• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Int J Methods Psychiatr Res. Author manuscript; available in PMC Dec 1, 2011.
Published in final edited form as:
PMCID: PMC2978808

Psychometric Properties of the Quick Inventory of Depressive Symptomatology in Adolescents

Ira H. Bernstein, Ph.D.,1,2 A. John Rush, M.D.,1,3,4 Madhukar H. Trivedi, M.D.,1 Carroll W. Hughes, Ph.D.,1 Laurie Macleod, R.N.,1 Bradley P. Witte, BS,1 Shailesh Jain, M.D. M.P.H.,1 Taryn L. Mayes, M.S.,1 and Graham J. Emslie, M.D.1



The clinician-rated (QIDS-C16) and self-report (QIDS-SR16) versions of the 16-item Quick Inventory of Depressive Symptomatology have been extensively examined in adult populations. This study evaluated both versions of the QIDS and the 17-item Children’s Depressive Rating Scale-Revised (CDRS-R) in an adolescent outpatient sample.


Both the QIDS-C16 and QIDS-SR16 were completed for the adolescents. Three different methods were used to complete the QIDS-C16: (a) adolescents’ responses to clinician interviews; (b) parents’ responses to clinician interview; and (c) a composite score using the most pathological response from the two interviews. Both classical and item response theory methods were used. Factor analyses evaluated the dimensionality of each scale.


The sample included 140 adolescent outpatients. All versions of the QIDS, save the parent interview, and the CDRS-R were very reliable (α ≥ 0.8). All four versions of the QIDS are reasonably effective and unidimensional. The CDRS-R was clearly at least two-dimensional. The CDRS-R was the most discriminating among low and extremely high levels of depression. The QIDS-SR16 was the most discriminating at moderate levels of depression. There was no relation between the QIDS scores and concurrent Axis III comorbidities.


The QIDS-C16 and the QIDS-SR16 are suitable for use in adolescents.

Keywords: Adolescent, depression, depressive symptom ratings, psychometrics, Quick Inventory of Depressive Symptomatology–Clinician-rated, Quick Inventory of Depressive Symptomatology–Self-report


Clinical depression, a common, disabling, and life shortening condition, occurs throughout the lifespan (Kessler et al., 1994; Weissman et al., 1991). Depressive disorders are a leading cause of morbidity and mortality in the pediatric age group, with prevalence of depressive disorders in juveniles ranging from 0.4 to 8.3% (Birmaher et al., 2007; Brent, 1987; Shaffer et al., 1996), and are greater in adolescents than children. These prevalence rates are comparable to adults, for whom the 12-month prevalence of major depressive disorder (MDD) is reported to be 6.7%±0.3% (Kessler et al., 2005).

Symptoms of MDD disrupt critical developmental processes that occur in adolescents, including social, emotional, cognitive, and even physical development. Depression in adolescents often leads to significant functional impairment in school or work, and entanglement in the legal system, and adolescents with depression are at increased risk for substance abuse, attempted and completed suicide, and for recurrent depression during adulthood. Furthermore, over half of children who are depressed will experience another depressive episode within 5 years (Birmaher et al., 2007). Given the prevalence of depression and the profound impact of the disorder on adolescent functioning (Fergusson and Woodward, 2002; Rao et al., 1995; Weissman et al., 1999), identification of depression has obvious public health benefits (Coyle et al., 2003; Olfson et al., 2003). Furthermore, the accurate measurement of depression helps to ascertain treatment benefits and guide care.

Presently, the field standard for assessing severity of adolescent depression in clinical trials is the 17-item Childhood Depression Rating Scale-Revised (CDRS-R) (Poznanski and Mokros, 1005), which has acceptable psychometric properties. However, because the CDRS-R is based on the Hamilton Rating Scale for Depression (HRSD) (Hamilton, 1960; 1967), it is not criteria-based, and the individual items carry different weight (e.g., sleep is scored on a 5-point scale, but depressed mood is scored on a 7-point scale). In addition, the CDRS-R does not have equivalent self- or parent report measures. Rather, it requires judgments derived from two separate clinical interviews with parents and the adolescent, which must then be synthesized. Furthermore, it is not in the public domain, which limits general usage. It has been noted that while systematic measurement of symptom change is important during treatment of depression, there are currently no easy-to-use, measurement-based research scales that are easily used in clinical practice (Trivedi et al., 2007).

The Quick Inventory of Depressive Symptomatology (QIDS) (Rush et al., 2000; 2003; Trivedi et al., 2004) is a rating scale that assesses the nine criterion symptom domains designated by the American Psychiatry Association Diagnostic and Statistical Manual of Mental Disorders - 4th edition (DSM-IV) (American Psychiatric Association, 2000) to diagnose a major depressive episode. A website containing information about this scale is supported by the Epidemiology Data Center, Graduate School of Public Health, University of Pittsburgh. Its URL is http://www.ids-qids.org/. The 16 items in the adult versions of the QIDS are a subset of the 30 items used in the Inventory of Depressive Symptomatology (Rush et al., 1996; 2000; Trivedi et al., 2004). The clinician-rated (QIDS-C16) and self-report versions (QIDS-SR16) each measure the nine criterion symptom domains (sleep, sad mood, appetite/weight, concentration/decision making, self view, thoughts of death or suicide, general interest, energy level, and restlessness/agitation) that define a major depressive episode by DSM-IV. The score for three domains (sleep, appetite/weight, and restlessness/agitation) is based upon the maximum score (most pathological) of two or more questions. The remaining domains are each rated by a single item. Each domain is scored from 0 to 3 reflecting increasing amounts of pathology, so the total test score can range from 0 to 27 (Rush et al., 2003). A 17th item dealing with irritability was added to create the adolescent version, but it was not used in the present research.

The QIDS-C16 and the QIDS-SR16 have been examined in several studies of adults (Rush et al., 2005; 2006). In particular, the QIDS-SR16 performs well against the QIDS-C16 and the 17-item Hamilton Rating Scale for Depression (HRSD17) in adults (Rush et al., 2006). These scales have not been evaluated in youth.

This study compared the CDRS-R, the clinician-rated QIDS (QIDS-C16), and a self-report version of the QIDS (QIDS-SR16) in adolescents seen through an outpatient psychiatric clinic.



All participants and their parents provided written informed consent/assent prior to enrollment. The study was approved by the Institutional Review Boards at the University of Texas at Arlington and the University of Texas Southwestern Medical Center at Dallas. Data were obtained during an initial or follow-up visit to the clinic on a convenience sample of consecutive adolescent outpatients obtained between December 2004 and September 2006. Eligible subjects were approached in the waiting area and informed of the study. Interested subjects were consented and interviewed either after the clinic appointment or during a separate appointment. Patients with neurological disorders, mental retardation, psychosis, or terminal illnesses, and those with acute substance/alcohol intoxication, judged clinically, were excluded. Clinical diagnosis was made by the treating psychiatrist through clinical interview.


Following consent/assent, participants and their parents were interviewed separately using the QIDS-C16 and CDRS-R. The QIDS-C16 is a clinician interview covering the nine criteria items of depression. After the adolescent and parent interviews, the clinician then creates a final composite score based on both interviews. For purposes of examining reliability and validity, we will refer to each method as follows: QIDS-C16(Adol) is the clinician rating based on the adolescent interview; QIDS-C16(Par) is the clinician rating based on the parent interview; and QIDS-C16(Comp) is the composite of the two interviews using the higher (more pathological) score for each of the nine domains from the QIDS-C16(Adol) and QIDS-C16(Par).

The QIDS also has a self-report version (the QIDS-SR16), which parallels its prior use in adults, in that the adolescent alone completed the questionnaire without clinician input or participation. All versions of the QIDS have a total score range of 0–27. Total scores ≤5 are indicative of no depression; scores between 6 and 10 suggest mild depression; scores from 11 to 15 suggest moderate depression; scores from 16 to 20 indicate severe depression; and total scores greater than 21 indicate very severe depression. These are the levels noted on the QIDS-A answer sheet.

The 17-item Children’s Depression Rating Scale-Revised (CDRS-R) was separately administered to participants and their parents. All clinician ratings (QIDS-C16 and CDRS-R) were obtained through clinical interview with the adolescent and their parent(s) by a highly experienced research nurse (author LM) who was trained in the assessment by a clinical psychologist (author CWH). The order of test administration was randomized.

Statistical Methods

Classical test theory (CTT) analysis estimated nine domain means and domain/total correlations (rit), scale means, and intercorrelations among the several versions of the QIDS and the CDRS-R. The scales were individually factored to infer their dimensionality using a component model and parallel analysis (Horn, 1965; Humphreys and Ilgen, 1969; Humphreys and Montanelli, 1975; Montanelli and Humphreys, 1976). Parallel analysis involves factoring random matrices having the same number of variables (9 for each version of the QIDS and 17 for the CDRS-R) and number of subjects (140) that the real data employed. The point at which the random and real data cross in the scree plots defines the dimensionality. Specifically, the first real principal component’s eigenvalue should exceed the first randomly generated principal component’s eigenvalue, but the reverse should be true of all subsequent eigenvalues for a unidimensional scale.

Samejima (Samejima, 1969; 1997) model item response theory (IRT) analyses were used to complement the CTT analyses. The CTT analysis is presented because of the familiarity of its basic concepts, i.e., the item mean as a measure (in this setting) of the tendency to endorse symptoms and the rit as a measure of how strongly a symptom relates to overall depression as indexed by the total scale score. In contrast, the analogous IRT measures, the intercept and slopes of the functions relating overall depression to item responses provide more explicit tests of differences between measures than CTT does (Nunnally and Bernstein, 1994). In the present case, three pairs of comparisons across conditions of measurement were made: (a) the QIDS-C16(Adol) and QIDS-SR16, (b) the QIDS-C16(Comp) and QIDS-SR16, and (c) the QIDS-C16(Adol) and QIDS-C16(Par). Each comparison involved assessing possible differences in symptom severity and possible differences in the relation of each domain to overall depression. For IRT, severity is inferred from the locations of what are known as boundary response functions. For the QIDS, which uses a four-point scale for each domain, there are three locations, which are the points respectively separating a response of “0” from responses greater than zero, responses of “0” or “1” from responses “2” or “3,” and responses of “0”, “1”, or “2” from a response of “3”. The relation of the domain to overall depression is inferred from the slopes of these three functions, which are assumed equal for a given QIDS domain. There are generally six CDRS-R locations since most items use a 7-point scale.

Tests compare a model in which the parameters of the two scales to be compared are allowed to vary freely with: (a) models in which one or more of the intercepts are constrained to equality, but the slopes are allowed to vary freely and (b) models in which one or more of the intercepts is allowed to vary freely, but the slopes are constrained to equality. The differences in fit may be expressed as a form of chi-square, symbolized G2, with df equal to the difference in number of parameters estimated. For example, the fit of the model in which all parameters are allowed to vary freely may be compared to the fit of the model in which the slope of QIDS-C16(Adol) domain 2 (sad mood) is constrained to equality with the slope of the QIDS-C16(Par) domain 2. If the value of G2 reaches significance, it can be assumed that sad mood relates to overall depression to a different extent in the two groups. Conversely, if this value is nonsignificant, it may be assumed that there is no such difference. Such tests may be made on individual domains (items) or on all domains taken collectively. Tests of slopes always were based upon 1 df, but tests of intercepts could be based upon 1 to 3 df dependent upon how frequently categories 2 and 3 were employed since the G2 test requires small expected values to be pooled.

We also obtained test information functions (TIF) for each of the four QIDS measures and the CDRS-R. TIF functions are somewhat similar to coefficients alpha in measuring the internal consistency of the scales, but TIF measures the ability of each scale to assess slight differences in level of depression as a function of amount of depression rather than as an overall measure.


The sample consisted of 140 self-referred outpatient participants (12–17 years of age) and their parents who were seen at the Child and Adolescent Psychiatric Clinic at UT Southwestern Medical Center at Dallas. The mean age was 14.4 years (standard deviation = 1.5 years). All participants were in the 6th to 12th grade. Overall, 71.4% were white, 15.7% African-American, 7.8% Hispanic/White, 1% Asian, and 4% of mixed ancestry. A total of 51% were female.

Based on clinical diagnoses (no structured interviews) rendered by child psychiatry fellows and adult psychiatric residents and reviewed by attendings, 81 (57.9%) were currently diagnosed with a depressive disorder (45.7% with MDD, 2.1% with depression NOS, 10% with adjustment disorder with depressed mood, and none with dysthymia). An additional 16 (11.4%) had an episode of major depression in the past. Of those with an MDD diagnosis, there was also a range of illness status, including still depressed, in partial remission (fewer than 5 symptoms of MDD), and remission (asymptomatic). A total of 43 (30.7%) had never been diagnosed with any depressive disorder. Similarly, QIDS-SR16 indicated a wide range of depressive symptoms. Fifty-nine adolescents (42.1%) reported no or minimal depressive symptoms (QIDS-SR16 ≤5); 38 (27.1%) reported mild depression; 23 (16.4%) reported moderate depression; 16 (11.4%) reported severe depression; and 4 (2.9%) reported very severe depression. Thus there was a wide range of depressive symptoms among the participants.

Depression Symptom Severity

Table 1 summarizes the CTT analyses for the QIDS (including individual interview scores for the adolescent and parent, composite score, and self-report). Sleep disturbance was the most common symptom, but sad mood and loss of general interest were most highly related to overall depression. The mean of the QIDS-C16(Comp) had the highest mean of the other versions of the QIDS.

Table 1
CTT Analysis of the QIDS-C16(Adol), QIDS-C16(Par), QIDS-SR16, and QIDS-C16(Comp)

Table 2 summarizes the CTT analysis of the CDRS-R. It is somewhat more difficult to compare CDRS-R items with one another since the items do not all have the same number of categories. This, however, applies minimally to the rit measures. Items on the QIDS with high item-total correlations (rit) have counterparts on the CDRS-R that also have a high rit, (e.g., sad mood).

Table 2
CTT Analysis of the CDSR17

Table 3 contains the observed intercorrelations among the scales (above diagonal), the coefficient alpha reliabilities (diagonal), and the disattenuated intercorrelations (below diagonal) among the scales. Because of the correlated error, three disattenuated correlations exceeded 1.0. Note that the QIDS-SR16 only correlates .68 with the CDRS-R but it correlates.81 with the QIDS-C16(Adol). The QIDS-C16(Comp) and the CDRS-R were highly correlated (r = .82). As expected, the QIDS-C16(Comp) correlated most with the QIDS-C16(Par) and the adolescent’s QIDS-C16(Adol). These latter two correlations were equal within rounding error (rs = .85). All measures were reasonably reliable, but reliability was lowest for the QIDS-C16(Par).

Table 3
Observed intercorrelations among measures (above diagonal), coefficients alpha reliabilities (diagonal), and disattenuated intercorrelations among measures (below diagonals)

Scale dimensionalities

Principal component analyses were conducted on each measure. Figure 1 shows that each of the four versions of the QIDS was unidimensional since the obtained eigenvalue for the first component exceeded the simulated value but the converse was true of the remaining components

Scree plots for the QIDS-C16(Adol), QIDS-C16(Par), QIDS-C16(Comp) and QIDS-SR16

In contrast, the CDRS-R (Figure 2) contains at least two and possibly three dimensions because the crossover of the real and simulated eigenvalues occurs between the third and fourth principal components. When examined, a three-factor solution was difficult to interpret because of the sparseness of salient variables (variables having large loadings) on the second and third factors. The results of a promax (oblique) rotation on the two-factor solution generated a modest correlation (−.27) with items 1–14 loading on the first factor and items 15–17 loading on the second.

Scree plot for the CDRS-R

IRT Comparisons

Because the QIDS-C16(Comp) interview was equivalent to the CDRS-R, further examination of the components of the QIDS-C16 interview (including the adolescent and parent interviews) as well as the adolescent self-report measure was conducted using IRT comparisons.

QIDS-C16(Adol) vs. QIDS-SR16

The slopes of the category response functions were steeper on the self-report version (QIDS-SR16) than on the clinical version [QIDS-C16(Adol)] in all nine domains, but the differences was significant only for domain 7 (general interest), G2(1) = 3.9, p<.05. Adolescents alone rated themselves as significantly more pathological than the clinician in two domains: sad mood (G2(2) = 15.9, p<.01) and appetite (G2(3) = 8.2, p<.05). The reverse was true in three domains: 4, 5, and 9 (concentration/decision making, self-view, and restlessness/agitation), G2(2) = 24.1, G2(3) = 20.6, and G2(3) = 47.1, p<.01.

QIDS-C16(Comp) vs. QIDS-SR16

The slopes of the category response functions were steeper on the QIDS-SR16 than with the QIDS-C16(Comp) on eight of nine domains (all but appetite). Thus, as in the above comparisons, the individual domains tended to be more discriminating when rated by the adolescents themselves, but the only significant difference was on domain 9 (restlessness/agitation) G2(1) = 4.5, p<.05. The composite led to higher estimates of pathology on five domains: 1, 5, 7, 8, and 9 (sleep, thoughts of death or suicide, general interest, energy level, and restlessness/agitation), G2(3) = 10.2, G2(3) = 88.1, G2(3) = 8.1, G2(2) = 18.8, and G2(2) = 66.2, ps < .01, .05, .01, .01, and .01.

QIDS-C16(Adol) vs. QIDS-C16(Par)

The slopes of the category response functions were steeper in eight of nine domains (all but general interest) for the QIDS-C16(Adol) ratings than the QIDS-C16(Par) ratings. Thus, the individual domains tended to be more discriminating when rated by the adolescents themselves than when rated by their parents alone, though statistical significance was achieved only for domain 3 (appetite), G2(1) = 5.9, p<.02. Two sets of locations differed: the adolescents’ self-views were rated more negatively when the information came from the adolescents alone than when it came from their parents alone (G2(3) = 8.6, p<.05), but the reverse was true regarding judgments of their restlessness/agitation (G2(3) = 9.3, p<.05).

Test Information Functions

Figure 3 contains the test information function for the various depression measures. This indicates that the CDRS-R is most sensitive to measuring depression (Θ, which is a standard designation for the abscissa) up to a depression level of approximately 1.0 (i.e., from no depression to mild-moderate depression). At this point, the QIDS-SR16, which performed poorly at the lower levels of depression, performed best by a slight margin. Past Θ = 2, i.e., at severe levels of depression, the CDRS-R again is most sensitive. The QIDS-C16(Comp) does slightly better than the QIDS-C16(Adol) up to Θ = −.5 (i.e., at lower to moderate levels of depression). Past this point, the QIDS-C16(Adol) performs slightly better. The QIDS-C16(Par) performs most poorly across all levels of depression (Θ).

Test information functions (TIF) for the QIDS-C16(Adol), QIDS-C16(Par), QIDS-C16(Comp), QIDS-SR16, and CDRS-R

Contributions of the QIDS-C16(Adol) and QIDS-C16(Par) to the QIDS-C16(Comp)

As noted above, the QIDS-C16(Adol) and QIDS-C16(Par) contributed nearly equally to the QIDS-C16(Comp). In addition, Table 4 contains the mean differences by domain between the QIDS-C16(Adol) and QIDS-C16(Par). Positive values denote that the adolescent’s response was more pathological than the parents’ report of the adolescent; negative values denote that the parents’ assessment was the more pathological. The reported p-values do not correct for the nine comparisons. Following Bonferroni correction, only the difference for domain 9 (psychomotor symptoms) remained significant. Thus, adolescents’ and parents’ evaluations of symptom severity were similar.

Table 4
Mean Differences by Domain: QIDS-C16(Adol) - QIDS-C16(Par)


All measures, including the QIDS-C16(Par) are of acceptable reliability, with the QIDS-SR16 and CDRS-R having the highest overall reliability. Based on IRT analyses, the QIDS-SR16 performs best at moderately high levels of depression, while the CDRS-R does best at low and very high levels. The QIDS-SR16 domains tended to be more discriminating than the other versions of the QIDS, but these differences were usually not significant. There were differences among the QIDS methods in terms of differences in tendencies to endorse symptoms, but the directions of difference were not consistent over domains. The results further indicated that the level of pathology reported by adolescent and parent was similar within domains. It is probably valuable to probe symptoms that adolescents report that the parent does not report and vice versa.

The QIDS measures were all unidimensional, meaning that the nine domains all measure the same thing -- depressive symptomatology. The CDRS-R, on the other hand, consisted of two factors of depression. One interpretation is that the first factor (29% of the total variance), which included items 1 through 14, represents symptoms (subjective aspects) of depression, whereas the second factor (14% of the total variance), items 15–17, represents signs (visible features) of depression. The two factors may thus be viewed as reflecting method variance that reflects with how depression is assessed rather than content variance that refers to different forms of depression.

One important finding was the reliability of the QIDS-SR. There has been question in the past about the reliability of self-reports in youth, and as such, most clinical trials of depression require a clinician-rated measure for primary outcome. In this study, the self-report was equally reliable, and even more so at moderate levels of depression. This finding is important both clinically and scientifically. First, the QIDS-SR is significantly more cost-effective, particularly for clinical practice — both due to use of a self-report (reducing clinician burden), but also in that it is free and available on the QIDS Web site (http://www.ids-qids.org/). Second, in some cases, involvement of parents may be limited (both in clinical and research settings), so use of measures that do not require parental input may be beneficial in some settings. On the other hand, the CDRS-R is somewhat more sensitive to slight differences in depression at the low and very high ends of this sample. The disattenuated correlation of .78 means that nearly half of the variance in each test is not shared with the other test. The CDRS-R is currently considered the gold standard outcome measure for pediatric depression, and contains an item on irritability, which is an important symptom in adolescent depression (no corresponding symptom on the QIDS).

Regarding the QIDS-C, it may also be a useful measure as a clinician rating. Similar to the CDRS-R, it requires obtaining information from both the adolescent and parent, and then creating a composite score. In this study, the fact that the QIDS-C16(Comp) mean exceeded the other clinical versions is a necessary outcome of choosing the more pathological response from the two individual interviews. While the composite score for the present study was created by taking the most pathological rating from the individual interviews, the composite score could also be created by using clinical judgment, as is done with the CDRS-R. The use of an adolescent self-report also complements the QIDS-C, which enhances its utility.

In sum, the present results indicate that both the QIDS-SR16 and the CDRS-R should be viewed as viable and useful, though additional research is needed. In sum, there are pros and cons associated with each of the measures and differences among them are not great. There is perhaps a slight advantage to using the CDRS-R in a less severely depressed population and the various versions of the QIDS with a more severely depressed sample.

There were three important limitations. First, the sample is only of moderate size. The use of parallel analysis rather than the more common “eigenvalues greater than one” rule to infer dimensionality compensates in part for the role of sample size. Second, the composite score for the QIDS-C was based on the most pathological response from the adolescent and parent interviews, allowing for limited clinical judgment to rate the final clinical rating. Outcome ratings generally require a synthesis of all available information, and it is possible that a composite score that requires clinical judgment of the reliability of the adolescents and parents may have yielded different results. Third, there was only one interviewer so inter-rater reliability could not be assessed. Finally, irritability, a DSM-IV sign of adolescent, but not adult, depression, is not included in the QIDS. It is therefore not clear from the present results how much of a role irritability might play in defining adolescent depression.

Although one might view the fact that around 30% of the sample had never been depressed as an additional limitation, this composition: (a) is perhaps typical of adolescent outpatient units in general, and (b) allowed us to evaluate the scales over a wider range of depression magnitudes than would have been the case with a sample that only consisted of patients with major depressive disorder. In turn, this allowed us to note the differences between the CDRS-R and the various versions of the QIDS that perhaps would not have been found within a purely depressed sample.


Using the CDRS-R as a standard, both the QIDS-C16 and the QIDS-SR16 are viable measures of adolescent depression. The addition of a brief, symptom domain based, reliable self-report for adolescent depression contributes to the assessment of depressive severity and as a potential outcome measure for monitoring response to treatment, both clinically and in research studies.


We appreciate the assistance of Catherine Karni, MD (Director) and the staff of the Children’s Medical Center Outpatient Division of Child and Adolescent Psychiatry for their administrative support.

This research was funded in part by the National Institute of Mental Health (NIMH), National Institutes of Health (MH-68852 to the University of Texas at Arlington, Ira H. Bernstein, Ph.D., PI; and MH-68851 to the University of Texas Southwestern Medical Center at Dallas, A. John Rush, M.D., PI).



Dr. Rush has received research support from the National Institute of Mental Health and Stanley Medical Research Institute; has been on the advisory boards and/or consultant for Advanced Neuromodulation Systems, AstraZeneca, Best Practice Project Management, Bristol Myers Squibb, Cyberonics, Forest Pharmaceuticals, Gerson Lehman Group, GlaxoSmithKline, Jazz Pharmaceuticals, Magellan Health Services, Merck & Company, Neuronetics, Novartis Pharmaceuticals, Ono Pharmaceuticals, Organon, Pamlab, Pfizer, Transcept Pharmaceuticals, Urban Institute, and Wyeth-Ayerst; has been on the speaker’s bureau for Cyberonics, Forest Pharmaceuticals, and GlaxoSmithKline; has equity holdings (excluding mutual funds/blended trusts) in Pfizer; and has royalty income affiliations with Guilford Publications and Healthcare Technology Systems. Dr. Trivedi has been a consultant and/or served on speakers bureaus for Abbott Laboratories, Inc.; Abdi Brahim, Akzo (Organon Pharmaceuticals Inc.); AstraZeneca; Bayer; Bristol-Myers Squibb Company; Cephalon, Inc.; Fabre-Kramer Pharmaceuticals, Inc. Forest Pharmaceuticals; GlaxoSmithKline; Janssen Pharmaceutica Products, LP; Johnson & Johnson PRD; Eli Lilly & Company; Meade Johnson; Neuronetics; Parke-Davis Pharmaceuticals, Inc.; Pfizer, Inc.; Sepracor; VantagePoint; and Wyeth-Ayerst Laboratories. He has received grant support from the Agency for Healthcare Research and Quality, Corcept Therapeutics, Inc.; Cyberonics, Inc.; National Alliance for Research in Schizophrenia and Depression; Merck; National Institute of Mental Health; National Institute on Drug Abuse; Novartis; Pharmacia & Upjohn; Predix Pharmaceuticals; Solvay Pharmaceuticals, Inc.; and Targacept. Dr. Hughes is a consultant for Biobehavioral Diagnostics Inc. Dr. Emslie has received research support from the National Institute of Mental Health, Biobehavioral Diagnostics Inc., Forest Laboratories, Organon, Shire, and Somerset; has been a consultant for Biobehavioral Diagnostics Inc., Eli Lilly, Forest Laboratories Inc, GlaxoSmithKline, Pfizer Inc., Shire, Validus Pharmaceuticals, and Wyeth Pharmaceuticals; and has been on the Speaker’s Bureau for McNeil. Ira H. Bernstein, Laurie Macleod, and Bradley Witte have no disclosures to report.


  • American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed, Text Revision. Washington DC: American Psychiatric Press; 2000.
  • Birmaher B. Longitudinal course of pediatric bipolar disorder. Am J Psychiatry. 2007;164:537–539. [PubMed]
  • Brent DA. Correlates of the medical lethality of suicide attempts in children and adolescents. J Am Acad Child Adolesc Psychiatry. 1987;26:87–91. [PubMed]
  • Coyle JT, Pine DS, Charney DS, Lewis L, Nemeroff CB, Carlson GA, Joshi PT, Reiss D, Todd RD, Hellander M. Depression and bipolar support alliance consensus statement on the unmet needs in diagnosis and treatment of mood disorders in children and adolescents. J Am Acad Child Adolesc Psychiatry. 2003;42:1494–1503. [PubMed]
  • Fergusson DM, Woodward LJ. Mental health, educational, and social role outcomes of adolescents with depression. Arch Gen Psychiatry. 2002;59:225–231. [PubMed]
  • Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23:56–62. [PMC free article] [PubMed]
  • Hamilton M. Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol. 1967;6:278–296. [PubMed]
  • Horn JL. An empirical comparison of various methods for estimating common factor scores. Educ Psychol Meas. 1965;25:313–322.
  • Humphreys LG, Ilgen D. Note on a criterion for the number of common factors. Educ Psychol Meas. 1969;29:571–578.
  • Humphreys LG, Montanelli RG., Jr An investigation of the parallel analysis criterion for determining the number of common factors. Multivariate Behav Res. 1975;10:193–206.
  • Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:617–627. [PMC free article] [PubMed]
  • Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, Wittchen HU, Kendler KS. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994;51:8–19. [PubMed]
  • Montanelli RG, Jr, Humphreys LG. Latent roots of random data correlation matrices with squared multiple correlations on the diagonal: a Monte Carlo study. Psychometrika. 1976;41:341–348.
  • Nunnally JC, Bernstein IH. Psychometric Theory. New York: McGraw-Hill; 1994.
  • Olfson M, Shaffer D, Marcus SC, Greenberg T. Relationship between antidepressant medication treatment and suicide in adolescents. Arch Gen Psychiatry. 2003;60:978–982. [PubMed]
  • Poznanski EO, Mokros HB. Children's Depression Rating Scale, Revised (CDRS-R) Manual. Los Angeles, CA: Western Psychological Services; 1995.
  • Rao U, Ryan ND, Birmaher B, Dahl RE, Williamson DE, Kaufman J, Rao R, Nelson B. Unipolar depression in adolescents: clinical outcome in adulthood. J Am Acad Child Adolesc Psychiatry. 1995;34:566–578. [PubMed]
  • Rush AJ, Bernstein IH, Trivedi MH, Carmody TJ, Wisniewski S, Mundt JC, Shores-Wilson K, Biggs MM, Woo A, Nierenberg AA, Fava M. An evaluation of the Quick Inventory of Depressive Symptomatology and the Hamilton Rating Scale for Depression: a Sequenced Treatment Alternatives to Relieve Depression trial report. Biol Psychiatry. 2006;59:493–501. Epub 2005 Sep 30. [PMC free article] [PubMed]
  • Rush AJ, Carmody TJ, Reimitz PE. The Inventory of Depressive Symptomatology (IDS): clinician (IDS-C) and self-report (IDS-SR) ratings of depressive symptoms. Int J Methods Psychiatr Res. 2000;9:45–59.
  • Rush AJ, Gullion CM, Basco MR, Jarrett RB, Trivedi MH. The Inventory of Depressive Symptomatology (IDS): psychometric properties. Psychol Med. 1996;26:477–486. [PubMed]
  • Rush AJ, Trivedi MH, Carmody TJ, Ibrahim HM, Markowitz JC, Keitner GI, Kornstein SG, Arnow B, Klein DN, Manber R, Dunner DL, Gelenberg AJ, Kocsis JH, Nemeroff CB, Fawcett J, Thase ME, Russell JM, Jody DN, Borian FE, Keller MB. Self-reported depressive symptom measures: sensitivity to detecting change in a randomized, controlled trial of chronically depressed, nonpsychotic outpatients. Neuropsychopharmacology. 2005;30:405–416. [PubMed]
  • Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, Markowitz JC, Ninan PT, Kornstein S, Manber R, Thase ME, Kocsis JH, Keller MB. The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biol Psychiatry. 2003;54:573–583. Erratum p. 585. [PubMed]
  • Samejima F. Estimation of latent ability using a response pattern of graded scores. Psychol Monogr. 1969;4:2.
  • Samejima F. Graded response model. In: van Linden W, Hambleton RK, editors. Handbook of Modern Item Response Theory. New York: Springer-Verlag; 1997. pp. 85–100.
  • Shaffer D, Gould MS, Fisher P, Trautman P, Moreau D, Kleinman M, Flory M. Psychiatric diagnosis in child and adolescent suicide. Arch Gen Psychiatry. 1996;53:339–348. [PubMed]
  • Trivedi MH, Rush AJ, Ibrahim HM, Carmody TJ, Biggs MM, Suppes T, Crismon ML, Shores-Wilson K, Toprac MG, Dennehy EB, Witte B, Kashner TM. The Inventory of Depressive Symptomatology, Clinician Rating (IDS-C) and Self-Report (IDS-SR), and the Quick Inventory of Depressive Symptomatology, Clinician Rating (QIDS-C) and Self-Report (QIDS-SR) in public sector patients with mood disorders: a psychometric evaluation. Psychol Med. 2004;34:73–82. [PubMed]
  • Weissman MM, Bruce ML, Leaf PJ, Florio LP, Holzer C. Affective disorders. In: Robins LN, Regier DA, editors. Psychiatric Disorders in America: The Epidemiologic Catchment Area Study. New York: The Free Press; 1991. pp. 1351–1387.
  • Weissman MM, Wolk S, Goldstein RB, Moreau D, Adams P, Greenwald S, Klier CM, Ryan ND, Dahl RE, Wickramaratne P. Depressed adolescents grown up. JAMA. 1999;281:1707–1713. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...