, Griffiths, Jorm, & Christensen, 2006; Blumenthal & Endicott, 1996; Lauber, Nordt, Falcato, & Rössler, 2003; Reavley et al., 2012; Smith & Shochet, 2011).

Limitations of this study include sample size and over‐representation of females thus limiting the generalisability of the results. Future research could expand on this study by including a larger sample size; include a better representation of males; focussing on students with mental health problems; and by directly assessing help seeking for sleep. Further, as university students are a part of the on‐line generation (Hinsch & Sheldon, 2013), they may be more likely to turn to this modality for their problems (Ryan et al., 2010). On‐line therapy is a growing field gaining increasing attention in the literature of university student intervention provision (Zochil, 2015). Therefore, future research could investigate on‐line platforms for seeking help.

The present cross‐sectional design limits conclusions on causality and directionality among the variables (Meltzoff, 2011) and causal modelling (Maxwell & Cole, 2007). Additionally, as increasing symptoms of common mental health problems (Stallman, 2010) and poor‐sleep quality (Lamberti, 2012) are experienced towards the end of semester as students’ academic demands increase, a longitudinal approach could assist in further explaining the temporal relations between the variables.

In conclusion, this study attempted to expand on previous mental health and help‐seeking research by examining these simultaneously with sleep and exploring how poor‐sleep quality may impact the relationship between depression, anxiety, and stress with help‐seeking intention within a university student population. Although the moderation effect was not supported, it advanced our knowledge of university student low intention to seek help for common mental health problems despite high scores of poor‐sleep quality. The negative relationship between students’ intention to seek help and mental health should be seriously considered in future mental health literacy campaigns and on campus mental health interventions.