|

Seven Points
for Implementing HIPAA Security
October 25, 2000
Here are suggestions for starting to deal with HIPAA in advance
of publication of the final rules on security and privacy:
- Start now, while you're figuring out how to bring top management
(the CEO and the board) actively into leadership of your enterprise's
HIPAA project, and organizing your HIPAA oversight or project
team.
- Designate a Security Officer and a Privacy Officer (two different
posts - neither should be in the General Counsel's office).
- Start budgeting now - refine iteratively.
- Understanding and implementing encryption technology is basic
to HIPAA. Contact PKI (Public Key Infrastructure) and other vendors
of encryption technology to learn about the options for encryption
available in the market.
- Contact your legacy system vendors and ask for details about
HIPAA system deployment and warranties. (Meanwhile, every new
contract for hardware, software, data communications, and related
services should contain a HIPAA covenant. It should state that
the vendor will negotiate with you in good faith, after the final
HIPAA security and privacy rules are published, to amend sales,
license, or maintenance agreements in order to allow you to comply
with HIPAA. Further, there should be a clause that allows you
to terminate the agreement if a satisfactory amendment for HIPAA
compliance cannot be negotiated.)
- Legal standards drive the technological and business process
choices, so get your counsel educated and involved in every step.
- Maintain a litigation perspective; think about privilege. It
will help your institution and your colleagues, who will feel
the ultimate impact of HIPAA.
Meeting HIPAA's challenges is not just an information technology
project, and it also is not just a compliance project. Rather, the
budgetary, cultural, business process, and technological changes
that HIPAA demands will make this an all-enterprise effort. To succeed,
it must be supported from the top.
Return to top of page
HIPAA
Privacy and Security Factual Questions
Fact Issues (each of which also requires knowing the applicable
legal standard):
- What sort of access control is required? Is it true that each
view of protected health information (PHI) in a patient's record
must be logged, showing who viewed what? Can a doctor look over
another doctor's shoulder and not log the access? Same if a nurse
is involved? What about doctors and medical students on grand
rounds? What are the penalties for not implementing proper logging
procedures? For not enforcing them vigorously? What kind of software
can perform this kind of logging? Is it available today on the
market?
- What protections must be implemented for terminals where PHI
will be accessed? Must nurses' stations be secure areas? If so,
how secure? What about monitors in ICUs and similar facilities?
What surveillance must be implemented for these areas?
- What kind of access procedures will satisfy HIPAA requirements?
Passwords alone? Passwords with SecureID or similar procedures?
Biometrics?
- What happens if a health care professional attempts to log on
to get access to medical records in an emergency, and makes a
mistake in their input of the password or other required information?
How many attempts can be allowed before the system locks them
out? What are the emergency access procedures? Must the doctor
or nurse then be physically identified by a member of the security
staff, in essence to revalidate that staff member? How will that
be done? How long might it take, especially in an emergency setting?
What if patient care suffers - what are the malpractice issues?
- How can the logon, identification, and other required attributes
of data protection, in transit and at rest, be satisfied without
resort to asymmetric encryption techniques?
- How can a medical professional or an administrator determine
what the "minimum necessary" disclosure is to an examining physician?
Where does one find the industry-standard guidelines to make these
judgments? Do they change in the face of a medical emergency?
What procedures are required (or prudent) for a hospital or a
physician practice to review these determinations after the fact?
What if there is a finding of excess disclosure? Can the professionals
involved in the original decision appeal? To whom? ? Must the
provider report the matter to HHS, and, if so, at what point?
What guidelines will HHS use to review the disclosure and its
review? What malpractice jeopardy is there for making too narrow
a disclosure?
- To what extent will the standard of care in the security industry
require that protected health information (PHI) be stored in encrypted
form? Are systems available on the market to perform this kind
of storage?
Return to top of page
Preliminary
Checklist for Security and Privacy Issues in HIPAA Implementation
Projects
November 9, 2000
NOTE: Because HHS representatives state explicitly that the final
privacy regulations will change substantially from those proposed,
this checklist concentrates in the interim on the security regulations.
It is of course possible that the final security rules will differ
substantially from those proposed.
1. Will the enterprise's systems
- create, store, manipulate, or transmit
- internally within the organization or externally
- data
- about a patient's
- past, present, or future
- physical or mental health condition, health treatment, or payment
for health care?
2. Are the systems
- clinical
- business
- administrative
- email?
3. Will the systems include
- backup
- storage
- archives
- transmission (internally or externally)?
4. Is the patient identifiable ("individually identifiable")?
5. Where will individually identifiable patient data be displayed,
stored, or transmitted in electronic form or in paper
derived from electronic form?
- Terminals: consoles, desktops, or laptops (at work, home, or
travel)
- registration
- clinical
- business
- research
- administration
- Storage
- Current data bases
- Archived data bases
- Current physical (paper) files (including mixed files, those
containing some pure paper records and some paper records
derived from electronic files)
- central
- distributed
- Archived physical (paper) files
- Networks, internal and external
- protected transmission path (VPN or comparable)
- public (Internet, public switched telephone network or comparable)
6. Have all personnel with access to system components been cleared
by an appropriate background check
- initial
- recurring (what is the interval in years)?
7. Are all components of the system physically secure?
- Can any affected terminals be viewed by
- the public
- any staff member who lacks authorization under
- "need to know"
- "minimum necessary disclosure"?
- Is each affected terminal under some form of surveillance?
- active (e.g., cameras with recording systems)
- passive (access logging)
- area surveillance (located in a secure area where ingress-egress
is surveilled)?
8. Are all access features of the system controlled sufficiently
to assure irrefutable identification and detect attempts at improper
access?
- Access controls
- password (PIN) plus smart card, SecureID, etc.
- biometrics
- Network protection
- firewalls, etc.
- intrusions detection
- anomalous event detection
- active and passive surveillance
- Comprehensive logging and audit trails
9. Will the system produce sufficient audit trails?
- Purposes:
- security incident procedures and response
- security configuration management (i.e., making changes
to the system so that no specific vulnerabilities are created)
- assessing compliance with self-reporting and colleague-reporting
requirements
- responding to patient requests for access records
- certification and accreditation
- cooperation with law enforcement so as to meet statutory
and administrative standards
- What data architecture is required
- existing industry practice for logging
- field-level logging?
10. If compliance is achieved in part by implementation of asymmetric
encryption (public key infrastructure or PKI)
- what is the system for managing digital certificates
- is the public key infrastructure interoperable with that of
business partners?
Return to top of page
Preliminary
Checklist for Computer and Telecommunications Systems Purchases
Our clients have come to believe that legal planning -- describing
a project, negotiating specifications, anticipating the inability
to fulfill technical or performance claims -- is as integral to
success as technical planning.
Negotiating a contract for a computer system, or for software,
is - or should be - a systems planning process as much as it is
an exercise in defining legal rights and duties. Inevitably, analysis
will go hand-in-hand with pre-installation tasks, and the purchaser
and vendor will learn significant new information about the purchaser's
needs and how they must be fulfilled. In addition, contract negotiations
are a means to identify problem areas, assign responsibilities and
develop a framework for ferreting out problems so that they can
be dealt with in a systematic way. In short, the aims of the contracting
process are to:
- Explain the project to all parties, and develop an implementation
plan;
- Furnish a process for discovering and accommodating the purchaser's
new or changing requirements during the development and installation
process;
- Supply a project management framework, including the means to
adjust responsibilities and schedules as obstacles arise and as
new needs are revealed;
- Preclude disputes, or specify a means for their quick resolution;
- Avoid litigation; and
- Establish a favorable litigation posture.
Consider the following questions when a project is initiated:
1. Will the system be crucial to business operations, or is it
a subsidiary or support system whose operation, while important,
can be foregone for a period of time?
2. Is it realistic to expect sustained parallel operation after
the system is installed, or will an immediate or rapid cut-over
to the new system be necessary? To what extent can the old system
be retained as a short-term backup?
3. Is there a need for an RFP? If so, to what extent will the RFP
contain preliminary criteria that will need to be refined either
(a) between issuing the RFP and awarding the contract or (b) after
the contract is negotiated and signed?
4. Are you buying both hardware and software? If so, will one vendor
stand behind - and service - the whole system, or will you have
to determine whether you are facing a hardware, software or telecommunications
problem in order to decide which vendor to call?
5. If you are buying hardware, has the equipment been installed
and used before in the same configuration, with the same communications
facilities and for the same throughput?
6. Does the project require the tailoring of existing (or "off
the shelf") software, or does it involve substantial software development?
7. What is the installation and implementation plan?
8. What is the installation and implementation schedule?
9. Who can change items 7 and 8, and by what process? How much
flexibility is necessary to be able to modify items 7 and 8 as the
project moves along?
10. How can system performance be described and tested? What performance
is satisfactory?
11. Who is authorized on behalf of the vendor to make representations
about the system's performance and capacity?
12. Who (or what group) is authorized on your behalf to agree to
the contractual description of system performance, and to agree
with the vendor to vary that description as the system is developed
(or tailored), installed, activated and tested?
13. Is installation, including debugging, to be done at a fixed
price or an hourly rate? Are installation and software development
commingled, and is it essential to combine them? Is it feasible
to separate them?
14. Will the project be installed and implemented in phases? If
so, how will the phases be defined?
15. Will you specify both installation and acceptance tests for
the project (or for each phase)?
16. How will "acceptance" be defined?
17. How can pricing be combined with installation and acceptance
benchmarks to reinforce contract aims?
18. What protection is offered under warranty, and how does that
relate to contracted maintenance?
19. If maintenance service is unsatisfactory, is a switch to third-party
maintenance feasible?
20. Who owns the copyright to the software? (Was an independent
contractor involved anywhere along the way?) How long can you use
it? Who can debug, enhance or create new versions of it? Will you
have the right to use enhancements and new versions?
21. Is software "lock-in" a problem and, if it is, how can it be
dealt with?
22. Must the system be protected by security measures? If so, is
the level of protection apparent, or is a threat analysis necessary?
What legal standards govern the level of security? What level of
security measures is realistic in terms of cost and the organization's
culture? Is that level at odds with the governing legal standard?
23. Will the system be used in applications with the potential
to create liability to third parties? If so, what prudently should
be done to minimize that potential?
24. How will a breach be identified? What about identifying problems
short of a breach? Will the opportunity to cure certain breaches
be permitted?
25. In the event of breach, what remedies will be effective? What
impact will proposed remedies have on the ability to conduct business?
26. When is the time to negotiate sensitive items? When does the
purchaser have maximum leverage, and when is the purchaser's bargaining
power gone?
27. How fair is the deal to both sides? Do both parties have incentives
to follow the plan, or are there incentives to slow down, back out,
or minimize performance
return to Publications
main page
|