As the original HIPAA standards become mandatory it will become possible to demonstrate a truth about standards that has so far been accepted only by the cognoscenti:
a) the benefits to society of interoperable national standards do not by themselves justify investments by individual organizations to restructure their systems and procedures to meet the standards, so therefore
b) it requires an externally-imposed mandate to allow individual organizations to override their own financial and structural constraints and do what benefits society.
The impact of HIPAA will further verify the principle of relying on modest advances of the existing standards that were developed by consensus standard organizations and have achieved substantial usage, rather than creating standards de novo.
In extending HIPAA to clinical standards, the principles are the same, but the stakes are higher. The benefits to society of providing fine-grained, structured standards for passing clinical data are profound. In a society where the government is a large payer, these standards support rational rationing, as opposed to the current situation, which includes a more substantial portion of rationing by political lobbying. In a society where consumers choose healthcare resources more freely due to the Internet and employer disassociation with health benefits, better data allows informed choices rather than the current situation, which is approaching "choice by Madison Avenue." In a society where a patient's data must be shared across geographical and organizational boundaries, the utility of computers for reduction of errors and medical management can only be achieved by institution of these standards.
At the same time, the costs of implementing such clinical standards will be substantial. The scope of the process that corresponds to developing implementation guides for HIPAA financial transactions will be broader because of the broader scope of clinical data. Healthcare providers that do not have computer-based patient records will have to acquire and deploy them. In many cases, healthcare providers that already have computer-based patient records will require significant upgrades to deploy systems with the capabilities of conforming to fine-grained structured standards. The vendors that produce many such systems will require several version cycles to adapt to the new standards, which they will be evolving.
It is therefore critical that the regulations that come about to support standardizing clinical data be evolutionary and provide for time for organizations to achieve returns on their investment as they pass through the evolutionary stages of adoption.
1. Draft Criteria for Selection of PMRI Message Format Standards
I would suggest the following restructuring of the criteria (in priority order):
1. Standards that improve the efficiency and effectiveness of the health system for delivering high quality care.
a. Unambiguous sharing of data and information (data comparability).
b. Data security and integrity.
c. Interoperability – this is the goal, but is extremely difficult to completely achieve.
2. Standards that work. Only select standards that are implemented and working in a wide variety of production clinical environments.
3. Cost effective standards. Select standards that will be the least costly to implement.
a. Market acceptance. Use what people are already using.
b. Public domain and commercial tools available for implementers.
c. Education and training available.
4. Sustainable standards
a. Stable organizational support
b. Follow ANSI open consensus process rules.
c. Manage change:
i. Flexibility to respond to new requirements
ii. Timely corrections and enhancements
iii. Preserve business knowledge in the face of rapidly changing technology.
2. Draft Questionnaire: PMRI Message Format SDOs
May we please have your comments and critiques on the following set of questions.
I think the proposed questions are appropriate.
I would suggest that the following questions be added:
One difficulty is understanding at what granularity to submit answers to the questions. For example, should HL7 submit one set of answers for its whole suite of standards, or submit answers for each kind of transaction domain? I would suggest that a separate questionnaire be submitted for each transaction domain. This will result in some redundancy between submissions from the same organization, but will lead to clarity in how the answers pertain to a specific transaction area. The proposed set of PMRI standards serve as an appropriate guide for transaction domains.
3. Proposed list of PMRI transactions to be considered for HIPAA standardization recommendations
May we please have your comments and critiques on the following list of PMRI transactions to be considered for HIPAA standardization in the first phase of PMRI standards selection.
The list of proposed transactions appears appropriate. I would suggest adding the following items:
The list of transactions should be prioritized. I think that the list as given is roughly correct:
1. ADT
2. Standard orders (excluding medication orders)
a. clinical laboratory
b. anatomic pathology
c. radiology
d. Etc.
3. Standard results (clinical laboratory, anatomic pathology, radiology).
4. Inpatient medication orders to pharmacy systems.
5. Clinical documents.
6. Chief complaint, problems, diagnoses.
7. Images.
8. Visual integration.
9. Data from bedside instruments and monitoring systems.
10. Orders for outpatient medications to pharmacy systems.
11. Procedure scheduling.
12. Charge capture information to billing systems.
4. Additional Comments or Critiques
1. Implementation guides will need to be developed as a first step for most clinical standards. Existing standards do not yet integrate clinical terminology in a tightly coupled manner. Some implementation guides already exist, i.e. immunizations. We will need implementation guides for:
a. Laboratory orders
b. Laboratory results
c. Medication orders
d. Microbiology culture results
e. Others
2. The best initial strategy may be to reward those who use the approved standards rather than mandate their use at the risk of criminal penalties. For example, the government could initially pay a per case or per transaction premium to institutions that submit data electronically using the approved standards. When the standards are broadly implemented, and proven to be effective, the use of standards could be mandated. Perhaps the process should specify that beta testing (proven case studies) be required; select the standard when beta testing is complete, and then lay out the transition plan specifying mandated start and end dates.
3. Based on readiness of standards and the potential value to society, standards should be adopted in the following contexts (in order of my priorities):
a. Reporting of clinical data to third party payers (as currently proposed for claims attachments)
b. Reporting to governmental departments, offices, and agencies
i. Infectious disease reporting to state and federal agencies
ii. Immunization information to state and federal agencies
iii. Tumor registries
iv. FDA for adverse drug events
v. FDA for clinical trials information
vi. HCFA chart review
c. Reporting of veterinary data to governmental departments and agencies
d. Reporting of clinical trials data to private companies
e. Reporting to national professional databases:
i. Cardiology
1. Myocardial infarctions
2. Open heart surgery
ii. Mother’s and newborns
1. Prenatal care
2. Neonatal care
3. Birth defects
iii. Endoscopy
iv. Others
f. Data exchange between health care enterprises
g. Data exchange within a single health care enterprise
4. New versions of standards or new standards that replace existing standards will need to be developed and adopted. There should be a well-defined process for prototyping and implementing new standards in production systems before general adoption.