A Procedure to Graph the Quality of Psychosocial Functioning Affected by Symptom Severity
Abstract
Assessment of the variations of clinical course to aid in diagnosis, assessment of patients’ functioning and to guide treatment planning has gained momentum in recent years. A specific scale is introduced to plot the temporal course to assist empirically-minded psychotherapists and researchers who treat the DSM-5 Disorders and who want to monitor the quality of the course of psychosocial functioning over time. A Timeline Course Graphing Scale to Chart the Quality of Psychosocial Functioning Affected by Symptom Severity (PFS) is described and accompanied by administration guidelines.
Introducing a Clinical Course Graphing Scale for Psychosocial Functioning
A clinical course graphing scale is described to provide a means to chart the course of changes in psychosocial functioning. The scale, Timeline Course Graphing Scale to Chart the Quality of Psychosocial Functioning Affected by Symptom Severity, illustrates the effects symptom severity exerts on psychosocial functioning. Before describing the scale and its administration procedures, several recent research efforts involving the assessment of clinical course are described.
Emerging Clinical Course Specifier Emphases in The Diagnostic and Statistical Manual of Mental Disorders (DSM) Since 1980
Taking note of clinical course or the manifestations of a patient’s psychological disorder profile over time has gained increasing importance since 1980. Evolving course descriptions in the DSMs published since 1980 reveal a growing emphasis on the long-term clinical course characteristics of disorders. For example, in DSM-III (APA, 1980), a brief clinical course description under the heading course was inserted within the text for each diagnostic category. The explosion of clinical research following the publication of DSM-III resulted in the elaboration of course information in the DSM-III-R (APA, 1987) under several headings: age-specific features; age of onset; and course. Beginning with DSM-IV (APA, 1994), and continuing with DSM-IV-TR (APA, 2000), attention to characterizing course features was evident in the specifier course graphs appearing in the mood disorder section (DSM-IV, p. 388 and the DSM-IV-TR p. 425).
In chronic depression research during the late 1980s and 1990s, differentiating acute/episodic major depression from chronic depression became an important diagnostic issue. To facilitate the differential diagnostic task and to enhance diagnostic accuracy, McCullough and colleagues (McCullough, 2001; McCullough, Kornstein, McCullough, Belyea-Caldwell, Kaye, Roberts, et al. 1996) proposed a visual course-graphing procedure. McCullough et al.’s (e.g., 1996) graphing methodology was used reliably in two clinical trials (Keller, McCullough, Klein, Arnow, Dunner, Gelenberg et al. 2000; Kocsis, Gelenberg, Rothbaum, Klein, Trivedi, Manber et al. 2009). The graphing procedure was inserted in a revised version of the Structured Clinical Interview for DSM-IV Axis I Disorders: Patient Edition (SCID-P) (First, Spitzer, Gibbon & Williams, 1995) and became the forerunner of the Graphing Scale for Psychosocial Functioning to be described below.
Recent prospective studies using clinical course methodology illustrated the trajectory of several unipolar and bipolar disorders (e.g., Birmaher, Axelson, Strober, Gill, Valeri, Chiappetta et al. 2006; Klein, Schwartz, Rose & Leader, 2000; Klein, Shankman & Rose, 2006; Judd, Akiskal, Schettler, Endicott, Maser, Solomon, et al. 2002; Judd, Schettler, Akiskal, Maser, Coryell, Solomon et al. 2003; Post, Denicoff, Leverich, Altshuler, Frye, Suppes et al. 2003). Except for the Post et al. (2003) investigation, which recorded the degree of psychosocial functional impairment among bipolar I patients over time on social, educational and occupational roles, no study specifically investigated the degree to which the extent of symptom severity affected the quality of psychosocial functioning. In this paper, a specific scale for graphing the level of psychosocial functioning affected by symptom severity is described. The scale is usable for both clinical psychotherapists and researchers wanting to plot the progress of psychosocial functioning of their patients to illustrate empirically the changes occurring as a function of treatment.
Development of the Psychosocial Functioning Scale (PFS). The PFS originated in 1994 (unpublished paper, McCullough & Roberts, 1994). The authors operationalized the DSM-III-R (APA, 1987) Global Assessment of Functioning Scale (GAF) to enhance rater agreement. The McCullough and Roberts’ scale used the DSM-III-R original 90-point GAF scoring range divisions, and clinical practitioners rated the degree of symptom disruption/interference from all mental disorders during the previous week affecting psychosocial functioning in familial, occupational and social domains. In 2012, the PFS Scale (McCullough, 2012) was further developed in teaching clinical psychology students in a Ph.D. program to rate the degree to which symptom severity adversely affected psychosocial functioning. The original McCullough & Roberts’-90-point scale was reduced to a 10-point scale.
World Health Organization Disability Assessment Scale (WHODAS) versus the Psychosocial Functioning Scale (PFS). Before continuing, it is necessary to make a brief comparison with the DSM-5 disability assessment scale (World Health Organization Disability Assessment Scale [WHODAS], which replaced the GAF Scale in DSM-IV/IV-TR.
The WHODAS in the DSM-5 presents in a self-report form and rates psychosocial functioning disability levels for the past 30 days by averaging disability-dimensional ratings across seven functioning domains. Clinicians calculate an average general disability score (GDS) by dividing the raw overall score by the number of all of the items on the WHODAS (n = 36).
Even though both scales seek to evaluate psychosocial/disability functioning, significant differences exist between the WHODAS and the PFS. The differences are as follows:
| (1) | The WHODAS is a self-report scale and the PFS is a clinician-rated instrument. | ||||
| (2) | The WHODAS is vulnerable to self-report biases (e.g., poor insight into psychosocial functioning, etc.). The PFS also relies on patients’ reports, but it is elicited and interpreted by a trained clinician. | ||||
| (3) | Compared to the WHODAS, the PFS is a simpler procedure and requires less time to administer. | ||||
| (4) | The WHODAS requires patients to rate multiple categories of functioning by using dimensional-rating scales, which produce cumulative disability scores. The PFS uses one continuous psychosocial/disability rating score that is operationalized by five symptom severity-level definitions. | ||||
| (5) | The WHODAS samples domain levels of functioning for the previous 30 days with single time-point dimensional rating scores. The PFS may be used to chart repeatedly the clinical course of psychosocial functioning levels from the onset of a disorder, throughout the treatment process and outcome, and during follow-up periods. | ||||
| (6) | The WHODAS assesses and interprets disability levels using dimensional scaling across several domains. The PFS charts psychosocial levels using one general psychosocial score multiple times that graphically illustrate the fluctuating influences of symptom severity. | ||||
| (7) | The WHODAS assesses physical disability and the PFS does not. | ||||
| (8) | The WHODAS describes disability functioning across multiple activity levels while the PFS does not. | ||||
| (9) | Perhaps the biggest difference is that the WHODAS assesses disability without trying to distinguish impairment due to symptoms or to other things while the PFS specifically seeks to assess impairment due to symptoms. It should also be noted that an important similarity of WHODAS and the PFS is that both instruments may be used to evaluate psychosocial/disability functioning with all the DSM-5 disorders. | ||||
Rater Reliability Testing with the PFS. The PFS has been reliability tested. Twelve clinical raters were trained to rate patients’ psychosocial functioning levels over time on the 0 to 10 scale operationalized to reflect no interference (0), mild interference (1-2), low-moderate interference (3-4), high-moderate interference (5-7) and severe interference (8-10). Each rater reviewed five case descriptions involving generalized anxiety disorder, persistent depressive disorder, panic disorder, and two social anxiety disorder cases. The raters then plotted the level of psychosocial functioning for each patient for the period that the patient functioned at that particular level. The rating task involved rating two to three intervals when functioning change occurred (time x interference level) per case. The ratings were compared to a gold-standard interval rating for that case established by the first author (JPM, and unknown to the raters). There was a total of 13 possible correct interval ratings for the five case scenarios. Each rater produced 13 interval ratings involving the five cases with no inter-rater disagreement concerning the number of intervals to be rated. Twelve kappa coefficients of agreement for the 13 intervals were then calculated for each rater, comparing each rater’s interval ratings to the gold-standard interval ratings (determined by the author of the rating scale).
The percentage of agreement among the 12 clinical raters with the criterion ratings ranged from 69% to 100%. Four raters obtained 100% agreement, four achieved 92% agreement with the criterion ratings, one rater obtained 85% agreement, two reported 77% agreement and one, 69% agreement. The overall mean for the percentage of rater agreement was 90%. Kappa coefficients ranged from .57-to-1.0 with nine rater coefficients falling in the .80 to 1.0 range, indicating good to very good agreement. Two raters obtained coefficients in the .60 to .79 range (kappa = .69 & .71, respectively), indicating good agreement. One rater obtained moderate agreement with a kappa of .57 (Cohen, 1960). When all twelve of the kappa reliability coefficients were averaged, the overall mean was .86.
PFS Scale: Procedural Steps
The PFS uses a timeline graphing procedure. Symptom interference over time is anchored on the y axis (vertical) while time is anchored on the x axis (horizontal). The longitudinal course of psychosocial functioning is illustrated moving from left (the present) to right (the past). The clinician works with a copy of the Scale PFS Rating Form (see Figure 1) and the Scoring Key (see Figure 2). Using the PFS Rating Form and relying on the patient’s self-report, the clinician then graphs the longitudinal course. As noted, the Scoring Key shown in Figure 2 is a 1 to 10 rating scale anchored by a Severe Level rating (10) and descending in scaled value to a No Interference Level (0). We have found that anchoring the Rating Form with marker dates and “recalled life events” (e.g., birthdays, anniversaries, holidays, etc.) facilitated memory recall.
Figure 1 Example of the rating form to be used in the session to rate the degree that one’s symptoms are affecting psychosocial functioning (rated with a number from 1-10) and working left (diagnostic interview)-to-right (past). An example patient’s symptoms are rated to be interfering with psychosocial functioning over a time-period extending from the present, July 10, 2013, back to April 2012 when the patient reported that she noticed that she was not doing or feeling well.
Figure 2 Operationalized PFS scoring key used to rate the course of symptom severity affecting levels of psychosocial functioning working from left (interview session)-to-right (past).
Step One: The first step entails asking the patient two questions: (1) “I want you to think about how much all your symptoms interfere right now with your daily functioning in your work, in your family, and in your social life? (2) “How long have your symptoms remained at this level?” The clinician, relying on the patient’s self-report, indicates the current level of interference with an “x” on the vertical axis. Working back in time, a “x” denotes the rating when the first change/shift point is pinpointed.
Step Two: The clinician asks two further questions after the current psychosocial interference level and its duration are pinpointed: (1) “What happened to the interference level at this change/shift point? Did it increase or decrease and how would you rate the level here?” (2) “How long did the symptom interference remain at this level?” Clinicians determine the longitudinal symptom interference level variations for the duration of the patient’s disorder(s) by continuing to repeat the step two questioning procedure.
Step Three: Empirical assessment of the improvement in psychosocial functioning over the process of treatment. The longitudinal course profile of psychosocial functioning may be plotted as many times as desired during treatment. Similarly, follow-up course graphing may also be completed using the same method—that is, working from the present time back towards to last data collection point.
Examples of the longitudinal course profiles of psychosocial functioning for several DSM-5 disorders are shown below. Refer to Figure 1 and Figure 2.
Conclusion
In recent years, clinical course methodology has received increasing emphasis in the diagnosis of mood disorder patients as well as in the assessment of other course manifestations such as the adverse effects symptoms exert on psychosocial functioning. A clinical course graphing scale, Course Graphing Scale to Chart the Quality of Psychosocial Functioning Affected by Symptom Severity, has been presented and described so that clinicians may continue to assess the effects of treatment as it reflects patients’ psychosocial functioning level. Steps to administer the PFS has accompanied the description of the instrument. Hopefully, the scale will assist empirically-minded psychotherapists and psychotherapy researchers to assess change in their patient’s psychosocial functioning over the course of treatment and beyond.
(2006). Clinical course of children and adolescents with bipolar spectrum disorders. Archives of General Psychiatry, 63, 175–183.Crossref, Medline, Google Scholar
(1060). A coefficient of agreement for nominal scales. Education and Psychological Measurement, 20, 37–46.Crossref, Google Scholar
(1995). Structured clinical interview for DSM-IV axis I disorders: Patient Edition (SCID-P). New York: Biometrics Research Department. New York State Psychiatric Institute.Google Scholar
(2002). The long-term natural history of the weekly symptomatic status of bipolar I disorder. Archives of General Psychiatry, 59, 530–537.Crossref, Medline, Google Scholar
(2003). Long-term symptomatic status of bipolar I vs. bipolar II disorders. International Journal of Neuropsychopharmacology, 6, 127–137.Crossref, Medline, Google Scholar
(2000). A comparison of nefazodone, the Cognitive Behavioral Analysis System of Psychotherapy, and their combination for the treatment of chronic depression. New England Journal of Medicine, 342, 1462–1470.Crossref, Medline, Google Scholar
(2000). Five-year course and outcome of dysthymic disorder: A prospective, naturalistic follow-up study. American Journal of Psychiatry, 157, 931–939.Crossref, Medline, Google Scholar
(2006). Ten-year prospective follow-up study of the naturalistic course of dysthymic disorder and double depression. American Journal of Psychiatry, 163, 872–880.Crossref, Medline, Google Scholar
(2009). Cognitive Behavioral Analysis System of Psychotherapy and Brief Supportive Psychotherapy for augmentation of antidepressant nonresponse in chronic depression: The REVAMP Trial. Archives of General Psychiatry, 66, 1178–1188.Crossref, Medline, Google Scholar
(2001). Skills training manual for diagnosing and treating chronic depression: CBASP. New York: Guilford Press.Google Scholar
(1996). Differential diagnosis of chronic depressive disorders. Psychiatric Clinics of North America, 19, 55–71.Crossref, Medline, Google Scholar
(1994). Revised GAF Scale to assess patients’ symptom influence on their psychosocial functioning. Richmond, VA: Virginia Commonwealth University. Unpublished manuscript.Google Scholar
(July, 2012). DSM-IV-TR Diagnostic Methodology and Diagnostic Decision-Guide Procedure: Fall Semester, Psychology 616. Richmond, VA: Virginia Commonwealth University. Unpublished manuscript.Google Scholar
(2003). Morbidity in 258 bipolar outpatients followed for 1 year with daily prospective ratings on the NIMH Life Chart Method. Journal of Clinical Psychiatry, 64, 680–690.Crossref, Medline, Google Scholar
