Journal of Gerontological Nursing


Sarah McDermott Keane, MN, RN, CS; Sharon Sells, RN


Depression in late life is treatable and frequently as chronic only because adequate therapeutic intervention is not initiated earlier.


Depression in late life is treatable and frequently as chronic only because adequate therapeutic intervention is not initiated earlier.

Depression is the most prevalent and mental disorder in later life. the frequency of depressive states may be as high as 25% for elderly living in institutional care settings,1 yet depression is frequently overlooked because its symptoms are often regarded as consistent with aging or confused with various physical illnesses. The institutionalized elderly themselves may accept a sense of dissatisfaction and unhappinness as being inevitable accompaniments of aging. The longer the depression is neglected, the more complex it becomes, generating a host of new problems such as complications of chronic illnesses or difficulties with behvior management.

The signs and symptoms of depression in elderly patients are difficult to differentiate from symptoms of normal aging. The lives of institutionalized elderly tend to he heavily laden with interpersonal losses, failing health, loss of social and economic resources, and loss of control.2 All of these situations can precipitate feelings of sadness, disappointment and helplessness.

Depression may be the first symptom of a serious physical illness. Medications used to treat diseases common in the elderly may often evoke depressive symptomatology. Very often, depression and physical illnesses coexist. Therefore, it is not surprising that treatable depressions are overlooked in elderly patients with physical illnesses, and that treatable physical illnesses are often not managed optimally in elderly patients diagnosed as depressed.3

Depression in late life is treatable and frequently is chronic only because adequate therapeutic intervention is not initiated earlier.4 Moreover, depression in the elderly has many similarities to depression in other age groups, particularly in its episodic nature, tendency to remit, and potential for favorable immediate outcome.5

One major problem in managing depression in the elderly concerns the adequacy of present diagnostic measures. Most existing depression rating scales have been developed and validated in younger populations; the reliability and validity of these scales in older populations has not been welldocumented. Therefore, this study was designed to assess the reliability of the Beck Depression Inventory (BDI)6 and the Zung Self-Rating Depression Inventory (SDS)7 in an older adult group; to compare the effectiveness of two self-report, single dimension depression scales as screening tools; and to identify a systematic nursing approach for recognizing depressive symptomatology in the elderly.


Sample Selection Criteria

The following criteria were used to determine eligibility for the study: aged 65 or older; ability to read, write, and comprehend English; ability to complete the instruments independently; and willingness to consent to participation in the study.

An exception was made to these criteria. Four patients who had visual or manual dexterity limitations but wished to participate in the group activity were administered the depression scales and the biographical questions with the assistance of the nurse investigator. Their results were not included in the aggregate data but were reported to their nurses.

Sample Characteristics

The study sample was a convenience sample of 68 elderly residents (60 women and 8 men) of a Southeastern nursing care facility providing skilled, intermediate, and retirement home levels of nursing care. Ages ranged from 70 to 97 years with a mean age of 82.74 (SD = 5.99). More than half (38) of the group were widowed (56%); 17 (25%) were single, never married; 11 (16%) were married; and 2 (3%) were divorced.





According to medical records, all subjects had at least one, but no more than two diagnosed medical illnesses. The primary diagnosis of 18 patients (26%) was osteoarthritis, 11 patients (16%) were diagnosed as having cerebrovascular accident, 10 patients (15%) as having arteriosclerotic heart disease, and the remaining 29 (43%) patients' diagnoses included a variety of chronic illnesses such as cancer, diabetes, pulmonary disease, and rheumatoid arthritis. The major medications prescribed included antihypertensives, antacids, analgesics, laxatives, vitamin and mineral supplements, and sedatives. No attempt was made to determine the medications actually taken on a regular basis by the patients.


Subject recruitment was accomplished by posting sign-up sheets on the activity boards of each floor within the facility's buildings. Morning and afternoon sessions were scheduled, thus allowing participants to select the time most convenient for their schedules, as well as acknowledging diurnal variations of depressive symptomatology expression. Sessions were held in the activity rooms of each of the buildings to provide a safe, familiar, and common environment.

Participants were informed of the purposes and nature of the project and the intended uses of the data so that informed consent could be obtained. Biographical questions included gender, age, marital status, religious affiliation, major illnesses, and activity participation. The nurse investigator had access to medical records to verify demographic and physiological data. Four group sessions devoted to completing the two depression scales and a biographical data sheet were conducted by the nurse investigator with the assistance of a geriatric staff nurse.


It is clear from the literature that many life changes may precipitate depression. Various screening scales, especially the BDI6 and the SDS,7 have been advocated in assessing the existence of depression.813 Furthermore, these depression scales include items dealing with loss of personal control, which is a major life issue for many elderly. Identification of the perceptions of the elderly is central to accurate assessment and the development of interventions enabling the elderly to regain a sense of control in decision making within their current living situations. Self-reporting scales also have the advantage of being sensitive to change and are therefore useful for assessing treatment' progress. Both the BDI6 (21 items) and the SDS7 (20 items) are widely used self-rating depression scales consisting of affec tive, psychological, and somati indicators related to the depressiv state. Both scales contain self-eval uative statements; respondents ar instructed to read and then select th one statement that best describes th way they felt during the past week.

The use of self-report scales such í the BDI6 and the SDS7 is limited t those who are able to complete th instruments; however, both are availabl in a parallel form that is interviewe assisted. Major criticisms of usin these two scales with the elderly cor cera the heavy weighting of somati items and the elderly's ability to full comprehend the task.

The BDI6 consists of 21 categories c behavioral indicators clinically relate to the depressive state. Each categor contains four self-evaluative statement that are severity graded ranging fron neutral (0) to maximum severity (3] Participants were instructed to rea each group of four statements and t< select the one that most appropriatel described the way they felt during th past week. If a participant reported tha more than one item was applicable ani could not decide between the two, th standard procedure of using the highe rating was followed.6 The BDI wa scored by summing the ratings; th higher the score, the greater the severit of depression.6 The range of possibk scores was 0 to 63; cutoff scores wen those recommended by Gallagher e al.12 These cutoff scores were used t< estimate severity: normal range = 10 oi below; mild depression = 11 to 16 moderate depression = 17 to 23; anc severe depression = 24 or greater.

The content validity of the BDI waj derived from the fact that the items wert based on clinical observations o! patients being treated for depression common attitudes and symptoms asso dated with depression, and the psychi atric literature on depression.14 Supporting evidence for concurrent validitj has been provided in studies that cone lated the BDI with other instrument designed to measure depression. It wa? found, for example, that the BDI cone lated significantly with clinicians' rat ings of depression (r = .61 to .73), with the Hamilton Rating Scale (r = .75 to .82), and with the SDS (r = .72 to .76). ,5'17

The SDS is a 20-item screening inventory of the client's current depressive symptomatology.7 Zung designed the scale to measure three basic dimensions of depression: physiological factors, psychological factors, and affect. Subjects were asked to indicate on a 4-point scale how often during the past week a specific statement was true with a range of possible scores from 20 to 80. The SDS is also scored ' by summing the ratings, with the higher scores indicative of more depressive symptomatology.7 Previous trials using the SDS with elderly subjects were consulted to determine cutoff scores.13·18

Blazer10 found correspondence between DSM-III diagnoses of major depressive episodes and increased SDS scores in elderly inpatients. Murkofsky et al19 found internal consistency, as measured by coefficient alpha, acceptably high (0.86) for a group of 48 wellfunctioning senior citizens with a mean age of 70.8, whereas McGarvey et al13 found internal consistency to be unacceptably low, particularly for their group of 107 old-old (73- to 88-yearold) subjects.

Validity of the SDS was demonstrated by Okimoto et al in their study of 55 patients, with a mean age of 69.4, who were attending a general medical clinic of a Veteran's Administration medical center.20 In addition to the SDS, patients were interviewed by a psychiatrist and 17 (31%) were found to meet DSM-III criteria for major ,depressive disorders. Employing the DSM-III as the criterion, the SDS was found to correctly classify 86% of the patients.


The first purpose of this study was to assess the reliability of the BDI and the SDS in a group of elderly persons. Utilizing this sample's responses, coefficient alpha reliabilities were sufficient to suggest that the BDI (0.79) and the SDS (0.75) are potentially useful screening systems with an elderly population, and psychometrically within an acceptable range.18

Using the BDI with persons over 60 years of age, Gallagher et al reported a coefficient alpha reliability of 0.76 for depressed elderly outpatients and 0.73 for the nondepressed normal elderly outpatients.21 This study's findings are also congruent with findings of Czirr and Gallagher,22 Gallagher et al,12 Lewinsohn and Ten,9 and Zung and Green,11 supporting the position that both the BDI and SDS scales are reliable measures of screening for depression in an elderly population.









Comparing the effectiveness of the BDI and SDS as screening tools for depression in the elderly was the second main objective of the study. Means and standard deviations for the scores on both the BDI and SDS are presented in Table 1 for the total sample and subsamples according to the type of care received.

The mean for the BDI was 6.9 (SD = 5.74); scores ranged from 0 to 26. Using a cutoff score of U resulted in 16% of the sample (n = 11) meeting the criteria for major or subclinical depression. The mean score of the SDS was 36.81 (SD = 8.59); scores ranged from 20 to 59 . For this scale, a cutoff score of 40 was used. According to this criterion, 31% of the sample (n = 21) was depressed.

The increased number of positive cases of depression found with the SDS is consistent with prior research findings that criticize the SDS for a large number of false positives.11 Three of four patients with a diagnosis of depression at the time of study participation were correctly identified as depressed by the SDS and all four were correctly identified by the BDI. The Pearson correlation coefficient of the BDI and SDS was +0.63. Utilizing this sample's responses, coefficient alpha reliabilities were sufficient to suggest that the BDI (0.79) and the SDS (0.75) are potentially useful screening instruments with an elderly population and psychometrically within an acceptable range.23

Overall, residents of the intermediate level group expressed higher mean scores on both scales than did skilled and retirement groups. Thus, those patients receiving intermediate levels of care tended to have higher depression scores than those in the other two groups.

Although residents of the intermediate level group expressed higher mean scores on both scales, a greater percentage of the retirement group were classified as depressed by both scales. The number of positive cases of depression according to group are depicted in Table 2.

An analysis of the rank ordering of the BDI revealed that the five items most disturbing to this sample included fatigue, loss of libido, work inhibition, sleep disturbance, and irritability. On the SDS, however, the five most distressing items were inability to do things, not feeling well even in the morning (diurnal variation), difficulty making decisions, feeling useless and unneeded, and loss of appetite. According to Blazer's comparison of symptoms assessed by various depression rating scales, this sample's responses indicate a greater intensity of physical symptoms on the BDI and a greater intensity of cognitive symptoms on the SDS.24

The third purpose of this study was to identify a systematic nursing approach for recognizing depressive symptomatology in elderly patients. The 11 subjects (10 females, 1 male) identified as depressed by the BDI were monitored with monthly assessments for 6 months. The BDI and a nursing assessment guide constructed and conducted by the primary nurse investigator comprised these assessments. Individualized assessment is considered a critical step in identifying the nature of symptoms and precipitating factors in the many depressions of the elderly (Table 3).25·20

The selection of the BDI was supported by the findings of Gallagher and Thompson that the BDI is sensitive to change and is useful in assessing changing patterns of depressive symptoms.27 It is also supported by the lack of current estimates of adequate reliability and validity for the SDS with old-old samples (over 70 years), as well as documented evidence of spuriously high identification of normal elderly as being depressed by the SDS.

Eight of these 11 subjects evidenced a cluster of depressive symptoms by both scale interpretation and nursing assessment for the entire 6 months. Initial symptoms were predominantly physiological followed by a combination of physiological symptoms and mood disturbance. Responses on the BDI identified target areas of concern, whereas the assessment guideline helped in the detailed description of the symptoms and the identification of factors leading to the depression. Within 6 months after initial testing, these eight subjects were either transferred to the next care level requiring more skilled nursing care or were deceased.

The remaining three subjects evidenced a changing intensity of core depressive symptoms, indicating the subjects were not clinically depressed, but were showing transitory symptoms common in the older age group. This descriptive data is indicative of the potential value of depressive screening scale responses in conjunction with assessment to establish the affective state of the patient. This information could also be pertinent for nursing home administrators in planning for staffing requirements and predicting patient acuity levels or behavioral problems.


Results of the present investigation suggest that both the BDI and SDS appear to be adequately reliable as clinical screening instruments with an elderly population. However, the SDS may result in a higher number of false positives. For this reason, the BDI might be a preferable screening instrument.

Some keys to the detection of depression are directly observable, and geriatric nurses can be instrumental in early detection because they have the greatest amount of contact with elderly patients. Findings of this study suggest that responses to the BDI, in conjunction with individualized nursing assessment, can facilitate identification of specific symptoms that should be further evaluated. Currently, the nursing staff at the facility where this study was conducted use printed index cards with depressive assessment guidelines to enhance their skills in early recognition of depressive symptoms.


It is important that the gerontological nurse working with elderly patients be aware that depression is the psychiatric symptom most often encountered in the and that depression is as distressing in this age group as it is in all others. Depression also represents a disorder that often is reversible with prompt and appropriate treatment.

By routinely using a systematic tool or guidelines, nurses could improve their ability to recognize depression or changes indicative of altering health status. Patient input may identify target areas for further exploration and enhance the elder's sense of control through active participation in care. The increased health risks and the clinical, financial, and emotional benefits of early detection, support the potential usefulness of a depression screening instrument used in conjunction with a nursing assessment guide.


  • 1. Ry PS. Depression, Stress, and Adaptations in the Elderly Rockville, MD: Aspen Publishers Inc.; Í986.
  • 2. Klerman G. Problems in the definition and diagnosis of depression in the elderly. In: Breslau L, Haug M, Eds. Depression and Aging: Causes, Care, and Consequences. New York: Springer Publishing; 1983:3-19.
  • 3 . Weissman M , Myers J. Affective disorders in a US urban community. Arch Gen Psychiatry. 1978; 35:1304-1307.
  • 4. Blazer D. Depression. Paradoxically, a cause for hope. Generations. 1986; 10(3):21-23.
  • 5 . Epstein L . Symposium on age differentiation in depressive illness. J Gerontol. 1976; 31(3):278-282.
  • 6. Beck AT, Ward CH. Mendelson M, et al. An inventory for measuring depression. Arch Gen Psychiatry. 1961; 10:561-571.
  • 7. Zung W. A self-rating depression scale. Arch Gen Psychiatry. 1965; 12:63-70.
  • 8. Spitzer R, Endicott J, Robins E. Research diagnostic criteria: Rationale and reliability. Arch Gen Psychiatry. 1978; 35:773-782.
  • 9. Lewinsohn P, Ten L. Selection of depressed and nondepressed subjects on the basis of self-report data. J Consult Clin Psychol. 1982; 50:590-591.
  • 10. Blazer D. The diagnosis of depression in the elderly. J Am Geriatr Soc. 1980; 28(2):52-58.
  • 11. Zung W, Green R. Detection of affective disorders in the aged. In: Eisdorfer C, Fann W, Eds. Psychopharmacology and Aging. New York: Plenum Press; 1973:127-131.
  • 12. Gallagher D, Breckenridge J, Steinmetz J, et al. The Beck Depression Inventory and research diagnostic criteria: Congruence in an older population. J Consult Clin Psychol. 1983; 51(6):945-946.
  • 13. McGarvey B, Gallagher D, Thompson L, et al. Reliability and factor structure of the Zung Self-Rating Depression Scale in three age groups. Essence. 1982; 5:141-151.
  • 14. Beck AT. Depression: Causes and Treatment. Philadelphia: University of Pennsylvania Press; 1972.
  • 15. Beck A . The Diagnosis and Management of Depression. Philadelphia: University of Pennsylvania Press; 1973.
  • 16. Beck A, Beamesderfer A. Assessment of depression: The depression inventory. In: Pichot P, Ed. Psychological Measurements in Pharmacology: Modern Problems of Pharmacopsychiatry. Vol. 7. Paris: Karger, Basel; 1974:151-169.
  • 17. Beck A, Beck R. Screening depressed patients in family practice: A rapid technique. Postgrad Med. 1972; 52(6):81-85.
  • 18. McNair D. Self-rating scales for assessing psychopathology in the elderly. In: Raskin A, Jarvik L, Eds. Psychiatric Symptoms and Cognitive Loss in the Elderly. Washington, DC: Hemisphere; 1979:157-168.
  • 19. Murkofsky C, Conte H, Plutchik R, et al. Clinical utility of a rapid diagnostic test series for elderly psychiatric outpatients. J Am Geriatr Soc. 1978; 26:22.
  • 20. Okimoto J, Barnes R, Veith R, et al. Screening for depression in geriatric medical patients. Am J Psychiatry. 1982; 139:799.
  • 21. Gallagher D, Nies G, Thompson L. Reliability of the Beck Depression Inventory with older adults. J Consult Clin Psychol. 1982; 50:152-153.
  • 22. Czirr R, Gallagher D. Assessing depression in older adults. In: Keller P, Rett K, Eds. Innovation in Clinical Practice: A Source Book. New York: McGraw-Hill Book Company; 1984.
  • 23. Nunnally J. Psychometric Theory. 2nd ed. New York: McGraw-Hill Book Company; 1978.
  • 24. Blazer D. Depression in Late Life. St. Louis: CV Mosby Company; 1982.
  • 25. Zarit S. Aging and Mental Disorders. New York: Free Press; 1980.
  • 26. Shinfuku N. Depression in old age. In: Ban T, Gonzalez R, Jablensky A, et al, Eds. Prevention and Treatment of Depression. Baltimore: University Parie Press; 1981:63-68.
  • 27. Gallagher D, Thompson L. Depression. In: Lewinsohn P, Teri L, Eds. Clinical Geropsychology: New Directions in Assessment and Treatment. New York: Pergamon; 1983:1-37.








Sign up to receive

Journal E-contents