Reducing the Risk of Late-Life Suicide Through Improved Depression Care

Professor, Psychiatry and Behavioral Sciences, Albert Einstein College of Medicine, Director, Division of Geriatric Psychiatry, Montefiore Medical Center

On September 14th, 2006, the United States Senate Special Committee on Aging conducted a hearing titled, “Generation at Risk: Breaking the Cycle of Senior Suicide,” in which chairman Smith called late-life suicide an epidemic.¹ Both Senator Smith and colleague Senator Kohl emphasized the contribution of mental illness, specifically depression, as a preventable cause of suicide among older Americans.^1,2 They both emphasized primary care settings as the logical avenue through which improvements in depression care could lead to reduced rates of suicide in old age. However, if late-life depression in primary care is the target, what characterizes effective interventions, and how effective are they? What follows is a review of studies which evaluated efforts to reduce suicidality through improved depression care in primary practice settings.

Introduction

Since the introduction of selective serotonin reuptake inhibitors (SSRIs), antidepressant prescriptions increased by 400%, but suicidal deaths declined by only 3%.³ The substitution of SSRIs, which are rarely lethal in overdose, for tricyclic antidepressants, which are highly lethal, may be responsible for the decline. Although antidepressant prescriptions provided by primary care physicians (PCPs) have increased, the number of psychotherapy visits has declined. Moreover, the largest increase in antidepressants has been among young adults, children, and adolescents.⁴ Suicide rates have declined among older Americans, but because of the increasing size of the aged population, the actual number of suicidal deaths in old age has not.⁵ If depression is the major cause of suicide in old age, simply increasing antidepressant prescriptions may not be sufficient to sustain the decline in suicide rates.

Depression Care management

A more aggressive approach to depression has emerged, incorporating principles of chronic disease management similar to that employed to improve care and reduce hospitalizations among people with diabetes or congestive heart failure. In primary care models of chronic disease management, nursing personnel collaborate with physicians and patients to improve adherence to evidence-based care. Specialist consultation from cardiologists or endocrinologists is available, but the majority of clinical interaction occurs between patient and nurse. For depression, the collaboration integrates a depression care manager (DCM), typically nurses with added psychiatric training, psychiatric social workers, or psychologists. Psychiatric consultation is available, but again, the majority of clinical interaction occurs between the patient and the Manager rather than with physicians. Comprehensive geriatric evaluation management clinics also employ interdisciplinary teams, including nurses, social workers, and psychologists, to provide chronic disease management including depression care.

Skultety and Zeiss⁶ found both comprehensive geriatric evaluation and the integrated models more effective than routine care in reducing depression in late life. However because the integrated models were designed specifically for depression and were more likely to focus interventions on severe depression, the models and their outcomes were difficult to compare. Gilbody and colleagues⁷ compared costs and benefits of physician education, collaborative care, and case management in studies of depression in primary practices. The authors found uniformly superior outcomes for collaborative care and case management but no benefit from education alone. However, incremental improvements in depression care were associated with increments in costs. In summary, a collaborative, interdisciplinary approach to depression in primary care settings is cost effective but does not reduce the total costs of care. Two of the eight studies reviewed by Skultety and Zeiss⁶ also attempted to reduce suicidality and are examined in greater detail.

PROSPECT

The Prevention of Suicide in Primary Care Elderly: Collaborative Trial (PROSPECT) integrated a DCM into primary care sites.⁸ Details of the methodology appear in Table 1. Through a screening process, patients with clinically significant levels of depressive symptoms were assessed by telephone and received an in-person interview with the Scale for Suicidal Ideation. Twelve percent of those screened, or >500 people, expressed a significant level of depressive symptoms. Two thirds of the sample exhibited major depressive disorder (MDD). PROSPECT defined subjects with a score >0 on the Scale for Suicidal Ideation as having suicidal ideas. As a result, the sample included people with “weak desire to live today” as expressing suicidal ideation. This threshold was set purposefully low in light of the finding by Brown and colleagues⁹ that psychiatric patients =55 years of age who scored =1 on the Scale for Suicidal Ideation exhibited a 15.5-fold increased risk of suicide compared to those who scored 0.

Among PROSPECT patients who reported suicidal ideation at baseline, 66.7% of intervention patients were free of suicidal ideation by 4 months compared to 58.7% of those receiving routine care (P=.34). By 8 months, the difference between intervention and routine care groups was both more substantial and significant (70.7% versus 43.9%, P=.005). However, by 12 months, the two groups were nearly identical (68.7% versus 65.8%, P=.89). Stated differently, after 12 months suicidal ideas among those receiving the intervention had declined 12.9% points (from 29.4% to 16.5%) compared to a decline of 3.0% points (from 20.1% to 17.1%) among the treatment as usual group (P=.01). Among the 202 patients with minor depression, intervention was superior to routine care only for the 23 with suicidal ideation (P=.03). Regarding the depression response (omnibus trend P=.003) and remission rates (omnibus trend P=.001), the patients receiving the intervention experienced statistically superior outcomes over 1 year of observation. However, only 54.8% met criteria for remission. Thus, nearly one third of depressed patients continued to express suicidal ideation and slightly more than half experienced a remission of depression.

IMPACT Study

In the Improving Mood: Promoting Access to Collaborative Treatment for Depression in Primary Care (IMPACT) study which, like PROSPECT, integrated a DCM, 13.3% of 1,801 depressed primary care patients randomized to routine care and 15.9% in the intervention group expressed passive or active thoughts of suicide. Details of the IMPACT intervention appear in Table 2. Patients in the intervention group exhibited significantly lower rates of active suicidal ideation at 6 months (7.5% versus 12.1%) and at 12 months (9.8% versus 15.5%) following baseline. Even after the DCMs were withdrawn, the beneficial effects remained significant for the intervention group at 18 months (8.0% versus 13.3%) and 24 months (10.1% versus 13.9%). The intervention was also significantly more effective than routine care in reducing passive thoughts of death at 6 months (27% versus 38%), 13 months (32% versus 51%), 18 months (38% versus 50%), and 24 months (41% versus 50%). In addition, among patients who received an emergent evaluation because they could not assure the DCMs or non-clinician outcomes evaluators that they would not act on suicidal impulses, significantly more were in the routine care than intervention group (7.7% versus 4.3%).¹⁰

In summary, both PROSPECT and IMPACT integrated specially trained mental health personnel into primary care settings. Both studies monitored adherence to antidepressant guidelines, providing advice to physicians and patients when deviations occurred. Psychotherapy and psychiatric consultation were readily available. However, both depressive symptoms and suicidal ideas remained in a sizeable minority of patients in both studies despite their interventions. These studies demonstrate the extent to which suicidality may be reduced when the effort is limited to sites of primary care only.

Reducing Suicidality Beyond Primary Care

Expanding the effort to reduce suicidality with a more broadly deployed approach to depression would be expected to provide added benefits. Paykel and colleagues’¹¹ “Defeat Depression Campaign” included pubic information efforts to reduce stigma and promote treatment as well as updating physicians on advances in diagnosis and treatment of depression. Suicide rates declined by 11.7%. A similar public education effort in Australia titled “beyondblue” resulted in an increase in public awareness.¹² Suicide rates were reduced and antidepressant prescriptions increased in the Island of Gotland through an intensive educational program for PCPs.¹³ However, none of these programs employed a comparison population such that the observed effects may have been due to events not associated with the intervention.

The Nuremberg alliance against depression (NAD) to reduce suicidality study used a broadly based multi-level intervention to compare suicidal deaths and suicide attempts over 2 years between Nuremberg and Wurzburg. As shown in Table 3, four levels of a comprehensive community and clinical intervention were directed at the population of Nuremberg while Wurzburg received observation only. The four levels of intervention demonstrate an effort to saturate the environment with a social marketing campaign to recognize and reduce depression and thereby suicidality. The campaign was targeted at PCPs and patients and their families, but also incorporated an array of communications media as well as “facilitators” to serve as potential case sentinels and informal community advisers for people who might be depressed. At both 12 and 24 months following the intervention, rates of suicide declined in both cities, although the reductions were statistically significant in neither. However, the decline in suicide attempts was significantly and substantially greater from baseline to 12 months (approximately 18.3%) and baseline to 24 months (approximately 26.5%) in Nuremberg but not in Wurzburg. The reductions in attempts were most noticeable for highly lethal means (shooting, hanging, and jumping) and among persons <70 years of age.¹⁴ Obviously, the advantage of a large-scale population-based study is its capacity to count suicidal behaviors and not just depressive symptoms. However, its size and design precluded the inclusion of DCMs.

Similar to the NAD, the Oregon Older Adult Suicide Prevention Plan seeks the broadest possible attack on the elements of suicidal risk in late life.¹⁵ Table 4 is a reduced list of the strategic objectives for Oregon’s clinical, community-based, and public health interventions. The introduction of integrated models of depression care into primary practice is but one element. Yet, the evidence supporting the integration of DCMs is among the strongest, and methods to sustain their presence in primary care beyond the research-funded initiatives of PROSPECT and IMPACT are critical.

Paying for the Model

The challenge of paying for the DCM arises largely from lack of adequate fee-for-service Medicare procedure codes and other financial incentives within capitated systems of care. However, reflecting on his experience with the IMPACT study, Jürgen Unützer, director of the IMPACT Coordinating Center, noted that “under Medicare Advantage there is actually a very strong incentive to document depression diagnosis and treatment, and a well run depression care management program can facilitate this and lead to significantly higher capitated payments for depressed patients. This has been a major motivator for one of the largest Medicare Advantage plans (offered by Kaiser Permanente) to roll an adapted version of the IMPACT model out to over 10 large regional medical centers in Southern California and they currently have over 20,000 patients in depression care management.” (Personal communication, November 20, 2006.)

Within fee-for-service settings, Medicare reimburses services such as psychiatric diagnostic interview, follow-up medication management or psychotherapy with medication management, and psychiatric consultation. However, psychotherapy without medication management is subject to 50% co-pay, which is a substantial disincentive to many patients unless they also have a medigap policy. Unützer added that “a master’s level clinical nurse specialist who bills largely fee-for-service insurance for IMPACT care delivered in two Duke [University Medical Center] affiliated primary care clinics, covers the majority of her salary with billings which include some direct billing to Medicare and some ‘incident to physician’ billing.” (Personal communication, November 20, 2006.)

However, DCM activities not covered by Medicare include communications other than those which occur directly, face-to-face with the patient. These include telephone monitoring of patient needs as well as telephone consultation with the patient’s PCP or psychiatrist. Thus, failure to reimburse for telephone communication is a barrier to widespread integration of DCMs into primary care settings. The failure is all the more problematic given findings from the Re-Engineering Systems for Primary Care Treatment of Depression study which employed a DCM whose sole contact with patients was via the telephone. Although not focused on older adults, the telephone intervention in the context of psychiatric oversight and specifically prepared PCPs was associated with a maintenance of minimal suicide risk among patients with dysthymia or MDD.¹⁶

Both the President’s New Freedom Commission on Mental Health¹⁷ and the National Business Group on Health¹⁸ have made strong recommendations that public (including federal) and private payers reimburse such evidence-based services. Bachman and colleagues¹⁹ have proposed a number of funding mechanisms and find the prospects of improved funding optimistic. Concerns for the economic impact of parent care on younger adults in the work force add to the incentive for improved geriatric mental health care.²⁰

Other Issues

There are additional issues relevant to detection and reduction of suicide risk among older adults. First is the controversy over the extent to which passive thoughts of death as assessed in PROPSECT and IMPACT predict risk of active suicidal ideation.²¹ In a study of patients with recurrent depressive disorder and either passive or active suicidal ideation, Szanto and colleagues²² found depressed patients with passive suicidal thoughts were as pessimistic about the future as those with active thoughts but denied intent to harm themselves. As treatment continued, active suicidal thoughts became passive before disappearing entirely, suggesting a continuum of risk and the utility of questions for passive thoughts of death.

The second issue relevant to depression and suicidality in primary care is chronic pain. In a recent review of pain and suicide, Tang and Crane²³ noted that the risk of suicidal death appeared to be at least doubled in chronic pain patients. The prevalence of suicidal ideation was 20% among people with chronic pain, with 5% to 14% reporting a prior suicide attempt. Eight risk factors were associated with suicidality and chronic pain, including the type, intensity, and duration of pain, and sleep-onset insomnia. Helplessness and hopelessness about pain, the desire for escape from pain, and catastrophic self-assessments were also associated. Coping by avoidance and deficits in problem-solving skills were prominent psychological processes relevant to the understanding of suicidality in chronic pain.

Interventions to ameliorate the cognitive risk factors associated with depression, pain, and suicidality have been developed. To site one example, Brown and colleagues²⁴ randomized 120 patients following a suicide attempt to case management or cognitive therapy plus case management. The cognitive therapy incorporated preventative techniques specific to the suicidal impulse (Table 5). After 18 months, participants in the cognitive therapy arm were 50% less likely to have re-attempted suicide and expressed less depression and hopelessness as well. Nonetheless, there were no significant differences in rates of suicidal ideation between the two groups. Moreover, 13 (24%) patients in cognitive therapy and 23 (41%) in case management made at least one subsequent suicide attempt.

Conclusion

The technology of depression care has advanced substantially with important implications for the reduction of suicidality in late life. The choice of medications singly, in combination, and in sequence,²⁵ modifications of psychotherapy,²⁴ and the integration of mental health services within primary care provide an evidence base upon which effective policy may be forged.¹⁹ Both public and private concerns have mobilized in recognition of the need. However, in order to continue the decline in late-life suicide rates, a national strategy similar to the Oregon Plan or the Nuremberg Alliance may be necessary. A social marketing campaign coupled with the financing of integrated models of depression care seems the next logical step. With the science in hand, anecdotes like Senator Smith’s public discussion of his son’s suicide or Senator Reid’s disclosure of his father’s suicide²⁶ may be necessary to provide the impetus for change.

Disclosure: Dr. Kennedy has received research support or honoraria from AstraZeneca, Eli Lilly, Forest, Janssen, and Pfizer.

References

1. Statement of Chairman Gordon H. Smith, US Senate Special Committee on Aging. A generation at Risk: Breaking the Cycle of Senior Suicide; Washington, DC; September 14, 2006. Available at: http://aging.senate.gov/public/_files/hr164gs.pdf. Accessed November 30, 2006.

2. Statement of Senator Herb Kohl, US Senate Special Committee on Aging Hearing. A generation at Risk: Breaking the Cycle of Senior Suicide; Washington, DC; September 14, 2006. Available at: http://aging.senate.gov/public/_files/hr164hk.pdf. Accessed November 30, 2006.

3. Mann JJ, Apter A, Bertolote J, et al. Suicide prevention strategies: a systematic review. JAMA. 2005;294(16):2064-2074.

4. McKeown RE, Cuffe SP, Schulz RM. US suicide rates by age group, 1970-2002: an examination of recent trends. Am J Public Health. 2006;96(10):1744-1751.

5. Kennedy GJ. Will more antidepressants mean fewer suicides in late life? Primary Psychiatry. 2005;12(1):26-29.

6. Skultety KM, Zeiss A. The treatment of depression in older adults in the primary care setting: an evidence-based review. Health Psychol. 2006;25(6):665-674.

7. Gilbody S, Bower P, Whitty P. Costs and consequences of enhanced care for depression; Systematic review of randomized economic evaluations. Br J Psychiatry. 2006;189:297-308.

8. Bruce ML, Ten Have TR, Reynolds CF 3rd, et al. Reducing suicidal ideation and depressive symptoms in depressed older primary care patients: a randomized controlled trial. JAMA. 2004;291(9):1081-1091.

9. Brown GK, Beck AT, Steer RA, Grisham JR. Risk factors for suicide in psychiatric outpatients: a 20-year prospective study. J Consult Clin Psychol. 2000;68(3):371-377.

10. Unützer J, Tang L, Oishi S, et al. Reducing suicidal ideation in depressed older primary care patients. J Am Geriatric Soc. 2006;54(10):1550-1556.

11. Paykel ES, Tylee A, Wright A, Priest RG, Rix S, Hart D. The Defeat Depression Campaign: psychiatry in the public arena. Am J Psychiatry. 1997;154(6 suppl):59-65.

12. Jorm AF, Christensen H, Griffiths KM. The impact of beyondblue: the national depression initiative on the Australian public’s recognition of depression and beliefs about treatment. Aust N Z J Psychiatry. 2005;39(4):248-254.

13. Rutz W, von Knorring L, Walinder J, Wistedt B. Effect of an educational program for general practitioners on Gotland on the pattern of prescription of psychotropic drugs. Acta Psychaitr Scand. 1990;82(6):399-403.

14. Hegerl U, Althaus D, Schmidtke A, Niklewski G. The alliance against depression: 2-year evaluation of a community-based intervention to reduce suicidality. Psychol Med. 2006;36(9):1225-1233.

15. Oregon Older Adult Suicide Prevention Plan. Available at: http://egov.oregon.gov/DHS/ph/ipe/esp/docs/plan.pdf. Accessed November 29, 2006.

16. Schulberg HC, Lee PW, Bruce ML, et al. Suicidal ideation and risk levels among primary care patients with uncomplicated depression. Ann Fam Med. 2005;3(6):523-528.

17. President’s New Freedom Commission on Mental Health. Available at: www.mentalhealthcommission.gov/reports/reports.htm. Accessed November 29, 2006.

18. National Business Group on Health. Available at: www.businessgrouphealth.org/prevention/depression.cfm. Accessed November 29, 2006.

19. Bachman J, Pincus HA, Houtsinger JK, Unützer J. Funding mechanisms for depression care management: opportunities and challenges. Gen Hosp Psychiatry. 2006;28(4):278-288.

20. MetLife Mature Market Institute. Caregiving Cost Study. Available at: www.metlife.com/WPSAssets/17239064071161116019V1FCaregiverCostStudy7.11.06.pdf. Accessed November 30, 2006.

21. Raue PJ, Brown EL, Meyers BS, Schulberg HC, Bruce ML. Does every allusion to possible suicide require the same response? J Fam Pract. 2006;55(7):605-612.

22. Szanto K, Reynolds CF, Frank E, et al. Suicide in elderly depressed patients: Is active vs. passive suicidal ideation a clinically valid distinction? Am J Geriatr Psychiatry. 1996;4(3):197-207.

23. Tang NK, Crane C. Suicidality in chronic pain: a review of the prevalence, risk factors and psychological links. Psychol Med. 2006;36(5):575-586.

24. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. 2005;294(5):563-570.

25. Kennedy GJ. The Sequenced Treatment Alternatives to Relieve Depression studies: how applicable are the results for older adults? Primary Psychiatry. 2006;13(11):33-36.

26. Congressional Press Releases. Reid calls for a national suicide prevention strategy. May 6, 1997. FDCHeMedia, Inc. No. 97-095.