insights_logo.GIF (5351 bytes)
committed.GIF (1460 bytes)




Current Events

BrdGov.GIF (512 bytes)

Awardss & Prizes

Press Releases







Managed Care and Women's Mental Health:
A Focus on Depression

December 1997 - Number Five

By Julianna S. Gonen, PhD

Mental health is essential both in its own right and in terms of how it affects physical health. Insurance coverage, however, has traditionally either neglected mental health needs or covered them at levels significantly lower than medical and other health needs, relegating mental health treatment to the status of a "step-child of medical plans."1 The shift to managed care, with its emphasis on comprehensive care and its structural innovations in care delivery, invites us to examine how these new systems will affect women's receipt of quality mental health care. We begin with a review of how mental health services have been approached and structured generally in managed care, and then move to focus specifically on depression, an ailment with a disproportionate effect on women.

Mental Health Benefits in Managed Care Plans

Under managed care, mental health services are now commonly combined with substance abuse services in a category called "behavioral health care." In this era of cost containment, the emphasis of managed behavioral health care has shifted from a primary emphasis on the underlying clinical illness, which generally meant longer-term intensive treatment, to a primary focus on improved functioning. This is a marked departure from traditional practice and has engendered intense concern among providers and advocates for the mentally ill. As with other health services, from preventive to acute care, providers under managed care are being asked to alter their philosophy of treatment from a purely individual-focused approach, as was possible in an era of third-party payment without management, to a more population-based approach, which requires an awareness of limited resources and the need for population-oriented programs to reduce overall demand.2

The Structure of Benefits and Services.   Some managed care organizations (MCOs) integrate behavioral health into overall health care services and include behavioral health providers within their networks. Others contract out to stand-alone behavioral health companies, "carving out" behavioral health services from general medical care. These behavioral health carve-outs have become firmly established in the market. In the past decade, increases in enrollment in these carve-outs have substantially exceeded those in general managed care. Even many indemnity insurance plans use managed behavioral health carve-outs. As of 1997, 149 million insured Americans with mental health coverage, in both indemnity and managed care plans, received their behavioral health care through carve-outs.3 Only about 15 percent of HMOs provide behavioral health services in-house.4 There are approximately 300 carve-out companies, but the 10 largest currently enroll nearly 80 percent of the covered lives in behavioral health carve-out programs.3

Whether providing these services within the regular MCO or carving them out to specialized networks is better for enrollees is a matter of debate, as we lack outcomes data that would enable assessments of differences in quality of care. Carve-out proponents assert that their specialized focus on behavioral health leads to higher quality and better management, while critics argue that this merely perpetuates the division between mental and physical health that has traditionally characterized the health care system and even served to stigmatize mental health services and those who require them.4 In some instances employers contract directly with mental health vendors, bypassing the general health plan altogether. Often enrollees access behavioral health services by calling the carve-out company directly, and the primary care provider is not involved in the process as would be the case when referring a patient for other specialty services. Although this may help protect the confidentiality of enrollees who would prefer that their general care provider not know that they are receiving mental health or substance abuse treatment, it runs counter to the goal of care coordination that is the major reason why MCOs use primary care providers. This lack of coordination may have significant consequences, particularly if an enrollee receives medications from a behavioral health provider that could interact with other prescriptions being managed by other providers.

Even without carving out mental health services, the constraints of managed care networks may pose problems. One recent study of primary care physicians' communication with mental health providers in fee-for-service and managed care plans found that those in managed care plans reported referring to unknown mental health providers, basing referral on factors such as location, asking patients to select their own mental health providers, and communicating with the mental health providers less frequently. The primary care physicians rated the quality of mental health providers, appropriateness of care, and amount of physician input into mental health care higher in fee-for-service than in managed care. The study authors concluded that mental health carve-outs could possibly worsen these findings.5

Much of the impetus for incorporating mental health services better into health benefits has come from health care purchasers. In areas with aggressive large employers and employer coalitions, more innovative forms of managed mental health care have emerged, as employers increasingly recognize the importance of mental health services in maintaining a healthy and productive work force.4 In many ways purchasers are ahead of the health plans themselves in recognizing the comorbidities and lost work days due to mental health disorders. Health care buyers often possess the clout, through their financial leverage, to exact changes from managed care companies that consumers and even physicians cannot.

Medicaid State Medicaid agencies have been at the forefront in the implementation of managed behavioral health programs for beneficiaries, and an increasing number of states are requiring Medicaid recipients to use managed behavioral health programs. Some (California, North Carolina, Massachusetts) use mental health carve-outs, while others (Missouri) include mental health in their comprehensive managed care programs. Since the Medicaid population as a whole tends to use more health care services, the advent of these new systems can impose additional stresses on an already vulnerable population, and in some cases cut them off from appropriate counseling, prescriptions, and providers.6 States have struggled with who should provide managed mental health services to Medicaid beneficiaries--the public agencies that have historically served this population, or private, for-profit behavioral health companies. "Treating the behavioral health needs of the Medicaid population is vastly different from caring for the privately insured, and some experts doubt the ability of most private, for-profit behavioral health MCOs to provide all the services required with quality and cost-effectiveness."7 On the other hand, state mental health authorities may have valuable lessons to learn from the mechanisms for managing care used by these private-sector organizations, such as contracting, utilization review, and monitoring.8 There have been reports of problems under Medicaid behavioral managed care programs, including slow payment of claims, denials of needed services, use of inadequate provider networks, use of drug therapy alone without counseling, and use of older, less effective pharmaceuticals. Rhode Island has gone as far as to discipline its Medicaid mental health vendor for denying care to enrollees.

The news is not all negative. Done properly, "managed care plans can expand access to mental health services for Medicaid patients by emphasizing increased use of preventive care and reducing costly and unnecessary hospitalizations."6 In Massachusetts, mental health expenditures were significantly reduced upon adoption of its managed care program, without apparent reductions in access or quality.9 States have and will continue to look to external indicators of health plan quality (see below) to determine which organizations should receive state contracts to provide behavioral health services to Medicaid beneficiaries.

Monitoring Quality.   The National Committee for Quality Assurance (NCQA), the primary accrediting entity for managed care, has been accrediting HMOs since 1991 under an evolving set of standards. With the increasing emphasis on behavioral health and the explosion of enrollment in specialized behavioral health plans, NCQA has developed special accreditation standards for managed behavioral health care organizations (MBHOs). Part of the impetus for the creation of these standards was a concern over the accountability of carved-out services that MCOs delegate to outside vendors. The new standards can apply to services provided within a general HMO or to behavioral health carve-out plans, and took effect in January 1997.10 Previously MBHOs could only be accredited by NCQA as a delegated entity under its regular MCO accreditation program. Although the regular MCO program does evaluate a health plan's mental health services, the review was not as rigorous as it is under the new mental health-specific standards. NCQA's goal is to incorporate the MBHO standards into its regular accreditation program, so that eventually even health plans that provide behavioral health services within their plan structure will be held to these more extensive standards. Among the major goals of the MBHO accreditation program are the coordination of behavioral health with medical care, the implementation of population-based quality improvement systems, and an emphasis on preventive behavioral health care. The new standards address the split between mental and physical health by requiring evidence that individual practitioners and providers are coordinating a patient's behavioral and medical care across delivery systems, and by encouraging the development of systematic data sharing to facilitate MBHO-MCO coordination, but in a manner that assures confidentiality. As of November 1997, seven specialty behavioral health plans have come forward to undergo review under the new standards, six for a "practice review" and one for an actual accreditation decision.

NCQA also promulgates the most widely used set of MCO performance measures, the Health Plan Employer Data and Information Set, or HEDIS. The current iteration of HEDIS contains 10 broad measures related to mental health, out of 71 total measures. Those with particular relevance for women and depression in managed care include availability of mental health/chemical dependency providers, and utilization of ambulatory services. Among the measures being pilot tested for possible future inclusion in HEDIS are continuation of depression treatment, appropriate use of psychotherapeutic medications, patient satisfaction with mental health care, and use of behavioral health services. Although there is a measure of screening for chemical dependency in the testing set, there is no measure of screening for depression.

Several competitors to NCQA have emerged in the past couple of years to challenge its hegemony in accreditation and performance measurement. The Foundation for Accountability (FAcct) is developing performance measures that are based more on outcomes than on process, in part as a response to a common criticism of HEDIS measures. FAcct selected six initial areas in which to focus its development of measures, including Major Depressive Disorder (MDD).11 This area includes several discrete measures, including experience of disease, remittance, follow-up/continuity of care, patient satisfaction, and lost work time and disability days due to MDD. The American Managed Behavioral Healthcare Association (AMBHA), the trade association for managed behavioral health care companies, has also developed a set of performance measures (PERMS) to assess the overall performance of managed behavioral health care delivery systems. Like NCQA's behavioral health accreditation standards, the PERMS measures are intended for use by both carve-outs and full-service MCOs.12

Public Policy: The Push for ParityOrganizations such as AMBHA and the National Alliance for the Mentally Ill (NAMI) have been advocating parity of coverage for mental health and medical services by insurance companies. Some degree of parity was achieved legislatively at the federal level in 1996, but the legislation does not require health plans to offer mental health benefits, and those that already do could in fact opt to drop mental health coverage rather than comply with the law's limited parity requirements. What the Mental Health Parity Act of 1996 does do is forbid health plans, if they establish annual payment limits or aggregate dollar lifetime caps, from having more restrictive limits or caps for mental illnesses than for physical illnesses. Since most plans do not have these annual and lifetime limits for other diagnoses, the equalization mandate is effectively a mandate to eliminate such limits on mental health diagnoses. The law stipulates, however, that this mandate shall not apply to group health plans whose costs increase by 1 percent or more as a result of this requirement. The mandate, which applies only to employer groups with 50 or more employees and does not include substance abuse services, becomes effective on January 1, 1998, and sunsets on September 30, 2001. Many states are considering or have passed similar legislation that would go further to mandate parity in copayments, deductibles, frequency of coverage (number of visits/days), and prescription drugs as well. These laws are not pre-empted by the more narrow federal law.

Women and Depression in Managed Care

Epidemiology and Comorbidities. Women face a different set of mental health risks than men. Overall rates of mental illness are approximately the same for men and women, but women suffer more from depression and anxiety disorders while men experience higher rates of substance and alcohol abuse and antisocial personality disorders.13 Major depression is the most common severe mental disorder among women, currently affecting approximately 7 million American women. The rates of diagnosed major depression among women are 57 per 1,000, and this rises to 100 per 1,000 if dysthymia, a milder but chronic form of depression, is included.13 Depression occurs most frequently in women aged 25 to 44, but there is some evidence that the peak onset age is decreasing, reflected in increased suicides and suicide attempts among young adults. Women have consistently demonstrated higher risk rates for most types of depression than men; the ratio on average is 2:1.14 This gender difference persists when other factors such as race, age, and socioeconomic differences are controlled for.15

A number of possible explanations exist for the significant gender disparity in the incidence of depression. One is that while women and men experience approximately the same number of negative life events, women tend to find them more stressful than men, perhaps in part because they are often burdened not only with their own care and well-being but with that of others as well. Specific potential explanations for gender differences in depression rates are generally grouped under biologic factors, reproductive-related events, or psychosocial factors.15

The evidence for a link between biologic factors and depression is indeterminate, but there is at least some evidence of depressed individuals sharing some abnormalities on biologic measures, and the postpartum period is often associated with an increase in depression. There is also some evidence of a genetic component, a possible link to endocrinologic changes that occur during women's childbearing years, and a vulnerability to and expression of depression related to short- and long-term effects of certain hormones.16 The various biologic factors being explored as possibly related to women's depression include the menstrual cycle, childbirth, menopause, infertility, legal induced abortions, and surgery and illness.15 Many of these factors also have important psychosocial components, which makes discerning the link to depression both more critical and more complicated.

Certain psychologic factors such as perceived control over one's life, social support, sense of personal accomplishment, and sense of independence have been correlated with depression. Other theories look more broadly at limits on women's roles in society and at women's social conditioning against assertiveness.15, 17 Violence, both physical and sexual, is also an enormous burden with both physical and psychological consequences. Some demographic factors, including younger age, less education and income, and unemployment and low socioeconomic status have also been found to be related to depression, albeit to a lesser extent. And while little is known about the direct effect of ethnic differences, to the extent that women from ethnic or racial minority groups are more likely to have many of the socioeconomic risk factors for depression, as well as bearing the additional burdens of discrimination, at least an indirect link may exist.15

Adolescent girls are also found to have higher rates of depression and depressive symptoms, including suicidal ideation, than adolescent boys. Factors correlated with depression in teenage girls include lack of parental support, low self-esteem (exacerbated by a cultural emphasis on thinness), low levels of attachment to family and peers, socialization against competence and assertiveness, and sexual abuse.15 Adolescent girls who are lesbian, bisexual, or even questioning their sexuality are also at particular risk for depression and suicide due to both internalized and societal homophobia. At the other end of the life span, much of the gender disparity in the incidence of depression disappears, but this is due to increases among men rather than decreases among women. Two of the most important risk factors for depression in older women are loss of a spouse or partner and loss of physical health; others include isolation, economic concerns, and role changes.

While depression causes significant morbidity alone, it also has important implications for other conditions in its sufferers. Those with mental disorders tend to use more general medical services and, because depression tends to modify the course of other illnesses, they also tend to have poorer health outcomes. This has implications not only for the well-being of individual patients, but also for the resultant costs to health plans, and casts considerable doubt on the common MCO practice of only covering limited mental health services for "acute" situations. Conversely, many patients with other disease states also suffer concurrently from depression, which often remains unrecognized and undetected.18, 19, 20

Detection and DiagnosisDespite the high morbidity associated with depression, the majority of episodes go undetected and therefore untreated, in large part because our health care system is based on a biomedical model. One study of 1,000 patients with depression found that those enrolled in managed care were properly diagnosed only 40 percent of the time, but those in fee-for-service did not fare much better, at 50 percent.21 In part because of the traditional division between physical and mental health and the stigma associated with obtaining mental health treatment, the majority of mental disorders are diagnosed and treated within the general medical sector,22 which can compromise the quality of care received.20 A paper commissioned by the Commonwealth Fund concluded that much of the care women receive for psychological problems takes place outside the hospital and even outside the professional mental health system, much of it being provided by general care physicians. These facts suggest the importance of focusing on primary care in order to improve the level of care women receive for depression, anxiety, and other frequently seen disorders. . . . Primary care providers have not been well trained to accept this role or provide this care, and this may pose both a problem and an opportunity as managed care expands even further the role of the primary care provider as the individual who will diagnose, treat, and refer women with symptoms of mental illness.23

There is clearly a need for mental health diagnosis and treatment (of uncomplicated cases) to be integrated into primary care with as much emphasis as screening, diagnosis, and initial treatment of medical conditions. The need for this integration engenders a role for both the medical education system and for MCOs as they engage in ongoing education of their providers. MCOs can improve the integration of mental health into women's primary care by providing in-service training for their primary care providers, by requiring continuing medical education credits in women's mental health, and by including appropriate women's health experts on practice guidelines committees.24

Because of the important role of primary care providers in detecting mental disorders such as depression, new guidelines have been promulgated to improve detection and diagnosis in the non-mental health sector. The federal Agency for Health Care Policy and Research (AHCPR) published its two-part guideline for Depression in Primary Care in 1993,25 but except for noting in the abstract that "female gender" is a major risk factor for depression, the guideline overall pays scant attention to gender differences in the causes and treatment of depression. Also in response to the recognition of primary care providers' role, the major primary care professional societies collaborated to develop a special version of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM) for primary care providers.26 PCPs had found the regular version too long, complicated and exhaustive, and sought a streamlined version for the primary care setting. While the introduction to the volume indicates that each section contains epidemiologic information, including sex ratios, for the disorder in question, in the section on depression there is no mention of the disproportionate prevalence in women.

Attention to mental health needs is critical in the arena of primary reproductive care as well. Many women consider their ob-gyns to be their primary care providers during their childbearing years,27 providing an impetus for these physicians to enhance their mental health diagnostic and even treatment skills.28 Ob-gyns must also contend with the relationship of premenstrual syndrome to other affective disorders, the risks involved in taking psychotropic medications during pregnancy, and distinguishing transient postpartum depression from underlying dysthymia or major depression.

Primary care competency in mental health has begun to attract research attention. A team from the Georgetown University Department of Psychiatry is studying several interventions with primary care physicians in an IPA-model HMO (most prior work has been done in staff-model plans in which monitoring and modifying physician practice is more feasible), including mailing guidelines, academic detailing, and the use of a treatment facilitator, aimed at improving the detection and management of depression in the primary care setting. Patients were surveyed to assess their attitudes about their primary care physician's ability to care for their depression. A third of respondents had received care from their primary care physician for mental health in the past 6 months, and while 70 percent felt comfortable discussing mental health issues with their physicians, only 39 percent felt their physician could manage patients with depression.29

Because they have begun to realize the high costs and morbidity associated with mental illness, some MCOs have begun to design and implement programs to enhance the detection and management of depression and other mental health conditions by their primary care clinicians. Group Health Cooperative of Puget Sound conducted two randomized trials of collaborative care models in one of the HMO's primary care clinics to improve the process and outcomes of care provided to depressed patients. The studies involved patient and provider educational and self-help materials, physician training, structured visits involving the patient and physician, case-by-case consultations with the depression specialist (psychiatrist or psychologist), surveillance of treatment adherence, and reminder phone calls to participating patients. Both studies improved satisfaction with care and adherence to treatment recommendations in patients with major and minor depression, and the two trials resulted in more favorable depressive outcomes in patients with major depression.30 Because further assessment by physicians is usually required to make a diagnosis after screening and diagnostic tests, Kaiser Permanente-Colorado has pilot tested a self-report tool called the Quick PsychoDiagnostics Panel, to provide severity scores and diagnoses without additional physician assessment in seven categories of mental disorders, including depression. Initial results indicate high patient acceptance and good convergent validity with the Zung depression scale.31 And Kaiser's Northern California plan is piloting a new model of treatment and follow-up for depressed patients newly treated with Prozac, Zoloft, or Paxil (three leading antidepressants) in a primary care setting. The new model of care includes physician training, telephone monitoring and support by clinic nurses, and peer support.32

TreatmentDepression can be treated through medication or psychotherapy, or both. The success rate for drug treatment of depression is between 60 and 80 percent, with success defined as some improvement. Medications alone, however, are usually not sufficient for effective treatment of depression, and for mild depression, the use of cognitive therapy first is recommended over medication. This is an area in which managed care practices have raised some concern. While managed care has helped to increase the number of Americans with mental health coverage, MCOs tend to favor drug therapy over psychotherapy,22 as even more expensive medications are less costly than sessions of therapy with a mental health professional.20 This omits half of the equation, as the combination of drugs and counseling have been shown to be most effective in the treatment of depression, as well as in other disorders that tend to disproportionately affect women such as eating disorders, panic attacks, and obsessive-compulsive disorders.33

Even with drug therapy, managed care plans are often not using the most effective drugs. The introduction of selective serotonin reuptake inhibitors (SSRIs) has revolutionized the biomedical treatment of depression. Many health plan formularies, however, include only the older, less expensive tricyclic antidepressants, despite evidence that the SSRIs are more effective in women and cause fewer side effects.19 These practices are counter even to the MCOs' own bottom-line interests of cutting costs, as patients taking tricyclics tend to require more physician visits, hospitalizations, and lab tests than those taking SSRIs.22, 34 Although the appropriate use of efficacious antidepressants has not been found to be high in either managed care or fee-for-service settings, one study has revealed that enrollees in managed care plans were twice as likely as those in fee-for-service to be prescribed inappropriate mild tranquilizers.21 An important and often overlooked additional factor for women is the interaction of antidepressant medications and exogenous hormones such as birth control pills. Such interactions require careful attention to dosage and side effects, and if an MCO has a restrictive formulary that covers only certain antidepressants, it is difficult for a prescribing physician to work out an appropriate regimen for each individual patient.

Managed care has brought into the open the question of who treats patients with mental disorders once they have been diagnosed. In the past there was a tendency for the most highly trained mental health professionals to treat the least difficult cases, due to a professional preference for individual psychotherapy and the increasing availability of reimbursement for such sessions. There was intense competition among providers for outpatient psychotherapy patients with the insurance or ability to pay for care in private offices, and fees charged were based on professional credentials rather than the nature or complexity of the treatment (i.e., psychiatrists commanded the highest reimbursements, followed by doctoral-level psychologists, social workers, and so forth.). Managed behavioral health care has challenged this by linking both referrals and compensation to the complexity of the work instead of to the provider's credentials.2 Thus social workers and psychologists provide most of the uncomplicated outpatient psychotherapy in these networks, with psychiatrists used primarily for comprehensive evaluations, consultation on biologic treatments, management of medications, and treatment of complex or difficult cases.2, 35 This reallocation of professional resources may in fact be an improvement if the match of providers to patients and their conditions is appropriate and not merely in the interest of cutting costs. Some studies, however, indicate that rather than employing a more efficient mix of providers, managed care plans are simply reducing women's access to mental health specialists.36

Outstanding Issues and Questions

Mental health needs are now beginning to gain the attention and respect they have long been lacking, with movements for coverage parity and to remove longstanding stigmas associated with mental disorders. Managed care may hold the potential to improve the detection and treatment of depression in women through better training of primary care providers, but the use of specialized carve-outs for the provision of behavioral health care may thwart the needed integration of mental health care into general medical care. Given the state of women's treatment for depression within managed care as outlined above, the following issues remain for exploration:

  • What is the effect of carving out behavioral health services on the quality of care women receive for depression in managed care?
  • Are there existing models of MCOs that have successfully instituted increased training for primary care providers in the detection and diagnosis of depression?
  • What is the impact of restrictive pharmaceutical formularies on the availability of antidepressant medications for MCO enrollees?
  • Which treatments for depression are the most cost effective for women?
  • What are the effects on women's health outcomes of managed care's emphasis on primary over specialty care, and on pharmacotherapy over psychotherapy?


1. Davis K. Patients v. The Bottom Line--HMOs and Mental Health. Briefing Note, The Commonwealth Fund, October 1996.

2. Shore MF and Beigel A. Sounding board--the challenges posed by managed behavioral health care. New Eng J Med 334:116;1996.

3. American Managed Behavioral Healthcare Association data.

4. Rovner J. Managing mental health: in-house or carve-out. In Managed Care Strategies 1997. New York, Faulkner and Gray, 1996.

5. Randel L. Physician communication with mental health providers in managed care: opinions of primary care physicians. Abstract, 1997 AAHP Research Conference.

6. Health Advocate (Newsletter of the National Health Law Program), No. 184 (Jan/Feb/Mar 1996).

7. Bell N. The Medicaid population: behavioral managed care's perilous new frontier. In 1997 Behavioral Managed Care Sourcebook. New York: Faulkner & Gray, 1996.

8. Essock SM and Goldman HH. States' embrace of managed mental health care. Health Affairs 14:3;1995.

9. Callahan et al. Mental health/substance abuse treatment in managed care: the Massachusetts Medicaid experience. Health Affairs 14:3;1995.

10. National Committee for Quality Assurance. 1997 Surveyor Guidelines for the Accreditation of Managed Behavioral Healthcare Organizations. Washington, DC: NCQA, 1997.

11. Foundation for Accountability. In Practice. Portland, OR: FAcct, 1996.

12. "PERMS 2.0 set for November release; tool will float test HMO indicators." Managed Behavioral Health News 3:36 (October 9, 1997).

13. Glied S and Kofman S. Women and Mental Health: Issues for Health Care Reform. The Commonwealth Fund, March 1995.

14. Robins LN, Regier DA (eds.) Psychiatric Disorders in America: the Epidemiologic Catchment Area Study. New York: Free Press, 1991; Kessler RC et al. Sex and depression in the national comorbidity survey. 1. Lifetime prevalence, chronicity and recurrence. J Affect Disord 29:85;1993.

15. Horton JA, ed. The Women's Health Data Book--A Profile of Women's Health in the United States (2nd ed.). Washington, DC: Jacobs Institute of Women's Health, 1995.

16. Halbriech U and Lumley LA. Journal of Affective Disorders (November 1993).

17. Ruble DN. Journal of Affective Disorders (November 1993).

18. Simpson SG. Depression in primary care: co-occurrence with medical illnesses. American Psychological Association conference, "Psychosocial and behavioral factors in women's health: research, prevention, treatment, and service delivery in clinical and community settings." September 19-21, 1996, Washington, DC.

19. Lewis-Hall FC et al. Fluoxetine vs. tricyclic antidepressants in women with major depressive disorder. J Women's Health 6:3;1997.

20. Sturm R and Wells K. How can care for depression become more cost-effective? JAMA 273:1;1995.

21. Donald-Sherbourne, Sturm, & Wells. Issues related to detection, treatment and outcomes of depression from the Medical Outcomes Study, 1996.

22. McFarland BH. Cost-effectiveness considerations for managed care systems: treating depression in primary care. Am J Med 97(suppl) 6a-47s (1994).

23. Leiman JM. Foreword. In Glied S and Kofman S. Women and Mental Health: Issues for Health Care Reform. The Commonwealth Fund, March 1995.

24. Hoffman E. Presentation at American Psychological Association conference, "Psychosocial and behavioral factors in women's health: research, prevention, treatment, and service delivery in clinical and community settings." September 19-21, 1996, Washington, DC; Meyer RE and Sotsky SM. Managed care and the role and training of psychiatrists. Health Affairs 14:3;1995.

25. Rush AJ, Golden WE, Hall GW, et al. Depression in Primary Care: Volume 1. Detection and Diagnosis, Clinical Practice Guideline, 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. 93-0550. April 1993.

26. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edition--Primary Care Version (DSM-IV-PC). Washington, DC:APA,1995.

27. The Commonwealth Fund's 1993 Survey of Women's Health found that a third of women use both a generalist primary care physician and an ob-gyn; an additional 16% used only ob-gyns as their regular source of care.

28. Association of Professors of Gynecology and Obstetrics. Depressive disorders in women: diagnosis, treatment, and monitoring. APGO Educational Series on Women's Health Issues. APGO, 1997; American College of Obstetricians and Gynecologists. Guidelines for women's health care. Washington, DC: ACOG, 1996; Stotland NL. Psychological Aspects of Women's Health Care--the Interface Between Psychiatry and Obstetrics and Gynecology. Washington, DC: American Psychiatric Press, 1993.

29. Gonzales J et al. Depression in primary care: challenges in recruiting, retaining, and enhancing participation by primary care physicians in an IPA. Abstract, AAHP 1997 Research Conference.

30. Bush T et al. Depression in primary care: description and results of two randomized trials of a collaborative model of care. Abstract, HMO Research Network Conference, Boston, 1997.

31. Beck A. Feasibility of an automated psychiatric diagnostic tool for primary care. Abstract, HMO Research Network Conference, Boston, 1997.

32. Hunkeler E et al. Implementing a new model of care for managing depression in the primary care setting of an HMO: physician training, nurse telephone follow-up, peer support. Abstract, HMO Research Network Conference, Boston, 1997.

33. Endicott J, Weissman MM, and Yonkers KA. What's unique about depression in women? Contemp Ob/Gyn (June 1997); Laurence L. Your mental health under managed care. Glamour, October 1996.

34. Selix R. The perils of restrictive drug formularies. Behavioral Health Management 17:4;1997.

35. Oss M. The managed behavioral health care industry: overview and future prospects. In Managed Care Strategies 1997. New York, Faulkner and Gray, 1996.

36. Glied S. The treatment of women with mental health disorders under managed care and fee-for-service insurance. Women and Health, 26(2) Forthcoming 1998; Sherbourne CD. "Issues related to detection, treatment, and outcomes of depression for the Medical Outcomes Study: what we learned about women." Commonwealth Fund, October 1996.