[Skip to Content]
[Skip to Content Landing]

Transitions of CareImprove Care and Reduce Costs

CME and/or other credits for this activity are no longer available.
How will this module help me use the SafeMed model in my practice?

  1. Four STEPS to develop and implement the SafeMed model.

  2. Answers to common questions about SafeMed.

  3. Tools to develop training materials and track progress.


Quiz Ref IDComplex patients, such as those with multiple chronic conditions, polypharmacy, and unmet social needs, are particularly at risk for serious drug therapy problems during transitions from hospital to home. The SafeMed care transitions model uses intensive medication reconciliation and home assessments to manage high-risk/high-needs patients in the primary care practice who tend to be heavy utilizers of inpatient and emergency services. The SafeMed model leads to decreased hospital and emergency department utilization, and improved outcomes for the target patient population.

The SafeMed model1 was originally developed at the University of Tennessee/Methodist Le Bonheur Healthcare and funded by a Health Care Innovation Award from the Centers for Medicare and Medicaid Services (CMS). The model was designed with the strengths of primary care in mind and relies on primary care-based team members, including physicians, pharmacists, nurses, and community health workers, to form a support network for high-risk/high-needs patients as they transition from the hospital to outpatient setting. These team members work closely with patients to forge strong relationships that make it easier to coordinate and manage their care. This module provides guidance, tools, and resources at no cost so you can adapt the SafeMed model for your practice.

Four STEPS to building a SafeMed care transitions team

  1. Develop your care transitions plan.

  2. Identify complex patients who are candidates for the program.

  3. Assemble and train your SafeMed team.

  4. Start the transition process and refine the plan over time.

Step 1 Develop your care transitions plan.

Start holding regular team meetings to plan for the implementation of SafeMed. Think about how to scale the SafeMed model to fit your practice's needs and to best take care of the patients who will benefit from your new program. You may want to consider working with multiple practices in your area to pool resources and create a SafeMed team as a shared community resource.

When you have a good idea of your practice's capabilities and how you intend to work with a partner, review and think about how to modify SafeMed job descriptions, protocols, and procedures to meet your target patients' needs.

Box Section Ref ID


Step 2 Identify complex patients who are candidates for the program.

Identify the most vulnerable patients who will receive the greatest benefit from intervention by the SafeMed team. If you have a smaller practice or only a few high-risk/high-needs patients, consider partnering with another practice to increase the size of your target population.

Example: Mr. S. was a 58-year-old Caucasian man with multiple chronic conditions and a history of depression and cocaine use. With the intensive support of all members of the SafeMed care transitions team at University of Tennessee/Methodist Le Bonheur Healthcare, he was able to get assistance with medications and homemaker services, meet his self-identified health goals, develop positive relationships with his providers, and subsequently avoid rehospitalization.

Box Section Ref ID


  • Which patients should my SafeMed program target?

    Quiz Ref IDA SafeMed approach is best suited for patients with:

    • High utilization, defined as two or more hospital admissions or one inpatient admission and two or more prior emergency department (ED) visits in the last six months prior to current admission.

    • Two or more of the following conditions: hypertension, diabetes, coronary artery disease, congestive heart failure, asthma, and chronic obstructive pulmonary disease

    • Polypharmacy, defined as six or more medications at one time or high-risk medications such as diabetes medications, pain medications, and blood thinners.

    These characteristics and criteria can be adapted to fit your patient population, and to include additional patients you deem most at risk, such as patients with documented medication nonadherence or those identified by health plans as being in the top five percent of predicted risk for high cost.

  • Are there any patients who aren't appropriate for SafeMed?

    The SafeMed model was designed to have maximum impact for a very specific group of patients who have complex health conditions and frequently visit the hospital. You may want to exclude patients for whom care transitions programs are not likely to be effective, such as those with:

    • Hospitalization related to cancer, pregnancy, or surgical procedure.

    • Severe mental illness, for example, recent psychosis or suicidal ideation.

    • Severe substance abuse disorder.

    • Life expectancy less than six months.

    • Homelessness or unstable residence.

    • Planned discharge to another location other than home, unless the patient will be discharged to a skilled nursing facility or rehabilitation center.

    • Severe cognitive difficulties and lack of caregiver assistance in program participation.

    Risk stratification can help you determine if these patients may be more appropriate for similar but less intensive services, for palliative, hospice or home care or for other community resources and programs.

Step 3 Assemble and train your SafeMed team.

Quiz Ref IDQuiz Ref IDSelect a leader to champion the effort, designate team leads, and hire any additional staff you need to make your care transitions plan work. The typical SafeMed team consists of three team leaders: a physician, nurse, and pharmacist, plus two community health workers (CHWs), one pharmacy technician, and one licensed practical nurse (LPN), medical assistant (MA), or health coach.

The SafeMed team will work with the rest of the primary care practice team and local pharmacies to help patients effectively manage their conditions and medications; therefore, team members should be knowledgeable about practice workflows and chronic disease symptoms, signs, medications, and treatment. All SafeMed team members should receive training in motivational interviewing, patient advocacy, transitions of care, and mental health issues.

The resources below can help you implement SafeMed within your practice as you develop your transitional care program.

Box Section Ref ID


Step 4 Start the transition process and refine the plan over time.

Quiz Ref IDUse your electronic health record (EHR) to identify high-risk/high-needs patients who meet your practice's criteria so that they can be flagged for the transitions team immediately in the event of an ED visit or hospital admission. Running a daily report will help you to identify eligible patients in real-time and begin the transition process.

Key performance metrics will indicate the impact of the care transitions team, and highlight areas for further development or improvement. Use your results to make rapid cycle or longer-term improvements to the process.

Box Section Ref ID


  • How do I inform patients that they qualify for transitional support services? What do I do if patients are resistant to the new approach?

    It is important to understand what type of transitional care services patients qualify for through their health plan. Some insurers may have their own recommended transitional care programs.

    You should not assume that patients will immediately sign up for the SafeMed program. The care transitions team must clearly explain the services and benefit to the patient. SafeMed experience in Memphis suggests that building rapport and enrolling patients while they are in the hospital is the most effective approach. It is easier to recruit patients when their primary care physician (PCP) supports and encourages their participation; the care transitions team will be able to introduce themselves as representatives of the patient's regular doctor. Use the modifiable recruitment materials in the module toolkit to help you enroll patients.

  • How long should patients participate in this program?

    The patients you are targeting will require intensive engagement early on to make a difference and should be followed closely for a minimum 45-day intensive period. After the first 45 days, invite patients to participate for an additional three months (for a total of four and a half months). This will help make sure that they receive the maximum benefit from the program. Selected patients can be followed on an ongoing basis as needed or re-enrolled if they are readmitted or revisit the ED.

  • What key performance indicators could we track?

    You may opt to monitor patients eligible for and participating in the program before and after implementation to track delivery of the following key program services:

    • Recruitment rate (percent of eligible patients who agree to participate).

    • Telephone follow-up rate within 30 days of discharge.

    • Home visit completion rate within 72 hours and within 30 days of discharge.

    • Primary care follow-up rate within 30 days of discharge.

    • Medication adherence for essential chronic disease medications.

    • Satisfaction of enrolled patients with the program. In addition, key utilization outcome measures can be tracked, including:

      • Emergency care utilization.

      • Overall hospitalizations.

      • Readmissions.

  • How long should home visits last and how often should they take place?

    The length of a home visit will vary based on the needs of the particular patient, but in general, home visits should not last longer than one hour. The chart below details the core hospital and home visit activities included in the SafeMed care transitions model.


The SafeMed model draws on existing primary care strengths and targets high-cost, high-risk/high-needs patients with a focus on medication and chronic disease management. By using proven methodology and team-based care approaches, practices can use SafeMed to simultaneously improve care and patient health while reducing costs.

Box Section Ref ID

AMA Pearls

Communication is essential

Interdisciplinary teamwork and patient-centered communication are the keys to good primary care practice-based transitional care.

Your outreach workers are the backbone of the SafeMed team

Hire outreach workers who excel in patient-centered communication.

Involve patients early and often

Engage your patients early on in continuous program improvement efforts so you can benefit from the honest perspectives of the people you intend to serve.

AMA Steps Forward logo

AMA STEPS Forward® presents actionable, practical toolkits and customizable resources that you can use to successfully implement meaningful and transformative change in your practice or organization. See How it Works

Article Information

About the Professional Satisfaction and Practice Sustainability Group: The AMA Professional Satisfaction and Practice Sustainability group has been tasked with developing and promoting innovative strategies that create sustainable practices. Leveraging findings from the 2013 AMA/RAND Health study, “Factors affecting physician professional satisfaction and their implications for patient care, health systems and health policy,” and other research sources, the group developed a series of practice transformation strategies. Each has the potential to reduce or eliminate inefficiency in broader office-based physician practices and improve health outcomes, increase operational productivity and reduce health care costs.

Disclosure Statement: Unless noted, all individuals in control of content reported no relevant financial relationships.

SafeMed Program.  (2015).  Retrieved from http://www.safemed.org
Coleman,  E.A. (2003).  Falling through the cracks: Challenges and opportunities for improving transitional care for persons with continuous complex care needs.  Journal of the American Geriatrics Society, 51(4), 549–555. doi: 10.1046/j.1532-5415.2003.51185.xGoogle ScholarCrossref
Coleman,  E.A., Parry,  C., Chalmers,  S., Min,  S.J. (2006)  The care transitions intervention: Results of a randomized controlled trial.  Journal of the American Medical Association Internal Medicine, 166(17),1822–1828. doi: 10.1001/archinte.166.17.1822Google Scholar
Makaryus,  A.N., Friedman,  EA. (2005).  Patients' understanding of their treatment plans and diagnosis at discharge.  Mayo Clinical Proceedings, 80(8), 991–994, doi: 10.4065/80.8.991Google ScholarCrossref
Kripalani,  S., LeFevre,  F., Phillips,  C.O., Williams,  M.V., Basaviah,  P., & Baker,  D.W. (2007).  Deficits in communication and information transfer between hospital-based and primary care physicians: Implications for patient safety and continuity of care.  Journal of the American Medical Association, 297(8), 831–841. doi: 10.1001/jama.297.8.831Google ScholarCrossref
Bell,  C.M., Brener,  S.S., & Gunraj,  N. (2011).  Association of ICU or hospital admission with unintentional discontinuation of medications for chronic diseases.  Journal of the American Medical Association, 306(8), 840–847. doi: 10.1001/jama.2011.1206Google Scholar
Tache,  S.V., Sonnichsen,  A., & Ashcroft,  D.M. (2011).  Prevalence of adverse drug events in ambulatory care:Aa systematic review.  Annals of Pharmacotherapy, 45(7-8), 977–989. doi: 10.1345/aph.1P627.Google ScholarCrossref
Ernst,  F.R., & Grizzle,  A.J. (2001).  Drug-related morbidity and mortality: Updating the cost-of-illness model.  Journal of the American Pharmaceutical Association, 41(2), 192 – 199. doi: 10.1016/S0003-0465(15)33027-5Google Scholar
Patient-Centered Primary Care Collaborative. (2015).  Project SafeMed – University of Tennessee Health Science Center.  Retrieved from https://www.pcpcc.org/initiative/project-safemed-university-tennessee-health-science-center
Center for Medicare and Medicaid Services. (2015).  Health Care Innovation Awards: Tennessee.  Retrieved from https://innovation.cms.gov/initiatives/Health-Care-Innovation-Awards/Tennessee.html
Surbhi,  S., Munshi,  K.D., Bell,  P.C., & Bailey,  J.E. (2016).  Drug therapy problems and medication discrepancies during care transitions in super-utilizers.  Journal of the American Pharmacists Association, 56(6), 633-642.e1. doi: 10.1016/j.japh.2016.07.004.Google Scholar
Harris,  L.J., Graetz,  I., Podila,  P.S.B., Wan,  J., Waters,  T.M., & Bailey,  J.E. (2015).  Characteristics of hospital and emergency care super-utilizers with multiple chronic conditions.  Journal of Emergency Medicine, 50(4), e203–e214. doi: 10.1016/j.jemermed.2015.09.002Google ScholarCrossref
Bailey,  J.E., Surbhi,  S., Bell,  P.C., Jones,  A.M., Rashed,  S., & Ugwueke,  M.O. (2016).  SafeMed: utilizing pharmacy technicians in a novel role as community health workers to improve transitions of care.  Journal of the American Pharmacists Association, 56(1), 73–81. doi: 10.1016/j.japh.2015.11.011Google ScholarCrossref
Nault,  C., Binkley,  B.L., Graff,  J.C., Surbhi,  S., & Bailey,  J.E. (2018).  How patient experience informed the SafeMed program: Lessons learned during a health care innovation award initiative to improve care for super-utilizers.  Healthcare: The Journal of Delivery Science and Innovation, 7(1), doi: 10.1016/j.hjdsi.2018.02.002.Google Scholar
Bailey,  J.E., Surbhi,  S., Wan,  J.Y., Munshi,  K.D., Waters,  T.M., Binkley,  B.L., Ugwueke,  M.O., & Graetz,  I. Effect of intensive interdisciplinary transitional care for high-need, high-cost patients on quality, outcomes, and costs: A quasi-experimental study.  Journal of General Internal Medicine. [Accepted for publication March5 , 2019.]Google Scholar
Surbhi,  S., Graetz,  I., Wan,  J.Y., Gatwood,  J., & Bailey,  J.E. (2018).  The effect of opioid use and mental illness on chronic disease medication adherence in super-utilizers.  Journal of Managed Care and Specialty Pharmacy, 24(3), 198–207. doi: 10.18553/jmcp.2018.24.3.198Google ScholarCrossref

Disclaimer: AMA STEPS Forward® content is provided for informational purposes only, is believed to be current and accurate at the time of posting, and is not intended as, and should not be construed to be, legal, financial, medical, or consulting advice. Physicians and other users should seek competent legal, financial, medical, and consulting advice. AMA STEPS Forward® content provides information on commercial products, processes, and services for informational purposes only. The AMA does not endorse or recommend any commercial products, processes, or services and mention of the same in AMA STEPS Forward® content is not an endorsement or recommendation. The AMA hereby disclaims all express and implied warranties of any kind related to any third-party content or offering. The AMA expressly disclaims all liability for damages of any kind arising out of use, reference to, or reliance on AMA STEPS Forward® content.


Name Your Search

Save Search

Lookup An Activity


My Saved Searches

You currently have no searches saved.


My Saved Courses

You currently have no courses saved.