The promise of Medical Home is that patient safety will improve; that the quality of care received by patients will be consistent and excellent, and that the cost of care will be reduced while the quality of care is maintained. There is a large and growing body of academic medical literature on the Patient-Centered Medical Home. While not a part of the peer-reviewed literature, SETMA alone has produced over fifty articles on Medical Home in the past thirty-one months.

All of us involved in Medical Home have been eager to have objective data on treatment outcomes and cost on Medical Home and particularly to have that information in contrast with medical practices not pursuing the principles of Medical Home.

In January 2011, SETMA was invited to participate in the Medical Home Study conducted by RTI International (RTI) with funding from the Centers for Medicare and Medicaid Services (CMS). The study compared patterns of care between clinical practices that have received National Committee for Quality Assurance (NCQA) recognition as a medical home and clinical practices with similar characteristics that have not received NCQA medical home recognition. The only compensation the 312 Medical-Home practices received for this participation was that RIT prepared reports summarizing information for each practice providing comparative information with two groups: a bench mark group of non-Medical Home practices and mean (average) performance of the NCQA-recognized, Medical-Home group.

RTI used Medicare fee-for-service (FFS) billing data as the information source for their study. The limitations of that methodology will be discussed later. For practices with multiple sites, a report was produced for each practice site.
Three data categories are presented:

1. Clinical quality of care measures - Summary information about selected quality of care measures, such as LDL-C, HbA1c, and influenza vaccination.
2. Coordination and continuity of care measures - Summary information for selected utilization measures, such as emergency room (ER) visits and hospitalizations for ambulatory care sensitive conditions, percentage of your Medicare FFS patients that had a follow-up visit within 2 weeks of a hospital discharge, percentage readmitted within 30 days of a hospital discharge, and rates of medical and surgical specialty use.
3. Medicare payments - Summary information on the share of care that you provide your Medicare FFS patients, total Medicare payments per beneficiary, and average Medicare provider payments by type of service.

The study analyzed patterns of care, health outcomes, and costs of care for Medicare fee-for-service (FFS) beneficiaries receiving healthcare services from clinical practices that are National Committee for Quality Assurance (NCQA)-recognized medical homes. In particular, the study was interested in determining if there are particular attributes of medical homes that are more favorably related to better outcomes of care. The information from these analyses will be used by CMS to help design Medicare and Medicaid medical home demonstrations.

SETMA's Results Against the Benchmark

Obviously, in January, 2011, our hearts where in our hands when we agreed to participate in this study. Who was to know how it would turn out? On June 30, 2011, SETMA received a communication from RTI, which in part stated:

"Thank you again for agreeing to provide feedback on a draft of the Patient-Centered Medical Home Study Practice Feedback Report. Your input will help us improve the report template before the final reports are prepared and distributed."

The key word was "draft," which meant one thing to RTI and another to SETMA. After reviewing the data in the "draft" report, I responded:

"I have reviewed your reports. Your data is seriously flawed. I am going to send you data on the same periods for the same measures. We use IBM's COGNOS Business Intelligence software which is very accurate. Look forward to our discussion."

RRI answered:

"...the data presented in the feedback report template are 'not' real data for your practice. The purpose of this exercise is for you to take a look at a mock-up of a template that would be similar to the one you would receive (and that would contain your real data), so that we can get your opinion on the format and content of the report. But the numbers shown in the tables and graphs that you have right now are completely made up just for illustrative purposes-just to show how the tables and graphs would look."

That part of our conversation was concluded when I said:

"I am laughing out loud. Smile. I had no idea."

With that "clarification" of what "draft" meant, we waited, not patiently, for the "real" results to arrive. They did on September 2nd. What a relief. The results were very good. They gave us confirmation of SETMA's Model of Care and of the efforts in which we have been engaged for 17 years.

Quality of Care

The following quality comparison was between SETMA and a benchmark which was developed from The benchmarks are from a predictive model that uses the comparison group performance and models the relationship between the outcomes and practice characteristics such as average health status of beneficiaries assigned to the practice,  size of practice, type of practice, etc.  As can be seen, SETMA outperformed the benchmark in all three areas of interest.

The data on influenza immunizations was worrisome, as even though we out performed the benchmark, the data we have on our practice is different. I wrote RTI and said:

"Patients have learned to get influenza immunizations at many different places, i.e., pharmacies, hospital, emergency department, the VA, other clinics, etc.  SETMA captures the overwhelming majority of these but they are not billed by us to CMS thus according to the process of this audit would not appear to be counted.

"The following are our results for FFS Medicare plus all Medicare (FFS, Medicare Advantage and Dual Eligible) during four periods:

• The first column is the same period you measured (July 1, 2009 to June 30, 2010) and is just the FFS Medicare.
• The second column is the same period you measured (July 1, 2009 to June 30, 2010) except it is all Medicare seen each clinic whether FFS or Medicare Advantage or Dual Eligible.
• The third column is the same group you measured for the year July 1, 2010 to June 30, 2011.
• The fourth column is all Medicare patients for July 1, 2010 to June 30, 2011 whether FFS or Medicare Advantage or Dual Eligible.

SETMA's Audited Influenza Immunization Statistics for two years and two different populations
to compare with the Statistics attached above from the CMS study

If a patient refused an influenza vaccine, they were included in the denominator but if they were allergic to the vaccine, they were excluded from the denominator.

The following is the answer which RTI International sent back:

"I have always had reservations about reporting influenza vaccination from Medicare claims data. And, your data shows why I am hesitant. We simply do not capture in our rates vaccinations provided to Medicare FFS beneficiaries that are not subsequently billed to Medicare. You clearly have a more robust system for capturing the actual rate of receipt among your patients." (Emphasis added)

Cost of Care

The area of our practice which we had never seen, as had no other medical practice, is the total cost to Medicare for the patients we see under CMS' Fee-for-Service.

Dividing our cost of $8134 by the benchmark's cost of $12,919, SETMA has a 37.04% lower cost than the benchmarks while our quality of care is higher than theirs.

Looking at the cost of care for Durable Medical Equipment (hospital beds, oxygen, etc), Hospice, Home Health and Skilled Nursing, the results are similarly good. While our Home Health use is slightly higher than the bench mark, if you add the Home Health and the Skilled Nursing Care, the benchmark is $1810 and SETMA is $1296 SEMTA is 28.4% lower than the benchmark.

One of the most important targets for improving care and controlling the cost of care is in decreasing preventable readmissions to the hospital. The following graph shows CMS' estimation of the potentially avoidable inpatient admission payments for SETMA and for the benchmark. The benchmark is $2259 and SETMA is $962. SETMA is 57.42% below the benchmark.

The finally part of the study which we will review today is the following graph which shows the trend of SETMA's performance.

SETMA will be the first to admit that we have a long way to go to reach our goals, but this interim judgment of our quality, cost and trending is encouraging and reassuring that our Model is sound. As with all of SETMA's work, the full reports for all three clinics are posted on our website under In-The-News, CMS Medical Home Feed Back Report Quality and Cost. For this report the information on SETMA II was used for space considerations. SETMA's transparency is for the benefit of our patients and our practice. Hopefully, it is an example to others. In another setting, you might say that SETMA is the Exhibitionist of the Medical Home world!