On Track Outcomes Program Client Feedback Form Manual Prepared for Beacon Health Options by Jeb Brown, PhD, Center for Clinical Informatics February 2014 For more information about this manual contact Jeb Brown, Ph.D. at [email protected], or call (801) 541-9720 For more information about the On Track Outcomes Program send email to [email protected]
37
Embed
Client Feedback Form Manual - Beacon Health Options...Client Feedback Form Manual Prepared for Beacon Health Options by Jeb Brown, PhD, Center for Clinical Informatics February 2014
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
On Track Outcomes Program
Client Feedback Form Manual
Prepared for Beacon Health Options by Jeb Brown, PhD, Center for Clinical Informatics
February 2014
For more information about this manual contact Jeb Brown, Ph.D. at
Global Distress Scoring and Interpretation............................................................................................... 8
Client Feedback Form Score Ranges for the Three Severity Levels ...................................................... 9
Monitoring Change over Time.................................................................................................................. 9
Case Mix Adjustment and Outcomes Benchmarking ............................................................................. 10
Reliability and Validity............................................................................................................................. 11
Section 4: CFF-Adult Specific Information.........................................................................................13
Reliability and Construct Validity ............................................................................................................ 13
Appendix A: Adult CFF Normative and Benchmarking Data ..............................................................28
Results of GLM prediction - adding variables incrementally ............................................................. 29
Beacon Health Options 3 Client Feedback Forms
Appendix B: Youth Self-Report CFF Normative and Benchmarking Data ............................................30
Results of GLM prediction - adding variables incrementally ............................................................. 31
Appendix C: Youth - Parent CFF Normative and Benchmarking Data .................................................32
Results of GLM prediction - adding variables incrementally .............................................................. 33
Appendix D: Child - Self CFF Normative and Benchmarking Data .......................................................34
Results of GLM prediction - adding variables incrementally .............................................................. 35
Appendix E: Child - Parent CFF Normative and Benchmarking Data ...................................................36
Results of GLM prediction - adding variables incrementally ............................................................. 37
Beacon Health Options 4 Client Feedback Forms
Section 1: Overview of the Client Feedback Form
The Client Feedback Forms (CFFs) used in Beacon’s On Track Outcomes Program are brief, reliable and
valid client completed questionnaires designed to measure improvement in symptoms, quality of social
relations, and functioning at work/school and other daily activities. Use of these or similar outcome
questionnaires has been shown to significantly improve treatment outcomes across large sample of
clinicians treating thousands of clients.
The questionnaires measure symptom severity and improvement across a broad range of problems, and
are not intended to be diagnostic or to substitute for clinical evaluation or other assessments that a
clinician may routinely conduct.
There are three versions of the CFF now available:
Adult version (age 18 and older)
Youth version (ages 13-17; completed by youth or adult)
Child version (under age 13; completed by youth or adult)
All versions of the questionnaires have high reliability (coefficient alpha => .87). Extensive factor
analyses demonstrate high construct validity, with items loading on the common factor found in the
most commonly used measures of treatment outcomes.
The Client Feedback Forms are unique among available outcome measures in that items asking for
feedback on the working alliance are included on every questionnaire. The use of alliance items has
been shown to reduce no shows and premature termination while contributing significantly to
improved outcomes.
Administration
The questionnaires were designed for routine use in clinical practice and can be completed by
most clients in less than two minutes, using a simple paper form.
When first introducing the questionnaire to a client, it is best practice to provide a brief
explanation of the reason for the questionnaire, and how it will be used as a routine part of
treatment. At many sites, this explanation is provided by office staff when the client checks in.
Following are a sample explanation scripts used by support staff:
Beacon Health Options 5 Client Feedback Forms
“Please take a moment to fill out the questionnaire. This will help you and your therapist
talk about how treatment is going.”
“We are really excited about the questionnaires. Research shows that therapists who use
questionnaires like these get much better outcomes.”
“Please answer as honestly as you can. This is important to your treatment, because it will
help your therapist understand how to help you.”
Research asking clients to provide feedback on their experience with the questionnaires
indicated a high level of honesty and willingness to complete the alliance items, especially if
they perceived that the clinician was interested in their responses. Of this group, 95% agreed
that the questionnaires were helpful in treatment.
For this reason it is recommended that the clinician acknowledges and value the client’s
willingness to give feedback. Taking a few moments at the start of the session to review the
questionnaire provides the clinician with a wealth of clinically relevant information while
acknowledging the value of the client’s time to complete the questionnaire.
Research of the past several years has revealed that clinicians who use the questionnaires with
a high percentage of their practice display significant gains in overall outcomes.
Outcomes Questionnaires and Clinician Feedback
A large body of research supports the proposition that routine use of outcomes questionnaires
combined with feedback to the clinician results in significant improvement in treatment
outcomes, as measured by pre-post change, percentage of patients improved, and reduced
dropout rates (Lambert, 2009; Goodman et al., 2013).
Among users of On Track Client Feedback Forms and related ACORN questionnaires, observation of frequency of measurement and use of the Decision Support Toolkit provides real world confirmation of results from clinical trials. Both the number of clients measured and the frequency at which the clinician views the data are strongly correlated with year-to-year improvement in treatment outcomes at the clinician level (r> .3; p<.0001; see Brown 2013 in references).
Beacon Health Options 6 Client Feedback Forms
Section 2: Development, Scoring, and Interpretation Development
The CFFs were developed for Beacon’s On Track Outcomes program as part of a suite of forms
developed cooperatively through A Cooperative Online Resource Network (ACORN). The
network consists of statisticians and researchers affiliated with the Center for Clinical
Informatics along with over 1,000 clinicians using client completed questionnaires in a wide
variety of clinical settings across the country.
The initial work on developing the ACORN questionnaire items was done in collaboration with
Warren Lambert, Ph.D., at Vanderbilt University. Dr. Lambert was instrumental in the
development of the Peabody Treatment Progress Battery (PTPB) for adolescents. Items from
the PTPB were utilized with permission, and additional items for adults, adolescents and
children were added using the same item format. Normative data on these items was collected
through the ACORN network of clinicians. Takuya Minami, Ph.D., of the University of
Massachusetts Boston, further assisted in the analyses of the psychometric properties of all of
the items.
The ACORN form development process takes advantage of the network’s ability to collect large
amounts of data and continuously test and refine items over time. Rather than focus on the
development of static forms, this process centers on the development of reliable and valid
items that can be combined flexibly. The psychometric properties of each item, as well as the
unique combination, are carefully evaluated with items for a specific questionnaire selected
based on the population to be measured and the needs of the various participating
organization. The result is a large item inventory with data from over 200,000 clients.
Evidence of the validity and reliability of the CFF was derived from data on tens of thousands of
administrations in both clinical and community settings. The development process included
item analyses to determine:
Item frequencies and distributions
Item correlations
Factor structure
Construct validity
Scale reliability
Sensitivity to change The CFFs consists of items well-suited for general use in outpatient settings. The questionnaires
are designed to be as brief as possible while retaining excellent psychometric properties. The
Beacon Health Options 7 Client Feedback Forms
Adolescent and Child versions may be completed either by the youth or by an adult who knows
the youth well.
Absenteeism/Presenteeism
The two absenteeism/presenteeism items on the adult form are based on the format and
wording of items in the Health and Productivity Questionnaire (HPQ), a questionnaire in the
public domain developed by Kessler and colleagues at Harvard University. Because the CFF is
designed for use by mental health professionals in a mental health setting, the HPQ wording
has been revised to focus on absenteeism/presenteeism specifically due to mental health
problems, rather than both mental and physical health problems encompassed by the original
HPQ questions.
Alliance Scale
All ACORN questionnaires also include items asking the client to provide feedback on their
experience of the prior session. The use of these so-called Alliance items ask for feedback on
elements of the therapeutic working alliance, such as agreement on treatment goals, and the
client’s perception of the quality of the relationship. Use of alliance measure has been
demonstrated to reduce treatment dropout and improve outcomes. Among the large sample
of clinicians using the CFFs, about 80% of clients complete these items routinely. These clients
have significantly better outcomes than those who fail to complete the items.
The client’s responses on these items may be easily influenced by their perception of the
clinician’s response. Clients may be reluctant to give anything other than perfect ratings in
order to avoid hurting the clinician’s feelings or out of fear of possible consequence to the
clinician if “Satisfaction” ratings are high.
It rests on the skill of the clinician to create a therapeutic environment that encourages honest
feedback on the alliance items. Skilled clinicians are about to use the alliance items as tools to
foster a strong collaborative working alliance with the client.
Consistent ratings of near perfect alliance are NOT associated with the best outcomes. Rather,
clients who provide meaningful feedback early in therapy are very likely to rate the alliance as
improved over time. This pattern, displayed by approximately one third of clients, is associated
with significantly greater improvement in treatment.
The best outcomes are associated with improvement on the Alliance Scale over the course of
the treatment episodes. This means that the patients with the best outcomes are also willing
to give feedback that the treatment encounters early in the treatment episode are less than
perfect, otherwise there is no room for improvement.
Beacon Health Options 8 Client Feedback Forms
Failing to complete the alliance items is associated with less improvement. If the client leaves
the items blank, this provides the clinician with an opportunity to initiate a discussion of how
the client is experiencing the treatment process.
Global Distress Scoring and Interpretation
The core global distress scale (GDS) is scored as the mean of all non-missing items on the form.
If more than 4 items are missing from the adult global distress scale or 6 items from the child or
youth scales, the questionnaire is not scored.
Scores can be divided into three severity ranges (Normal, Moderate, Severe) based on
normative data from clinical and community samples. The cut-off scores for each range are
included with the information on the individual questionnaires.
In the case of the adult questionnaire, the cut-off score for the Normal Range was determined
by collecting a sample of over 1,000 individuals from the community who had never sought
mental health service combined with a sample of over 75,000 individuals receiving mental
health services. The cut-off score was calculated using the method proposed by Jacobson &
Truax (1991) as represented by this formula:
C = (SD1)(mean2)+(SD2)(mean1)
SD1+SD2
A score in the normal range means that the score is in a range typical of respondents from a
community sample that have never sought mental health services. Seventy-five percent (75%)
of a community sample and 25% of a clinical sample will fall into this range. Mental health
clients with scores in this range tend to not show improvement with treatment.
A score in the moderate range of distress is characteristic of individual seeking mental health
services. About 50% of a clinical sample will fall into this range. Mental health clients with
scores in this range tend to show significant improvement within a few sessions of therapy, and
most complete treatment with a good outcome in fewer than 8 sessions.
Twenty five percent (25%) of a clinical sample will have scores in the severe range, while fewer
than 10% of a community sample will fall in this range. Clients with scores in this range are
highly likely to show rapid improvement with psychotherapy, but may need more sessions to
realize the full benefit of treatment.
Beacon Health Options 9 Client Feedback Forms
In the case of child and adolescent questionnaires, cut-off scores were estimated by the 25th
and 75th percentile. Practicality prevented collection of large enough community samples.
However, the cut-off scores are comparable to those reported for other similar measures with
the OQ-45 for adults and YOQ-64 for children and youth. Items were selected for the Client
Feedback Form in order to assure comparability to the OQ-45 and YOQ-64. Analysis of archival
data for the OQ-45 and YOQ-64 confirms that the Client Feedback Forms produce results very
similar to these OQ measures.
Client Feedback Form Score Ranges for the Three Severity Levels
Form Normal Range Moderate Range Severe Range Adult CFF 0 to 1.5 1.6 to 2.5 2.6 to 4.0 Adolescent CFF
(Youth completed) 0 to 1.0 1.1 to 1.8 1.9 to 4.0
Adolescent CFF (Parent/Adult completed)
0 to 1.2 1.3 to 1.9 2.0 to 4.0
Child CFF (Youth completed)
0 to 1.0 1.1 to 1.8 1.9 to 4
Child CFF (Parent/Adult completed)
0 to 1.2 1.3 to 1.9 2.0 to 4.0
Monitoring Change over Time
The ability for the clinician to monitor client change as the treatment unfolds is one of the
features of On Track that leads to improved outcomes. Identification of clients who are “off
track” assists the clinician in preventing early drop-out in treatment.
The On Track Decision Support Toolkit provides graphs for each case. The actual client scores
are compared to a “predicted score” at each assessment point. The predicted score is
determined using a statistical prediction technique known as General Linear Modeling. The
method takes advantage of all of the normative data for other clients completing multiple
questionnaires at multiple points in treatment.
The predicted score at each assessment is computed using the initial CFF Global Distress Score,
the assessment number, and the number of weeks that have passed since the initial
assessment to determine the expected assessment at each measurement point. The actual
score can them be compared to the predicted score in order to determine the extent that the
clients current score deviates from the expected score. Clients who scores are significantly
higher than expected at classified as “off track”.
Beacon Health Options 10 Client Feedback Forms
Case Mix Adjustment and Outcomes Benchmarking
Simply measuring pre-post change on an outcome questionnaire provides little information
without some basis for comparison or benchmarking. The On Track program is based on a long
history of research using real world effectiveness data as well as meta-analyses of clinical trials
to establish efficacy.
At the most basic level, change scores on the CFF questionnaires are converted to effect size,
based on dividing the pre-post global distress change score by the standard deviation of the
global distress scores at intake. A simple effect size is calculated by dividing the pre-post change
score by the standard deviation of the outcome measure at intake. An effect size of one means
that the client improved one standard deviation on the measure.
In order to make results comparable to results from clinical trials, effect size is only calculated
for cases with intake scores above the clinical cutoff score. In a general outpatient population,
75% of cases will be in the clinical range. Another reason for excluding non-clinical range cases is that these who enter treatment
reporting little distress do not show improvement on average. This would have the effect of
artificially lowering effect sizes and making comparisons to results from published studies
invalid.
The second manner in which outcomes are benchmarked is by using a large and diverse
normative sample of over 90,000 outpatient treated at hundreds sites around the country. The
sample represents a wide range of ages and ethnic groups, includes those covered by
commercial insurance, EAP programs, Medicaid/Medicare, as well as self-pay and other sources
of funding. As such, the sample is highly representative of patients seen in outpatient general
practice.
In order to establish a benchmark, the statistical procedure known as General Linear Model was
applied to establish which variables collected at intake predicted the final global distress score at
the end of the episode.
Employing this model, it us evident that the first global distress score is the strongest predictor
of subsequent scores, including the final score. The intake score alone accounts for 30% to 50%
of the variance in final scores, depending on the length of treatment.
A second predictor in naturalistic data such as On Track is the session at which the first
questionnaires were administered. If the first assessment is at intake or no later than the
second appoint, measured change is larger than if the first assessment is later. In many cases,
the session number for the first assessment is unknown. In order to take session number into
account, each episode is classified as Early Assessment (session 1 or 2), Later Assessment, or
Beacon Health Options
11 Client Feedback Forms
Unknown. These three categories are included in the General Linear Model as a class variable,
but explain less than 1% of additional variance.
A third potential predictor is diagnosis. Inclusion of diagnosis adds minimally to the prediction
of final score, typically explaining less than .005% of additional variance. In some instances,
such as the youth self-reported version of the CFF, diagnosis is a non-significant predictor.
Diagnosis is not currently included in the On Track benchmarking models.
The prediction of the final (benchmark) score for the On Track program is based on predictive
formulas. These formulas take the form of a simple regression formula: Last Score = First Score
* slope + intercept). The regression formulas are coded from each version of the questionnaire,
with regression formulas differing based on session number at first assessment. Sections 4-6
contain form-specific analyses, including the specific formulas used for each CFF form and
session number.
The appendices provide detailed normative information for each questionnaire with regards to
mean intake scores, final scores, change scores, number of assessments, and average time pre-
post (reported in weeks) for the entire sample, including a breakout by diagnostic group. A
second set of normative information is provided for only those cases with intake scores in the
clinical range.
Each patient’s actual final score is compared to the benchmark target to determine to what
extent the patient reported more or less improvement than comparable patient in the
complete normative sample. The results of the General Linear Model analyses from which the
formulas are derived are provided in the appendices.
The general methodology for benchmarking outcomes has been published in a series of peer
reviewed journal articles, though in these publications the benchmark scores are calculated
using a multivariate GLM. These publications are available upon request (Minami et al. 2007;
Minami et al. 2008a and 2008b; Minami et al. 2011).
Reliability and Validity
Reliability of the global distress scale is measured using Cronbach’s alpha, which is a measure of
internal consistency. This is consistent with classical test theory, which seeks to develop single
factor scales with internal consistency.
Reliability for the CFFs are as follows:
Adult version = 0.89
Youth version = 0.87
Child version = 0.90
Beacon Health Options 12 Client Feedback Forms
Validity of the scales is estimated primarily as construct validity, addressing whether the
questionnaire measures a single construct or factor. Prior research indicates that items on most
outcomes questionnaires used in behavioral health correlate highly with one another and
measure a single construct, generally referred to as “Global Distress.”
For example, Brophy et al. (1988) found that the SCL-90 subscales all load on a common factor, and likewise correlate highly with similar scales from other measures. Miller et al. (2003) found that the Outcome Rating Scale correlates highly with the OQ-45.
Enns et al. (1998) performed factor analyses on the Beck Depression Inventory and the Beck Anxiety Inventory. To quote from this study:
" [T]he parameter estimate was very high (0.784) and a unidimensional, single-factor model of negative affectivity approached the criteria for good fit. It was concluded that the Beck Anxiety and Depression Inventories assess distinct anxiety and depression phenomena to a limited extent when used in a clinically depressed sample."
Recent research by the ACORN collaboration further investigates the relationship between
items assessing states of emotional well-being and high life satisfaction with measures of
psychiatric symptoms and lost productivity (Brown & Minami, 2013). This work demonstrates
that measures of well-being and life satisfaction likewise correlate highly with the same
common factor as symptoms and lost productivity.
The existence of a global distress factor and the fact that multiple outcome questionnaires,
including widely used measures of depression, are all found to be correlated with one another
provide strong evidence of the construct validity of patient self-report outcome measures
designed to measure and assess global subjective distress.
The estimate of reliability and validity for the Alliance Scale is complicated by the fact that the
responses are not normally distributed. Rather, approximately 50% of clients report the alliance
is virtually perfect all of the time, with little variance from week to week. This pattern is the
most common, but is not associated with better than average outcomes. Less than perfect
Alliance Scores at the start of the treatment are associated with better outcomes if the client
remains engaged in treatment. To a large extent the utility of the Alliance Scale is dependent on
the skill of the clinician to elicit and utilize frank feedback from the client. The following sections describe the psychometric properties of each version of the Client
Feedback Form, including detailed information on item analyses.
Beacon Health Options
13 Client Feedback Forms
Section 4: CFF-Adult Specific Information Reliability and Construct Validity
The Adult Client Feedback Form consists of 10 items which assess symptoms of depression,
anxiety, social relationships, and functioning in work and other daily activities. Three additional
items assess problems related to substance abuse. In addition, the forms contain three items
which ask for feedback on the client’s last session experience. Use of these Alliance items is
associated with better outcomes.
The adult form also includes items asking about prior treatment and presence of chronic
illnesses. The variables may be used when performing calculations for case mix adjustment.
The following assessment of the factor structure and reliability of the Adult questionnaire is
based on a sample of 86,185 clients completing the questionnaire at the start of a treatment
episode.
Factor analyses (principal components and varimax rotation) reveal that these items related to
depression, anxiety, interpersonal problems, and impaired functioning in work, school and
other daily activities all load on a common factor labeled Global Distress. Table A displays the
results of this analysis.
Beacon Health Options 14 Client Feedback Forms
CFF-ADULT Factor Analysis (N=86,185)
Factor 1 in principle components is the Global Distress factor. Note that all GDS items have positive factor loadings of .54 or greater.
Varimax rotation does little to alter factor structure. Global Distress Scale contains a single factor.
Factor 1 in principle components is the Global Distress factor. Note that all GDS items have positive factor loadings of .45 or greater.
Rotated Factor 1 reflects symptoms of depression and anxiety. This accounts for more of the variance with Youth completed forms than Factor 2, which reflects attention and behavioral problems. Factor 3 is substance abuse.
Factor Method:
Principle Components
Factor Method:
Varimax Rotation
Scale
Item wording Factor
1
Factor 2
Factor 3
Factor 1
Factor 2
Factor 3
GDS/Symptoms …eat a lot more or a lot less than usual?
Factor 1 in principle components is the Global Distress Scale. Note that all but one GDS items have positive factor loadings of .49 or greater. A single item, shyness, is at .25
GDS/Symptoms …have a hard time controlling your temper?
0.742
-0.058
-0.229
0.684
0.210
0.221
GDS/Symptoms …worry about a lot of things? 0.485 0.605 0.033 0.079 0.752 0.174
GDS/Symptoms
…have a hard time sitting still?
0.627
-0.103
0.482
0.221
0.145
0.752
GDS/Symptoms …have a hard time paying attention?
0.672
-0.081
0.444
0.266
0.189
0.741
GDS/Social …have a hard time waiting your turn?
0.699
-0.223
0.223
0.468
0.093
0.601
GDS/Social … get into fights with family members and/or friends?
0.674
-0.045
-0.388
0.720
0.295
0.050
GDS/Social …feel nervous and/or shy around other people?
0.3450
0.576
0.093
-0.044
0.659
0.154
GDS/Functioning …lose things you need? 0.555 0.046 0.456 0.122 0.248 0.664
GDS/Social …argue with adults? 0.707 -0.213 -0.330 0.192 0.608 0.158
GDS/Social …annoy other people on purpose?
0.668
-0.266
-0.187
0.694
0.077
0.254
GDS/Social …think that you don't have any friends?
0.493
0.432
-0.037
0.192
0.608
0.158
Variance Explained by Each Factor
6.162
1.888
1.152
3.850
2.732
2.620
Final Commonality Estimate
9.202401
Beacon Health Options 24 Client Feedback Forms
Child CFF Factor Analysis (Responder=Parent or other adult, N=6,114)
Factor 1 in principle components is the Global Distress Scale. Note that all but one GDS items have positive factor loadings of .49 or greater. A single item, shyness, is at .25
Factor structure is same as for Child completed, but with different order in factors. Rotated Factor 1 reflects behavioral problems. Rotated Factor 2 reflects problems with
attention. Rotated Factor 3 reflects internalizing symptoms of depression and anxiety. Combined Rotated Factors 1 and 2 for measure of externalizing symptoms.
Enns MW , Coxa BJ, Parker B JDA, & Guertinc JE (1998) Confirmatory factor analysis of the Beck Anxiety and Depression Inventories in patients with major depression Journal of Affective Disorders 47 195-200.
Goodman JD, McKay JR, DePhilippis D (2013) Progress Monitoring in Mental Health and
Addiction Treatment: A Means of Improving Care. Professional Psychology: Research and
Practice 44(4) 231-246.
Henry W. (1997) Harvard School of Public Health College Alcohol Study, Ann Arbor, MI: Inter-
university Consortium for Political and Social Research.
Jacobson, NS., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12-19.
Lambert MJ (2009) Yes, It Is Time for Clinicians to Routinely Monitor Treatment Outcome. In
Miller, S. & Hubble, (M. Eds.), Heart and Soul of Change (2nd ed.). American Psychological
Association Press: Washington.
Miller DD, Duncan Bl, Brown J et al. (2003) The Outcome Rating Scale: A preliminary study of
the reliability, validity , and feasibility of a brief visual analog measure. Journal of Brief Therapy
Minami T, Brown GS, McCulloch J, Bolstrom B. (2011) Benchmarking therapists: Furthering the benchmarking method in its application to clinical practice. Quality & Quantity. 46:1699-1708.
Minami T, Wampold BE, Serlin RC, Hamilton EG, Brown GS, & Kircher JC. (2008a) Benchmarking the effectiveness of psychotherapy treatment for adult depression in a managed care environment: A preliminary study. Journal of Consulting and Clinical Psychology. 76, 116-124
Minami, T, Serlin, RC, Wampold, BE, Kircher, JC, & Brown, GS (2008b) Using clinical trials to benchmark effects produced in clinical practice, Quality and Quantity 42:513-525.
Minami, T., Wampold, B. E., Serlin, R. C., Kircher, J. C., & Brown, G. S. (2007) Benchmarks for psychotherapy efficacy in adult major depression, Journal of Consulting and Clinical Psychology, 75, 232-243.
SAMHSA (1998) National Household Survey on Drug Abuse.
Adult CFF - Cases with intake scores in the clinical range, multiple assessments within the episode and care, and the first assessment at session 1 or 2.
N
First GDS Score Last GDS Score Pre-post change # CFF per episode Pre-post weeks
Youth Self Report CFF: Intake scores in the clinical range, multiple assessments within the episode and care, and the first assessment at session 1 or 2.
N
First GDS Score Last GDS Score Pre-post change # CFF per episode Pre-post weeks
Youth Parent Report CFF: Intake scores in the clinical range, multiple assessments within the episode and care, and the first assessment at session 1 or 2.
N
First GDS Score Last GDS Score Pre-post change # CFF per episode Pre-post weeks
Child Self Report CFF: Intake scores in the clinical range, multiple assessments within the episode and care, and the first assessment at session 1 or 2.
N
First GDS Score Last GDS Score Pre-post change # CFF per episode Pre-post weeks
Child-Parent Report CFF: Intake scores in the clinical range, multiple assessments within the episode and care, and the first assessment at session 1 or 2.
N
First GDS Score Last GDS Score Pre-post change # CFF per episode Pre-post weeks Mean (sd) Mean (sd) Mean (sd) Mean (sd) Mean (sd)