Some recent research findings question the construct validity of brief single-domain measures of common presenting problems, such as depression and anxiety. For example, results from a factor analysis of the PHQ-9 and GAD-7 scale in patients with traumatic brain injury suggest these tools measure 1 psychological construct that includes depression and the cognitive components of anxiety (eg, worry)13—a finding consistent with those of other tools.14 Similarly, a larger study of 7763 BH patients found that a single factor accounted for most of the variance of the 2 combined measures, with no set of factors meeting the exacting standards used to develop multidimensional routine outcome monitoring.15 These findings suggest that the PHQ-9 and GAD-7 largely overlap and are not measuring different aspects of health as most practitioners believe (eg, depression and anxiety).
In commonly used assessments, multiple-factor analytic studies with high standards have supported the construct validity of domain-specific subscales, indicating that the various questions tap into different constructs of psychological health.14,16,17
Beyond multiple domain–specific indicators, multidimensional routine outcome measurements provide a global total score that minimizes Type I (false-positive conclusion) and Type II (false-negative conclusion) errors in tracking patient improvement or deterioration.18 As one would expect, multidimensional routine outcome monitoring generally includes more items than single-domain measures; however, this comes with a trade-off. If there are specificity and sensitivity concerns with an ultra-brief single-domain measure, an alternative to a core multidimensional routine outcome measurement is to aggregate a series of single-domain measures into a battery of patient self-reports. However, this approach may take longer for patients to complete since they would have to shift among the varying response sets and wording across the unique single-domain measures.
In addition, the standardization/normalization of multidimensional routine outcome monitoring likely makes interpretation easier than referring to norms and clinical severity cutoffs for many distinct measures. Furthermore, increased specificity enhances predictive power and allows BHPs to screen and track other conditions besides depression and anxiety. (It is worth noting that there are no known studies that have looked at the difference in time to administer or ease of interpretation of multidimensional routine outcome monitoring tools vs multiple single-domain measures.)
Two multidimensional routine outcome monitoring tools that cover a comprehensive series of discrete symptom and functional domains are the Treatment Outcome Package12 and Counseling Center Assessment of Psychological Symptoms.16 These tools, which include subscales beyond general depression and anxiety (eg, sleep, substance misuse, social conflict), take 7 to 10 minutes to complete and provide outcome results across 12 symptom and 8 functional dimensions. As an example, the Treatment Outcome Package has good psychometric qualities (eg, reliability, construct and concurrent validity) for adults,12 children,14,19 and adolescents,19 and can be administered through a secure online data collection portal. The Counseling Center Assessment of Psychological Symptoms has demonstrated high construct validity and good convergent validity.16 These assessments can be administered in paper or digital (eg, electronic medical record portal, smartphone) format.20
Continue to: CASE SCENARIO