The NetWellness PreAnesthesia Interview PAI:

1
The NetWellness Pre-Anesthesia Interview (PAI)
A Free, Web-Based, Multilingual, Customizable,
System Gareth S Kantor1 MD, Allison Rand RN1,
Mark McCuistion BS2, John Gage MD3 1Case Western
Reserve University University Hospitals of
Cleveland, Cleveland, Ohio, USA 2University of
Cincinnati, Cincinnati, Ohio, USA 3State
University of New York at Stony Brook, New York,
USA http//www.netwellness.org/interview/anesthesi
a
PATIENT EDUCATION Clicking on the Help icon
next to each question displays patient
educational material which explains the reasons
for asking the question, and, where appropriate,
the meaning of the question (Fig 5). Original
text information, written by the authors, is
displayed. The text is supplemented by hyperlinks
to other parts of the NetWellness site and, as a
result of future development, will also contain
hyperlinks to resources outside NetWellness.
INTRODUCTION An evaluation that can be
completed by the patient prior to surgery is
particularly important in the field of
anesthesiology since the patient almost
invariably meets the anesthesiologist only
briefly before the procedure.   Time
constraints, and stress on the patient, make it
difficult to acquire a complete medical history,
to adequately inform the patient about the
procedure, and to obtain genuine informed consent
for anesthesia.   Interactive computer-based
patient self-assessment tools, such as online
questionnaires and interviews, can facilitate
acquisition of the medical history, enhance
patient education, and improve subsequent care
1,2.   Computer-patient interviewing can
improve the quality of the doctor-patient
relationship because a comprehensive initial
medical history allows both individuals to focus
on the issues most relevant to the patient's care
when they meet. Because less time is wasted on
unimportant areas, there is also the potential
for increased clinical productivity.   We have
developed a free, web-based patient
self-assessment tool the Pre-Anesthesia
Interview (PAI) (http//www.netwellness.org/interv
iew/anesthesia). We believe the potential
benefits of this system justify its use and
further study to determine optimal design and
implementation.
INTERVIEW ASSESSMENT The final set of questions
in the PAI is an optional patient feedback survey
(Fig 6) The patient is asked whether the
interview was satisfactory in various aspects
clarity, presentation, overall value, etc. The
printout has a series of questions for the
anesthesiologist or other anesthesia provider
(Fig 7). This survey is to be separated from the
Interview results, completed, deposited in a
collection box at collaborating hospitals, and
then mailed or faxed for entry in the database
and subsequent reporting.
ABOUT NETWELLNESS The Pre-Anesthesia Interview
prototype was developed using the netwellness.org
web site. NetWellness is a unique non-profit
model for delivering health care information, at
no cost to the consumer, via the World Wide Web.
It draws on and transfers to consumers and
communities the wealth of medical expertise of
three respected American medical schools. DATA
PRIVACY AND CONFIDENTIALITY   The NetWellness
web site is secured to protect patient privacy
and confidentiality, and the Interview system is
designed to comply with US Federal health data
regulations (HIPAA). The system does not gather
any information (name, address, date of birth,
social security number, etc) that could be used
to associate clinical data with the actual
identity of the patient taking the PAI. Users
are directed to the NetWellness Privacy Policy at
the beginning of the interview. After data
collection is complete, the system produces a
printout. The printout contains no patient
identifiers the patient must sign and date the
printout to identify the source of the
information. A Private Interview Number is
generated which the patient can use to return to
the site to retrieve or update results.

Fig 5. ANESTHESIOLOGIST SURVEY Fig 6. PATIENT
FEEDBACK
DESIGNING THE INTERVIEW The questions are
adapted from a paper-based pre-anesthesia
evaluation questionnaire used at University
Hospitals of Cleveland. They cover all relevant
areas of the preoperative history, divided into
15 sections. Validated questions from scientific,
peer-reviewed studies are used wherever
possible. The Interview Builder (Fig 1) allows a
domain expert (e.g. physician) to create the
interview without knowing anything about
programming, database structure, or HTML. A
secure database is used for storing all
questions, responses, instructions and
educational material. This architecture
facilitates design, maintenance and reporting.
Because the database is web-accessible, content
is easily updated. Designers in different centers
can collaborate using the Interview Builder.
The Interview instructions are checked for
appropriate reading level, using the
Flesh-Kincaid score, a readability test. The goal
is a Flesh-Kincaid score at or below an 8th grade
equivalent. The patient interviewee needs basic
web browsing skills. Some typing is required, for
instance when describing prescription drugs and
their doses, allergies, or previous surgical
procedures. The eventual goal is to substitute
menu-based choices for such free-text entry.
RESULTS The PAI is available now, in French and
English, and is free! Please try it!! The address
is http//www.netwellness.org/interview/anesthesi
a The preliminary response from patients and
clinicians has been positive. The number of
patients interviewed is small and we have yet to
analyze the data. Further development is ongoing
(see Discussion).
DISCUSSION Previous computer-patient
preanesthesia interview systems have failed
commercially 3. We believe there is
nevertheless a large unmet need for such a system
especially at low cost (or free) over the
internet. An interview system must be easy to
use, fit workflow, protect health data privacy,
enhance patient education, support clinical
decision-making and be compatible with other
computerized information systems. USABILITY An
interview system should undergo usability testing
for ease of use. Forcing a response to each
question ensures a complete interview, one of the
main benefits of using a computer. It should be
possible to look up, or even download
prescription data into the record rather than
making the patient enter these data. The
Interview printout must highlight important
clinical findings rather than making the
practitioner review masses of paper to discover
those key findings. A PDA version of the
interview might be useful also.
WORKFLOW Enhancing perioperative workflow is
challenging. We must decide how the Interview can
be used as a triage tool e.g. to help decide
whether a patient needs to be evaluated by an
anesthesiologist before, or on, the day of
surgery. We must also determine how, and in what
form, the results of the interview can be
transmitted to the hospital for timely review.
The data must then, ideally, be integrated with
other clinical information databases and data
repositories. DATA PRIVACY Health information
privacy is of paramount importance. The PAI
complies with regulatory requirements by avoiding
the collection of personal identifiers, but the
most useful systems must permit this data to be
stored, retrieved and transmitted. EDUCATION
INSTRUCTION The PAI presents an excellent
opportunity to effectively educate instruct
patients preoperatively. Educational material can
be presented in different ways, based on the
patients choice, and can include visual images,
diagrams, audio and video clips. Preoperative
instructions can cover preoperative fasting,
preoperative medications, and facility-specific
instructions (maps, appointment times, contact
information). OUTCOME STUDIES Computer-patient
interviewing presents an enormous opportunity for
anesthesia outcome studies which are difficult to
perform without such tools. There is also the
opportunity to integrate clinical decision
support tool, such as preoperative testing and
preoperative triage guidelines, and calculation
of clinical risk profiles using validated
algorithms (e.g. cardiac risk, risk of
postoperative nausea, DVT risk, etc.) A POTENTIAL
GLOBAL RESOURCE The PAI is universally available
at no cost via the internet. To facilitate use
outside large institutions in developed
countries, the design must be simple enough for
rapid transmission over low bandwidth internet
connections. We plan a Spanish translation, and
welcome volunteers who wish to translate to other
languages.  
Fig 1. INTERVIEW BUILDER
Fig 2 and 3. QUESTIONS AND RESPONSES
Fig 4. INSTRUCTIONS, SECTION SELECTOR PROGRESS
BAR
INTERVIEW CONTENT Most questions require a
simple Yes or No response (single click).
Dont Know, Dont Understand or Skip
options are also offered. They give more
information to the physician than an omitted
response, and help the designers reword a
question or redesign the PAI as necessary.
Positive responses usually lead to follow-up
questions that explore the issue in more depth.
For example, answering Yes to the question
Have you ever had chest pain? leads to inquiry
about the cause of the chest pain (Fig 2, 3). A
failure to answer is recorded and reported, but
the system does not compel a response. This may
limit the systems potential for providing more
complete data than a paper questionnaire and will
be changed in the next phase of this project. An
indicator bar graphically charts the patients
progress towards completion. Patients can skip to
different parts of the interview rather than
answer each series of questions sequentially (Fig
4). At the end of the interview, the patient can
review the responses, change them, or supply
missing or skipped answers. The patient then
prints the interview results. The output
reproduces the questions in their entirety one
per line and then shows the patients response.
A question that has not been answered has a blank
in the response area.
REFERENCES 1 Bachman JW. The patient-computer
interview a neglected tool that can aid the
clinician. Mayo Clin Proc. 20037867-78 2
Vitkun SA, Gage JS, Anderson DH. Computerization
of the preoperative anesthesia interview.
International Journal of Clinical Monitoring and
Computing 19951271-76. 3 Lutner RE, Roizen
MF, Stocking CB. The automated interview versus
the personal interview Do patient responses to
preoperative health questions differ?
Anesthesiology 199175394-400