34_Attachment_J-24_HA_Policy_Memo_for_MHS_Care_Quality_Assurance_Data_Transparency.pdf

OFFICE OF THE ASSISTANT SECRETARY OF DEFENSE

WASHINGTON. DC 20301-1200

OCT 202010

HEAl.TI.f AFFAIRS

MEMORANDUM FOR ASSISTANT SECRETARY OF THE ARMY (M&RA)

ASSISTANT SECRETARY OF THE NAVY (M&RA)

ASSISTANT SECRETARY OF THE AIR FORCE (M&RA)

DIRECTOR OF THE JOINT STAFF

SUBJECT: Policy Memorandum for Military Health System Health Care Quality

Assurance Data Transparency

References: (a) Executive Order 1341O-Promoting Quality and Efficient Health Care in Federal Government Administered or Sponsored Health Care Programs

(b) Department of Defense (DoD) 6025.13-R-Military Health System Clinical Quality Assurance Program Regulation

(c) DoD 6025.18-R-Department of Defense Health Information Privacy Regulation

(d) 5 United States Code 552a-Privacy Act of 1974

(e) Title 10, Section 1102. Confidentiality of Medical Quality Assurance

Records: Qualified Immunity for Participants

(f) Public Law 104-1 9 I-Health Insurance Portability and

Accountability Act

The Department of Defense (DoD) Military Health System (MHS) will have readily available and transparent relevant quality assurance (QA) information provided to its beneficiaries, enrollees, and providers in an understandable manner and in collaboration with similar initiatives in the private sector and non-Federal public sector.

Transparency in health care quality is defined in accordance with reference (a).

"Each agency shall implement programs measuring the quality of services supplied by health care providers to the beneficiaries or enrollees of a Federal health care program.

Such programs shall be based upon standards established by multi-stakeholder entities identified by the Secretary or by another agency subject to this order. Each agency shall develop its quality measurements in collaboration with similar initiatives in the private and non-Federal public sectors."

Although individual event health care QA data may not be released by MHS, health care QA data may be released publicly when stated as aggregate statistical data.

Measures will be approved for release by the Assistant Secretary of Defense for Health Affairs. Currently, the Healthcare Effectiveness Data and Information Set (HEDIS®)

Attachment J-24 HA Policy Memo for MHS Care Quality Assurance Data Transparency

TQMC HT9402-15-R-0005, Section J, Attachment J-24 Page 1 of 4 reubanks Typewritten Text reubanks Typewritten Text reubanks Typewritten Text

HA POLICY

10-008 reubanks Typewritten Text like measures and ORYX® performance measures are approved for such release, providing the data is released in a manner to meet the definition of aggregate statistical information in attachment (1) and de-identified protected health information (PHI) in reference (c). The Service Surgeons General will establish policy regarding the release of individual Military Treatment Facility aggregate QA data.

Aggregate statistical information, such as HEDIS®-like measures and ORYX® performance measurement data, are an assembled collection of numerical facts and other information or data derived from various DoD health program activities and stated as a statistical number. Aggregate statistical information is numerical data that constitute all the data in pre-defined common demographic grouping that have been assembled in order to facilitate its interpretation. In accordance with references (b), (c), (d), (e), and (f), aggregate statistical data derived from medical records and medical QA information may be released outside of DoD when authorized by the appropriate authority. See attachment

(1) for guidelines on the release of aggregate statistical information.

MHS is committed to being patient-centered and providing quality health care.

Providing QA data creates opportunities to save lives and improve the services provided to all of our beneficiaries. Questions may be directed to the Director of Clinical Quality, Office of the Chief Medical Officer, TRICARE Management Activity, Falls Church, Virginia, (703) 681-0064.

George Peach Tay l::rr.,<t:

Deputy Assistant Secretary of Defense

(F orce Health Protection and Readiness) Performing the Duties of the

Assistant Secretary of Defense (Health Affairs) cc:

Service Surgeons General

Attachment:

(1) Guidelines for the Release of Aggregate Statistical Quality Assurance Data

Attachment J-24 HA Policy Memo for MHS Care Quality Assurance Data Transparency

TQMC HT9402-15-R-0005, Section J, Attachment J-24 Page 2 of 4 reubanks Typewritten Text

HA POLICY

10-008 reubanks Typewritten Text

Guidelines for the Release of Aggregate Statistical Data

The following guidelines have been developed in order to ensure the privacy of the individual’s health and treatment records, as well as prevent the inadvertent release of information that would allow the identification of the individuals involved when the released information is combined with other publicly available data. These guidelines recognize that there may be instances when a third party possesses “insider” information or knowledge about a patient or individual provider based on their personal relationship (family member, neighbor, close friend) with that patient or individual provider and that these persons may be able to add their “insider” knowledge to the information released by Department of Defense (DoD) to the general public, and thereby identify their friend as a member of the demographic grouping. However, absent this insider information the guidelines outlined below provide protections against others being able to positively identify a member of a demographic grouping. DoD has balanced the risk to those about whom “insider” information is known by acquaintances with the overall good of releasing information about the Military Health System’s quality and costs of care to our beneficiaries and the general public. The following guidelines must be followed when releasing aggregate statistical information:

(1) The data may only be released publicly as aggregate statistical data.

Aggregate statistical data are expressed in the form of a number, including whole numbers, fractions, or percentages. Numerical data that are derived from records within the DoD quality assurance program must also be in such demographic groupings that the release of the information would not lead to the identification of the patient or the individual provider involved in providing care.

(2) The data must be de-identified as the term is used in reference (c) (Chapter 8, paragraph C8.1.3). This specifically includes the identifiers listed in paragraph C8.1.3.3 of this reference relating to the individual involved. The term de-identified data does not include the geographic location or name of the treating facility or the period covered for the care upon which is being reported. Thus, it would be permissible to release aggregate statistical data for a particular Military Treatment Facilities (MTFs) by name or by group of MTFs by region or some other similar subdivision, as well as by period, such as month(s) or year(s).

(3) The release of aggregated data may include several types of demographics. This means that if the grouping of aggregated data includes several types of demographics, such as age, sex, race, Active Duty status, rank, or Service, the population or number of persons meeting all of the demographics in the grouping must be three or greater. To illustrate—assume that DoD wants to release information relating to how many persons in a demographic grouping that did not receive aspirin within 30

Enclosure

Attachment J-24 HA Policy Memo for MHS Care Quality Assurance Data Transparency

TQMC HT9402-15-R-0005, Section J, Attachment J-24 Page 3 of 4 minutes of presenting to the emergency room (ER) with symptoms of acute myocardial infarction (AMI). Additionally, it wants its groupings to include those patients who possess the following demographics: age 50 or above, Caucasian, male, on Active Duty who are seen at the MTF East Coast, during the month of February for symptoms of AMI in the ER. In this example, the “numerical data” is being presented as a fraction with the numerator being the total of those persons in the entire demographic set that did not receive aspirin and the denominator is the entire number of persons who fit all the demographic fields. In order to release any information about how many persons did not receive the aspirin then there must be three or more individuals who fit all of the demographics in that grouping, i.e., three or more Caucasian males aged 50 or older on Active Duty who are seen at the MTF East Coast in the month of February, and who presented to the ER with symptoms of AMI. If there are not at least three persons in this group, then no release can be made. It is noted that if the denominator contains too many demographic fields, the number may be too small for release, and one or more demographic fields may have to be removed and the aggregate statistical data on a subset of the original demographic fields could then be released. For example, if there were not three persons who met all of the demographics in the original example above but there were three or more Caucasian males aged 50 or older, who were seen at MTF East Coast during February with symptoms of AMI in the ER by removing the demographic requirement to be on Active Duty, then this subset of the original demographic fields would be sufficiently large to release the aggregate statistical information. Having a denominator equal to three has been shown to provide adequate safeguards to provide reasonable assurances that the identity of any particular patient can not be ascertained just by the grouping.

(4) If the above threshold cannot be met, the aggregate statistical data may not be released; however, a notation may be made that it is not being released because the number is too small to assure that the privacy and the identity of the persons involved can be maintained. It is noted that even if the numerical data is equal to 1 or 0 (the ratio, percentage or combination equals the whole (one) or none (zero) of the grouping to be released), then this aggregated statistical data may still be released as long as there are at least three people in the groupings as discussed above. Using the example from the above paragraph—assume we wanted to release the statistic on what percentage of persons received aspirin within 30 minutes of presenting with these symptoms of AMI.

If our grouping consisted of three individuals and if all of them received aspirin then this would mean that 3 of 3 or 100 percent could still be released. Conversely if none received aspirin then the number would be 0 of 3 or 0 percent and this too could be released.

Attachment J-24 HA Policy Memo for MHS Care Quality Assurance Data Transparency

TQMC HT9402-15-R-0005, Section J, Attachment J-24 Page 4 of 4