Recognition and Management of Depression in Skilled-Nursing and Long-Term Care Settings: Evolving Targets for Quality Improvement

Recognition and Management of Depression in Skilled-Nursing and Long-Term Care Settings: Evolving Targets for Quality Improvement

Recognition and Management of Depression in Skilled-Nursing and Long-Term Care Settings Evolving Targets for Quality Improvement Vicki L. Boyle, R.N.,...

103KB Sizes 0 Downloads 3 Views

Recognition and Management of Depression in Skilled-Nursing and Long-Term Care Settings Evolving Targets for Quality Improvement Vicki L. Boyle, R.N., B.S.N., Canopy Roychoudhury, Ph.D. Renee Beniak, R.N., B.S., L.N.H.A., Lisa Cohn, M.S. Albert Bayer, M.D., Ira Katz, M.D., Ph.D.

Objective: Depression is a common disorder associated with suffering, morbidity, and mortality in nursing home residents. It is treatable, and improving the quality of treatment can have a major impact. Methods: MPRO, Michigan’s Quality Improvement Organization, initiated a quality-improvement project in 14 nursing facilities to improve the accuracy of assessments, targeting, and monitoring of care. Electronic Minimum Data Set (MDS) data and medical-record abstraction results were combined to form the analytic dataset. Results: Findings from the baseline phase demonstrated that, according to medical and administrative records, 26% of newly admitted nursing home residents had symptoms of depression that were apparent at admission, and an additional 12% were recognized early in their stay. Eighty-one percent of residents with depression were receiving treatment on admission to the facility, and 79% of those with depression recognized by Day 14 were treated by then. Conclusions: These data demonstrate progress toward improving the initiation of treatment for depression in nursing homes; however, there are still opportunities for improving the quality of care and, especially, the quality of assessments. The authors recommend the addition of the Geriatric Depression Scale to the federally mandated MDS for cognitively intact patients. There could also be mechanisms to ensure that providers and facilities follow recommended practice guidelines. Initiating treatment with antidepressant medications should be followed with monitoring of residents to identify those who still have depressive symptoms and to modify or intensify their treatment. (Am J Geriatr Psychiatry 2004; 12:288–295)

Received February 4, 2003; revised October 24, 2003; accepted November 16, 2003. From MPRO, Michigan’s Quality Improvement Organization, Farmington Hills, Michigan (VLB,CR,RB,LC,AB) and the Department of Psychiatry, University of Pennsylvania, Philadelphia, PA (IK). Send correspondence to Vicki L. Boyle, R.N., B.S.N., Project Manager, MPRO, 22670 Haggerty Road, Suite 100, Farmington Hills, MI 48335-2611. e-mail: [email protected] 䉷 2004 American Association for Geriatric Psychiatry

288

Am J Geriatr Psychiatry 12:3, May-June 2004

Boyle et al.

T

he past 10–15 years have seen profound changes in rates of treatment for depression in nursing home residents. In the early to mid-1980s, only 10% of residents with a known diagnosis of depression ever received treatment,1 and the undertreatment of depression was one of the problems cited in the 1986 Institute of Medicine report2 that called for increased federal regulation of the country’s nursing homes. Since that time, nursing home reform legislation mandated a number of federal regulations designed to improve the quality of care for nursing home residents. These include requirements for use of a specific resident assessment instrument, the Minimum Data Set (MDS),3 which includes an evaluation of depressive symptoms, and guidelines for the use of psychiatric medications. These federal initiatives, along with the availability of safer and better-tolerated antidepressant medications as well as the dissemination of scientific findings on the malignancy of late-life depression and the efficacy of treatment, led to dramatic increases in treatment rates. According to a report from the Centers for Medicare & Medicaid Services (CMS), from the time that nursing home reform regulations were implemented until 1997, there was a 97% increase in antidepressant prescription use for all residents, from 12.6% to 24.9%.4 More recent 2001 data from CMS’s Online Survey, Certification, and Reporting (OSCAR) database demonstrates that 36% of all of the nation’s nursing home residents were receiving an antidepressant,5 and a study of residents from a convenience sample of nursing facilities in Philadelphia found rates in long-term care residents of 45%.6 Depression and its treatment were included in the MDS-based quality indicators for nursing home care that were developed in the early to mid-1990s by the Center for Health Sciences Research and Analysis (CHSRA) at the University of Wisconsin and were adopted for use by CMS.7 One of these indicators was the prevalence of depression, and another, the prevalence of depression not treated with antidepressant medications. Although it is clear that the recognition and treatment of depression is an important part of the overall quality of care in nursing homes, there may be questions about these indicators. One question is about the reliability and validity of case-recognition. Although the approaches used in the MDS have been validated, some research has suggested that differences between facilities in reported rates of depression may reflect variability in the process of

Am J Geriatr Psychiatry 12:3, May-June 2004

evaluation. This research also suggested that for cognitively-intact residents and those with mild-to-moderate cognitive impairment, the accuracy of assessments might be improved through the use of a standardized self-report instrument, such as the Geriatric Depression Scale (GDS).8 Another question is about the second indicator. It was developed when there were widespread concerns about the undertreatment of depression. However, in the current environment, where use of antidepressants is widespread, there may be questions about whether an indicator that encourages the initiation of treatment is sufficient, or whether there is a need to place an equal emphasis on the components of care that determine outcomes after treatment has been initiated. Motivated by these considerations, Michigan’s Quality Improvement Organization (MPRO) developed a CMS-supported quality improvement initiative in two phases. The first phase was designed to determine current practices related to the recognition, evaluation, and treatment of depression in skillednursing and long-term care settings. The second phase was to implement interventions to increase the accuracy of assessments and the effectiveness of treatment. This article reports the following findings from the first phase on: 1) the occurrence of depressive symptoms in newly-admitted residents to skillednursing facilities (SNFs) and long-term care facilities; 2) the rates of treatment for newly-admitted residents with depression, both at admission and during the initial stages of the nursing home stay; and 3) the processes of care and the outcomes of treatment. Also, because improving assessments by facilitating use of standardized assessment instruments is one of the components of the intervention planned for the second phase of this project, the article also reports findings on 4) the current use of the GDS.

METHODS This article reports on baseline findings from a project implemented by MPRO, a CMS-supported qualityimprovement organization, as part of an initiative designed to improve the quality of care for depression in nursing homes. Consistent with the goals, context, and conduct of the project, the methods used for data acquisition and the human-subject protections were those typical of quality-improvement activities,

289

Depression in Long-Term Care rather than either patient-oriented or health-services research. All data were derived from administrative or clinical records; the GDS, when conducted, was administered by facility staff. Fourteen facilities were recruited from a list of SNFs, including for-profit and not-for-profit, government/county, and Eden娂 homes that agreed to collaborate in this study. Each of the facilities indicated that they used the GDS (and not other screening or rating instruments) during clinical care. The inclusion criteria for residents within these facilities were the following: 1) a Medicare admission date between 1/1/2001 and 6/30/2001, and 2) the availability of both 5- and 14-day MDS assessments, one of which must have been the admission assessment. The exclusion criteria were MDS assessments indicating: 1) schizophrenia; 2) comatose state; 3) severely impaired daily decision-making; 4) making self understood rarely or never; and 5) understanding others rarely or never. Sampling was based on a fixed proportion of residents from each facility. All eligible residents with stays of 30 days or longer were selected (N⳱637). Also, 208 residents were randomly selected proportionately from each facility, to give an overall sample size of 845. The indicators described in Table 1 were designed to address questions about the rate of depressive symptoms in new admissions, rates of treatment, outcomes of care in longer-stay residents, and current use of the GDS using only MDS and medical record data. Quality Indicators (QI) are established in quality-of-care literature,7 whereas the test indicators (TI) are experimental. The Quality and Test Indicators were developed by the study team with expert consultation from a panel appointed by CMS and the National Institute of Mental Health (NIMH). Data abstractors used a tool developed by MPRO based on MedQuest娂 to collect medical record data not captured in the MDS. MPRO also developed a set of abstraction rules that provided field-by-field enumeration of the values that may be entered, data location in the source document, and aids in finding and recording data. A contingent pool of nurse-abstractors was trained in the use of the tool and subsequently tested for accuracy and interrater reliability. Their accuracy score was 94%, and their interrater reliability score was 92%. A total of 818 records were successfully abstracted. Residents were classified as depressed if they had

290

symptoms of depression documented in the MDS or chart, whether or not diagnoses were recorded or treatment was administered. For residents to be considered symptomatic, they had to exhibit sad mood plus two or more of the following: negative statements, agitation or withdrawal, waking with an unpleasant mood, not being awake for most of the day, being suicidal or having recurrent thoughts of death, or weight loss. Treatment was considered to be any of the following: antidepressant medication, or psychological therapy (individual or group) by any licensed mental health professional. This study used an MDS-based symptom score to track the resident’s mental condition over time. Symptoms considered were negative statements, repetitive physical movement, no activity, reduced social interaction, resisting care, unpleasant mood in morning, mood persistence, suicidal thoughts, or recurrent thoughts of death. On the basis of the MDS, scoring for each item was as follows: not exhibited in 30 days (0); exhibited up to 5 days/week (1); and exhibited 6–7 days/week.2 The scale score was the sum of the item scores. The baseline score was the higher of the scores at the 5-day and 14-day assessment. The electronic MDS data and the medical record abstraction results were combined to form the analytic dataset. For each indicator, information relevant to coding the numerator and denominator was derived from all available sources of information. SAS Version 8.1 was used to analyze the data.

RESULTS The study sample was primarily female (69%). Of the women, 17% were age 75 or younger, 46% were 76– 85, and 37% were 86 or older. Among the men, 24% were age 75 or younger, 51% were 76–85, and 25% were 86 or older. Of the 818 residents, 313 (38%) had depressive symptoms by Day 14. Of these, 213 (68%) were admitted with a diagnosis of depression. For 100 (32%), depression began or was first recognized after admission to the facility. Among the 213 residents with depressive symptoms at the time of admission, 193 (91%) were on some form of treatment when they entered the facility (Figure 1). Of the 154 residents in our sample with a length of stay of 60 days or more, 69 (45%) were depressed by Day 14, and 53 (77%) of the depressed residents were

Am J Geriatr Psychiatry 12:3, May-June 2004

Boyle et al. TABLE 1.

Results of indicators

Indicators

N

D

Rate

QI-1a: Proportion of newly-admitted residents who are assessed for depression with the Geriatric Depression Scale (GDS) by Day 7 of stay N⳱Those in the numerator with assessment by Day 7 D⳱All residents who meet inclusion/exclusion criteria QI-1b: Proportion of newly-admitted residents who are assessed for depression with the Geriatric Depression Scale (GDS) by Day 14 of stay N⳱Those in the numerator with assessment by Day 14 D⳱All residents who meet inclusion/exclusion criteria QI-2a: Proportion of depressed residents who receive depression treatment by Day 7 of stay N⳱Those in the numerator who receive depression treatment by Day 7 D⳱Residents with symptoms or diagnosis of depression by Day 7 QI-2b: Proportion of depressed residents who receive depression treatment by Day 14 of stay N⳱Those in the numerator who receive depression treatment by Day 14 D⳱Residents with symptoms or diagnosis of depression by Day 14 QI-3: Proportion of residents assessed positive for depression by the Geriatric Depression Scale (GDS) by Day 14 who are also assessed positive for depression by the MDS assessment N⳱Those in the numerator with symptoms or diagnosis of depression noted in either the 5or 14-day MDS D⳱Residents assessed positive for depression by GDS by Day 14 TI-1: Proportion of residents assessed positive for depression and receiving depression treatment by Day 14 whose symptom score decreased N⳱Those in the numerator who had a reduction in symptom score between the baseline and 60-day MDS D⳱Residents with symptoms or diagnosis of depression by Day 14 who received treatment by Day 14 and had a 60-day MDS completed TI-2: Proportion of residents assessed positive for depression and receiving depression treatment by Day 14 whose symptom score increased or did not change between the baseline MDS assessment and the 60-day MDS assessment who were re-assessed by Day 60 of stay N⳱Those in the numerator with a reassessment completed by Day 60 of stay D⳱Residents who: had symptoms or a diagnosis of depression by Day 14, received treatment by Day 14, had a completed 60-day MDS, had an increase or no change in symptom score between baseline MDS and 60-day MDS TI-3a, TI-3b, TI-3c, TI-3d: Proportion of those residents assessed positive for depression by Day 14 whose symptom score decreased between the baseline MDS assessment and the 60-day MDS assessment who had: N⳱Those in the numerator who had a) medication, b) therapy, c) both, d) neither D⳱ Residents who: • had symptoms or a diagnosis of depression by Day 14 • received treatment by Day 14 • had a completed 60-day MDS • had decrease in symptom score between baseline MDS and 60-day MDS a. Medication b. Other therapy c. Both d. Neither TI-4a, TI-4b, TI-4c, TI-4d: Proportion of those residents assessed positive for depression by Day 14 whose symptom score increased or did not change between the baseline MDS assessment and the 60-day MDS assessment who had N⳱Those in the numerator who had a) medication, b) therapy, c) both, d) neither D⳱ Residents who: had symptoms or a diagnosis of depression by Day 14, received treatment by Day 14, had a completed 60-day MDS, had an increase or no change in symptom score between baseline MDS and 60-day MDS a. Medication b. Other therapy c. Both d. Neither

71

818

9%

92

818

11%

203

269

75%

248

313

79%

26

42

62%

19

53

36%

32

35

91%

22 1 2 1

26 26 26 26

85% 4% 8% 4%

32 2 8 2

44 44 44 44

73% 5% 18% 5%

QI: quality indicator; TI: test indicator; D: denominator.

Am J Geriatr Psychiatry 12:3, May-June 2004

291

Depression in Long-Term Care treated. However, only 18 of the 53 residents (34%) actually exhibited an improvement in symptom scores between baseline and the 60-day assessment (Figure 1). Table 1 provides a synopsis of the specific quality and test indicators. Of the residents whose condition improved, 85% were on medication, 4% on FIGURE 1.

Sample assessment and outcome

Breakdown of diagnosed with depression by Day 14 Rx prior to admission (N=193)

Total Sample (N=818)

No Rx prior to admission (N=20)

Diagnosed prior to admission (N=213)

Newly diagnosed (N=100)

Diagnosed with depression by Day 14 (N=313)

Discharged/ died by Day 60 (N=159)

Depressed by Day 60 (N=69)

Treated (N=53)

No improvement (N=35)

292

No depression diagnosed by Day 14 (N=505)

Complete 60-Day assessment (N=154)

Not depressed by Day 60 (N=85)

Not treated (N=16)

Improved (N=18)

therapy, 8% on both, and 4% on neither form of treatment (Test Indicator 3a-d). For those who did not improve, the corresponding proportions were 73%, 5%, 18%, and 5% (Test Indicator 4a-d). Although the use of therapy was greater in those who did not improve (23% as opposed to 12%), the difference was not statistically significant (Fisher’s exact test: p⳱0.345). Of the 35 residents whose condition did not improve, 32 (91%) were reassessed by Day 60. However, only 13 (41%) had a change in medication, adjustment of dosage, or augmentation and documentation of the change in medication as an element in their reassessment process. Although each of the nursing homes that participated in this project stated that they used the GDS to evaluate depressive symptoms, their staff administered it to only 9% of residents by Day 7 of their stays and to 11% by Day 14 (Table 1, QI 1a,b). Thus, its use is highly selective; however, the methods used in this project are not informative about when it is used. Nevertheless, findings do suggest that it can be of value. Of the residents assessed positive for depression by Day 14 by the GDS, only 26 (62%) were also assessed positive by the MDS (Table 1, QI3). Although there are multiple sources of potential bias in the sample of residents for whom the GDS was administered, there must be concerns about the low apparent sensitivity of the MDS.

DISCUSSION This article reports findings from the baseline phase of a quality-improvement initiative conducted by MPRO. They are of general interest because they demonstrate that the problems in care delivery evolve over time. The findings confirm the general principle that quality-improvement must be a continuous process, requiring both specific interventions designed to improve performance in validated quality-indicators, as well as monitoring to evaluate the continued salience of indicators in light of the multiple factors that modify and shape care over time. The most general conclusion to be drawn from this project must be that the most significant targets for quality improvements at this time are very different from those that were relevant when the current indicators were developed.

Am J Geriatr Psychiatry 12:3, May-June 2004

Boyle et al. Occurrence of Depression in Newly-Admitted Residents Thirty-eight percent of the admitted residents sampled in this study were positive for depressive symptoms by Day 14 of their nursing home stay. Thus, depression remains a highly significant problem among those admitted to a nursing home. The high occurrence of symptoms leading to suffering, together with the evidence for their associations with increased morbidity, care needs, and mortality, confirm the importance of an ongoing focus on improving the quality of depression treatment in nursing facilities. Although the high rate represents a call for facilitating treatment, the fact that the depressive symptoms are recognized and documented in the clinical and administrative records stands as a testament to the sensitivity of the facilities’ staff and to the value of federal initiatives, including those mandating the use of the MDS. It is important to note that, for 68% of the residents found to have depression during the initial phases of their nursing home stays, the problem had already been recognized at the time of their admission. This finding supports the view of depression as a chronic disease, or a potentially chronic complication of the other medical and neurological disorders that led to nursing home placement, rather than a reaction to the stresses of nursing-home life. Rates of Treatment for Newly-Admitted Residents With Depression On average, nursing facilities initiated or continued treatment on 75% of their residents with depressive symptoms by Day 7 and 79% of their residents by Day 14 of their nursing home stay. These proportions stand in sharp contrast to older scientific findings, the Institute of Medicine Report, and even some more recent papers that have called attention to the undertreatment of depression in nursing homes.10 With respect to the quality indicators developed by CHSRA, the finding that the majority of residents with depression were being treated at the time of their admission raises questions about the extent to which the indicators can serve as measures of the quality of care within the nursing facility, especially for short-stay patients. Furthermore, the 38% rate of symptoms in spite of high rates of treatment demonstrates that the assessment of quality should not be limited to whether or not treatment is initiated. It

Am J Geriatr Psychiatry 12:3, May-June 2004

should also consider subsequent components of care. From another perspective, the finding that 78% of the treatments provided to depressed residents by Day 14 of their stays had been initiated before their admissions, suggests that a comprehensive approach to improving the quality of care for late-life depression must attend to the continuity of care across treatment contexts, as well as to what happens within any single care setting. Outcomes of Treatment and Processes of Care Of the residents who remained in the facilities, only 34% of those receiving treatment for depression had symptom scores that decreased by Day 60. This raises questions about why more of the patients being treated did not improve. To address this question, we looked at the processes of care described in the medical records for those residents who received treatment. Although 91% of the residents who did not improve were reassessed by their physicians, 59% of the reassessments included only a confirmation of diagnosis, and/or an evaluation of treatment response. In these cases, treatment was not changed, nor were residents referred to mental health professionals. The most likely explanation for the low rates of improvement in spite of treatment lies in the findings on the downstream processes of care. Key recommendations from the Agency for Health Care Policy and Research Practice Guidelines for the Diagnosis and Treatment of Depression in Primary Care10 remain useful for older patients, in general,11 and for nursing home residents.12 They call for reevaluations of those receiving treatment by approximately 6 weeks, with modification or intensification of treatment for those who have not shown evidence of therapeutic responses. Our findings demonstrate that treatment was modified over a 60-day period in only a minority of non-responders. The Minimum Data Set (MDS) as a Data Source and the Potential Added Value of the Geriatric Depression Scale (GDS) The findings and conclusions discussed above are based on data derived from the MDS and medical records. Although there has been substantial research documenting the value of the MDS, questions have been raised specifically about the validity of the MDS assessments for mood and behavioral symptoms.

293

Depression in Long-Term Care Self-report instruments such as the GDS are widely recognized as valid tools for assessing depressive symptoms in elderly subjects. Of specific relevance to its use in the nursing home, there is evidence that, when it is administered verbally, it remains useful in individuals with mild-to-moderate degrees of dementia.13–15 Of the 14 facilities studied, 13 stated that they used the GDS as a screening tool. However, the study found that the GDS was infrequently and inconsistently used. Apparently, facilities relied heavily on the MDS, including the “Mood and Behavior” section, as well as other information, such as weight change and activity-pursuit patterns for assessment. Supplementing the use of the MDS with consistent use of validated ratings such as the GDS in residents who are cognitively intact or mildly/moderately impaired, might allow facilities to recognize and evaluate depressive symptoms with a high degree of sensitivity and specificity. On a facility level, ratings of depression derived from use of the GDS in more intact patients can serve as a tool for calibrating the MDS assessments, confirming or disconfirming conclusions about the presence or absence of clinically significant depression. In fact, use of the GDS in this way could serve a function in quality-improvement initiatives, helping staff to be more confident and accurate in their evaluations. Limitations If there are, in fact, limitations on the accuracy of the MDS assessments for depression in certain nursing homes, there must be concerns about the validity of the findings reported here. However, in its design, this initiative worked to minimize these concerns by utilizing data from both the MDS and from medicalrecord abstractions. Nevertheless, there may be validity limitations in quality-of-care judgments based on the assessments conducted by the staff within specific facilities. These limitations apply not just to the findings reported here, but also to the federal evaluation of the quality of care that facilities deliver. Another limitation to the findings is related to the attrition of residents over the 60 days after admission. The universe for this study included all newly-admitted Medicare residents who had a 60-day stay, including those admitted for subacute as well as longterm care. The 60-day stay admission cohort represented only 18.8%. Thus, our conclusions about

294

treatment outcomes are based on a limited sample. However, from another perspective, the high rates of discharge raise a number of concerns that are relevant to both clinical care and public policy. For many individuals who are admitted to nursing homes, depression begins before admission and/or continues after discharge. If there are concerns about the quality of care within any one clinical setting, there must be even greater concerns for those individuals whose treatment for depression (or other chronic conditions) is fragmented by transitions between settings.

CONCLUSIONS AND PRACTICAL RECOMMENDATIONS This quality-improvement initiative has identified concrete strategies that can advance the goals of improved screening, assessment, and management of depression in skilled-nursing and long-term care settings. Also, our findings suggest that merely detecting, diagnosing, and initiating treatment of residents with depression is not enough. More attention should be given to the importance of reassessment and the need to intensify or modify the treatment plan until an adequate response has been achieved, and resolution of signs and symptoms is well documented. More generally, our findings suggest that there has been dramatic progress over recent years in the extent to which treatment for depression is initiated, with almost universal treatment of symptomatic, newlyadmitted residents. However, only a minority of these individuals appeared to benefit from treatment. Thus, we have made substantial progress in a qualityimprovement cycle. Accordingly, it may be time to revise the domains of care that are included in quality-indicators. They should shift from a major emphasis on initiating treatment to improving guideline adherence for downstream care. Additional research will be useful in developing strategies to ensure continuity of care to people across treatment settings. We thank all of the 14 skilled nursing homes that participated in the project and were willing to share their data with us. Disclaimer: The analyses upon which this publication is based were performed under Contract number 500-99-

Am J Geriatr Psychiatry 12:3, May-June 2004

Boyle et al. MI02 titled “Utilization and Quality Control Peer Review Organizations for the State of Michigan,” sponsored by the Centers for Medicare & Medicaid Services, Department of Health and Human Services. The content of this publication does not necessarily imply endorsement by the U.S. Government of trade names, commercial products, or organizations. The authors assume full responsibility for the accuracy and completeness of the ideas presented. This ar-

ticle is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare & Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore, required no special funding on the part of this contractor. Ideas and contributions to the authors concerning experience in engaging with issues presented are welcomed.

References 1. Heston LL, Garrard J, Makris L, et al: Inadequate treatment of depressed nursing home elderly. J Am Geriatr Soc 1992; 40:1117– 1122 2. Institute of Medicine, Committee on Nursing Home Regulation: Improving the Quality of Care in Nursing Homes. Washington, DC, National Academy Press, 1986 3. Morris JN, Hawes C, Fries BE, et al: Designing the national resident assessment instrument for nursing homes. Gerontologist 1990; 30:293–307 4. Health Care Financing Administration: A Report to Congress: Study of Private Accreditation (Deeming) of Nursing Homes, Regulatory Incentives, and Non Regulatory Incentives, and Effectiveness of the Survey and Certification System. Washington, DC, 1998 5. www.medicare.gov/nhcompare/Search/Related/SpecialNote.asp. 6. Datto CJ, Oslin DW, Streim JE, et al: Pharmacological treatment of depression in nursing home residents: a mental health services perspective. J Geriatr Psychiatry Neurol (in press) 7. Zimmerman DR, Karon SL, Arling G, et al: Development and testing of nursing home quality indicators. Health Care Financing Review 1995; 16:107–127

Am J Geriatr Psychiatry 12:3, May-June 2004

8. Brink TL, Yesavage JA, Lum O, et al: Screening tests for geriatric depression. Clinical Gerontologist 1982; 1:37–43 9. Brown MN, Lapane KL, Luisi AF: The management of depression in older nursing home residents. J Am Geriatr Soc 2002; 50:69– 76 10. Clinical Practice Guideline: Depression in Primary Care, Vol 2: Treatment of Major Depression, Number 5. Rockville, MD, U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 93-0551. April 1993 11. Alexopoulos GS, Katz IR, Reynolds CF III, et al: Expert Consensus Panel: Pharmacotherapy of depressive disorders in older patients. Postgrad Med Special Report, October, 2001 12. American Medical Directors Association, Clinical Practice Guidelines for Depression. Columbia, MD, American Medical Directors Association, 1996 13. Parmelee PA, Lawton MP, Katz IR: Psychometric properties of the Geriatric Depression Scale among the institutionalized aged. Psychol Assess 1989; 1:331–338 14. McGivney SA, Mulvihill M, Taylor B: Validating the GDS depression screen in the nursing home. J Am Geriatr Soc 1994; 42:490– 492 15. Espiritu DA, Rashid H, Mast BT, et al: Depression, cognitive impairment, and function in Alzheimer’s disease. Int J Geriatr Psychiatry 2001; 16:1098–1103

295