Abstract
This study examined the relationship between internalizing behaviors (measured as anxious/depressed, withdrawn, and somatic problems) and lifestyle-related, modifiable factors as substance use, sleep, and healthy behaviors in a sample of 191 college students in training for social work and other helping professions. The results indicate that participants with lower scores on internalizing behaviors had fewer sleep disturbances, higher scores on healthy behaviors and lower substance use. The implications for social work education, research, and practice are considered, including exploring the relationships between internalizing behaviors and different components of a healthy lifestyle, such as exercise, sleep, and substance use.
The prevalence of internalizing symptoms such as depression, anxiety, withdrawal, and somatic complaints (Achenbach and Rescorla, 2001; American Psychiatric Association, 2013) is staggering. Globally, 28 percent of individuals with anxiety disorder experience severe role impairments at home and at work, in relationships, and in social life (Ruscio et al., 2017). Nationally, every fifth US adult lives with an anxiety disorder, 9.5 percent experience symptoms of depression (Kessler et al., 2003), and 56 percent share symptoms of anxiety and depression (Clark, 1989). According to the Centers for Disease Control and Prevention (CDC, 2010), the highest prevalence of current depression was found in Mississippi (14.8%). Prior research found significant correlations between depression and other student health problems and the lack of health-promoting behaviors (Doom and Haeffel, 2013). The aim of this study is to expand the knowledge base on the correlates of a wide range of internalizing problems in a sample of social work, counseling, and other helping professional majors at a public university in America’s Deep South.
Currently, there is a dearth of information on the health behaviors and health status of students majoring in helping professions such as social work. Healthy behaviors can encompass a range of actions such as physical exercise and diet, social support, health responsibility, stress management, and life appreciation (Chen et al., 2003). Healthy behaviors tend to reduce stress (Ng and Jeffery, 2003), mortality (Kvaavik et al., 2010; Stringhini et al., 2010), and medical multimorbidity (Loprinzi, 2015).
Sleep is related to the aforementioned healthy behaviors. Children who had poor sleep as early as ages 3–8 were at statistically significant risk of involvement with alcohol, cigarette, and marijuana use in adolescence and more externalizing and internalizing symptoms (Pieters et al., 2015; Wong et al., 2009). Poor sleep has been associated with disordered patterns of eating (Chardon et al., 2016), depression (Fernandez-Mendoza et al., 2016), anxiety disorders (Kahn-Greene et al., 2007; Roberts, 2017), and poor academic performance (Eliasson et al., 2002). Chen et al. (2006) also found that sleep was positively associated with healthy behaviors.
Previous studies indicate that substance abuse may be related to internalizing problems. Although these conditions often co-occur developmentally, internalizing symptoms can be detected as early as age 3 (Burlaka et al., 2015) while substance use typically begins around adolescent age (Burlaka, 2017; Kessler, 2004). However, Homman et al. (2017) found that for male students, alcohol problems first predicted internalizing problems and then internalizing problems would predict alcohol problems, whereas for female college students, a unidirectional relationship was found with alcohol problems predicting internalizing behavior problems.
In addition to gender, ethnicity can influence the development of internalizing problems among college students. For example, Riolo et al. (2005) reported much higher rates of major depression among White individuals compared to Mexican Americans and African Americans. These results were consistent with findings from the National Survey of American Life, in which the lifetime prevalence of major depressive disorder was 17.9 percent for Whites, 12.9 percent for Caribbean Blacks, and 10.4 percent for African Americans (Williams et al., 2007). In addition, being of a non-White race has been a risk factor for depression in several studies reviewed by Musliner et al. (2016).
Age is another important factor that can impact the development of internalizing problems. Previous research has shown that levels of depressive symptoms peak during late adolescence and then decrease in early adulthood (Adkins et al., 2009). However, high levels of depression and anxiety can persist into adulthood for some individuals. A meta-analysis of 25 studies revealed that these individuals tend to share such characteristics as being female and having a lower socioeconomic status (Musliner et al., 2016).
The research on mental health characteristics of the future helping professionals, such as social workers, is limited (Horton et al., 2009). Horton et al. (2009) reported that 34 percent of social work students who participated in the study reported high scores on depression. In another study, Ting (2011) found that 50 percent of social work students have reached a clinical cutoff for depressive symptoms. Prior research has documented that future helping professionals avoid seeking help because of stigma, fear, lack of trust, concerns with confidentiality and quality of care, and the lack of culturally appropriate services (Burlaka et al., 2014b; Ting, 2011). Other barriers to service utilizations were low availability, inconvenient location, and inconvenient hours of mental health services (Burlaka et al., 2014a).
Given the previous literature, we hypothesized that students with higher scores on health-promoting behaviors, including quality sleep (Beiter et al., 2015), regular exercise, healthy diet (Doom and Haeffel, 2013), and lower scores on substance use (Hilt et al., 2017; Homman et al., 2017), would report fewer internalizing symptoms; also, that students who are African American (Riolo et al., 2005; Shim et al., 2012; Williams et al., 2007) and of older age (Riolo et al., 2005) would have lower scores on internalizing problems.
Methods
Participants
This study of internalizing problems is part of a comprehensive study investigating risk and protective factors associated with student wellbeing and success. Data were collected between May 2017 and May 2018, during proctored small-group lab sessions with the help of trained research assistants. Institutional Review Board approval and a Certificate of Confidentiality from the National Institutes of Health (NIH) were received prior to data collection, and all participants signed informed consent forms and Family Educational Rights and Privacy Act (FERPA) releases.
The study used a cross-sectional sample of 191 predominantly young adult participants aged 19–57 (M = 26.45, SD = 8.61), who were currently attending Bachelor or Master of Social Work, health promotion, or counseling programs at a public university in the US rural south. The majority of participants (95%) were females, 45 percent European American, 50 percent African American, and 5 percent Asian American, Hispanic American, or other ethnicity.
Measures
Sociodemographic characteristics
All participants answered questions about age, sex, and race.
Internalizing problems
We used items from the Adult Self-Report (ASR; Achenbach and Rescorla, 2003) to measure internalizing problems. The ASR measures behavioral, emotional, and social problems among adults aged 18–59. Participants answered questions on a three-point scale from 0 = not true to 2 = very/often true. The internalizing problems scale includes 18 Anxious/Depressed items (e.g. ‘I worry about my future’, ‘I feel worthless or inferior’), 9 Withdrawn items (e.g. ‘I would rather be alone than with others’), and 12 Somatic items (e.g. ‘I feel dizzy’, ‘I have rashes or other skin problems’; possible range 0–78). The alpha coefficients were .90 for the Anxious/Depressed subscale, .66 for the Withdrawn subscale, .83 for the Somatic Complaints subscale, and .92 for the Internalizing Behaviors Scale.
Substance use
For the purposes of this study, we measured substance use with two items from the ASR (Achenbach and Rescorla, 2003): ‘In the past 6 months, about how many times per day did you use tobacco (including smokeless tobacco)? __ times per day’, and ‘In the past 6 months, on how many days were you drunk? __ days’. Participants’ answers were then dichotomized to reflect their tobacco use and/or alcohol intoxication during the past 6 months (0 = did not use tobacco or alcohol, 1 = used tobacco or alcohol).
Healthy behaviors
Participants answered 21 items from the Adolescent Health Promotion Scale (AHPS; Chen et al., 2014) measuring nutrition (e.g. ‘I choose foods without too much oil’), social support (e.g. ‘I talk about my concerns with others’), life appreciation (e.g. ‘I usually think positively’), health responsibility (e.g. ‘I watch my weight’), stress management (e.g. ‘I make an effort to determine the source of my stress’), and exercise habits (e.g. ‘I exercise rigorously for 30 minutes at least 3 times per week’). Participants indicated the frequency with which they engaged in those behaviors on a 5-point Likert-type scale, ranging from 1 ( ‘never’) to 5 ( ‘always’), with the possible range being from 21 to 105, with higher scores indicating better health promoting behaviors. The reported Cronbach’s alpha for the AHP-SF was .91 (Chen et al., 2014). In this study, the internal consistency alpha was .86.
Sleep disturbances
Ten questions from the Pittsburg Sleep Quality Index (PSQI; Buysse et al., 1989) measuring sleep disturbances were used in this study. Participants reported on circumstances such as coughing or snoring, having to go to the bathroom during the night, not being able to breathe comfortably, feeling pain, or having bad dreams that prevented them from having a good sleep in the past 30 days (e.g. ‘How often have you had trouble sleeping because you wake up in the middle of the night or early morning?’) on a 4-point Likert-type scale (0 = ‘not during the past month’, 3 = ‘3 or more times per week’). We used the row mean function to compute the Sleep Disturbances subscale from 10 items. Hence, participants’ answers ranged from 0 to 3, with higher scores indicating more disturbances. Previous studies reported good validity and reliability of the PSQI with Cronbach’s α = .80 (Backhaus et al., 2002; Carpenter and Andrykowskia, 1998). In our study, the internal consistency of the PSQI was .64.
Statistical analyses
All analyses were conducted using the statistical software Stata, version 14.2 (StataCorp, 2013). Multivariate ordinary least-squared regression analysis was used to examine the relationship between internalizing behaviors (the outcome variable) and health-promoting behaviors, sleep disturbance, and substance use, controlling for gender and age. Results of this analysis are presented in Table 2. Statistical significance was set at α = .05. Standardized betas were computed in the regression model to compare the strength of influence of different independent variables on internalizing problems.
Results
Every sixth student in this sample (16%, n = 32) met a borderline cutoff for internalizing behavior problems (t-score ⩾ 60); among them, 12 percent (n = 25) of participants were in the clinical range (t-score ⩾ 63; Achenbach et al., 2003). The average score on health behaviors was 70.14 (SD = 12.95), and the mean score on sleep disturbances was .98 (SD = .52). The average score on substance use was .45 (SD = .50). Intercorrelations between study variables are presented in Table 1.
Intercorrelations between study variables.
p < .05; **p < .01; ***p < .001.
The ordinary least squares linear regression results (Table 2) suggested that students were less likely to report internalizing behaviors if they had fewer sleep disturbances (B = .42, p < .001), had higher AHPS health behavior scores (B = –.26, p < .001), were older (B = –.17, p < .01), African American (B = –.14, p < .05), and reported lower substance use (B = .12, p < .05).
Relationship between internalizing behaviors, sleep disturbances, healthy behaviors, substance use, age, sex, and race (N = 191).
p < .05; **p < .01; ***p < .001.
The sex of participants was not statistically associated with internalizing behavior problems. The model explained 32 percent of the variance in ASR internalizing scores, F(7, 183) = 12.25, p < .001.
Discussion
This study begins to fill a gap in the social work knowledge about the correlates of internalizing problems among young adults who are in training to become helping professionals. Every sixth student in our study reported a borderline internalizing behavior cutoff score. This result suggests that depression, anxiety, and somatic symptoms might be very common among those in training to become helping professionals. Among all predictor variables, we found the strongest association between internalizing behaviors and sleep disturbances.
Consistent with previous studies on sleep and anxiety, social phobia, and depression in other socio-cultural contexts, students in the US rural south with more sleep disturbances were at risk of experiencing internalizing symptoms (Fernandez-Mendoza et al., 2016; Kahn-Greene et al., 2007; Roberts, 2017; Stein et al., 1993). Future research needs to focus on campus-based treatment options as well as program policies that could help social work students adopt more productive sleep strategies.
As predicted, students who reported a higher frequency of healthy behaviors had fewer internalizing behaviors (Doom and Haeffel, 2013). Heart disease and other non-communicable diseases remain the leading cause of death globally (Global Burden of Diseases [GBD] 2016 Causes of Death Collaborators, 2017), and healthy eating and exercise are directly linked with such diseases. In addition, the finding that such healthy behaviors as better stress management can reduce internalizing problems is particularly important for young adults because of the higher risk of suicide in this age group related to psychological distress (Kim and Burlaka, 2018).
In our study, older students were less likely to suffer from internalizing psychopathology. This finding is consistent with earlier research on adolescents (Burlaka et al., 2017) and adults (Musliner et al., 2016). It is possible that older students initially had higher levels of internalizing problems and over time were able to get help and recover. In addition, it is possible that older students were exposed to more social work clinical courses that offered helpful tools for these students to cope with anxiety and depression. A follow-up study with social work students needs to be implemented to discern the relationship between age and internalizing behaviors.
Consistent with previous research (Riolo et al., 2005; Williams et al., 2007), African American students in this study had lower scores on internalizing problems. Two explanations can be offered for this finding. First, feelings of closeness to one’s family and friends have been linked with decreased depression (Shim et al., 2012). It is not uncommon for African Americans to be supportive and unselfish in friendships and in family relationships (Harris and Molock, 2000). In addition, these data from the Deep South may be indicative of the beneficial role of church involvement and spiritual support and coping previously reported for the African American people who live in the coterminous United States (Krause and Hayward, 2012). Indeed, the church environment, praying, and listening to gospel music may provide for many individuals a viable tool to deal with complicated life situations. These mechanisms are likely to alleviate stress and decrease depression and anxiety. The close interactions and support vehicles can have an impact on the progression of mood disorders similar to the ones in the professional therapeutic relationships. Although not tested in this study, another explanation for the lower internalizing scores among African American participants can stem from socially desirable responding and underreporting that have been found in other research (Bardwell and Dimsdale, 2001).
Substance use was another statistically significant predictor of internalizing behaviors in our study. This finding is consistent with Homman et al.’s (2017) recent research with students. It is possible that some students tend to cope with life and academic stressors by smoking and consuming alcohol (Churakova et al., 2017). However, avoiding dealing with the root causes of problems may further exacerbate internalizing behaviors in the long run. Longitudinal results from the Ontario Child Health Study (Georgiades and Boyle, 2007) indicated that tobacco use has led to deteriorated functioning, poorer physical health, lower life satisfaction, lower income, and higher depression.
Conclusions and implications
This study is directly relevant to social workers’ education and to the social work profession. When discussing the role of the reflexive practice in social work practice, education, and research, Ruch (2002) challenges the monopolistic role of the empirical, scientific process as such that tends to impose ‘only one external, objective truth’ (p. 202) in the of construction of knowledge and development of skills. Such an approach underplays the role of experiential and intuitive knowledge. The Social workers’ overall competence typically includes tacit knowledge (knowledge based on personal experiences), explicit knowledge (knowledge taught formally), and personal characteristics (e.g. stress toleration and sense of humor) that may enhance or hamper task performance (Koskinen, 2003). This article underlines the importance of the two dimensions – tacit knowledge and personal characteristics – that need to be actively integrated in social work education and practice. It is critically important to help social work students recognize and integrate the lived experiences of internalizing behaviors such as anxiety and depression.
The National Association of Social Workers (NASW) Code of Ethics (NASW, 2017: Section 4.05) highlights the importance of acknowledging personal impairments such as personal problems, substance abuse, or mental health difficulties, which may interfere with workers’ professional judgment and performance. It is then the ethical responsibility of social workers to immediately seek help. The findings from our study may also be useful for social work programs that might focus on helping students recognize the impact of healthy behaviors on mental health in their own life as well as the importance of self-care and seeking help. Likewise, internalizing behaviors that include excessive anxiety, depression, worry, fear, and withdrawal (Beauchaine and Hinshaw, 2008) can pose a significant risk of undermining students’ effective functioning in the classroom, practicum, and later when practicing the social work profession. The study findings underscore the importance of enriching social work curricula with elements integrating students’ emotional awareness, strategies, and tools to reduce the detrimental effects of anxiety and depression, as well as stigma associated with help-seeking.
Limitations
Although this study contributes to the literature on factors associated with internalizing problems among students in social work and other helping professions, it is important to mention some limitations. First, the study used a cross-sectional design. Without longitudinal data, it cannot be determined whether individuals’ healthy behaviors, less interrupted sleep, and limited substance use led to low levels of internalizing problems, or whether not having substantial internalizing problems resulted in choosing healthier lifestyles. Another limitation is that we relied on self-reported data obtained through a single administration of a multiple-item survey. Although this method is common in cross-sectional literature, it may introduce common method biases (Podsakoff et al., 2003), such as trying to choose behaviors that are regarded as healthy, selecting answers across constructs under the impact of a certain mood, or choosing responses under the influence of the item context.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for the preparation of this article was provided in part by the School of Applied Sciences at the University of Mississippi.
