The Revised Mental Health Inventory-5 (MHI-5) as an ultra-brief screening measure of bidimensional mental health in children and adolescents

Highlights

• There are very few instruments for bidimensional mental health in children and adolescents.

• The Revised MHI-5 shows good psychometric properties similar than previous studies in different cultures.

• The EFA showed the same bidimensional mental health factor structure (well-being and distress) than the original longer version (MHI).

• The Revised MHI-5 is a useful instrument to detect anxiety and depression symptoms in children and adolescents.

• This is the first study, based on bidimensional mental health model, to report data on the psychometrics of the Revised MHI-5 for children and adolescents.

Abstract

The Mental Health Inventory-5 (MHI-5) is a brief, valid, and reliable international instrument for assessing mental health in adults. The aim of the present study is to examine the psychometric properties of the MHI-5 in children and adolescents. A sample of 595 students (10–15 years old) completed the MHI-5 Spanish version adapted for this study, as well as another measure of anxiety and depression symptoms, and a clinical interview as a gold standard. The overall coefficient obtained indicate good internal consistency. A unique factor solution explaining a 53.70% and a two-factor structure explaining 69.20% of the total variance were obtained. The correlations with total and subscale scores of anxiety and depression were significant. A ROC analysis showed good properties as a screening test to predict anxiety and depressive diagnoses in children and adolescents. The Revised MHI-5 presents two essential changes: a simplified 4-point response format and a new factor solution including distress and well-being. These outcomes show that the Revised MHI-5 is a brief, valid, and reliable measure to bidimensionally assess mental health and screening emotional disorders in children and adolescents.

Introduction

The concept of mental health has evolved over the years. World Health Organization defined health as a global state of complete physical, mental, and social well-being, and considering the absence of disease as well as the presence of well-being (WHO, 1948). However, through the years, studies on mental health have mainly focused on the development of concepts that explained psychopathology in terms of psychological distress, neglecting positive health aspects like well-being (Ryff and Keyes, 1995).

Most of the scientific literature published in the past century were associated with negative moods, and only around the 15% of them focused on well-being or similar variables (Salanova, 2008). On the contrary, over the past two decades, the relevance of positive psychological states and their relationship and impact on the evolution of several diseases have increased (Taylor et al., 2000). The reason for this rise is that well-being and positive psychology seem to be a cornerstone for prevention of and recovery from illness (Vázquez et al., 2009).

Therefore, these issues underscore the importance of assessing both mental health sides, according to the Bidimensional Mental Health Model (BMHM; Greenspoon and Saklofske, 2001), that recommends using mental health instruments comprising both facets.

The Mental Health Inventory (MHI; Veit and Ware, 1983) is a validated 38-item instrument developed to assess psychological well-being and distress in the general population, comprising both facets of the BMHM in its structure. The MHI has a brief version called MHI-5 (Berwick et al., 1991), that comprises the five items from the original items pool that better reproduce the total score based on the MHI.

The MHI-5 is as effective as the comprehensive version, and it has the advantage of allowing a faster assessment (Berwick et al., 1991) and it seems sufficiently brief, easy to complete, valid, and reliable for use with different subgroups and in different cultures, including the United States (Ware et al., 1993), Norway (Strand et al., 2003), Denmark (Bültmann et al., 2006), Portugal (Pais-Ribeiro, 2001), Sweden (Sullivan and Karlsoon, 1998), and other European countries (e.g., Bray and Gunnell, 2006).

In contrast with other measures of mental health, this brief instrument has significant advantages. For example, according to Kelly et al. (2008), the MHI-5 performs remarkably well against the longer Mental Health Component Summary (MCS; Ware et al., 2000). Additionally, similar to the General Health Questionnaire (GHQ-12; Goldberg and Williams, 1988), the MHI-5 detects mental health problems, and the latter has the advantage that it can be used not only in mental health surveys, but also in general health and quality of life surveys (Hoeymans et al., 2004). Another benefit of the MHI-5 is that Strand et al. (2003) found it to be a better measure as compared to three versions of the Hopkins Symptom Checklist (SCL-25, SCL-10, SCL-5 versions; Derogatis, 1983).

The interest in the utility of the MHI-5 has grown considerably in recent years, not only because of its briefness, but also because it has exhibited high sensitivity in detecting depressive and anxiety disorders diagnoses in the general population, according the DSM-IV (Rumpf et al., 2001). Further, it is highly accurate in detecting some diagnoses like major depression or panic disorder in primary care patients (Means-Christensen et al., 2005).

Furthermore, the MHI-5 has the extra edge of assessing both psychological well-being and distress, which makes it suitable for using with a non-psychiatric population (Marques et al., 2011b). Some studies have divided the MHI-5 into two forms, MHI-a (for anxiety) and MHI-d (for depression), and it has been noted that both specific versions are as suitable as the 5-item questionnaire for screening. Consequently, the MHI-d and MHI-a can be used for assessing depression (Mitchell and Coyne, 2007) and anxiety (Kroenke et al., 2007), respectively. These data support the recent findings reported by Thorsen et al. (2013) that the MHI-5 was a better predictor of sick leave in the general population as compared to the Major Depression Inventory (MDI; Bech et al., 2001). Thus, the MHI-5 is a valuable instrument for screening depressive disorders in the general population, having high sensibility and specificity (Cuijpers et al., 2009).

Despite the increasing interest in the MHI-5 as a brief and simple mental health instrument for the general population and its use in different countries, there is little research on its use with youth population. Therefore, only one study reports the properties of the MHI-5 with children and adolescents (Marques et al., 2011b). Specifically, in this study with a sample of Portuguese adolescents, the MHI-5 obtained an internal consistency of 0.82, item-total correlations between 0.78 and 0.81, a single factor solution that explained 59.88% of the total variance, communalities from 0.60 to 0.73, and an external validity between 0.41 and 0.56 with other positive mental health measures. These data show that the MHI-5 is a valid and reliable measure for assessing mental health in children and adolescents. The same team carried out a longitudinal study in which MHI-5 was applied three times (time 1, time 2 = 1 year later, and time 3 = 2 years later), showing good estimations of reliability (0.82, 0.83, and 0.82, respectively), as well as correlation coefficients of 0.49 and 0.47 one year and two years later (Marques et al., 2011a).

In the present study, we examined the psychometric properties and the potential utility of a revised version of the MHI-5 to assess mental health in Spanish students. Thus, it is the second study to report data on the validation of the MHI-5 in children and adolescents. As a result, we expected to replicate and find similar psychometric properties to those reported in the study by Marques et al. (2011b).