Rating Scales
Andrea K. Maikovich-Fong PhD
Alison Leary PhD
Brent Collett PhD
Kathleen Myers MD, MPH
Rating scales are instruments that provide rapid assessment of behaviors or psychological dimensions, and that yield numerical scores with relatively straightforward scoring and interpretative guidelines. They can be used without extensive training in both research and clinical applications. As psychiatric disorders do not yet have identified genes, serum tests, or functional imaging profiles, rating scales have assumed an important role in identifying criterion symptoms for making a diagnosis, establishing the severity of a disorder, and tracking symptoms over time. This chapter introduces several rating scales pertaining to childhood disorders that are frequently treated in general psychiatric practice but that are also readily applicable to other clinical settings. From the large number of available scales, those presented in this chapter were selected on the basis of their frequency of use in clinical practice and the adequacy of their psychometric properties. Whenever possible, emphasis is on scales that are available in the public domain. Information on obtaining these scales is included at the end of the chapter. To avoid redundancy, the individual scales are not also cited in the bibliography, although relevant other citations are included.
Functioning of Rating Scales
Advantages and Disadvantages of Rating Scales
Myers and Winters have reviewed the role and functioning of rating scales in child and adolescent mental health research and clinical practice. An abbreviated overview is provided in Table 3-1. One of the most common ways in which people describe a child’s difficulties is via comparisons. For example, parents might express concerns that their younger daughter is more aggressive than their older daughter was at that age, teachers might report that a child’s moods are more labile than those of his classmates, and a teenager might worry that she has a harder time focusing at school than her friends do. One of the overarching advantages of rating scales is that they allow for multiple such comparisons in structured, scientific ways. First, ratings scales can be used to compare youth to other youth. For example, a youth’s score on a selfreport scale is often derived by comparing his or her answers to those of some comparison group—ideally a large, representative sample of same-aged, same-gendered peers. On the Child Behavior Checklist (CBCL) (described later in text), a teenaged female will have a “clinically significant” internalizing score if her self-reported symptoms of internalizing problems are significantly above those of same-aged females from a nationally representative sample. If a caregiver or teacher is the informant, then this adult’s answers are compared to the answers from a sample of parents or teachers of same-aged youth in order to derive scores. Second, rating scales can be used to compare a youth’s functioning to his or her own functioning at previous points in time. For example, if a child experiences a major stressor, such as invasive treatment for cancer or the loss of a parent, comparisons between pre- and poststressor scores can assist with understanding how the event affected the youth’s functioning from
multiple peoples perspectives. Rating scale scores taken at different points in time can also help establish a trajectory for a child’s functioning and show, for example, whether the youth is experiencing increasing or decreasing symptoms. Increasing symptoms, even if not yet at the “clinically significant” level, may suggest that the child is at risk for later problems and that early intervention is warranted. When used to establish trajectories, rating scales can provide a cost-efficient means of documenting evidence-based treatment effects. Third, rating scales can be useful in comparing a youth’s functioning across different settings. Youth often function differently at home and at school, for example, so comparing a teacher’s perspective to that of a primary caregiver can be critical to understanding the pattern and nature of a youth’s difficulties. The Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) requires evidence of impaired functioning in multiple settings in order to raise symptom clusters to the level of diagnosis. Therefore, multiple perspectives help to establish the contextual variation and context of the youth’s symptoms and support formal diagnoses.
multiple peoples perspectives. Rating scale scores taken at different points in time can also help establish a trajectory for a child’s functioning and show, for example, whether the youth is experiencing increasing or decreasing symptoms. Increasing symptoms, even if not yet at the “clinically significant” level, may suggest that the child is at risk for later problems and that early intervention is warranted. When used to establish trajectories, rating scales can provide a cost-efficient means of documenting evidence-based treatment effects. Third, rating scales can be useful in comparing a youth’s functioning across different settings. Youth often function differently at home and at school, for example, so comparing a teacher’s perspective to that of a primary caregiver can be critical to understanding the pattern and nature of a youth’s difficulties. The Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) requires evidence of impaired functioning in multiple settings in order to raise symptom clusters to the level of diagnosis. Therefore, multiple perspectives help to establish the contextual variation and context of the youth’s symptoms and support formal diagnoses.
TABLE 3-1 Essentials of Rating Scales in Data Collection | ||||||||
|---|---|---|---|---|---|---|---|---|
|
Rating scales can also be useful because they allow for a systematic investigation of a youth’s perspective on his or her own functioning. Studies suggest that adults, including caregivers, are not always aware of the extent and exact nature of a child’s symptoms, particularly “internalizing” symptoms such as depression or anxiety. Youth may also be more willing to disclose distress on a written rating scale than in a face-to-face interview. This may be particularly true among certain groups of youth, such as sexually abused males.
There are also several disadvantages to using rating scales. Most of these stem from false assumptions about rating scales or a poor understanding of their role in an overall assessment. For example, problems arise when physicians, psychiatrists, and other evaluators assume that an elevated score on a measure automatically equates to a diagnosis. The data collected with rating scales should be considered only in conjunction with other evaluation procedures and clinical judgment. Similarly, the text printouts generated by computer scoring systems should be
treated with caution. When selecting a scale, it is important to evaluate its developmental suitability, particularly when working with special populations (e.g., those with developmental delays or very young children). A scale is not necessarily suited for all children, even if they fall into the age range for which the scale was designed. If, for example, a developmentally delayed 8-year-old is given a self-report measure designed for 8-year-olds, he or she may not understand the task at hand and misinterpret items and/or circle answer choices randomly, yielding an invalid score.
treated with caution. When selecting a scale, it is important to evaluate its developmental suitability, particularly when working with special populations (e.g., those with developmental delays or very young children). A scale is not necessarily suited for all children, even if they fall into the age range for which the scale was designed. If, for example, a developmentally delayed 8-year-old is given a self-report measure designed for 8-year-olds, he or she may not understand the task at hand and misinterpret items and/or circle answer choices randomly, yielding an invalid score.
Unfortunately, many popular scales do not have sufficient data to determine how well the scale discriminates clinical groups, how reliable scores are, and how suitable the scale may be for a particular youth or clinical population. For example, if a scale has been developed with a school-based sample, it may not function as well with a clinic-based sample. This is a particular problem when it comes to using scales with minority and immigrant youth, as many of the most popular scales were validated on samples of Caucasian children. Simply translating a scale into another language does not ensure that the meaning of a score will equate across cultures.
Rating scales rely on the reports of youth and those individuals in their environment who know them well, and are subsequently only as useful as the informants are honest and perceptive. Individual and situational factors can affect a youth’s score on rating scales. Youth who seek social acceptance may underreport symptoms (“denial” or “lying”), whereas those who feel overwhelmed or who are seeking access to diagnosis-based services may overreport symptoms (“faking” or “malingering”). Similarly, on observer-rated scales, adult respondents may convey their own distress and frustration with the youth by exaggerating symptoms, or may minimize their child’s problems in an effort to protect him or her. Caregivers who are involved in custody disputes, or who are being monitored by child protective services, may have other reasons for inflating or deflating their child’s scores, and caution should be used when interpreting scores from scales completed by parents under circumstances such as these.
Finally, it is important to be aware that there is typically low agreement among adults who rate a youth in different settings, such as teachers and parents, and only moderate agreement among adults in the same environment, such as mothers and fathers. Such disparities reflect both subjective differences in reporters’ perceptions and objective variations in youths’ behavior across settings. Not surprisingly, correlations between youths’ self-reports and adults’ reports are low. These disparities highlight the need to gather information from multiple informants across settings.
Psychometric Properties of Rating Scales
Rating scales are a means of measuring a construct, and all measurement is subject to error. Examining the psychometric properties of a measure is one way to examine this error in order to determine whether a measure is appropriate for use. Several psychometric properties are relevant to rating scales. This chapter briefly introduces these properties. More information is available in the resources listed in the “Suggested Readings” section at the end of this chapter.
Reliability refers to the consistency with which a scale’s items measure the same construct the same way, every time. There are several types of reliability to consider. Internal reliability, or internal consistency, represents the degree to which individual items are consistent with each other. Items that are not internally consistent detract from the scale. Test—retest reliability, or stability, assesses whether a scale is stable over time. If the construct measured has not changed, then repeated measurements should be similar. This might be more difficult to determine for a “state” construct that is expected to wax and wane, such as suicidality, than for a “trait” construct, such as hyperactivity. Interrater reliability represents the agreement, or concordance, between different informants. As noted above, even well-regarded scales may have relatively low interrater reliability.
Validity indicates whether the scale accurately assesses what it was designed to assess. There are several types of validity. Content validity assesses whether the scale’s items represent the construct being measured. Criterion-related validity refers to whether there is a relationship between the rating scale and other types of criteria or outcomes, such as other rating scales, verbal report, clinical data, etc. There are two subtypes: predictive validity assesses whether the scale correlates with an event that will occur in the future (such as whether a high score on a suicidality scale predicts later suicide attempts); concurrent validity assesses whether the scale correlates with an event assessed at the same time the scale is administered (such as whether a self-report anxiety scale correlates with clinicians’ ratings of children’s anxiety). Often, it is of interest to determine whether a scale effectively differentiates a group of youth who have been diagnosed with a condition using some “gold standard” evaluation from those without the diagnosis. This is a type of criterion validity, sometimes termed discriminant validity. Finally, convergent validity refers to whether the measure relates to, or correlates with, other measures that are designed to assess the same construct, whereas divergent validity refers to whether the measure does not correlate with measures that are supposed to measure some other construct. For example, ideally scores on a new measure of teen depression would more highly correlate with scores on other established measures of teen depression (demonstrating convergent validity) than with measures of teen anxiety (demonstrating divergent validity).
Psychometric properties and other characteristics of rating scales should be matched to the intended application. For example, screening requires high sensitivity (i.e., a low number of “false negatives”) and a relatively brief instrument to reduce respondent burden. Monitoring requires good stability and a response format that is sensitive to response variation. Finally, the cutoff score that would indicate a clinically significant score varies in part due to factors like a youth’s developmental status, culture, and clinical status, but such information is often not available. Thus, as mentioned earlier in text, caution is needed when using scales that were developed with groups that differ from the youth in question. Even when a scale has strong psychometric properties, it is often appropriate to use more than one scale to tap various aspects of a problem.
Broadband Rating Scales
Broadband scales assess youths’ functioning across broad dimensions of behavior and symptoms, and are not designed to focus on one specific domain. They have high utility for initial evaluation and serve as guides for identifying problem areas that might warrant further evaluation. Broad coverage is important as a youth referred for one concern often has other problems needing attention which are not mentioned during an initial evaluation. Despite their utility, these scales suffer some limitations. To minimize respondent burden without sacrificing coverage of a broad range of problems, these scales often contain few items per subscale. Further, because broadband scales are often lengthy and cover multiple domains, they are not as useful for some applications, such as treatment monitoring. In short, these scales are best used to identify problems needing further evaluation with interview, observation, or narrowband scales.
The Child Behavior Checklist is a commercially available scale from the Achenbach System of Empirically Based Assessment (ASEBA). These scales have been the “gold standard” among broadband rating scales for over two decades. They include multiple versions for different reporters and age groups, including the CBCL 11/2—5 and Caregiver-Teacher Report Form (C-TRF) for preschoolers, the parent report (CBCL) and Teacher Report Form (TRF) for youth 6 to 18 years old, and the Youth Self-Report (YSR) for youth 11 years and older. Although details vary by version, there are approximately 140 items that take 15 to 30 minutes to complete. These scales were updated in 2001 with new normative data and modifications to item content and subscale structure. In 2007, the authors updated the CBCL 6-18, TRF, and
YSR computer scoring profiles to offer norms relevant to several different cultures, which is helpful to avoid misinterpreting culture-based behavior.
YSR computer scoring profiles to offer norms relevant to several different cultures, which is helpful to avoid misinterpreting culture-based behavior.
The CBCL, TRF, and YSR include subscale scores for several specific problem areas, as well as composite scores for internalizing, externalizing, and total problems. The items can also be scored using factors that approximate the diagnostic criteria of the DSM-IV. There are also items to assess youths’ adaptive functioning in the home, community, and school. The structure of the CBCL 11/2—5 and C-TRF is comparable, with a few differences in subscales for this developmental stage, and the inclusion of a screen for communication deficits. Ivanova and colleagues have provided strong evidence that the basic eight-syndrome structure of the CBCL is upheld in 30 international societies across continents.
One aspect of the CBCL warrants comment. The CBCL subscale labels are sometimes misleading and should not be taken at face value. For example, the Aggressive Behavior subscale contains items about oppositional and defiant behaviors with only a few items actually describing physical aggression. Scores on the Thought Problems subscale can be affected by various cognitive problems, and elevated scores are not equivalent to a thought disorder. It is therefore important to review a respondents responses to ascertain what he or she is really reporting. It is also important to inquire further about unusual responses to ensure understanding, particularly with younger children or lower-functioning caregivers. Overall, the CBCL has high utility due to its rapid coverage of a wide range of problems in various settings, the assessment of adaptive functioning, cross-cultural normative data, and its extensive use in the research literature. Computer scoring increases ease of use.
The Early Childhood Inventory-4 (ECI-4), Child Symptom Inventory-4 (CSI-4), Adolescent Symptom Inventory-4 (ASI-4), and Youth’s Inventory-4 (YI-4) by Gadow and Sprafkin comprise another commercially available series of broadband scales for children aged 3 to 18 years. They have the advantage of being based on DSM-IV diagnostic criteria for the most common disorders of childhood and adolescence, as well as less common disorders such as schizophrenia, reactive attachment disorder, and somatization disorder. Parent and teacher forms exist for all ages, and a self-report form exists for youth aged 12 to 18 years (YI-4). Thus, the scales provide easy comparison of symptoms endorsed over time and across informants. There are approximately 100 items per scale that take 15 to 20 minutes to complete.
Two scoring procedures are available. The Symptom Severity procedure simply sums the items endorsed, which can then be compared to normative data. Kamphaus and Frick have cautioned, however, that the moderate size and diversity of the scales’ normative sample limit the utility of this dimensional scoring approach. The Symptom Count procedure is more commonly used and allows clinicians to identify whether the child or adolescent is exhibiting the sufficient number of clinical symptoms (i.e., rated as occurring “often” or “very often”) necessary to consider a DSM-IV diagnosis. However, the scales do not consider age of onset of symptoms or functional impairment, both of which are needed to make a formal diagnosis.
Many clinicians and investigators prefer this series of scales, as item responses can help to focus the diagnostic interview on the most likely problematic categories and related comorbidities. Another strength is the inclusion of disorders that are severe but rarely covered in other broadband scales. Thus, these scales may be helpful with children who present with more severe symptomatology. The psychometric properties of the scales vary by age, informant, and disorder, but substantial data guide users toward the most effective uses of these scales.
Narrowband Rating Scales: Externalizing Symptoms
Collett and colleagues have discussed the use of externalizing scales in child and adolescent psychiatry. Externalizing symptoms are observable by others, and include behaviors such as hyperactivity, aggression, and oppositionality. Youth displaying these behaviors are typically referred for services because of the challenges they pose to others. Given that youth tend to
underestimate their externalizing symptoms, adults are generally considered to be the optimal respondents. Ratings are generally obtained from multiple adults in order to ascertain varied perspectives and to assess the ecological aspects of youths’ behaviors (i.e., if aggression occurs both at home and at school). Many of these scales were developed for school-aged boys, and suitability is less clear for younger and older youth, and for girls.
underestimate their externalizing symptoms, adults are generally considered to be the optimal respondents. Ratings are generally obtained from multiple adults in order to ascertain varied perspectives and to assess the ecological aspects of youths’ behaviors (i.e., if aggression occurs both at home and at school). Many of these scales were developed for school-aged boys, and suitability is less clear for younger and older youth, and for girls.
Multiple available scales purport to measure attention-deficit hyperactivity disorder (ADHD), the most common, and best studied, externalizing disorder.
The Conners’ Rating Scale-Revised (CRS-R), and its recently updated version—the Conners’ Rating Scale-3 (CRS-3)—are commercially available ADHD-rating scales that have been the prototypical ADHD scale for two decades. The 2008 CRS-3 includes normative data, an assessment of executive functioning, a measure of impairment, and a validity scale. The coverage of executive functioning is especially useful given the overlap between ADHD and executive dysfunction. The CRS-R and the CRS-3 cover core ADHD subtypes in addition to comorbid problems, such as oppositional-defiant disorder (ODD) and conduct disorder (CD). They can be used with children aged 3 through 17 years (parent and teacher forms) and 12 through 17 years (self-report form). There are short and long form options for each informant that vary from 39 to 115 items and take 5 to 20 minutes to complete. The CRS-R and CRS-3 are strong choices for comprehensive assessment given the multiple indices, normative base, and strong psychometrics. The high sensitivity, particularly for the parent report version, makes these scales a good choice for screening. There is also an abbreviated version that has been useful in monitoring medication treatment. Relative disadvantages include the scales’ somewhat poorer functioning of the comorbidity indices, and poor discrimination between ADHD, ODD, and CD. The CRS-3 does not yet have a body of literature supporting its use.
The Swanson, Nolan, and Pelham-IV Questionnaire (SNAP-IV) is an ADHD-rating scale that is available in the public domain. The SNAP was the first of several scales to utilize DSM symptoms of ADHD in a rating scale format that can be completed by both parents and teachers. The scale was the primary outcome measure in the Multisite Multimodal Treatment Study of Children with ADHD (MTA). The short version of the SNAP-IV includes the core DSM-IV-derived ADHD subscales of inattention, hyperactivity/impulsivity, as well as ODD, along with summary questions in each domain. The longer version (90 items, which take 20 minutes to complete) includes these core subscales along with items selected from other scales measuring ADHD and associated features. Finally, the SNAP-IV contains 40 items that have been extracted from DSM-IV-based criteria for several other disorders, such as internalizing symptoms, and motor disturbances. Thus, this scale incorporates multiple dimensions into a single scale. Clinicians and investigators often adapt selected indices or subscales from the SNAP-IV to their specific applications. Scoring information for the SNAP-IV is conveniently provided free on a website.
Representative normative data are not available, and cutoff scores are based on a study of 5- to 11-year-olds from low socioeconomic status and predominantly Hispanic heritage. Recent studies by Bussing and colleagues in 2008 identified two ADHD factors, an ODD factor and acceptable reliability, as well as race differences in teacher reports, which merit further examination. Internal consistency appears good to excellent.
Advantages include the SNAP-IV’s basis in the DSM-IV and its coverage of other problems to gain a brief assessment of comorbidity. Several of the additional SNAP-IV indices (e.g., the SKAMP) have been useful to assess functioning. Its free, online availability and scoring make it readily available. However, the collapsed age and gender data preclude optimal interpretation of an individual’s scores. There is no adolescent data or self-report version.
The Eyberg Child Behavior Inventory and the Sutter-Eyberg Student Behavior Inventory-Revised (ECBI and SESBI-R) are commercially available, well-established scales assessing externalizing behaviors corresponding to diagnoses of ADHD, ODD, and CD. The ECBI is completed by parents and the SESBI-R by teachers. Both versions use the same format with item
overlap. The ECBI contains 36 items and the SESBI-R 38 items, and both take 5 to 10 minutes to complete. Respondents rate each item on two dimensions: an intensity scale (I) assesses behavior frequency, and a problem scale (P) assesses reporters’ perceptions of whether the behavior is problematic. This format is clinically useful, as respondents may rate behaviors as problematic even if they occur at a normative rate, thus indicating a low threshold and/or inappropriate expectations. Conversely, a respondent may rate problem behaviors as frequent but not problematic, reflecting a high threshold. In 2008, Butler and colleagues found support for the ECBI discrepancy hypothesis; that is, discrepancy in elevated scores on the I and P scales is associated with problematic parenting, such as intolerance of misbehavior or permissive parenting.
overlap. The ECBI contains 36 items and the SESBI-R 38 items, and both take 5 to 10 minutes to complete. Respondents rate each item on two dimensions: an intensity scale (I) assesses behavior frequency, and a problem scale (P) assesses reporters’ perceptions of whether the behavior is problematic. This format is clinically useful, as respondents may rate behaviors as problematic even if they occur at a normative rate, thus indicating a low threshold and/or inappropriate expectations. Conversely, a respondent may rate problem behaviors as frequent but not problematic, reflecting a high threshold. In 2008, Butler and colleagues found support for the ECBI discrepancy hypothesis; that is, discrepancy in elevated scores on the I and P scales is associated with problematic parenting, such as intolerance of misbehavior or permissive parenting.
The ECBI and SESBI-R have been extensively used to discriminate clinical samples of disruptive youth and to assess the efficacy of interventions, such as Hutchings and colleagues’s 2006 study of the relation of parenting programs for high-risk preschoolers to crime rates. Although norms are available for youth up to 16 years of age, these scales are most suitable for younger children, not for older children with major CDs. These measures are widely used across ethnic and racial groups in the United States. There are no major disadvantages.
Rating Scales Assessing Internalizing Symptoms
Myers and Winters have reviewed rating scales that assess youths’ internalizing symptoms, specifically depression and anxiety. These symptoms are manifestations of psychological distress that may not be readily observable by others, such as anxiety and depression, and it is important to assess these symptoms via youth self-report; ideally, relevant adults complete parallel parent-report and/or teacher-report forms. Youth generally endorse more symptoms than these adults appreciate, and interrater reliability is often low. Clinician-rated scales integrate youths’ and adults’ responses and may provide greater accuracy in treatment studies. Given that youths’ feelings of depression, anxiety, or suicidality wax and wane, it is difficult to know whether observed change over time is due to a real clinical change or random error in the scale unless test—retest reliability has been established. Finally, many internalizing symptom scales detect symptoms of other internalizing disorders; for example, depression-rating scales generally detect anxiety and suicidality, making scales with good discriminative and divergent validity especially valuable.
Rating Scales Assessing Mood Symptoms
Depression-rating scales have many challenges. Many measure distress rather than depression. Also, it can be difficult to discriminate clinically depressed youth from their nondepressed peers, as depressive symptoms are common in both clinical and community samples.
The Beck Depression Inventory-II (BDI-II) is a commercially available scale developed for adults that has become the most popular depression-rating scale for adolescents over 13 years. It has been used in multiple diverse applications. The BDI-II contains 21 items that take 10 minutes to complete. This measure assesses the same aspects of depression in adolescents that it does in adults: cognitive, behavioral, affective, and somatic. The BDI-II discriminates outpatient depressed teens from those with anxiety and CDs. This scale is also useful for monitoring treatment process. The BDI-II has been translated into several languages and used with various ethnic populations, but caution is warranted as cutoff scores vary with culture and are not always available. One of the scale’s weaknesses is that it lacks an adult-report form to provide a contextual perspective.
The Children’s Depression Inventory (CDI) is a commercially available scale that was developed by Kovacs as a downward extension of the BDI. It can be used with youth 7 to 17 years old. It is the most frequently studied and utilized measure of juvenile depression. Its 27 items take 15 minutes to complete. Scoring yields five subscales: negative mood, interpersonal problems, ineffectiveness, anhedonia, and negative self-esteem. The CDI functions well
psychometrically and has good predictive validity. It has been translated into several languages and has been used extensively across cultures. However, sensitivity, specificity, and discriminative validity are suboptimal and the CDI appears to measure distress, rather than depression. Nonetheless, the CDI enjoys a broad literature supporting its use.
psychometrically and has good predictive validity. It has been translated into several languages and has been used extensively across cultures. However, sensitivity, specificity, and discriminative validity are suboptimal and the CDI appears to measure distress, rather than depression. Nonetheless, the CDI enjoys a broad literature supporting its use.
The Children’s Depression Rating Scale-Revised (CDRS-R) is a commercial, clinicianrated scale patterned by Poznanski and Mokros on the Hamilton Depression Rating Scale. It can be used with youth aged 6 years and older. It consists of 17 items that take 15 to 20 minutes to administer. The CDRS-R is unique in that it integrates information from both the child and the parent, incorporates behaviors observed during the interview, and includes several items not specific to depression. Thus, the measures construct of depression differs somewhat from DSM-IV. Jain and colleagues have recently found solid psychometric properties for the CDRS-R. There is some evidence of discriminative validity, and impressive interrater reliability supports the alleged benefit of clinician-rated scales over lay scales. The short form facilitates screening and repeated assessment. Wagner and colleagues have found it to be sensitive to treatment. Use of the CDRS-R in combination with self-reports and global ratings offers comprehensive, yet efficient, assessment.
The Moods and Feelings Questionnaire (MFQ) is a scale in the public domain developed by Angold and colleagues to screen for depression in youth aged 8 to 18 years. It is available as a child-self-report and a parent-report form. The 33 items of the long form (MFQ) take 15 minutes to complete and the 13 items of the short form (SMFQ) take 5 minutes to complete. The MFQ covers symptoms of major depressive disorder and dysthymia as specified by the Diagnostic and Statistical Manual, Third Edition, Revised (DSM-III-R), but includes other symptoms of clinical significance as well. This construct of depression approximates DSM-IV criteria sufficiently to remain relevant. Several studies, including that by Daviss and colleagues in 2006, demonstrate solid psychometric properties, including high internal consistency for all forms, strong test—retest reliability, and good concordance with depressive diagnoses derived from standardized diagnostic interviews. Furthermore, the MFQ discriminates between pediatric and psychiatric patients and depressed and nondepressed youth with good sensitivity and specificity. The MFQ has two major advantages. First, it is based on a clear diagnostic construct of depression, which distinguishes it from other depression-rating scales. Second, the MFQ can be readily downloaded from the author’s website free of charge.
The Mania Rating Scale (MRS) is an older, brief scale available in the public domain to assess mania. It was developed for completion by the clinician after he or she collects data from the patient and relevant other individuals, typically on inpatient units. It consists of 11 items that take around 15 minutes to administer. The MRS has been increasingly used with youth diagnosed with bipolar disorder (BPD), but the lack of consensus regarding the diagnostic criteria for juvenile BPD impedes interpretation of these studies. In 2004, Youngstrom and colleagues developed a parent-report version that shows good concordance with the clinicianadministered MRS. The MRS has shown some ability to discriminate BPD from ADHD, other disruptive behaviors, and depression. Several investigators including Wagner and colleagues have found that the MRS is sensitive to treatment with mood stabilizers. Overall, however, the MRS has not been adequately examined to establish its utility for clinical applications with youth, despite its increasingly wide usage.
Rating Scales Assessing Suicidality and Hopelessness
Winters and colleagues have reviewed scales relevant to suicidality. The term suicidality encompasses suicidal ideation, suicide attempts, and completed suicide. Suicidal ideation and attempts are relatively common, but completed suicide is not. It is therefore difficult to attain strong predictive validity with these scales, and the goal is for scales with high sensitivity to identify youth in need of further assessment. Hopelessness strongly predicts suicidality, and scales measuring hopelessness are often used instead of suicide-rating scales.
The Suicidal Ideation Questionnaire (SIQ) by Reynolds consists of two very popular, commercially available scales that are widely used to assess suicidality, one for high school students and one for middle school students. Each contains 15 to 30 items that take 10 to 15 minutes to complete. These scales are most commonly used as part of suicide prevention efforts in schools, although they are appropriate to clinical settings as well. Both scales measure the intensity and frequency of suicidal ideation during the past month. Standardization was conducted with large normative samples, a major strength of the measure. Psychometric properties are quite good over various samples and across cultures. Large differences in scores between suicide attempters and other youth, combined with good sensitivity and specificity, suggest discriminative validity. These scales have been used to elucidate the relationship of suicidality to hopelessness, depression, and loss.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
