6. Date of publication: 7. Age range for the assessment: 8. Length of time to administer: 9. Description of test items: 10. How are the items answered (e.g., Likert scale, yes/no 11. How are the items scored and what do the raw scores mean? 12. What subscales are present? 13. In your own words, describe the reliability, validity, norming methods 14. What are the limitations of the assessment? 15. Why did you choose this assessment? (write 1-2 paragraphs in narrative form) 16. How will this assessment help your client? (write 1-2 paragraphs in narrative form) DESCRIPTION. The Clinical Assessment of Depression (CAD) is a 50-item self-report instrument designed to assist with the assessment of depression in children, adolescents, and adults, ages 8 to 79 years. Whereas the CAD is meant to be general enough to assess depressive symptomatology across the lifespan, it is also purported to address the unique characteristics associated with depression in children/adolescents and adults. The test authors developed this instrument so that it would be more technically sound, comprehensive, and useful than currently available scales for depression. The CAD is also meant to assist in intervention planning and the monitoring of treatment progress.

The CAD includes a Total Scale score that reflects an overall level of depressive symptomatology and four symptom-based scales including Depressed Mood, Anxiety/Worry, Diminished Interest, and Cognitive and Physical Fatigue. In addition, the CAD contains Inconsistency, Negative Impression, and Infrequency validity scales. The test authors also suggest that users examine six critical item clusters that provide additional information on Hopelessness, Self-Devaluation, Sleep/Fatigue, Failure, Worry, and Nervousness.

Based on a Flesch-Kincaid reading level analysis, the CAD is written at a third-grade reading level. No alternative language versions of the CAD are available. Administration of the CAD may be accomplished by persons with little training; however, they should be supervised by a qualified professional. Test materials include the CAD Rating Form, which includes answer and scoring sheets, and the Score Summary/Profile Form sheet. Clients are asked to circle Strongly Disagree, Disagree, Agree, or Strongly Agree to items based on "how you have been feeling lately" (professional manual, p. 10). The CAD may be completed in approximately 10 minutes.

Scoring of the CAD may be accomplished by hand or hand-entered for computer scoring. Hand scoring entails separating the answer sheet from the scoring sheet and then transferring the circled score items to the appropriate scale boxes. These scale scores are then tabulated and transferred to the Score Summary sheet where they are converted to T scores, percentiles, and 90% confidence intervals based on tables within the manual. The Inconsistency score also must be tabulated on the Score Summary sheet. The Negative Impression and Infrequency scores are simply transferred from the Scoring Sheet to the Score Summary sheet. Finally, T scores are transferred to the Profile Form and plotted. Scoring program software is available from the publisher and provides basically the same information for the user.

In regard to interpretation, the test authors recommend that properly licensed or certified personnel follow a five-step process that includes consideration of the CAD Total Scale score, the symptom scales, individual items, follow-up clinical interview data, and other related information. They go on to say that the Total Scale score is the "best overall measure of the client's general affectivity" (professional manual, p. 17). However, they suggest that test users consider closely the validity scales, symptom scales, and critical item clusters when interpreting the total score.

DEVELOPMENT. The test authors developed items for the CAD following a review of the literature and existing instruments, an assessment of current diagnostic criteria, and consultation with colleagues. Various versions were developed and piloted, and eventually a 75-item instrument showed moderate-to-high correlations with the Multiscore Depression Inventory, the Beck Depression Inventory, and the Children's Depression Inventory. Analyses of variance demonstrated a significant main effect for age and gender, but no main effects for race/ethnicity. There were small effect sizes for the difference between the child and adolescent groups and between the child and adult groups; however, differences across ages were not deemed of developmental importance. Slightly larger effect sizes were seen for gender. Additional content and factor analyses resulted in the current 50-item version with its total scale and four symptom scales labeled Depressed Mood (23 items), Anxiety/Worry (11 items), Diminished Interest (6 items), and Cognitive and Physical Fatigue (10 items).

TECHNICAL. The standardization sample included 1,900 children, adolescents, and adults between the ages of 8 and 79 years in 22 states. Samples are fairly similar to the U.S. population in most respects; however, the CAD normative sample is skewed toward being better educated and from the Midwest. T-scores and percentiles are provided for the CAD total scores and for the four subscales by four normative age groups (i.e., 8-11 years, 12-17 years, 18-25 years, and 26-79 years).

Internal consistencies (coefficient alpha) and standard errors of measurement for the CAD Total Scale and symptoms scales were computed by age, gender, and race/ethnicity for the total sample (n = 1,900). By age group, coefficient alphas for the CAD Total Scale ranged from .96 (8-11 and 18-25 years) to .97 (12-17 and 26-79 years). Alpha coefficients for the four symptom scales range from .78 (Diminished Interest scale for the 8-11 age group) to .96 (Depressed Mood scale for both the 12-17 and 18-25 age groups). By gender and race/ethnicity, alpha coefficients were nearly identical for the CAD Total Scale scores; however, mild variations in alphas existed on several symptom scale scores.

For the clinical sample (n = 378), alpha coefficients for the CAD Total Scale scores ranged from .97 (8-18 years) to .98 (19-79 years and total clinical sample). Coefficients for the symptom scales ranged from .85 (Diminished Interest for the 8-18 age group) to .97 (Depressed Mood for the ages 19-79 and total clinical sample groups). The clinical sample was composed of mainly Caucasian females who apparently have major depression (n = 48), dysthymia (n = 33), and a mixed clinical presentation (n = 108). No information is provided within the manual on factors such as how participants were classified, inpatient versus outpatient status, and so forth.

Corrected test-retest reliabilities for the CAD Total scale and symptom scales were computed for child/adolescent (n = 40) and adult (n = 59) samples over a mean time interval of 17.53 days (range of 7 to 36 days) for the former and 13.27 days (range of 1 to 51 days) for the latter. Coefficients for the CAD Total Scale ranged from .81 (child/adolescent) to .87 (adult). Corrected coefficients for the symptom scales ranged from .64 for Diminished Interest (child/adolescent) to .89 for Anxiety/Worry (adult).

In terms of content validity, the test authors state that this is ensured by the CAD's inclusion of items representing symptoms associated with depression within the literature and those noted within diagnostic criteria such as the DSM-IV. In support of criterion-related validity, a 75-item version of the CAD was shown to correlate .74 with the Multiscore Depression Inventory and .68 with the Beck Depression Inventory in a sample of adults, and .80 with the Children's Depression Inventory in a sample of children/adolescents.

The 50-item CAD was administered to nonclinical control (n = 67) and combined clinical (n = 189) samples. The nonclinical youth sample is skewed toward being older adolescents (mean age = 17.5 years), and the nonclinical adult sample is skewed toward being younger adults (mean age = 24.7 years). Both groups are composed of mainly Caucasians (93.3% and 97.3%, respectively). Within the combined clinical sample, the mean age of the youth group is 13.1 years and the mean age for the adult group is 37.1 years. Whereas the race/ethnicity breakdown for the youth group is appropriate, over 94% of the adult combined clinical sample is Caucasian and none are African American. In addition, the child/adolescent nonclinical and adult combined clinical groups are predominantly female.

Correlations between the CAD Total Scale score and the Beck Depression Inventory-II (BDI-II) were .71 (child/adolescent), .87 (adult), and .86 (overall) in the combined clinical group and .67 (child/adolescent), .70 (adult), and .69 (overall) in the nonclinical group. Correlations between the CAD Total Scale score and the Reynolds Adolescent Depression Scale (RADS) were .64 (child/adolescent), .90 (adult), and .83 (overall) in the combined clinical group and .82 (child/adolescent), .83 (adult), and .85 (overall) in the nonclinical group. CAD symptom scale correlations with the BDI-II ranged from .42 (Anxiety/Worry) in nonclinical adults to .85 (Depressed Mood) for adults in the combined clinical group. Correlations with the RADS ranged from .49 (Diminished Interest and Cognitive and Physical Fatigue) in the child/adolescent combined clinical sample to .87 (Depressed Mood) in the adult combined clinical sample.

A University of Western Kentucky study with 122 mainly Caucasian adolescents found the CAD and its symptom scales correlating from .64 (Diminished Interest) to .75 (Depressed Mood) with the BDI-II and from .71 (Cognitive and Physical Fatigue) to .86 (Depressed Mood) with the RADS. Total scale correlations were .77 with the BDI-II and .88 with the RADS.

Evidence for the validity of the CAD based on its internal structure is provided within the test manual. Item-with-total scale correlations ranged from .41 to .74. Intercorrelations among the symptom scales ranged from .68 to .82 in a sample of 700 8-17-year-olds and from .68 to .81 in a sample of 1,200 18-79-year-olds. Intercorrelations among the symptom scales for the total standardization sample (n = 1,900) ranged from .68 to .81. A confirmatory factor analysis was performed, and results favor a four-factor model including Depressed Mood, Anxiety/Worry, Diminished Interest, and Cognitive and Physical Fatigue. No information is provided within the manual supporting the use of the critical item clusters, particularly the claim that their content is "especially sensitive to individuals who may be at risk for harming themselves" (p. 7). Neither is there evidence to support the test authors' contention that the CAD may be useful in planning interventions or for monitoring treatment progress.

COMMENTARY. The CAD was developed in order to provide "a more technically sound, comprehensive, and useful scale" (professional manual, p. 5) for assessing depression in children, adolescents, and adults. It is also meant to be useful in planning interventions and monitoring treatment progress. The test authors' overall goal was to publish an instrument that is an improvement over existing scales.

So, how does the CAD measure up? First, in terms of its technical soundness, the CAD is an instrument with strong Total Scale internal consistency and relatively strong symptom scale internal consistency. Test-retest reliability is slightly less than desirable; however, given the variable nature of the affective states being measured it is not problematic. The CAD demonstrates reasonable internal consistency and correlates well with other measures of depression including the BDI-II and the RADS. One concern lies in the normative samples and their skewness toward better-educated, Caucasian females and the general lack of descriptive information regarding the clinical samples. Despite this, the CAD appears to meet the test authors' objective of developing an empirically supported instrument that is appropriate for measuring depression across age groups.

In terms of comprehensiveness, the CAD includes 50 items designed to be sensitive to depressive symptomatology. However, the CAD items only completely cover four (i.e., depressed mood, diminished interests/pleasure, fatigue/loss of energy, and concentration difficulties/indecision), partially cover three (i.e., sleep issues, feelings of worthlessness/guilt, and psychomotor retardation [not agitation]), and fail to cover two (i.e., changes in weight/appetite and thoughts of death or suicide) of the nine DSM-IV-TR (APA, 2000) criteria for major depressive episode. Other instruments (e.g., BDI-II, Zung Self-Rating Depression Scales) do as well or better in this respect. Also, an accurate diagnosis of depression is made difficult because the CAD asks respondents to rate how they have been feeling "lately" versus over the same 2-week period as required by the DSM-IV-TR criteria.

In terms of its usefulness, most would likely agree that the CAD is an easily administered and scored instrument for depression. Its true utility lies in its ability to assess depression across the life span and in its validity scales that may assist the clinician to determine various response sets.

Finally, with regard to the CAD being useful in planning interventions and monitoring treatment progress, the jury is still out. Although, intuitively, one suspects that scores from the various scales on the CAD would be helpful in planning interventions and monitoring progress, no data are provided within the test manual to support these assertions. Additional studies must be conducted in these areas.

SUMMARY. The CAD is a self-report instrument designed to assess depression in children, adolescents, and adults. The test authors should be given credit for developing a technically sound instrument that may assess depression across the ages. The incorporation of validity scales may also be useful for clinicians. CAD problems include limited descriptions of the normative clinical samples, which makes interpretation difficult, and the lack of data within the test manual regarding the use of the CAD in planning interventions or monitoring treatment. Instruments that are more comprehensive in their DSM-IV-TR coverage and possess a wealth of data associated with their use include the original BDI, the Zung Self-Rating Depression Scale, and the Reynolds Adolescent Depression Scale-2nd Edition. Although more age restricted, these scales may be deemed more appropriate for clinical use until additional studies are conducted on the CAD.

REVIEWER'S REFERENCE

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., TR). Washington, DC: Author.

Review of the Clinical Assessment of Depression by JODY L. KULSTAD, Adjunct Professor, Seton Hall University, South Orange, NJ:

The Clinical Assessment of Depression Scale (CAD) is a relatively brief, comprehensive, psychometrically sound measure of depression in individuals from childhood to late adulthood. Closely linked to DSM-IV-TR criteria, the 50-item CAD provides a diagnostically driven assessment of major, minor, and subclinical depressive episodes. The primary purpose of the CAD is for clinical use, allowing clinicians the ability to not only diagnose, but also to monitor symptoms and severity across the individual's lifespan. In addition, the CAD can be used in epidemiological or outcome research studies to measure depression longitudinally or cross-sectionally.

The CAD is a pencil-and-paper self-report measure of depression. Test materials include the 88-page professional manual, The CAD Rating Form (a carbonless rating form that includes the answer sheet on top and the scoring sheet underneath), and the CAD Profile Form/Score Summary. The CAD is intended for individuals from age 8 to 79. Each of the 50 items addresses diagnostic criteria or clinical manifestations of depression across the lifespan. Content includes age-sensitive symptoms (e.g., irritability in childhood) as well as hallmark criteria (e.g., anhedonia) for depression. Items are written in first person, with response options including strongly disagree, disagree, agree, and strongly agree.

The CAD provides a total score assessing overall level of depression as well as four symptom scales: Depressed Mood (23 items), Anxiety/Worry (11 items), Diminished Interest (6 items), and Cognitive and Physical Fatigue (10 items). The CAD also provides validity scales as well as six critical item clusters measuring Hopelessness, Self-Devaluation, Sleep/Fatigue, Failure, Worry, and Nervous.

ADMINISTRATION, SCORING, AND INTERPRETATION.

Administration. The CAD takes approximately 10 minutes to complete and can be administered by individuals with training in administering and scoring self-report measures. Although specialized professional training is not necessary, administration and scoring should be done under the supervision of a qualified professional. The measure can be administered in individual or group settings, but the location needs to be as private as possible and free from distraction. The test manual recommends rapport be established prior to administering the CAD, as this may motivate the individual to respond in a more open and accurate manner. Individuals respond to the 50 items regarding how they have been feeling lately. Respondents should be strongly encouraged to carefully read and answer all items, selecting only one response per item. If the respondent changes his or her mind about a response, he or she must mark an "x" over the incorrect response and circle the intended response. Responses should not be erased, as this will appear as a smear on the scoring sheet. Though the instrument is written at an overall third-grade reading level, examiners should be cautious about assuming reading ability and should monitor the administration. For some, the examiner may need to administer the assessment orally. Also, because the CAD is available only in English, use with individuals who are non-English-speaking or speak English as a second language should be undertaken only with great caution, and the examiner should ensure comprehension of the items prior to scoring and interpretation.

Scoring. After the respondent has completed the CAD, the examiner should review the form for missing responses or multiple responses to one item. The CAD can be hand scored easily or can be computer scored using the CAD Scoring Program, available from the publisher.

When hand scoring, remove the perforated edge to separate the answer sheet from the scoring sheet. The individual's responses and information will show on the scoring sheet, with boxes to the left for the four symptom scales and to the right for the validity scales. To calculate the total raw scores and raw symptom scale scores, follow these steps. First, transfer the individual's circled response value to the box to the left of the item. When all items are recorded in the appropriate box, sum the boxes and record the subtotal in the box provided at the bottom of the column. Then, transfer the column totals for Items 1-25 (left-hand page) to the corresponding boxes provided at the bottom of the columns on the right-hand page. Sum the two subtotal boxes at the bottom of the column to arrive at a Total CAD raw score and a total raw score for each of the four symptom scales. Then, transfer all five raw scores to the Score Summary Sheet. Raw scores may be converted to T scores using the age-based Raw Score Conversion tables provided in the appendices of the professional manual. In addition to T scores, percentiles and 90% confidence intervals are recorded.

To score the validity scales, there are three steps to follow. Step 1 entails recording a "1" in the Negative Impressions (NI) and Infrequent Response (IF) boxes, where available, to the right of the respondent's circled score, for each item marked a "4." For example, if the respondent indicated a 4 on Item 1, then the person scoring the form would record a 1 in both the NI and IF boxes, because both boxes are shown. However, if the respondent indicated a 4 on Item 2, a "1" would be recorded in the NI box only, because that is the only box shown. As with calculating the total and scale scores, the next step is to tabulate the total NI and IF scores from the left-hand page and record the score in the box provided on the right-hand page. Step 3, then, is to calculate a total score for the NI and IF scales and then transfer that total score to the Score Summary form.

The third validity scale, the Inconsistency scale (IN), is tabulated differently than the previous scores. Only 10 pairs of items (20 items total) are used to tabulate the IN score, so careful attention must be paid when transferring the scores from the scoring sheet to the Score Summary form. First, on the scoring sheet, in some of the boxes to the left of the respondent's circled number, there is an "I" in the right side of the box. These are the items used for the IN scale. For each box that shows an "I," one records the value in that box on the Score Summary form. When all 20 values have been recorded on the form, one then subtracts the lower value from the higher value and records the difference in the "total" column. Then, sum each total value for an overall inconsistency score.

Interpretation. Although the CAD can be administered and scored by nonprofessionals, the professional manual cites professional training and clinical experience as necessary for interpreting the CAD. Interpretation follows a five-step protocol.

First, omitted items or multiply marked items can invalidate an individual's CAD. As such, the reason for their occurrence needs to be discussed with the respondent. The clinician/researcher needs to determine whether this impacts the overall validity of the responses. Regardless of the reason, if more than 10% of any scale's items are omitted or mismarked, the scale is invalid. The only exception is Diminished Interest, which requires responses for all items in order to be valid. Then, the validity scales are reviewed. Using the Score Summary form, the clinician compares the IN, NI, and IF scores to the classification table at the bottom of the form. Classifications of Atypical or Very Atypical raise a red flag and need further exploration. High IN scores may suggest unreliable or careless responding. High NI scores may suggest a highly negative response set that requires close review to determine specific areas of concern. High IF scores could indicate a fake good or fake bad profile.

Once validity is established, the CAD Total Score (TS) is examined. With the exception of the validity scales, all scores are reported as T scores with scores of 60 or above suggesting clinical relevance. T scores between 60 and 69 reflect Mild Clinical Risk (MCR), between 70 and 79 reflect Significant Clinical Risk (SCR), and scores 80 or above reflect Very Significant Clinical Risk (VSCR). Thus, Total Scores (TS) above 60 reflect varying levels of depression. Though scores falling below 60 on the TS reflect "No Risk" (NR), scale scores and critical item clusters may evidence specific problem areas needing intervention so evaluation of the individual's protocol should continue beyond the TS assessment.

Symptom Scales are interpreted in a similar way, along the same classification parameters. Elevations in each of the areas result in a classic presentation of a depressed individual: depressed, anxious, unmotivated and disinterested, and fatigued. Elevations on all four would tend to reflect high CAD Total Scores, but it is important to remember that each individual may manifest depression differently and so each scale needs to be carefully examined for symptom pattern.

The fourth component of the interpretation is the critical item clusters. The goal of exploring these six clusters is twofold: to determine if there are any symptom endorsements that require immediate attention, and to glean information useful for planning therapy. Because there are no cutoffs to use in interpreting the clusters, the clinician should review the individual's responses to each item to determine specific areas of client concern.

Finally, the clinician should review each item in the CAD to ensure that no clinically significant information needs to be explored further or acted on. The manual recommends following up the administration, scoring, and interpretation with a clinical interview to provide further information, and then to consider the results of the CAD and the interview in light of any other data obtained.

DEVELOPMENT. The CAD was developed to fill a void in the assessment of depression. According to the professional manual, the CAD was designed to address the psychometric and content-related shortcomings of existing measures. A major benefit was the development of a single form for use with children to older adults. Content was developed based on review of the professional literature, existing measures, consideration of age-related depressive symptoms, and the DSM-IV-TR diagnostic criteria. This process generated a pilot version consisting of 175 items across 16 domains, that demonstrated a full scale coefficient alpha of .99. The scale was subsequently reduced to 130, then 80, then 75 items and was administered to a wide range of respondents, including children and adolescents, all resulting in total scale coefficient alphas of .96 and higher. Following further factor and content analyses, the CAD was reduced to the current 50 items. No information on these analyses was included in the test manual.

TECHNICAL. The CAD was standardized on a sample of 1,900 individuals age 8 to 79 obtained from a range of settings (e.g., schools, clinicians, community organizations). The sample was a close match to the 2001 U.S. Census, with only educational level higher than that of the Census values. Results from ANCOVA analyses evidenced no effect of demographic variables on variance in CAD scales. Despite only minor mean differences between age groupings, derived scores were developed with associated T scores, percentiles, and 90% confidence intervals.

Reliability. The CAD evidences reliability through internal consistency, standard error of measurement, and test-retest reliability. Coefficient alphas for the CAD Total Scores across age groups were .96 and higher, and were .97 for both genders. Symptom scale reliabilities ranged from .78 (Diminished Interest, 8-11-year-old age group) to .96 (Depressed Mood, 18-25-year-old age group), with all but one above .80. Lowest values were found for the 8-11-year-old sample. Standard error of measurement for the CAD Total Score and Symptom Scales is between 2 and 5 T score points, which indicates that a person's obtained score is likely to be very similar to his or her true score. The CAD stability reliabilities, obtained from all age groups (interval ranging from 7 to 36 days for the child/adolescent group and 1 and 51 days for the adult sample), was .84 overall and ranged from .74 to .84 (uncorrected) and .81 to .87 (corrected), with higher stability values for the adult sample. It is important to note, even though the values dip below the desired .90 criterion, the mean score differences were one T score point or less in all cases.

Validity. The CAD evidences validity via content, criterion, and construct validity. All information comes from the CAD professional manual. Content validity is evidenced through scale development. The test authors based item content on professional literature, accepted diagnostic criteria, and other widely used scales that assess depression. Criterion validity is evidenced through moderate to high correlations with other widely used instruments as assessed on the various versions of the CAD during the scale development process. More recent comparisons (Bowers, 2004; Tinsley, 2004) also showed high correlations with existing instruments (i.e., Beck Depression Inventory-II [BDI-II] and Reynolds Adolescent Depression Scale [RADS]). Concurrent validity for the final version was obtained by comparing clinical and nonclinical samples across age ranges, where clinical samples scored significantly higher than the mean and most age groups in the nonclinical sample scored below the normative mean. Construct validity is shown through both internal structure and confirmatory factor analysis (CFA). High coefficient alpha values suggest that the items are measuring a similar construct. This is supported by moderate to strong scale intercorrelations and moderate item-with-total scale correlations. CFA results show the CAD to be a multidimensional scale assessing one overall construct, Depression, and four secondary constructs (Depressed Mood, Anxiety/Worry, Diminished Interest, and Cognitive/Physical Fatigue).

COMMENTARY. Depression is one of the leading psychiatric illnesses across the lifespan. It is only in more recent times that we as a society have begun to fully appreciate the extent to which this disorder affects the young and the old. As a result, our assessment of depression has been more limited. Although there are some very good scales available, such as the Beck Depression Inventory (T7:275), Children's Depression Inventory (17:41), and Hamilton Rating Scale for Depression (T7:2161), to name a few, all have limitations. Most are age specific and some are theoretically driven, paying less attention to the multidimensional nature of depression. This is where the CAD shows its greatest strength. The CAD offers a single form, which can be used with children up to older adults, and addresses basic symptom presentation needed for diagnosis but also aspects of depression that may be more unique to the developmental level or specific situation of the individual. The CAD's applicability across the lifespan is a real benefit; however, this asset must be viewed with some caution, as the manifestation of the depression varies so greatly across developmental levels. This is seen in the reliability and validity studies, which show that values tended to be lower in the younger groups. Even though they were somewhat lower, the values were still good and suggest that the CAD is a psychometrically sound instrument for use with individuals from age 8 to 79. Although the CAD appears to offer a very attractive alternative to other depression measures, it is unclear why it is not being seen more in the professional literature. A literature search for studies using the CAD yielded no results. However, this could be more related to the relative newness of the CAD than any statement of its value in research or clinical settings. That being said, this review is limited by the fact that all information was obtained from the professional manual and the publisher's website.

SUMMARY. The CAD offers clinicians a brief, psychometrically sound, comprehensive method for assessing depression in clients. Its multilayered content and interpretation possibilities, combined with its ease of use, make the CAD a good option for the busy clinical professional as well as researchers.