Depression in Children and Adolescents by aoonto


 Depressionin Children and Adolescents
With Sickle-Cell Disease
Yih-Ming     Yang, MD; Manuel Cepeda, MD; Clinton Price, MD; Arvind Shah, PhD; Vipul Mankad,               MD

Objective: To look at the role of a standardized screen-           Measurements and Results: The mean Children's De-
ing test for assessing depression in children and adoles-          pression Rating Scale\p=n-\Revisedtotal scores were 27.1 and
cents  with and without sickle-cell disease and to com-            22.1 in study and control groups, respectively (P=.0073).
pare findings with this instrument with clinical evaluations       Eleven (29%) of 38 children in the study group had scores
by child psychiatrists. In addition, to suggest the preva-         indicating a high risk for clinical depression. Four (12%)
lence of clinical depression in children and adolescents           of 34 children in the control group were in this category.
with sickle-cell disease.                                          The four items accounting for most of this significant dif-
                                                                   ference were excessive fatigue, physical complaints, self-
 Design: Nonrandomized, sequential sample of sub-                  esteem, and morbid ideation. This contrasted with the clini-
jects, convenience control sample, prevalence study us-           cal evaluation by a child psychiatrist in a clinical interview
 ing the Children's Depression Rating Scale\p=m-\Revisedand       in which 13% of study subjects and 15% of control sub-
an  interview assessment by a child psychiatrist based on         jects had evidence of clinical depression (P=.85).
criteria from the Diagnostic and Statistical Manual of Men-
tal Disorders, Revised Third Edition.                              Conclusion: Excessive fatigue and physical complaint
                                                                   factors contributed to a high false-positive rate when the
Setting: Comprehensive Sickle Cell Center outpatient clin-         Children's Depression Rating Scale\p=n-\Revisedwas used to
ic and ambulatory Resident Practice Clinic at the Univer-          screen for depression among patients with sickle-cell dis-

sity of South Alabama Children's Medical Center, Mobile.           ease. Based on the clinical interview by a child psychia-

                                                                   trist, the actual prevalence of clinical depression was not
Study Participants:     Patients aged 6 through 18 years           increased in children with sickle-cell disease compared
with homozygous sickle-cell disease (hemoglobin SS) served         with those without this chronic illness.
as study subjects. Subjects of the same age and race who
did not have sickle-cell disease served as controls.               (Arch Pediatr Adolesc Med. 1994;148:457-460)

                                                           disease mainly af¬      pairment.1'2 This is borne out by studies
                                               fects African Americans in          that suggest that children and adoles¬
                                              the United States and is a           cents  with sickle-cell disease are likely to
                                              chronic illness usually be¬          have psychosocial maladjustment and
                                              ginning in early childhood           emotional problems.3 5
                                 and characterized by acute episodic crisis              Childhood depression is a medical
                                 events. Complications include pain cri¬           problem that can have serious conse¬
                                 ses, infections, splenic sequestration cri¬       quences if it is unrecognized and un¬
                                 ses, aplastic crises, acute chest syndrome,       treated.6·7 It has been observed that chil¬
                                 and central   nervous system events. Pa¬          dren and adolescents with sickle-cell
 From the   Departments of
 Pediatrics (Drs Yang and        tients generally exhibit chronic hemo¬
 Mankad) and Psychiatry          lytic anemia and multiple organ damage.
 (Drs Cepeda and Price) and      Their growth is impaired, sexual matura¬
 the Comprehensive Sickle Cell   tion is delayed, psychologic function is af¬
 Center (Drs Yang, Cepeda,
                                 fected, and interaction with peers may be
                                                                                           See   Subjects and Methods
 Price, Shah, and Mankad),                                                                         on   next   page
 University of South Alabama     impaired. As with any chronic disease,
 College of Medicine, Mobile.    there is a high risk of psychosocial im-

                                      Downloaded from on May 20, 2010
     SUBJECTS           AND METHODS                                   tice Clinics were located. Interviewers included a faculty child
                                                                      psychiatrist (M.C.), a senior resident in child psychiatry (C.P.),
                                                                      and    twopediatrie nurse clinicians. The psychiatric evalua¬
     SUBJECTS                                                         tion   was conducted by a child psychiatrist or a senior resi¬
                                                                      dent in child psychiatry or both. The clinical psychiatric evalu¬
     Thirty-eight children with homozygous sickle-cell disease        ation was discussed between the psychiatrists for concurrence.
     (hemoglobin SS) aged 6 through 18 years who were fol¬            For a diagnosis of clinical depression to be made, both child
     lowed up at the University of South Alabama Comprehen¬           psychiatric interviewers had to agree that the subject met the
     sive Sickle Cell Center Outpatient Clinic served as the study    Diagnostic and Statistical Manual of Mental Disorders, Re¬
     group. Thirty-four children (patients and siblings of pa¬        vised Third Edition" criteria for dysthymia (depressive neu¬
     tients from the Resident Practice Clinics at the University      rosis) or major depression. This study was conducted from
     of South Alabama Children's Medical Center and cousins,          late 1990 through early 1992.
     siblings, and neighbors of study subjects) made up the con¬            A detailed medical history was obtained from the medi¬
     trol group. None of the subjects in the control group had        cal records of the Comprehensive Sickle Cell Center. Clini¬
     sickle-cell disease or other chronic disorders. Any subject      cal data, including outpatient visits and hospital days for
     who was actually ill at the time of the interview was ex¬        sickle-cell pain crisis events and other sickle-cell compli¬
     cluded from the study. All subjects were African Ameri¬          cations, were collected during the 3-year period from 1989
     cans. Participation was voluntary and no patient or con¬         through 1991. Clinical severity of illness among patients
     trol subject who was asked to participate declined. This study   with sickle-cell disease was assessed by summing the num¬
     was approved by the Institutional Review Board of the Uni¬       ber of major pain crisis events that required outpatient or
     versity of South Alabama. Written consent was obtained           emergency department visits and the total days of hospi¬
     from at least one parent of all subjects.                         talization for pain crisis or other sickle-cell complications,
                                                                      such as infection, stroke, or acute chest syndrome.
     METHODS                                                                 Statistical analysis was done using the statistic (a mea¬
                                                                      surement of interrater agreement), two-group Student's t test,
     The CDRS-R10 was administered to all subjects in examina¬         a two-sample median test, 2 analysis, one-way analysis of
     tion rooms at the Children's Medical Center where both the       variance (ANOVA), and multiple comparisons. The analy¬
     Sickle Cell Center Outpatient Clinic and Resident Prac-           sis was performed using SAS software (SAS Ine, Carey, NC).

 disease have higher depression scores using psychologi¬               (Table 1 ). Forty-three (60%) of the 72 subjects (23 con¬
 cal screening compared with those children who have no                trol subjects and 20 study subjects) were interviewed by
 sickle-cell disease or chronic illness. It has been sug¬              two raters. For those subjects interviewed by two raters,

 gested that depression is more prevalent in patients with             the interrater agreement for the CDRS-R total score was
 sickle-cell disease.81' However, the true prevalence of clini¬        good ( =0.66).
 cal depression (based on clinical examination of the pa¬                   The mean CDRS-R total scores were 27.1 and 22.1
 tient and not on use of a screening instrument alone) in              in the study and control groups, respectively (P=.0073)
 children and adolescents with sickle-cell disease is not              (Table 1). As shown in Table 2, four of the 17 subscore
 known.                                                                items on the CDRS-R showed a significant difference be¬
       This report presents observations on the preva¬                 tween the study and control groups. These subscore items
 lence of clinical depression, the characteristic clinical mani¬       were    excessive   fatigue (P=.0009), physical complaints
 festations of depression, and the relationship of severity            (P=.0024), self-esteem (P=.0036), and morbid ideation
 of sickle-cell disease to depressive symptoms in children             (P=.0009).
 who were screened with the Children's Depression Rat¬                       A total CDRS-R score of 30 or higher was consid¬
 ing Scale-Revised (CDRS-R) and assessed by a child psy¬               ered to indicate a high risk for depression.1214 Eleven (29%)
 chiatrist at the University of South Alabama Compre¬                  of 38 study subjects and four (12%) of 34 control sub¬
  hensive Sickle Cell Center, Mobile.                                  jects scored in the high-risk category. This difference was
                                                                       significant (P=.0088). Five (13%) of 38 study subjects
                            RESULTS                                    and five (15%) of 34 control subjects had evidence of clini¬
                                                                       cal depression when evaluated by the child psychiatrist
 Thirty-eight subjects aged 6.3 through 18.4 years (me¬                (Table 3). The prevalence of clinical depression was not
 dian age, 11.0 years) and 34 subjects aged 6.5 through                significantly different between the two groups (P=.85).
 18 years (median age, 11.7 years) in the study and con¬                     Within the study group, the false-positive rate when
 trol groups, respectively, were interviewed and included              screened with the CDRS-R was 18%. The false-negative
 in the analysis. There was no significant difference in age           rate was 0% and the correct classification rate was 84%
 between subjects in these two groups (P=.6392)                         (Table 4). In the control group, the child psychiatrist's

                                      Downloaded from on May 20, 2010
 Table 1. Age Distribution and CDRS-R
 Scores in Study and Control Groups*
                                                                                             clinical interview identified all subjects scoring 30 or higher
                                                                                             as actually having clinical depression at the time of the
                                                                                             interview (0% false-positive rate). One subject in the con¬
                                            Study Group      Control Group                   trol group with a CDRS-R score below the cutoff of 30
                                                                                             was clinically depressed (Table 3). The correct classifi¬
 Age,       y              Median              11.0              11.7              .6392
                                                               6.5-18.0                      cation rate was 97%.
                           Range             6.3-18.4
 CDRS-R Score              Mean               27.1                22.1             .0073            The clinical severity of illness did contribute to the
                           Range               17-66             17-39                       CDRS-R score. There seemed to be a positive trend be¬
                           SEM                  1.47              1.3                        tween clinical severity of sickle-cell disease and CDRS-R
                                                                                             total scores.
     CDRS-R indicates Children's Depression Rating Scale-Revised.                                   Two items on the CDRS-R, excessive fatigue and
                                                                                             physical complaints, are common clinical manifesta¬
 Table 2. Mean Scores                  on   the CDRS-R*                                      tions of sickle-cell disease. Comparison analysis of the
                                                                                             CDRS-R between the study and control groups with these
                                                Study          Control                       two items excluded was performed. The significant dif¬
 Item                                          Subjects       Subjects
                                                                                             ference between total CDRS-Rscores remained (P=.0461).
 Schoolwork                                      1.5             1.46             .7997
                                                                                                    Pearson's correlation coefficient between the CDRS-R
 Capacity to have fun                            1.28            1.06             .1498
                                                                                              total score and age was .13 (P=,2751). Twenty-one (55%)
 Social withdrawal                                               1.15             .5108
                                                 1.38            1.28             .5181       of 38 study subjects and 17 (50%) of 34 control subjects
                                                                                  .1049      were 11 years or younger. There were no statistical dif¬
 Appetite       or   eating patterns             1.50            1.24
 Excessive fatigue                               2.47            1.53             .00091      ferences in the total CDRS-R scores between the younger
 Physical complaints                             2.86            1.99             .0024f      and older ages for either study subjects (P=.281) or con¬
 Irritability                                    1.56            1.47             .7059       trol subjects (P=.765). When the control subjects were
 Guilt                                           1.29            1.07             .0948       divided into three groups (siblings of study group pa¬
 Self-esteem                                     1.78            1.22             .0036f      tients, patients and siblings of patients seen in the Pedi¬
 Depressed feelings                              1.53            1.44             .7168       atrie Acute Care Clinic, and cousins or neighbors of study
 Morbid ideation                                 2.21            1.40             .00091      group subjects) and compared simultaneously using one¬
 Suicidal ideation                               1.19            1.34             .4198       way ANOVA, no significant difference between these three
 Weeping                                         1.64            1.28             .1119       subsets of subjects was seen (P=.276).
 Depressed affect                                1.37            1.21             .3100
 Tempo of speech                                 1.19            1.03             .0897
 Hyperactivity                                   1.10            100              .1031
     CDRS-R indicates Children's Depression Rating Scale-Revised.                            Our study found that children and adolescents with sickle-
 tSignificantatP=.05.                                                                        cell disease have higher scores than control subjects when
                                                                                             screened for depression using the CDRS-R. Although nine
     Table 3. Clinical Interview and CDRS-R Results                                          (26%) of 34 control subjects had siblings with sickle-
     for Study and Control Subjects*                                                         cell disease, this was not believed to introduce bias since
                                                                                             no significant difference was seen between their CDRS-R
                                 No. (%) of                        No. (%) of
                                                                                             scores and those of other control subjects.
                               Study Subjects                    Control Subjects
                                   (n=38)                                (n=34)                    Four items on the CDRS-R accounted for the higher
                                -1                                 -1                        CDRS-R scores in children with sickle-cell anemia. They
                           Clinical Depression                 Clinical   Depression
                                                                                             were excessive fatigue, physical complaints, morbid ide¬
     Score                Present             Absent          Present             Absent     ation, and self-esteem problems. There was a significant
     a=30                  5(13)               6(16)           4(12)                0        association between CDRS-R scores and the clinical se¬
     <30                   0                  27(71)           1(3)               29(85)     verity of sickle-cell disease. Both excessive fatigue and
     *   CDRS-R indicates Children's         Depression Rating Scale-Revised.                physical complaints are common medical symptoms of
                                                                                             sickle-cell disease. Self-esteem problems have been as¬
                                                                                             sociated with many chronic illnesses and are not unique
     Table 4. Sensitivity, Specificity, and                                                  to those with sickle-cell disease. The specific self-esteem
     Predictive Value of CDRS-R*                                                             problems most often voiced by patients with sickle-cell
                                                                                             disease involved dissatisfaction with physical appear¬
                                            Study Subjects          Control       Subjects
                                                                                             ance, ie, a body image problem. The clinical interview
     Sensitivity,    %                           100                          80
                     %                            82                         100              helped to explain the higher morbid ideation score. Al¬
     Specificity,                                                                             most all of the study group subjects knew friends or fam¬
         Predictive value, %                      46                         100
                                                                                              ily members who had died (usually from a complication
         CDRS-R indicates Children's          Depression Rating Scale-Revised.                of sickle-cell disease). They had talked about sickle-cell

                                                             Downloaded from on May 20, 2010
 disease and death. Children often mentioned that they
 had thought about their own death as a result of sickle-
 cell disease and also worried about a parent dying. These
                                                                                      It is important to recognize and seek treat¬
                                                                               for this problem in young patients with sickle-cell

 four symptom categories do not alone make a clinical di¬
 agnosis of depression even though they are commonly                Accepted for publication December 20, 1993.
 associated with depression.                                              This work was supported in part by Comprehensive
       The clinical psychiatric interview revealed that the         Sickle Cell Center grant HL-38639-03 from the National
 prevalence of clinical depression in children with sickle-         Institutes of Health, Bethesda, Md.
 cell disease is not increased compared with that of chil¬                Reprint requests to the Department of Pediatrics, Uni¬
 dren without chronic disease. Our results clarified that           versity of South Alabama Medical Center, 2451 F'tllingim
 although childen with sickle-cell disease may have in¬             St, Mobile, AL 36617 (Dr Yang).
 creased somatic complaints, worries about death, and self-
 esteem problems compared with their peers, these symp¬                                              REFERENCES
 toms alone do not meet the threshold criteria for diagnosis
 of depression as a medical disorder. It is conceivable that          1. Gortmaker SL, Walker DK, Weitzman M, Sobol AM. Chronic conditions, so-
 the experience of recurrent acute pain episodes and the                 cioeconomic risks, and behavioral problems in children and adolescents. Pediatrics.
 chronicity of the illness in patients with sickle-cell dis¬               1990;85:267-276.
                                                                      2. Pless IB, Pinkerton P. Chronic Childhood Disorder:     Promoting Patterns of Ad-
 ease lead to psychosocial adjustment problems that mani¬
                                                                           justments. London, England: HO Klimpton Publishers; 1975.
 fest as low self-esteem and a propensity for an increase             3.    Morgan SA, Jackson J. Psychological and social concomitants of sickle-cell
 in morbid ideation. However, the increase in these prob¬                  anemia in adolescents. J Psychol. 1986;11:429-440.
 lems does not lead to an increased prevalence of clinical            4.    Hurtig AL, White LS. Psychosocial adjustment in children and adolescents with
                                                                            sickle-cell disease. J Psychol. 1986;11:422-427.
 depression.                                                          5.    Conyard S, Krishnamurthy M, Dosik H. Psychosocial aspects of sickle-cell ane-
      The CDRS-R was shown to be of moderate benefit                        mia in adolescents. Health Soc Work. 1980;5:20-26.
 as a screening device for identifying patients who need              6.    Seigel WM, Golden NH, Gough JW, Lashley MS, Sacker IM. Depression, self-
 psychiatric assessment for clinical depression. The high                  esteem, and life events in adolescents with chronic diseases. J Adolesc Health
                                                                         Care. 1990;11:501-504.
 incidence of false-positive findings in the study group (a           7. Doglan JI. Depression in children. Pediatr Ann. 1990;19:46-50.
 finding that was not present in the control group) em¬               8. Morin C, Waring EM. Depression and sickle-cell anemia. South Med J.          1981;
 phasizes that care must be taken in using this scale since                74:766-768.
 several clinical characteristics of sickle-cell disease con¬         9.   Leavell SR, Ford CV. Psychopathology in patients with sickle-cell disease. Psy-
                                                                           chosomatics. 1983;24:23-37.
 tributed disproportionately to scores on the CDRS-R                 10.   Poznanski EO, Freeman LN, Mokros HB. Children's Depression Rating Scale\x=req-\
 screening instrument.                                                     Revised. Psychopharmacol Bull. 1985;21:979-1124.
        Our study suggests that children with sickle-cell dis¬       11.   American Psychiatric Association, Committee on Nomenclature and Statistics.
 ease   may experience symptoms that may be misidenti-                     Diagnostic and Statistical Manual of Mental Disorders. Revised Third Edition.
 fied as clinical depression. Several symptoms commonly                    Washington, DC: American Psychiatric Association; 1987.
                                                                     12.   Poznanski EO, Cook SC, Carroll BJ, Corzo H. Use of the Children's Depression
 seen  in depression are also a part of the clinical picture               Rating Scale in an inpatient psychiatric population. J Clin Psychiatry. 1983;
 of sickle-cell disease. The findings of the child psychi¬                 44:200-203.
 atric evaluation suggest that the true prevalence of clini¬         13.   Poznanski EO, Cook SC, Carroll BJ. A depression rating scale for children.
                                                                           Pediatrics. 1979;64:442-450.
 cal depression is not increased in patients with sickle-            14.   Shanahan KM, Zolkowski-Wynne J, Coury DL, Collins EW, O'Shea JS. The children's
 cell disease compared with their peers. However, almost                   depression rating scale for normal and depressed outpatients. Clin Pediatr.
 one child in seven with sickle-cell disease was clinically                1987;26:245-247.

                                    Downloaded from on May 20, 2010

To top