Our health care system is broken, fragmented, and costly. Its performance compares poorly with that of other industrialized countries, in part because of policies that favor disproportionate reimbursement for curative care over prevention, hospital over community-based care, number of patient encounters over health outcomes, and denial of care to the sickest over access to care for all.1–3
Physicians may be well positioned to play a major role in challenging such deleterious health policies. They are on the frontline of health care delivery, where they witness the impact of health policy decisions, and they are granted considerable trust and respect in their communities. However, most physicians are ill prepared to affect health policy, and their voice in health policy debates is muted.4 The socioecological factors that may exacerbate the medical conditions of marginalized patients seem insurmountable to many physicians. For many, changing health policy is a task remote from their busy daily medical practice, or it is a task best left to their professional organizations or to other sectors of society. Also, physicians may shy away from investing time in such tasks that have little immediate impact or that place them outside their professional comfort zone.
This attitude is reinforced during medical education. Medical students and residents receive little exposure to health policy in their training, many presume health policy refers only to state and national endeavors beyond their reach, and many presume that changing health policy requires a long-term time commitment beyond the scope of their busy training schedules. Curriculum offerings that could bring health policy to life, such as rotations in community health, are usually relegated to “elective” status or to final-year time slots, leaving little opportunity for them to truly affect trainees' long-term attitudes and actions.5 Finally, hospital wards—the place where residents and clinical medical students spend most of their training time—have not been tapped as settings for public health and health policy learning.
Attempts to remedy the lack of community health focus in medical training have emerged in many forms, including the Medical School Objectives Project,6 the Clinical Prevention and Population Health Curriculum Framework devised by the Healthy People Curriculum Taskforce,7 and an Eight-Part Toolkit devised by Trevena et al8 to help medical schools integrate population health into their curriculums. These recommendations are broad, and each school can adapt portions of them to their own needs and curricular opportunities. Health policy is an aspect of each set of recommendations, but the related recommended focus is usually on knowledge of the subject rather than action.
Recently, an opportunity to introduce health policy in action to students and residents in a hands-on manner presented itself to medical educators at the University of New Mexico School of Medicine. The dean and curriculum committee approved a requirement that all medical students graduate with a public health certificate, beginning with the class matriculating in 2010.9 The certificate is available to medical students at the university, but it has never before been a required component of the school of medicine curriculum. In preparation for this expanded curriculum requirement, interested medical students and residents have piloted each component of the certificate, and some students and residents have completed the full, 16-credit course.
To prepare a larger number of faculty to deliver the certificate courses to all medical students, a group of school of medicine faculty, residents, and students took the certificate's Principles of Public Health course to see how such a traditional public health course might be adapted to medical school learners. Aspects of health policy in that course motivated the faculty to apply the principles of health policy instruction in the customary venue of medical student and resident education—the wards. Whereas health policy is a broad topic encompassing laws, regulations, governance, and measured outcomes related to the health of the public, we chose to focus on hospital policy as an introductory exercise to the topic because it was easily accessible to our clinical learners. We sought to answer the question of whether hospital wards, the dominant learning setting for residents and medical students, could serve as an effective venue for learning to influence public health policy in general and hospital policy in particular. In this article, we describe the methods and outcomes of this pilot.
Integrating Hospital Health Policy into the Clinical Curriculum
Those who participated in the pilot were family medicine residents and medical students rotating on the family medicine inpatient service for four-week blocks during the 2007 calendar year at The University of New Mexico Hospital. Two attendings supervised this pilot, both of whom attended the Principles of Public Health course. One (S.B.) taught a modified version of this course to all first-year residents, and the other (A.K.) implemented health policy aspects of the course on ward rounds. The participating family medicine residents and students had no direct exposure to the original course. The patients were primarily indigent, about half were Hispanic or Native American, and approximately a third carried diagnoses of alcohol and substance abuse. Many were homeless.
During ward rounds, the learners were asked three questions after presenting their patients: (1) How could this admission have been prevented? (2) What might explain the fact that many in the community have the same health problem but are not receiving care at all? and (3) What hospital or health system policies facilitate or hinder appropriate diagnosis and treatment of patients with this condition? These questions focused learners' attention on the potentially policy-related aspects of their patient cases.
The team of learners was asked to select, from among the policy-related problems raised by these ward encounters, those they would like to address. Problems ultimately selected for action shared several characteristics: they motivated residents or students to intervene; they had policy implications; they could be addressed within the limited time frame of a resident's or student's schedule; and necessary faculty assistance was readily available. It is important to recognize that learners were not limited in their topic choice—any policy-related problem that presented itself was open for intervention. This gave learners the freedom to effect change in the best ways they were able, and it also allowed them to tackle problems that they were able to address within a limited time period. This selectiveness can be seen as a limitation to the pilot, but it is in keeping with the program's intent to provide learners an introduction to the broad topic of health policy through the more limited scope of hospital policy.
Once the team chose a problem for intervention, participating team members were encouraged to gather relevant background data about the policy-related problem in question. This process invariably included online literature searches done on ward computers, analysis of patient data housed in particular hospital departments, and interviews with relevant hospital administrators, providers, and other personnel. For resident participants, all relevant hospital departments were notified of the residents' policy projects at project initiation, and the residents' solicitation of their department's participation was encouraged as an important step in obtaining financial, patient-care, and personnel data and in understanding patient-care policies. Hospital personnel were courteous and helpful toward achieving departmental participation in the data-gathering process. Once the team had gathered sufficient background data about their identified problem, they collaborated to propose an action plan, discuss it with the faculty, and then implement it. Updates on the project were discussed on daily rounds so that health policy became a topic integrated into inpatient learning for all members of the ward team. Finally, participants monitored, recorded, and disseminated outcomes of the intervention via one of several channels: e-mail, newsletter, grand rounds, or presentation to hospital administration.
Policy Change in Action
In the first six months of the pilot, there were 10 substantive hospital policy projects and interventions that were implemented by family medicine residents and medical students. Four examples follow.
Medications for discharged homeless patients
Problem. The family medicine inpatient service at the hospital accepts about half the admissions referred by the Health Care for the Homeless Clinic in Albuquerque. However, the rate of readmission of homeless patients discharged from the hospital far exceeded that of other discharged patients.
Data. Interviews with readmitted patients revealed that many of the prescribed discharge medications were either unaffordable or not on the formulary of the Health Care for the Homeless outpatient pharmacy. The family medicine service had no knowledge of the content of that formulary.
Intervention. A resident (J.C.) led her ward team on a “field trip” to the Health Care for the Homeless clinic, meeting with their administrators, providers, and pharmacists. The team discovered the Health Care for the Homeless formulary was online and received permission to access it. They were also informed of the schedule and policies of the clinic in dispensing these medicines.
Outcomes. The Health Care for the Homeless formulary now appears on the resident Web site. It is now hospital policy that all ward teams discharging homeless patients limit their prescriptions to the Health Care for the Homeless pharmacy's formulary, thereby improving adherence to prescribed medication regimens. Residents were also informed of the pharmacy's hours of operation and procedure for procuring medications so that they could better inform discharged patients.
After-hours pharmacy services for uninsured patients
Problem. Two uninsured patients were admitted to the family medicine service with preventable medical conditions (diabetic ketoacidosis in one and chronic obstructive pulmonary disease exacerbation in the second). Because both were county residents eligible for indigent care funding, they could only obtain affordable, reduced-cost medications at the university hospital's pharmacy. However, both found it difficult to fill their prescriptions because the pharmacy had no evening or weekend hours. Both patients worked and felt they couldn't afford to leave work and lose pay to obtain their medications.
Data. A resident (M.D.) and two medical students (D.D., R.A.) reviewed literature on pharmacy access for indigent patients, interviewed the university hospital pharmacy director and staff to understand barriers to service for this population, and visited a community health center in the city of Albuquerque to learn alternative ways of providing after-hours access to pharmacy services for their indigent patients.
Intervention. The resident and students developed a report on their findings for the hospital administration that concluded with a recommendation that the university hospital pharmacy change its policy on pharmacy hours of operation by extending those hours to evenings and/or weekends. They argued that increasing such access would more than pay for itself through a reduction in costly and preventable after-hours emergency room visits for prescription refills and costly and preventable hospitalizations.
Outcome. The university hospital agreed to open its pharmacy every Saturday from 9 am to 5 pm.
Advocacy for expanded social services
Problem. The inpatient ward team was frustrated over the slowness of hospital discharges and consequent pressure from emergency room attendings to move admitted patients, who often were held for days after admission, from the overcrowded emergency room to ward beds. The team blamed the social workers for the ward discharge delays, so a meeting was arranged between the family medicine inpatient service and the hospital's social services department. The goal of this meeting was for each side to understand the challenges that the other side faced when trying to discharge patients. To start, the social service department members explained that, on any given day, they were understaffed by two to five social workers. They also expressed frustration about the hospital's new system for assigning social workers to ward teams that admitted patients to different wards rather than to specific wards. Although this new arrangement made patient care easier for the physicians, it proved to be more time-consuming for the social workers. This new system created conflict and a communication barrier. Finally, the most difficult patients to discharge were those uninsured ones who needed transfer to skilled nursing facilities that refused transfer for financial reasons. These patient cases left the already-shorthanded social work staff to work doubly hard to cajole nursing homes into taking these needy patients.
Data. To gauge the magnitude of this problem, a resident (P.M.) studied what are referred to as patient avoidable days (PADs) during a month's worth of admissions to his team. There were 35 patients in his sample. Of the 311 hospital days generated by these patients, 172 were for bona fide acute care, but 139 (44% of total bed days) were PADs. In this case, the PADs were the days that patients were no longer ill but were boarding while awaiting acceptance by a skilled nursing facility or rehabilitation hospital. The cost to the hospital of the PADs for these 35 patients totaled $147,752. Not one of these patients had private insurance, and most were indigent, so the estimated financial impact on the hospital and on public payers of PADs throughout the hospital system during a year's time is enormous.
Intervention. The resident presented his data to the clinical leadership of the university hospital and medical school. Rather than adding to the pervasive criticism of the overworked and undermanned social services department, he advocated an expansion of resources for social services, arguing that such an expansion would more than justify its cost through reduction of PADs for uninsured patients, increasing bed availability with many of those beds being filled by insured patients.
Outcomes. The social service department was allocated two additional social work positions, and that department held a luncheon at which the family medicine ward team, among others, were recognized for their “above and beyond” support of the social needs of the hospital's inpatients.
Group certification of residents and faculty to prescribe outpatient buprenorphine treatment for opiate-dependent patients
Problem. Approximately 40% of the inpatients on the family medicine service have been diagnosed with alcohol or substance abuse problems, and such abuse is suspected to be an underlying factor in many admissions for other primary diagnoses. The population of New Mexico in general, and of some of its rural counties in particular, suffers some of the highest rates in the nation of intravenous narcotic abuse and consequent hepatitis C and other complications. The success of primary care and hospital discharge referrals to behavioral treatment centers remains low, and the availability of methadone treatment is far below the demand. Better means must be found for primary care clinicians to help patients with this prevalent problem. Yet, although the institution requires that all residents be certified, for example, in cardiopulmonary resuscitation, there is no such training requirement for treating opiate dependency.
Data. Some of the residents and attendings participated in a local conference on the successful use of buprenorphine/naloxone (Suboxone) sublingual treatment prescribed by primary care physicians in the outpatient setting in the community. Few primary care physicians are trained and certified in its use, and at the university's pharmacy Suboxone was only approved for inpatient use. Training to be certified to prescribe buprenorphine, a schedule III partial opiate agonist, consists of an eight-hour training course required by the U.S. Food and Drug Administration for opioid agonist maintenance treatment in office-based practice. One of the residents (V.J.) volunteered to pursue the introduction of this certification for family medicine residents and faculty. While interviewing stakeholders in the community, including public health and addiction specialists, she discovered that some physicians feared that after receiving certification, their practices would quickly be dominated by opiate-dependent patients.
Intervention. The resident proposed group training and certification of family medicine residents and faculty to prescribe buprenorphine. To overcome providers' fears of being overwhelmed in their practices, and to spread the expertise in this type of treatment across the practice, the number of opiate-dependent patients assigned to each resident and faculty member would be limited. Such group training of residents has not been previously reported in the literature. The resident presented the data, met with residents and faculty individually, and worked with addiction specialists to offer two eight-hour certification training opportunities in May 2007. The course was funded by Project ECHO, a University of New Mexico program which uses teleconferencing between specialists and primary care providers to develop community capacity to treat common chronic, complex diseases.10
Outcomes. The resident was successful in registering 28 residents and faculty in the buprenorphine certification course. She was successful in recruiting at least two faculty and more residents from each of the three university-linked, family medicine resident training sites. In addition, she was successful in working with addiction specialists in winning approval from the university hospital's pharmacy and therapeutics committee to place Suboxone on the outpatient formulary at a reduced cost for indigent patients.
Lasting Effects of Hospital Health Policy Education
Another downstream effect of the resident and medical student experiences in health policy has been that training in health policy was incorporated as the centerpiece of the department's graduate and undergraduate Health Resources and Services Administration Title VII health professions education grant submissions in 2007. Both proposals were funded and will enable expansion of these health policy education efforts to all medical students and all family medicine residents at the university, beginning in 2008. Health policy has been incorporated into the public health certificate curriculum planned for all medical students, to be incorporated into the students' required eight-week family medicine clerkship. Impressed with the ability of residents and students to influence hospital policy, department faculty donated $5,000 to a special fund for the family medicine department to facilitate students' and residents' completion of such policy projects. Residents and students will be able to apply for up to $500 per project to support needs such as making posters for presentations and to travel to meetings to present their projects.
Time allocation for the health policy projects was integral to the ward experience, so it did not seem onerous to the residents. Residents traditionally are asked to research and report back the next day on clinical questions that arise on rounds. Similarly, residents were asked to pursue and report back on policy issues. One resident summed up his time commitment to this pilot:
I spent 3.5 hours spread out over 2 weeks to do the data collection (i.e., chart reviews), another two hours of meetings with the medical school's inpatient medical director and the hospital's chief financial officer. I then spent two hours analyzing and organizing the data, and 1.5 hours making the PowerPoint presentation. Total time was 9 hours spread over 6 weeks.
Policy projects are summarized on the department's residency Web site and have been cited as an important attraction for recruits to the residency.11
Hospital wards can be a stimulating venue for introducing health policy. Such an introduction should not be mistaken for a quick fix to a complex learning process, but it should be recognized as an accessible gateway to the process in the midst of a time-demanding learning environment. An approach toward revealing the public health and policy issues underlying the medical problems of admitted patients is required of the attendings and learning teams on the wards. Even when learners are on time-intensive inpatient services, they willingly find time to engage in health policy projects when they are emotionally engaged in the policy problems and actions in question and see the direct connections between policy and patient care.
The case studies from our own experience document the policy changes initiated by learners on the wards of one teaching hospital. However, true health policy should be chosen on the basis of an identified community need and should be created and adapted to meet that need. Although it is an open question whether our pilot experience will lead learners to look beyond hospital policy to wide-reaching health policy that affects the entire community, we were impressed that for at least one of the pilot projects (advocacy for after-hours pharmacy hours), the residents and students were responding to a growing community concern expressed through local advocacy groups. Further, a true measure of whether a policy change has worked is whether it has improved the health or access to care of the patients it was intended to affect. The brevity of our learners' projects obviated an impact evaluation at this level and serves as a limitation of this pilot.
Our recommendations for more immediate assessments of this curriculum include tracking the implementation of policy changes as well as the associated budgetary and personnel allocations within the University of New Mexico Hospital system. Assessment at the learner level is being incorporated into the policy and advocacy curriculum being developed with support from the Health Resources and Services Administration residency training grant. This includes knowledge of and comfort with identifying and addressing policy issues within the context of training as well as the residents' perception of the applicability of their policy training to future practice situations. The individual residents associated with the 10 projects that developed from our pilot can form the core of a focus group addressing questions of feasibility of policy work during inpatient rotations and other training settings. As residents progress through the program and begin practice, they will be surveyed regarding incorporation of policy and advocacy knowledge and skills into their actual practice, and their recommendations for improvement and expansion of the curriculum.
The timing of our approach to health policy education is ideal, because the Accreditation Council for Graduate Medical Education, in conjunction with the Residency Review Committee, now requires that all resident trainees complete a scholarly project during their training. Our experience demonstrates that commonly confronted patient problems seen on the wards can serve as a fertile ground for generating health-policy-related scholarly endeavors addressing this requirement. At a minimum, the program has raised participants' awareness that even on a hospital ward, patient health outcomes are the product of more than just a physician–patient bedside interaction.
Despite the barriers they face, physicians play important roles in community health through community participation, political involvement, and collective advocacy.12 Stephens13 believes that family medicine has an obligation to be a force for needed change in the health care system: “Family practice, in its advocacy for distributive justice in medical care that is humane, merciful, moral, personal, and cost-effective, has a necessary relationship to politics, economics, ethics, and social change.”
Health policy is made more accessible to residents and students if the issues that generate policy discussions are made a part of their daily learning environment, if learners can intervene to improve those policies within a limited time frame, and if faculty mentors are available to guide their interventions.