Psychosomatics
Journal Home Search Current Issue Past Issues Subscribe All APPI Journals Help Contact Us
 
Psychosomatics 47:376-384, October 2006
doi: 10.1176/appi.psy.47.5.376
© 2006 Academy of Psychosomatic Medicine
Quicksearch
Advanced Search
Or Search All APPI Journals
This Article
* Abstract Freely available
* Full Text (PDF)
* Alert me when this article is cited
* Alert me if a correction is posted
* Citation Map
Services
* Email this article to a Colleague
* Similar articles in this journal
* Similar articles in PubMed
* Alert me to new issues of the journal
* Add to My Articles & Searches
* Download to citation manager
* reprints & permissions
Citing Articles
* Citing Articles via HighWire
* Citing Articles via Google Scholar
Google Scholar
* Articles by Guo, Y.
* Articles by Nemeroff, C. B.
* Search for Related Content
PubMed
* PubMed Citation
* Articles by Guo, Y.
* Articles by Nemeroff, C. B.
Related Collections
* Depression
* Syndromes Secondary to General Medical Disorders

The Diagnosis of Major Depression in Patients With Cancer: A Comparative Approach

Ying Guo, Ph.D., Dominique L. Musselman, M.D., Amita K. Manatunga, Ph.D., Natalie Gilles, M.P.H., Kathryn C. Lawson, M.S., Maryfrances R. Porter, M.S., J. Stephen McDaniel, M.D., and Charles B. Nemeroff, M.D., Ph.D.

Received February 2, 2005; revised September 2, 2005; accepted September 29, 2005. From the Dept. of Psychiatry and Behavioral Sciences, Emory Univ. School of Medicine; Dept. of Biostatistics, Rollins School of Public Health, Emory Univ., Atlanta; and the Dept. of Psychology, Univ. of Virginia School of Medicine, Charlottesville, VA. Address correspondence and reprint requests to Dr. Nemeroff, Dept. of Psychiatry and Behavioral Sciences, Emory Univ. School of Medicine, Woodruff Research Memorial Building, 101 Woodruff Circle, Suite 4000, Atlanta, GA 30322. e-mail: cnemero{at}emory.edu


  ABSTRACT

 
TOP
 ABSTRACT
INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 REFERENCES
 
Depressive symptoms not only impair quality of life in cancer patients but constitute an independent risk factor for increased mortality. In order to accurately and efficiently identify depression in cancer patients, the authors developed a biostatistical strategy to identify items of the 21-item, observer-rated Hamilton Rating Scale for Depression (Ham-D) that would optimize the diagnosis of depression among cancer patients. Exhibiting a relatively high sensitivity and specificity, our most optimal diagnostic tool contained six Ham-D items (late insomnia, agitation, psychic anxiety, diurnal mood variation, depressed mood, and genital symptoms). This study may serve as a prototype to generate valid instruments accurate for the diagnosis of major depression in other populations of cancer patients.


  INTRODUCTION

 
TOP
 ABSTRACT
 INTRODUCTION
METHOD
 RESULTS
 DISCUSSION
 REFERENCES
 
The diagnostic process of an ill patient with comorbid major depression and a medical illness, especially cancer, has historically been an area fraught with controversy.1 Various diagnostic schema have been proposed in an effort to accurately diagnose major depression in the medically ill patient. In brief, researchers have advocated either the "inclusive approach," that is, counting every depressive symptom,2 the "exclusive approach," that is, disallowing depressive symptoms common to medical disorders,3 or the "substitutive approach," in which depressive symptoms such as social avoidance are substituted for neurovegetative symptoms of depression.4 Moreover, recent reports have advocated the use of abbreviated, structured interviews,5 or patient self-report scales for the screening6 or diagnosis of mood disorders.7 Calculation of the positive predictive value (PPV) of a depression rating scale has provided a measurement of the accuracy of the scale as a diagnostic tool in patients with cancer.8–11 Receiver operating characteristic (ROC) analyses may assist in setting a "cut-off" score of an observer- or self-rating instrument, above which optimal sensitivity, specificity, and PPV are obtained.12,13 Last, factor-analytic techniques may help to identify the symptom clustering of mood, cognitive, or somatic symptoms suffered by cancer patients.14 Indeed, although numerous other studies have utilized short screening tests in medically ill patients, nearly all of these studies have utilized heterogeneous populations of medically ill persons, without specifically focusing upon the depressive symptom psychopathology of patients with cancer.13,15–17 The challenges of the diagnosis of major depression in the cancer patient are heightened by the depressive symptoms that may be induced by neoplastic progression and/or its treatment, for example, anorexia, fatigue, diminished concentration, sleep disturbance secondary to pain, and so forth, whereas other depressive symptoms may be more specific to major depression per se.1,18,19 However, the critical importance of the diagnosis of major depression has been recently underscored by increasing evidence that depressive symptoms not only impair quality of life in cancer patients but are an independent predictor of mortality,20–24 although discordant reports exist.25–29

We sought to develop a diagnostic tool with high sensitivity and good specificity in order to diagnose major depression in patients with cancer. We hypothesized that use of four groups of individuals, that is, patients with major depression without cancer, normal-comparison individuals, cancer patients with major depression, and cancer patients without major depression, would allow discrimination of which items of the Hamilton Rating Scale for Depression (Ham-D)30 might be predictive of the presence of major depression in patients with cancer. For the sake of sensitivity, we would want to choose Ham-D items that are endorsed by patients with cancer and major depression. To improve specificity, we would exclude Ham-D items that are unduly endorsed by the nondepressed group of cancer patients. In Figure 1, Ham-D items are accordingly categorized into four types. The "Positive" items are most desirable in terms of both sensitivity and specificity and will be included in the diagnostic tool. On the other hand, the "Negative" items perform poorly in either respect and need not be considered. Whether the "Common" or "Neutral" items should be included in the tool are investigated using logistic-regression models.


Figure 1
View larger version (24K):
[in this window]
[in a new window]
 

FIGURE 1.  Relationship of Ham-D Items Endorsed by Cancer Patients With and Without Major Depression



  METHOD

 
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
Study subjects with cancer were recruited from outpatients and inpatients at Emory University-affiliated hospitals via referral from medical providers and/or advertisements for "cancer patients with and without depression." Outpatients with major depression and normal-comparison subjects were recruited from Emory and the surrounding community by advertisements or word-of-mouth. After a complete description of the study had been given, 111 potential subjects provided written, informed consent and were then screened in person for the presence of psychiatric disorders, substance abuse, medical illness, and medication intake. Subjects were excluded if they had evidence of bipolar disorder (mania, hypomania, or cyclothymia), schizoaffective disorder, schizophrenia, delirium, or dementia, current alcohol/substance abuse or dependence, or pregnancy. Subjects were between the ages of 18 and 80, had a Mini-Mental State Exam score of at least 24,31 spoke and read English well enough to complete study questionnaires, and had no untreated endocrine, cardiovascular, hematologic, hepatic, renal, or neurological disease. Subjects were not currently ingesting any medications implicated in causing depression and were not receiving any psychotropic medications for the treatment of symptoms of depression or anxiety. The normal-comparison subjects were without any current or past personal history of a psychiatric disorder except for past alcohol/substance abuse or dependence. The groups of depressed subjects without cancer and the normal-comparison subjects were without significant medical illness on the basis of physical exam, blood and urine analyses, ECG, and chest X-ray.

Sixty-two cancer patients were screened, of which 29 subjects were excluded because of withdrawal from the study (N=16), current alcohol dependence (N=4), inability to complete the Ham-D items (N=3), possible brain metastasis (N=1), or being without neoplasm discovered after oncologic evaluation (N=5).

Of 24 subjects screened for the depressed-comparison group, 11 were excluded because of psychiatric disorders (bipolar disorder [N=2], current alcohol abuse or dependence [N=2], or numerous Axis I diagnoses [N=1]); others were excluded because of undiagnosed medical disorders (N=3), ingestion of a medication that could cause depression (N=1), marked cognitive dysfunction (N=1), and lack of depression (N=1).

Twenty-five potential participants were screened for the normal-comparison group, of which 13 were excluded for the following reasons: a history of depression (N=3), withdrawal from the study (N=2), current alcohol abuse (N=2), pregnancy (N=1), failure to complete all questionnaires (N=3), ongoing treatment with an antidepressant (N=1), and insufficient language skills to complete the study (N=1).

Of those screened, 33 cancer patients, 13 patients with major depression without cancer, and 12 normal-comparison subjects completed the study (Table 1). Thirteen (81%) of the 16 depressed cancer patients in the final study sample were recruited from Emory-affiliated hospitals; two (13%) were recruited through advertisement or word-of-mouth, and one (6%) was missing data on site of recruitment. Similarly, of the nondepressed cancer patients who completed the study (N=17), 16 (94%) were recruited from the hospital, and one (6%) was recruited from advertisement/word-of-mouth. In the depressed comparison group (N=13), seven (54%) were recruited through word-of-mouth; five (38%) were recruited from Emory University Hospital; and one (8%) was missing data on recruitment. Most of the normal-comparison subjects (N=7; 58%) were recruited via word-of-mouth; three (25%) were recruited from the Emory-affiliated hospitals where they worked; and two (17%) were missing data on their recruitment site.


View this table:
[in this window]
[in a new window]
 

TABLE 1. Demographic and Clinical Characteristics of Normal-Comparison Subjects, Comparison Subjects With Major Depression, and Cancer Patients With and Without Major Depression, N (%)



Procedures and Methods
The Structured Clinical Interview for DSM-III-R (SCID)32 and the dimensional 21-item Ham-D30 were administered to all study participants by a single rater who was either a master’s-level nurse or a fourth-year psychiatry resident. Final psychiatric diagnoses were provided by consensus of the research team, comprising the aforementioned individuals and two board-certified psychiatrists.

Of the 16 subjects in the depressed cancer group, the majority (N=12; 75%) fulfilled diagnostic criteria for major depressive disorder-only; 2 subjects (13%) had both major depressive disorder and dysthymia ("double depression"); 1 subject (6%) fulfilled diagnostic criteria for adjustment disorder with depressed mood; and another (6%) had adjustment disorder with mixed anxiety and depression. The depressed control group comprised 13 subjects, of whom 11 (85%) exhibited an ongoing episode of major depressive disorder-alone; and 2 subjects (15%) fulfilled diagnostic criteria for both major depressive disorder and dysthymia.

Statistical Analyses
We compared demographic and clinical characteristics of the four study groups. One-way analysis of variance (ANOVA) was used to compare the continuous variables. For categorical variables, we used Pearson’s chi-square test; Fisher’s exact test was used when the cell frequencies were small. Comparison of the 21-item Ham-D scores of the four study groups was performed with one-way ANOVA.

The Cochran-Armitage trend test was used to determine whether an increased score of each Ham-D item was associated with an increased probability of depression of individuals without, and with, cancer (Table 2). Compared with the commonly-adopted Person chi-square test, the Cochran-Armitage trend test is more powerful in detecting the trend of probability of depression with the increase in Ham-D item score. The sign and value of the Z statistic of the test reflects the direction and magnitude of the trend.


View this table:
[in this window]
[in a new window]
 

TABLE 2. Increases in Ham-D Items and the Probability of Depression in Patients With and Without Cancer



To further determine those Ham-D items that might be "falsely" elevated because of the cancer, we compared Ham-D items between cancer patients without major depression against normal-control subjects by use of the Mantel-Haenszel chi-square test (Table 3).


View this table:
[in this window]
[in a new window]
 

TABLE 3. Increases in Ham-D Items in Cancer Patients Without Major Depression Versus Normal-Comparison Subjects



The Ham-D items of Table 2 and Table 3 were then classified as "Positive," "Common," "Negative," or "Neutral" Ham-D items, as depicted in Figure 1. Utilizing logistic regression, five candidate models indicating the probability of depression of patients with cancer were then constructed using the total scores of the Ham-D items of the aforementioned categories. That is, the predictor in Model I consisted of the total score of "Positive" Ham-D items. The predictor in Model II was the total score of the "Positive" and "Common" Ham-D items. Model III used the total score of the "Positive" and "Neutral" Ham-D items. Model IV used the total score of the "Positive," "Common," and "Neutral’ items. Finally, Model V utilized the total score of all 21 items of the Ham-D. The four candidate models (Models I–IV) and Model V were then compared by three criteria: –2 log-likelihood, rescaled generalized coefficient of determination, or what is commonly known as the rescaled R2, and the area under the receiver operating characteristics (ROC) curve. We used the –2 log-likelihood instead of Akaike Information Criterion (AIC) or Schwarz Criterion (SC) for model comparisons because all five models have one predictor, which is the total score of the corresponding Ham-D items. The –2 log-likelihood and rescaled R2 are both based on the likelihood of the fitted logistic-regression model and measure the goodness of fit of the model. A more desirable model is indicated by lower –2 log-likelihood and larger rescaled R2 values. The ROC curve is a plot of the sensitivity of a diagnostic test versus its false-positive rate for all possible "cut-off" scores of the diagnostic test considered by the model. As illustrated by Figure 3, the closer the ROC curve is to the upper left corner, the higher the accuracy of the diagnostic test. The area under the ROC curve is the most popular measure of the accuracy of a diagnostic test and ranges between 0 and 1, with 0 indicating total inaccuracy and 1 representing perfect accuracy.


Figure 3
View larger version (28K):
[in this window]
[in a new window]
 

FIGURE 3.  ROC Curves for Model II and Model V



  RESULTS

 
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
DISCUSSION
 REFERENCES
 
The demographic and clinical characteristics of the study participants are presented in Table 1. The study groups did not differ significantly in age (p=0.34), race (p=0.10), or gender, although there was a trend for a higher proportion of women among the depressed cancer patients than in the other three study groups (p=0.07). The stage of cancer was similar between the cancer patients with, and without, major depression (p=0.89), as was the type of cancer (p=0.07). However, the mean 21-item Ham-D scores of the four study groups were significantly different (p<0.0001).

As detailed in Table 2, the probability of depression in the subjects without cancer increased significantly with greater scores in nearly all (18 out of the 21) Ham-D items. The Ham-D items on weight loss, insight, and obsessive and compulsive symptoms were not different in the patients with major depression without cancer than in the normal-control subjects. In the cancer patients, 6 of the 21 Ham-D items were significantly associated with an increased probability of major depression: depressed mood (Z-score=2.89; p=0.004), late insomnia (Z-score=2.30; p=0.022), agitation (Z-score=2.64; p=0.008), psychic anxiety (Z-score=3.36; p=0.001), genital symptoms (Z-score=2.40; p=0.017), and diurnal variation (Z-score=2.00; p=0.046).

As shown in Table 3, there were nine Ham-D items significantly endorsed more often by nondepressed cancer patients than normal-comparison subjects: depressed mood (p=0.018), middle insomnia (p=0.048), diminished work and activities (p=0.0002), retardation (p=0.007), gastrointestinal symptoms (p=0.007), general somatic symptoms (p=0.013), genital symptoms (p=0.007), weight loss (p=0.007), and obsessive and compulsive symptoms (p=0.050).

Utilizing the results of Table 2 and Table 3, the Ham-D items were then inserted into Figure 2. The white circle contains the Ham-D items endorsed by depressed cancer patients from Table 2; the darkest circle contains the Ham-D items endorsed by nondepressed cancer patients from Table 3. Ham-D items significantly endorsed by cancer patients with major depression, and not by the nondepressed cancer patients, that is, the "Positive" Ham-D items, were late insomnia (Item 6), agitation (Item 9), psychic anxiety (Item 10), and diurnal variation in mood (Item 18). Ham-D items significantly endorsed by both cancer patients with and without major depression, that is, the "Common" Ham-D items, were depressed mood (Item 1) and genital symptoms (Item 14). Ham-D items significantly endorsed by nondepressed cancer patients, and not by the depressed cancer patients, that is, the "Negative" Ham-D items, were middle insomnia (Item 5), work and activities (Item 7), retardation (Item 8), gastrointestinal symptoms (Item 12), general somatic symptoms (Item 13), weight loss (Item 16), and obsessive and compulsive symptoms (Item 21). The Ham-D items not endorsed by either the depressed or nondepressed cancer patients, that is, the "Neutral’ items, were feelings of guilt (Item 2), suicide (Item 3), early insomnia (Item 4), somatic anxiety (Item 11), hypochondriasis (Item 15), insight (Item 17), depersonalization and derealization (Item 19), and paranoid symptoms (Item 20).


Figure 2
View larger version (47K):
[in this window]
[in a new window]
 

FIGURE 2.  Ham-D Items Significantly Endorsed by Cancer Patients With and Without Major Depression


In Table 4, we compared the five models by use of the –2 log-likelihood, rescaled R2, and area under the ROC curve. Models I through IV were all superior to Model V (comprising all 21 items of the Ham-D). Among the four candidate models (Models I–IV), Model II was optimal, according to –2 log-likelihood, rescaled generalized coefficient of determination (or the rescaled R2), and the area under ROC curve. Figure 3 shows the ROC curves of Model II compared with Model V.


View this table:
[in this window]
[in a new window]
 

TABLE 4. Comparison of the Five Logistic-Regression Models



On the basis of observations from Table 4 and Figure 3, the four "Positive" items and two "Common" items of the Ham-D would comprise the diagnostic tool for identifying the presence of depression among cancer patients. Our next step for the diagnostic tool was to choose an optimal cut-off score, which would provide at least 80% sensitivity as well as specificity. As shown in Table 5, the most optimal cut-off score was 6, with a sensitivity of 81.3% and a specificity of 87.5%.


View this table:
[in this window]
[in a new window]
 

TABLE 5. Sensitivity and Specificity of Cut-Off Scores, From "Positive" and "Common" Ham-D Items




  DISCUSSION

 
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
REFERENCES
 
Our objective was to develop a diagnostic tool with both high sensitivity and specificity for the optimal diagnosis of depression among cancer patients. Although not originally constructed for use in medically ill patients, the Ham-D continues to have broad use in clinical trials for the treatment of depression, most recently in persons with heart disease.33,34 Other depression rating scales, for example, the Beck Depression Inventory, have also been successfully utilized in the diagnosis of major depression in patients with diabetes35 and recent myocardial infarction.34

Our final model, Model II, contained "Positive" and "Common" Ham-D items endorsed by cancer patients with major depression. Inclusion of the "Neutral’" Ham-D items (as in Models III and IV) diminished the instrument’s accuracy in predicting the presence of major depression in patients with cancer. Of note is the finding that the symptom of anhedonia (included within Ham-D Item 7 "work and activities"), historically cited as a predictor of depression in the heterogeneous populations of medically ill persons,36,37 is not included within our proposed instrument. Other research groups have also reported that this item is endorsed both by depressed and nondepressed patients with serious medical disorders,17,36 including cancer patients receiving chemotherapy15 and palliative treatment.37 Although other investigators have utilized psychiatric inpatients with depression36 to help distinguish depressive symptom patterns between medically ill patients (without and with depression), neither anhedonia, nor "low positive affect," distinguished the depressed medically ill persons from their nondepressed medically ill counterparts. In our study, however, depressed mood remained a useful item in our final model.

Our study’s major limitations include a small sample size of cancer patients, with a heterogeneity of cancer diagnoses. Indeed, a more sensitive, and specific, diagnostic tool would likely have arisen from a larger, more homogeneous study population recruiting normal-comparison subjects and patients with major depression similar in age and gender to cancer patients diagnosed with a single type of neoplasm, receiving a uniform anti-cancer regimen. However, given the lack of a "gold standard" for the presence of major depression, even in depressed patients without cancer,38 this study may serve as a prototype in which similar biostatistical strategies may generate diagnostic instruments accurate in the diagnosis of major depression in cancer as well as other major medical disorders. Although our analyses were performed on a relatively small sample of patients with cancer, our research design included individuals without cancer, that is, patients with major depression and normal-comparison individuals, to identify which Ham-D items would be useful for the detection of major depression in cancer patients. Our final model exhibited greater sensitivity and specificity than the 75% sensitivity and 75% specificity of the 14-item, self-rated Hospital Anxiety and Depression Scale,12 a scale specifically designed for medically ill patients. Of note is that our most optimal model contained only six Ham-D items, with a relatively low cut-off score of 6, congruent with other reports that accurate identification of major depression in medically ill patients can be achieved with relatively few questions.13,16,17,39–41 This observer-administered, six-item instrument may support patient "paper-and-pencil" self-report questionnaires, especially if used as a confirmatory follow-up screening instrument to identify those patients who might benefit from antidepressant treatment alone, or in combination with other types of psychosocial interventions. Another potential use of the instrument would be to identify subjects for research protocols where brief tools are used initially.15,42 Validation of this study’s modified Ham-D questionnaire will certainly require additional study of other groups of patients with cancer, especially those patients who may be currently treated with psychotropic medications. Finally, whether cancer patients identified by this (or other) modified dimensional instruments consistently exhibit neurobiologic alterations observed in previous studies of cancer patients with comorbid depression is unknown and is worthy of further investigation.43,44


  ACKNOWLEDGMENTS

 
The authors thank the study participants; the physicians, physicians’ assistants, and nursing staff of the Divisions of Medical Oncology, Radiation Oncology, and Surgical Oncology of Emory University School of Medicine; the Winship Cancer Institute; and the Division of Radiation Oncology at Grady Memorial Hospital for assistance with this study.

This study was supported by NIMH grants MH-49523, MH-01399, and MH-00680, grant DK-07298 from the National Institute of Diabetes and Digestive and Kidney Diseases, and NIH grant MO1-RR00039.


  REFERENCES

 
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Bukberg J, Penman D, Holland JC: Depression in hospitalized cancer patients. Psychosom Med 1984; 46:199–212[Abstract/Free Full Text]
  2. McDaniel JS, Musselman DL, Porter MR, et al: Depression in the cancer patient: a commentary on diagnosis, biology, and treatment. Arch Gen Psychiatry 1995; 52:89–99[Abstract]
  3. Plumb MM, Holland J: Comparative studies of psychological function in patients with advanced cancer, I: self-reported depressive symptoms. Psychosom Med 1977; 39:264–276[Abstract/Free Full Text]
  4. Endicott J: Measurement of depression in patients with cancer. Cancer 1984; 53:2243–2248[CrossRef][Medline]
  5. Leopold KA, Ahles TA, Walch S, et al: Prevalence of mood disorders and utility of the PRIME-MD in patients undergoing radiation therapy. Int J Radiat Oncol Biol Physiol 1998; 42:1105–1112[CrossRef][Medline]
  6. Kugaya A, Akechi T, Okuyama T, et al: Prevalence, predictive factors, and screening for psychologic distress in patients with newly diagnosed head and neck cancer. Cancer 2000; 88:2817–2823[CrossRef][Medline]
  7. D’Antonio LL, Long SA, Zimmerman GJ, et al: Relationship between quality of life and depression in patients with head and neck cancer. Laryngoscope 1998; 108:806–811[CrossRef][Medline]
  8. Kathol RG, Mutgi A, Williams JBW, et al: Diagnosis of major depression in cancer patients according to four sets of criteria. Am J Psychiatry 1990; 147:1021–1024[Abstract/Free Full Text]
  9. Golden RN, McCartney CF, Haggerty JJ Jr, et al: The detection of depression by patient self-report in women with gynecologic cancer. Int J Psychiatry Med 1991; 21:17–27[Medline]
  10. Silverstone PH: Poor efficacy of the Hospital Anxiety and Depression Scale in the diagnosis of major depressive disorder in both medical and psychiatric patients. J Psychosom Res 1994; 38:441–450[CrossRef][Medline]
  11. Hosaka T, Awazu H, Aoki T, et al: Screening for adjustment disorders and major depression in otolaryngology patients using the Hospital Anxiety and Depression Scale. Int J Psychiatry Clin Pract 1999; 3:43–48
  12. Razavi D, Delvaux N, Farvacques C, et al: Screening for adjustment disorders and major depressive disorders in cancer inpatients. Br J Psychiatry 1990; 156:79–83[Abstract/Free Full Text]
  13. Goring H, Baldwin R, Marriott A, et al: Validation of short screening tests for depression and cognitive impairment in older medically ill inpatients. Int J Geriatr Psychiatry 2004; 19:465–471[CrossRef][Medline]
  14. Passik SD, Lundberg JC, Rosenfeld B, et al: Factor analysis of the Zung Self-Rating Depression Scale in a large ambulatory oncology sample. Psychosomatics 2000; 41:121–127[Abstract/Free Full Text]
  15. Middelbow T, Ovesen L, Mortensen EL, et al: Depressive symptoms in cancer patients undergoing chemotherapy: a psychometric analysis. Psychother Psychosom 1994; 61:171–177[CrossRef][Medline]
  16. Leon AC, Portera L, Walkup JT: The development and evaluation of the brief depression screen in medically ill disability claimants. Int J Psychiatry Med 2001; 31:389–400[CrossRef][Medline]
  17. Parker G, Hilon T, Bains J, et al: Cognitive-based measures screening for depression in the medically ill: the DMI-10 and the DMI-18. Acta Psychiatr Scand 2003; 105:419–426
  18. Chochinov H, Wilson KG, Enns M, et al: Depression in the terminally ill. Am J Psychiatry 1994; 151:537–540[Abstract]
  19. Cohen-Cole SA, Kaufman KG: Major depression in physical illness: diagnosis, prevalence, and antidepressant treatment: a 10-year review, 1982-1992. Depression 1993; 1:181–204[CrossRef]
  20. Watson M, Haviland JS, Greer S, et al: Influence of psychological response on survival in breast cancer: a population-based cohort study. Lancet 1999; 354:1331–1336[CrossRef][Medline]
  21. Faller H, Bulzebruck H, Drings P, et al: Coping, distress, and survival among patients with lung cancer. Arch Gen Psychiatry 1999; 56:756–762[Abstract/Free Full Text]
  22. Loberiza FR Jr, Rizzo JD, Bredeson CD, et al: Association of depressive syndrome and early deaths among patients after stem-cell transplantation for malignant diseases. J Clin Oncol 2002; 20:2118–2126[Abstract/Free Full Text]
  23. Stommel M, Given BA, Given CW: Depression and functional status as predictors of death among cancer patients. Cancer 2002; 94:2719–2727[CrossRef][Medline]
  24. Hjerl K, Andersen EW, Keiding N, et al: Depression as a prognostic factor for breast cancer mortality. Psychosomatics 2003; 44:24–30[Abstract/Free Full Text]
  25. Kaplan GA, Reynolds P: Depression and cancer mortality and morbidity: prospective evidence from The Alameda County Study. J Behav Med 1988; 11:1–13[CrossRef][Medline]
  26. Zonderman AB, Costa PT Jr, McCrae RR: Depression as a risk for cancer morbidity and mortality in a nationally representative sample. JAMA 1989; 262:1191–1195[Abstract]
  27. Bleiker EM, van der Ploeg HM, Hendriks JH, et al: Personality factors and breast cancer development: a prospective longitudinal study. J Natl Cancer Inst 1996; 88:1478–1482[Abstract/Free Full Text]
  28. Tross S, Herndon J II, Korzun S, et al: Psychological symptoms and disease-free and overall survival in women with Stage II breast cancer: Cancer and Leukemia Group Br J Nat Cancer Inst 1996; 88:661–667
  29. Teno JM, Harrell FE Jr, Knaus W, et al: Prediction of survival for older hospitalized patients: The Help Survival Model: Hospitalized Elderly Longitudinal Project. J Am Geriatr Soc 2000; 48:S16-S24
  30. Hamilton M: A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23:56–62[Medline]
  31. Folstein MF, Folstein SE, McHugh PR: "Mini-Mental State:" a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189–198[CrossRef][Medline]
  32. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987
  33. Glassman AH, O’Connor CM, Califf CM, et al: Sertraline treatment of major depression in patients with acute MI or unstable angina. JAMA 2002; 288:701–709[Abstract/Free Full Text]
  34. Freedland KE, Skala JA, Carney RM, et al: The Depression Interview and Structured Hamilton (DISH): rationale, development, characteristics, and clinical validity. Psychosom Med 2002; 64:897–905[Abstract/Free Full Text]
  35. Lustman PJ, Clouse RE, Carney RM, et al: Proceedings of the National Institute of Mental Health Conference on Mental Disorders in General Health Care Settings. Rockville, MD, NIMH, 1987, pp 127-129
  36. Clark DA, Cook A, Snow D: Depressive symptom differences in hospitalized, medically ill, depressed psychiatric inpatients and nonmedical controls. J Abnorm Psychol 1998; 107:38–48[CrossRef][Medline]
  37. Lloyd-Williams M, Friedman T, Rudd N: An analysis of the validity of the Hospital Anxiety and Depression Scale as a screening tool in patients with advanced metastatic cancer. J Pain Symptom Manage 2001; 22:990–996[CrossRef][Medline]
  38. Parker G, Kalucy M: Depression comorbid with physical illness. Curr Opin Psychiatry 1999; 12:87–92[CrossRef]
  39. Silverstone PH: Concise Assessment for Depression (CAD): a brief screening approach to depression in the medically ill. J Psychosom Med 1996; 41:161–170
  40. Chochinov H, Wilson KG, Enns M, et al: Screening for depression in the terminally ill. Am J Psychiatry 1997; 154:674–676[Abstract]
  41. Whooley MA, Avins A, Miranda J, et al: Case-finding instruments for depression: two questions are as good as many. J Gen Intern Med 1997; 12:273–281
  42. Fisch MJ, Loehrer PJ, Kristeller J, et al: Fluoxetine versus placebo in advanced cancer outpatients: a double-blinded trial of The Hoosier Oncology Group. J Clin Oncol 2004; 21:1937–1943[CrossRef]
  43. Sephton SE, Sapolsky RM, Kraemer HC, et al: Diurnal cortisol rhythm as a predictor of breast cancer survival. J Nat Cancer Inst 2000; 92:994–1000[Abstract/Free Full Text]
  44. Spiegel D, Giese-Davis J: Depression and cancer: mechanisms and disease progression. Biol Psychiatry 2003; 54:269–282[CrossRef][Medline]



This article has been cited by other articles:


Home page
 PsychosomaticsHome page
M. R. Levy
Cancer Fatigue: A Neurobiological Review for Psychiatrists
Psychosomatics, July 1, 2008; 49(4): 283 - 291.
[Abstract] [Full Text] [PDF]

This Article
* Abstract Freely available
* Full Text (PDF)
* Alert me when this article is cited
* Alert me if a correction is posted
* Citation Map
Services
* Email this article to a Colleague
* Similar articles in this journal
* Similar articles in PubMed
* Alert me to new issues of the journal
* Add to My Articles & Searches
* Download to citation manager
* reprints & permissions
Citing Articles
* Citing Articles via HighWire
* Citing Articles via Google Scholar
Google Scholar
* Articles by Guo, Y.
* Articles by Nemeroff, C. B.
* Search for Related Content
PubMed
* PubMed Citation
* Articles by Guo, Y.
* Articles by Nemeroff, C. B.
Related Collections
* Depression
* Syndromes Secondary to General Medical Disorders

Get information about faster international access.

Privacy Policy

Copyright © 2006 Academy of Psychosomatic Medicine. All rights reserved.

Home | Search | Current Issue | Past Issues | Subscribe | All APPI Journals | Help | Contact Us

American Psychiatric Publishing, Inc. Academy of Psychosomatic Medicine
1000 Wilson Boulevard, Suite 1825, Arlington, VA 22209-3901 * 800-368-5777 * appi at psych.org