U.S. Department of Health and Human Services

Physician Practices in Nursing Homes: Final Report

Executive Summary

Cari Levy, M.D., Anne Epstein, Ph.D., Lori-Ann Landry, M.S.W., and Andrea Kramer, M.D.
University of Colorado, Health Sciences Center

Jennie Harvell, M.ED., and Charlene Liggings, M.P.H., M.P.A.
U.S. Department of Health and Human Services

April 4, 2006

---

This report was prepared under contract#HHS-100-03-0028 between U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and the University of Colorado. For additional information about this subject, you can visit the DALTCP home page at http://aspe.hhs.gov/_/office_specific/daltcp.cfm or contact the ASPE Project Officer, Jennie Harvell, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, S.W., Washington, D.C. 20201. Her e-mail address is: Jennie.Harvell@hhs.gov.

The opinions and views expressed in this report are those of the authors. They do not necessarily reflect the views of the Department of Health and Human Services, the contractor or any other funding organization.

---

INTRODUCTION

The Assistant Secretary for Planning and Evaluation (ASPE) contracted with the University of Colorado Health Sciences Center (UCHSC) to examine and describe models of physician practices in nursing homes; the financing arrangements and payment rates associated with these models; and determine what is known about the impact of physician practice models on the quality of care received by nursing home patients and on the quality of care provided in nursing homes.

In the first phase of this project, the UCHSC completed a literature review related to physician practice patterns in nursing homes, including the: requirements related to such practices; extent to which medical schools prepare physicians to practice in nursing homes; level of physician practice and specialization in providing services in nursing homes; barriers to physicians practicing in nursing homes and innovative physician practice arrangements in nursing homes.

The second phase of the study, which is presented in this report, was designed to further explore several issues that were not adequately addressed in the literature. Using information learned during the first phase, discussions were held in nursing homes and with key stakeholders in certain areas of the country to better understand the use and impact of various physician practice arrangements in nursing homes. This report summarizes discussions with stakeholders throughout the country in an attempt to identify obstacles and promising approaches to successful physician practices in nursing homes; synthesizes themes that recurred during the discussions; presents issues for further consideration; and offers potential areas for future research on physician practice in nursing homes.

METHODS

A purposeful sample of eight facilities was selected for case studies of physician practice models. The objective of the purposeful sampling methodology was to select varied examples of physician practice models in a range of markets and facilities. No attempt was made to select a representative sample from which to draw statistical inferences.

Market, Facility and Practice Models of Interest

The literature review identified multiple market, facility and practice model characteristics related to access to and quality of nursing home physician care. Market characteristics included: (1) managed care penetration, (2) rural location, (3) geographic region, and (4) litigious environment. Facility characteristics included: (1) ownership and (2) size. Practice model characteristics included various types of physician specialization in nursing home care including practices that: (1) emphasize geriatrics and teaching, (2) limit physician practice to only nursing home patients, (3) integrate mid-level providers, (4) limiting the number of physicians in the care of patients in one facility (including the use of closed panel physician groups), and (5) include a variety of other physician practice models.

Stakeholders of Interest

Stakeholders targeted for discussions in each nursing home included: (1) a physician, (2) the nursing home administrator, (3) the medical director, (4) the director of nursing, (5) a social worker, and (6) one stakeholder for each facility who was identified by other respondents as an employee who would add valuable information to the discussions because of their unique perspective on physician care in the facility. Once a facility agreed to participate, a representative from each of the vocational disciplines was contacted by e-mail, telephone or fax to arrange individual stakeholder discussions.

Creation and Administration of the Discussion Guides

Discussion guides (Appendix A) were created to assist interviewers gather information from stakeholders regarding physician practices in the selected nursing homes. The discussion guides focused on four areas: (1) qualifications and training of the practicing physicians and medical directors; (2) practice models and patterns used by physicians; (3) access to care associated with specific practice models; and (4) the quality of care provided in the practice models used in the facility. Six versions, one for each vocational discipline, of the discussion guides were created so that each respondent was asked a set of questions relevant to their vocational discipline.

RESULTS

Three themes emerged from the stakeholder discussions regarding physician practice in nursing homes including: (1) physician specialization in nursing home care; (2) benefits of mid-level practitioners; and (3) importance of information transfer.

Physician Specialization in Nursing Home Care

The form of physician specialization in nursing homes varied, but most stakeholders clearly valued models of care that permitted physicians to specialize in the care of nursing home residents. While the stakeholders did not have precise definitions for nursing home specialists, they generally defined physician specialists as those who spend a substantial portion of time in the delivery of nursing home care or have the majority of their patient care caseload in nursing homes.

The incentives to the physicians who specialize in nursing home care were described as: (1) a reduction in overhead expenses associated with maintaining an office practice among those who practiced exclusively in nursing homes; (2) an ability to develop improved knowledge of regulations in nursing homes and effectively meet the needs of the nursing homes; (3) scheduling flexibility; and (4) the opportunity for long-term relationships with patients and families. Stakeholders felt that if medical students had a greater exposure to nursing home care during medical training, more physicians would select this career path. Based on the literature review, early exposure to the benefits of specializing in nursing home care can increase the number of physicians specializing in nursing home care.¹

Benefits cited by stakeholders regarding specialist nursing home physicians included: (1) greater accessibility of physicians to patients, family, and nursing home staff; (2) improved knowledge of and sensitivity by physicians to challenges faced by nursing homes (e.g., regulations regarding use of anti-psychotics); and (3) enhanced medical management of common syndromes faced by nursing home residents (e.g., falls, urinary incontinence, agitated behaviors associated with dementia). The literature review also suggested that selected outcomes are better among patients of physicians specializing in geriatric medicine.¹

Nursing home regulatory requirements are unique and unlike the requirements in other health care arenas. The respondents associated lack of knowledge about these regulations with negative effects on physician communication with the nursing staff, medication ordering practices and compliance with regulatory visits. Several respondents found the American Medical Directors Association (AMDA) certification program a valuable resource in the pursuit of information on various regulatory requirements such as the use of physical and chemical restraints and required intervals for physician visitations. Sources identified in the literature review supported early exposure during training as a strategy to increase physician experience and familiarity with nursing homes.^{2, 3, 4}

The stakeholders’ concept of the best model of specialization differed; some stakeholders favored a mid-level provider-based model (e.g., using nurse practitioners (NPs) as is the practice in the Evercare model), others favored a closed physician practice model and still others preferred a model with physicians practicing independently in multiple facilities. The closed models provided a salary to the physicians in contrast to other models in which physicians obtained payments via billing Medicare Part B for physician services rendered to individual patients. In the literature review, selected outcomes (e.g., response to emergencies, hospitalization rates, satisfaction) were superior in mid-level provider and closed models compared to the traditional independent physician practice model.^{5, 6, 7}

The stakeholders differed in their opinion of how physicians should be compensated for the delivery of specialized nursing home care. Stakeholders who practiced in a model with physicians salaried by the nursing home agreed that this model afforded the physicians more time to devote to the care of the patients, but they recognized that payment to physicians under the salaried model could be less than in a model in which physicians bill Medicare Part B for their services.

Differences in opinion also were evident with regard to the ideal mixture of practice across settings of care. For example, one stakeholder felt it was essential for nursing home physicians to practice in outpatient clinics and hospitals and then follow patients from these settings to the nursing home to provide continuity of care. Others felt an office or hospital practice detracted from the physicians' ability to focus on nursing home patients because the physicians were often unavailable for calls during office hours and tended to see their nursing home patients as a last priority in evenings and on weekends. The non-physician stakeholders consistently reported difficulty communicating with physicians who care for only a very few nursing home residents because of the competing demands of physicians’ clinic and hospital-based practices.

Nursing facility staff members uniformly reported a preference for physician practice models that involve regular and frequent presence in the facility. These physicians were described as having a better understanding of the pressures faced by nursing homes and improved relationships with the nursing staff and interdisciplinary teams. Social workers valued models that allowed the physicians to become nursing home specialists and respond quickly to emergency calls. Staff members, and nursing home residents and their families, uniformly were pleased with working with nursing home physician specialists. No patterns in preferences for specific models of care were observed consistently across facility types.

Although the evidence is limited, there is some literature to support that specializing in the care of patients in one care setting improves quality of care and reduces costs. For example, patients who are cared for by physicians who specialize in the hospital setting, have lower mortality rates and reduced hospital costs.² The literature also suggests that physicians who specialize in nursing home care are on-site at nursing homes more frequently, have quicker response times to emergencies, lower hospitalization rates, and reduced use of medications.^{7, 8}

Stakeholders also alluded to the difficulty researchers may have in determining whether or not specialization results in better care. For example, if one utilizes a higher frequency of lab tests as an indicator of good medical care, there may be factors other than clinical acumen guiding the volume of tests. On one hand, a non-nursing home specialist may order more tests because they are not in the facility to perform an assessment and must instead rely heavily on lab tests to substitute for a clinical examination. Alternatively, a nursing home specialist physician may order more tests than non-nursing home specialists, in part, because he/she are concerned about lawsuits if tests are not ordered to support their clinical assessments. Similarly, one stakeholder commented that more visits from practitioners increased the volume of orders but it is unknown whether or not an increase in orders results in improved care. Thus, future investigations regarding the quality of care provided by specialist versus non-specialist nursing home physicians should use caution in the design of outcome variables.

Increasing Use of and Desire to Work with Mid-Level Practitioners

The literature review and most respondents indicated that mid-level practitioners have the potential to increase the quality of care provided to nursing home residents and provide an important service in nursing home physician practice models. Respondents indicated that operating an efficient, large nursing home medical practice is not possible without the use of mid-level practitioners. Mid-level practitioners were described as enabling physicians to provide more efficient care because of their ability to respond quickly to urgent care needs.

Current Medicare policy permits mid-level practitioners to alternate visits with physicians. However, a concern expressed by several respondents was that physicians relied too heavily on the mid-level practitioners and participated less in the care of patients when a mid-level practitioner was involved.

An additional concern expressed by respondents and reinforced by the literature was that while the overall supply of NPs has increased in the past decade, there are not enough geriatric-trained mid-level practitioners to meet the demand for these practitioners in nursing homes.^{8, 9} Reasons for this shortage are not well understood, but they may parallel the reasons cited in the literature and offered by respondents regarding the lack of geriatric-trained physicians (e.g., minimal exposure to nursing homes during training).

Information Exchange -- Knowledge About Nursing Home Operations

Stakeholders and the literature review suggest the need for enhanced communications and/or information exchange between physicians and nursing home staff, across settings of care and across providers of care (e.g., communications among physicians, nursing homes, and laboratories).¹⁰

Issues and requirements that, in some instances, are unique to the nursing home setting and in other instances common across the health care continuum characterize medical service delivery in nursing homes. Persons treated in nursing homes often are severely functionally and cognitively impaired and/or medically complex, requiring intervention by interdisciplinary teams with substantial family/informal caregiver involvement.^{11, 12, 13} The literature provides evidence that transitions to and from nursing home care are a common occurrence and a major source of medical errors in relation to medication administration, advanced care directives, allergies, and delivery of essential services.¹⁴ An AMDA survey of 3,000 sequential admissions to skilled nursing facilities from 25 different hospitals found the following: 22% of transfers had no formal summary of information; legible summaries were available only 56% of the time; secondary diagnoses were missing in 30% of transfers; test results were omitted in 31%-67% of transfers, advance directives and code status were absent in 81% of transfers; and a legible phone number for the transferring physician was present in only 33% of transfers.¹⁵

The stakeholders also expressed similar concerns with respect to transitions from hospitals to nursing homes. They noted that transferring accurate medical information from one care setting to another is time-consuming and often inaccurate. This is because physicians often are not caring for patients across care settings, and there is no standardized process or standard set of information transferred across care settings to communicate health information. As observed by the stakeholders, nursing homes often are separated geographically from hospitals, diagnostic services, and physician offices, creating communication barriers that contribute to medical errors.

A common theme that emerged from discussions with stakeholders was the need for physicians to receive accurate and complete information regarding nursing home patients. Information transfer between hospital and nursing home stays were described as cumbersome and wrought with inaccurate and incomplete transfer of information. In most cases, the nursing facilities did not have computerized health information beyond the software used to record information for the Minimum Data Set. All stakeholders agreed that electronic medical records would vastly improve quality of care by reducing the inefficiencies involved with duplicating documentation that has occurred in other care settings and tracking down information.

ISSUES REQUIRING FURTHER CONSIDERATION AND FUTURE RESEARCH

The case presentations (Appendix B) provide examples of facilities with varied physician practice models and the issues related to each of these models. Several of these models offer promising approaches for improving physician care in nursing homes. This information could be useful when considering ways to promote effective and efficient medical service delivery to nursing home residents, to researchers in designing future studies examining physician practices in nursing homes, and to nursing homes and physicians in designing and implementing effective efficient physician practice models in nursing facilities.

Issues Requiring Further Consideration

Several overarching issues that arise from the literature review and stakeholder discussions that merit further consideration include availability of mid-level providers in long-term care (LTC). The literature review and stakeholder informants consistently suggested that mid-level providers are useful in extending medical services to nursing home residents and in promoting higher quality medical care to nursing home residents. However, these sources also discussed the limited supply of these mid-level providers. Other issues included physician training in geriatrics and nursing home care; and the need for incentives for enhancing physician care through pay-for-performance.

Availability of Mid-Level Providers in Long-Term Care

One strategy that has demonstrated success in attracting advanced practice nurses to LTC is the teaching nursing home concept in which schools of nursing establish affiliations with LTC facilities providing an advanced practice degree program for nurses with an emphasis in LTC. Funding specialty and advanced practice internship and residency programs for post degree recipients may also increase the availability of advanced practice nurses to nursing home care and help to provide a career ladder for nurses interested in positions of greater responsibility.

Physician Training in Geriatrics and Nursing Home Care

Strategies to stimulate physician practice in nursing homes may include modifications of the Medicare Graduate Medical Education program to provide direct and indirect funds to teaching hospitals in exchanges for placing an emphasis on training in nursing home care and/or geriatrics. Another approach may be to modify Medicare physician payment rules/methods to encourage physicians to provide services in nursing homes. Public Health Service training grants may also be structured in a manner that encourages medical education in geriatrics and nursing home care. Favorable student loan repayment programs for physicians who devote a significant proportion of their practice to nursing home care may also be established. Medical education could also incorporate exposure to nursing home patients for all internal medicine and family practice physician training programs.

The literature indicated few physicians are trained to provide care in nursing homes yet the literature and stakeholder informants indicate that when appropriately trained in the care of nursing home residents, physicians have an ability to effectively manage the frail, elderly population and improve outcomes. Stakeholders suggest that physicians who specialize in nursing home care are more proficient in understanding nursing home regulations and therefore are better able to respond to the medical needs of patients. However, the limited supply of physicians who are adequately trained to care for nursing home patients remains a challenge.

Incentive Payments for Medical Practices in Nursing Homes

Increasingly, purchasers of health care are interested in value-based purchasing. Recently legislation has been enacted to establish various pay-for-performance incentive programs to promote specific outcomes for physician practices. However, pay-for-performance incentive programs do not address clinically relevant outcomes for the frail institutionalized elderly.

Future Research

Currently, research is not available on features identified by stakeholders as important components of nursing home physician practice models. Specifically, the literature does not quantitatively or qualitatively address: (1) the efficacy of nursing home specialization, (2) effective strategies for information exchange between nursing homes and physicians or (3) methods to study physician payment incentives in nursing homes. Study designs are proposed for these topics. Recommendations related to specific research activities that would advance our knowledge related to physician practice in nursing homes include: (1) physician specialization in nursing home care and its impact on quality and costs; (2) health information transfer using technology across providers and settings when caring for nursing home patients; and (3) pay-for-performance incentives for physicians providing nursing home care.

Physician Specialization in Nursing Home Care

To study the effect of physician specialization in nursing homes an investigation could determine the extent to which quality indicators, influenced by physician practice, and avoidable hospitalizations are associated with the degree to which: (1) nursing homes are staffed by physicians who spend the majority of their practice caring for nursing home care patients (facility-level); and (2) a physician's practice is devoted to caring for nursing home residents (physician-level).

Health Information Transfer

Research could identify the issues that support or create barriers for information exchange and strategies that could facilitate information exchange, including electronic information exchange. The investigation would involve a literature review, stakeholder discussions and case studies examining how LTC organizations receive information, what information they receive when they serve patients that are treated by physicians and in hospitals that have the capacity to exchange health information electronically, and the factors that promote or inhibit electronic information exchange. ASPE is funding a study on health information exchange with post-acute and LTC settings.

Pay-for-Performance

A study could be developed under either Section 646 of the Medicare Modernization Act to establish a pay-for-performance demonstration program or, under an expanded demonstration program, to create physician incentive payments to promote high quality medical management by physicians and/or physician practice groups that include physician extenders on behalf of medically fragile nursing home patients. In developing pay-for-performance measures for nursing home care, the unique characteristics of nursing home residents and the environment of the nursing home must be taken into consideration. Such measures will apply specifically to the care of nursing home residents and not to a physician's outpatient population.

A study to pay for physician performance in skilled nursing facilities and nursing homes could test the effects of incentives for physicians (and their mid-level providers) to play a more active role in the care of their skilled nursing facility and nursing home residents. Process performance measures for physician and mid-level provider care in nursing homes could include timeliness of visits, responsiveness to phone calls from nursing home staff, and transfer of necessary information between a physician and nursing home staff. Outcome performance measures would need to target aspects of nursing home care over which the physician has the most influence. For example, use of unnecessary or inappropriate medications, untreated depression and pain, and rates of potentially avoidable hospitalizations might be considered.

Medication data available under the Medicare Part D program could be used to evaluate physician prescribing for nursing home and skilled nursing facility residents and whether unnecessary medications or inappropriate medications were used. Hospitalization for potentially avoidable causes, such as urinary tract infections, respiratory infections, sepsis, wound infections, and conditions such as congestive heart failure where monitoring and early response might avoid the need for hospitalizing a resident, are potential performance measures for physician care. However, physicians are not solely responsible for hospitalization of nursing home residents, and thus these measures should be risk adjusted for resident as well as facility characteristics that research has found to be associated with preventable hospitalization rates (such as nursing home staffing).

Following standard treatment approaches for problems such as pain and depression also can be used as physician performance measures. If such conditions have been identified, treatment is appropriate. However, for some conditions following standard treatment guidelines typically used in outpatient medicine may not be the most appropriate treatment intervention for the elderly nursing home patient.

REFERENCES

1. Schols J, Crebolder H, van Weel C. Nursing home and nursing home physician: The Dutch experience. Long-Term Care Around the Globe: JAMDA, May/June 2004; 207-212.

2. Meltzer D, Manning WG, Morrison J, et al. Effects of physician experience on costs and outcomes on an academic general medicine service: Results of a trial of hospitalists. Annals of Internal Medicine, 2002; 137:866-874.

3. Katz PR, Karuza J, Kolassa J, Hutson A. Medical practice with nursing home residents: Results from the National Physician Professional Activities census. Journal of the American Geriatrics Society, 1997; 45:911-917.

4. Warshaw GA, Bragg EJ, Shaull RW. Geriatric medicine training and practice in the United States at the beginning of the 21st century. New York, NY and Cincinnati, OH: Association of Directors of Geriatric Academic Programs and the Institute for Health Policy and Health Services Research. 2002; 1-113.

5. Kane RL, Flood S, Keckhafer G, Rockwood T. How EverCare nurse practitioners spend their time. Journal of the American Geriatrics Society, 2001; 49:1530-1534.

6. Ackermann RJ, Kemle KA. The effect of a physician assistant on the hospitalization of nursing home residents. Journal of the American Geriatrics Society, 1998; 46:610-614.

7. Karuza J, Katz PR. Physician staffing patterns correlates of nursing home care: Initial inquiry and consideration of policy implications. Journal of the American Geriatrics Society, 1994; 42:787-793.

8. Fama T, Fox PD. Efforts to improve primary care delivery to nursing home residents. Journal of the American Geriatrics Society, 1997; 45:627-632.

9. Bureau of Health Professions, Division of Nursing, National Advisory Council on Nurse Education and Practice. Report to the Secretary of Health and Human Services. Federal Support for the Preparation of the Nurse Practitioner Workforce through Title VIII. 1997; 19-20.

10. Kramer A, Bennett R, Fish R, Lin CT, Floersch N, Conway K, Coleman E, Harvell J, Tuttle M. Case Studies of Electronic Health Records in Post-Acute and Long-Term Care. Available at http://aspe.hhs.gov/daltcp/reports/ehrpaltc.htm. August 18, 2004.

11. Coleman EA, Kramer AM, Kowalsky JC, et al. A comparison of functional outcomes after hip fracture in group/staff HMOs and fee-for-service systems. Eff Clin Pract, 2000; 3:1-11.

12. Kramer AM. Rehabilitation care and outcomes from a patient’s perspective. Med Care, 1997; 35:JS48-JS57.

13. Kramer AM, Kowalsky JC, Lin M, Grigsby J, Hughes R, Steiner JF. Outcome and utilization differences for older persons with stroke in HMO and fee-for-service systems. J Am Geriatr Soc, 2000; 48:726-734.

14. Coleman EA, Min S, Chomiak A, Kramer AM. Posthospital care transitions: patterns, complications, and risk identification. Health Serv Res, 2004; 39:1449-1465.

15. Foley C. Falling Through the Cracks. Symposium at the American Geriatrics Society Annual Conference, Baltimore, MD. http://www.amda.com/caring/february2003/foundation.htm. 2003.

The Full Report is also available from the DALTCP website (http://aspe.hhs.gov/_/office_specific/daltcp.cfm) or directly at http://aspe.hhs.gov/daltcp/reports/2006/phypracfr.htm.