October 2009

Grant Results


In 2007 and 2008, Mercer Human Resource Consulting (now simply Mercer) joined with a local union and a health system in Atlantic City, N.J., to pilot test a new primary care delivery model called the Ambulatory-Intensive Caring Unit. The unit aims to improve care for patients with chronic serious illnesses while lowering health care costs and the cost of health insurance by helping patients insured by the union's health plan better manage their own health conditions.

Key Results

  • The Atlantic City-based model, the Special Care Center, opened in August 2007; by the end of 2008, the center was serving approximately 800 mostly low-income patients with chronic illnesses. Two surveys found a generally positive reception to the center among patients.
  • Project staff produced two reports to assist in the development and evaluation of the Special Care Center:
    • Interviews with prospective Special Care Center patients, documenting their needs.
    • A plan for an evaluation of the center's progress in five key areas.

The Robert Wood Johnson Foundation (RWJF) provided $194,560 for technical assistance and expert consultation to support the development of the Special Care Center.

 See Grant Detail & Contact Information
 Back to the Table of Contents


A major contributor to the rising number of uninsured people in the United States is the high and growing cost of health benefits coverage. The problem is particularly acute for low-wage workers and their employers. In 2006, the average family premium for health insurance exceeded the total amount earned at the federal minimum wage for the first time.

One strategy for lowering costs is to provide better quality care to the segment of health plan enrollees who are at the highest risk—that is, the 20 percent of enrollees who generate approximately 60 percent of a health plan's spending in a given year.

With funding from the California Health Care Foundation, Mercer (formerly Mercer Human Resource Consulting) designed a new primary care model called the Ambulatory-Intensive Caring Unit (A-ICU). The model pairs high-performing clinical teams with high-risk patients—those with chronic illnesses or socioeconomic issues that contribute to high health care usage.

The aim is to prevent higher "downstream" costs related to traditional primary care, specialty care and hospital admissions, by implementing these cost-saving features:

  • "First Floor" care provided by trained nonphysicians. These "health coaches," supervised by A-ICU nurses and physicians, help patients manage their own care.
  • "Second Floor" care provided by physicians, medical assistants and nurse practitioners. The team uses an electronic health record called E-clinicalworks, on-the-spot telephone consultations with specialists and selective in-sourced specialist services to reduce the costs of primary care visits. For example, a behavioral health specialist visits the site weekly to work with patients in need of such services.
  • "Third Floor" specialist referrals. Using data from the partner insurer, the A-ICU selects the most cost-effective and high-quality specialists for referral.

Mercer's analysis suggests that the A-ICU could achieve a net savings of 38 percent over a high-risk enrollee's expected subsequent year's health care costs.

 Back to the Table of Contents


The Pioneer Portfolio supports innovative ideas and projects that may lead to breakthrough improvements in the future of health and health care. The grant described in this report was issued under the Pioneer Portfolio.

 Back to the Table of Contents


Mercer (formerly Mercer Human Resource Consulting), a global human resources organization that provides strategic guidance on health insurance coverage, pilot tested the Ambulatory-Intensive Caring Unit (A-ICU) model in Atlantic City in partnership with a local union and a health system:

  • The union, UNITE HERE, represents predominantly low-wage workers in the area (largely in the casino and gaming industry).
  • Atlanticare is the dominant local health system, serving approximately 30,000 residents.

RWJF funding supported the Mercer/Atlanticare/UNITE HERE partnership to obtain expert advice on:

  • The design and evaluation of the clinic.
  • The best use of the community health workers.
  • The selection criteria and effective recruitment strategies to reach out to the patients most in need of the Special Care Center's services.

RWJF funding also supported:

  • The union's participation in a national collaborative of providers and payers at multiple sites working to adapt, build and pilot test the A-ICU model.
  • An on-site physician with expertise in the A-ICU model.

 Back to the Table of Contents


Mercer (formerly Mercer Human Resource Consulting) reported these results:

  • Project staff conducted home visits in March and April of 2007 with over 20 members of the union who were likely candidates for the Special Care Clinic. The interviews yielded important information about the patients' interest in a Special Care Center, their current experiences of care, their need for and access to specialty services and how the Special Care Center might meet the needs of these patients. See Findings for details of their responses.
  • Mercer and its partners opened the Special Care Center in August 2007; by the end of 2008, the center was caring for some 800 patients. The center adapted the Ambulatory-Intensive Caring Unit (A-ICU) model to fit the particular needs of the patients and the community. For example, it did not fully implement "Third Floor" specialist referrals as part of the cost-savings plan because of concerns about physician resistance and wariness about competition between Atlanticare and the center.

    Instead, the Special Care Center networked with hospitalists from Atlanticare's two hospitals to identify informally the higher-quality, more cost-effective specialists. Mercer hired an analysis contractor, who by the end of 2009 will identify appropriate specialists more formally.

    See the Appendix for more details about the design of the Special Care Center.
  • Two surveys found a generally positive reception to the center among patients who received care:
    • A telephone survey of 102 Special Care Center clients conducted by a local union (Local 52) found that all but one (1 percent) respondent would recommend the center to friends.
    • In a phone survey conducted by PRC, a large national patient survey vendor that serves hospitals and health systems, more than 80 percent of respondents gave the center "excellent" scores for overall quality of care.
  • Project staff produced an evaluation plan to track the success of the center. The center will be evaluated in five key areas: patient experience, patient functional status, staff and physician satisfaction, clinical process and outcomes and cost. The plan outlines specific measures and tools to be used to gauge results in each of the areas.


Mercer reported these findings from home visits with over 20 members of the union who were considered prospective candidates for the Special Care Center:

Respondents' Current Experiences With the Health Care System

  • Although most patients had a primary care physician and initially said they were satisfied with their care, more extended interviewing revealed dissatisfaction with factors such as rude treatment, long waits and lack of information.
  • Most patients reported having to wait 30 to 45 minutes for an appointment, and some routinely waited between one and two hours. Doctors rarely spent more than 15 minutes with them.
  • Although patients reported that the doctors had answered their questions about medications, they had more questions later and there were gaps in their knowledge about the drugs they were taking.
  • Access to specialists was found to be poor, with patients waiting up to six months to see an endocrinologist.
  • Routine eye care was poor, largely because patients perceived the out-of-pocket cost to be too high.
  • Some primary care providers appeared to pay much more attention than others to chronic disease management. Certain physicians' patients all knew their hemoglobin A1C levels, whereas others didn't even know what A1C levels were.

Areas in Which the Special Care Center Could Help

  • Medication management. Most patients took multiple medications, which were kept haphazardly. The report suggested that medication lists be trimmed and patients be given pillboxes to track them.
  • Better management and monitoring of chronic illnesses, such as depression and diabetes.
  • More intensive diet interventions, rather than simply telling people to "eat better."

 Back to the Table of Contents


  1. To test a new model for health care delivery, make sure the people you're working with understand delivery system redesign. Very few physicians are trained in the design of health care delivery systems. "Doctors know how average clinical care works, so when you talk about design, they think it's not that important and figure they'll muddle through." A lack of understanding of how the model worked contributed to the failure of the first medical director, according to the project director. (Project Director/Milstein)
  2. To improve outcomes for the sickest patients, you need to help them change their whole lives. "We went into the project believing that if you get the technical side of care right and let people know that you care about them, results would improve," Project Director Milstein said. "But for people who are the sickest on a chronic basis and incurring most of the costs, it's often not just about health care. These people are severely challenged due to social isolation, behavioral health impairment, or other factors, and you have to think about changing their lives if you want to ask them to do the things they need to do to stay out of health trouble." (Project Director/Milstein)
  3. To work effectively with people at the bottom of the socioeconomic pyramid, clinicians must work harder to gain their trust. "Hotel workers making $12 an hour, cleaning pots in the back of the kitchen, are people whose day-to-day experience in life is not very well grasped by health professionals, even well-meaning doctors, nurses and social workers," Milstein said. "For some, trusting powerful people who represent authority—including people in white coats—is difficult when you've been at the bottom of the societal ladder. Day-to-day life is often one of being let down and treated badly. Trust building, such that they depend on you and stop going to prior doctors who were in some cases providing less intensive care, takes a lot of clinician work and investment." (Project Director/Milstein)
  4. Using primary care to radically reduce health care spending and improve quality requires a lot of cooperation from two physician groups: ER doctors and hospitalists. "You need to be able to rely on ER physicians to alert you when they're seeing your patient and and not admit unnecessarily," Milstein said, "and on hospitalists to help you understand what it was about the prior care that failed and caused the patient's admission to the hospital." (Project Director/Milstein)

 Back to the Table of Contents


Mercer (formerly Mercer Human Resource Consulting) hopes to have enough patient participation months to evaluate the effectiveness of the program in lowering costs of care by summer 2010.

The Special Care Center expanded its home visit program to patients to include visits from both health coaches and physicians.

Plans are under way to expand the clinic's scope in order to serve patients who are not members of unions under the UNITE HERE umbrella, including three additional payers/unions: Boeing, Lowes and the UFCW National Health and Welfare Fund.

UNITE HERE has launched a similar program in Las Vegas, working with a risk-bearing physician group rather than a hospital or health system.

 Back to the Table of Contents



Exploring low-cost health care delivery alternatives for low-wage earners in Atlantic City, N.J.


Mercer Human Resource Consulting (San Francisco,  CA)

  • Amount: $ 194,560
    Dates: February 2007 to December 2008
    ID#:  056351


Arnold Milstein, M.D., M.P.H.
(415) 743-8803

Web Site


 Back to the Table of Contents


Appendix 1

The Design of the Special Care Center in Atlantic City, N.J.

In Atlantic City, the Ambulatory-Intensive Caring Unit (A-ICU) is called the Special Care Center and is housed on one floor of a modern health care office building built with federal and state gaming revenue. UNITE HERE and Atlanticare jointly share the costs of operating the clinic.

The Special Care Center is staffed by two physicians (one who serves as medical director); a registered nurse (RN); a social worker who manages the clinic; and a team of community health workers, known as health coaches, supervised by the RN.

Patients visiting the Special Care Center receive a thorough, hour-long initial examination and history taking from a physician, with a health coach in the room serving as a medical assistant. The health coaches are responsible for independently meeting with patients to:

  • Make sure they understand their plans for managing their own care.
  • Monitor their ability to keep up with that plan.
  • Give advice and assistance to those who are struggling to manage their illnesses.

The coaches make home visits, as necessary, to a subset of the patient population that has particular difficulty managing their own health. These 50 to 60 patients are unable to manage their care plan alone because of factors such as secondary mental illness, extreme social isolation, low functional intelligence and/or low income and few resources.

The Special Care Center did not fully implement "Third Floor" specialist referrals as part of the cost-savings plan because of concerns about physician resistance and wariness about competition between Atlanticare and the center. Instead, the Special Care Center networked with hospitalists from Atlanticare's two hospitals to identify informally the higher-quality, more cost-effective specialists. Mercer hired an analysis contractor, who by the end of 2009 will identify appropriate specialists more formally.

Patient Recruitment
Staff of the Special Care Center recruited patients using four main methods:

  • Predictive modeling software that identified participants within UNITE HERE's claims data who had the highest predicted health care spending, and who also had a chronic illness deemed manageable by the Special Care Center's resources. These illnesses included hypertension, diabetes, coronary artery disease, chronic obstructive pulmonary disease/asthma, depression and congestive heart failure.
  • Direct appeals to UNITE HERE membership through the union's newsletters and other publications, seeking people with severe chronic illnesses who could benefit from more intensive management.
  • Direct contact with physicians who treated many union members.
  • Referrals from the two Atlanticare hospitals in the area. The case management departments referred patients seen in the emergency room or hospitalized with a target illness who were deemed a significant risk for further ER or hospital use.

 Back to the Table of Contents


(Current as of date of this report; as provided by grantee organization; not verified by RWJF; items not available from RWJF.)


Milstein A and Gilbertson E. "American Medical Home Runs." Health Affairs, 28(5): 1317–1326, 2009. Abstract available online.

 Back to the Table of Contents

Report prepared by: Gina Shaw
Reviewed by: Kelsey Menehan
Reviewed by: Marian Bass
Program Officer: Anne Weiss