November 2001
Volume 65 |
Number 11
|
| |
|
| Outcomes Measures
in Pain Medicine |
David P. Martin, M.D., Ph.D.
Committee on Pain Medicine
How well do we help our patients who have chronic pain? Are our
treatment interventions effective? Answers to these questions
are not always straightforward, but if we do not measure outcomes,
we cannot improve them. Increasingly, regulatory pressures will
require a more formal approach for assessing pain and measuring
outcomes. For example, the Joint Commission on Accreditation of
Healthcare Organizations will require assessment of pain as the
“fifth vital sign” for all patients admitted to the
hospital. In the future, pain physicians may need to show objective
evidence of the efficacy of their interventions to justify reimbursement
by third-party payers. Additionally, the ability to document good
outcomes and patient satisfaction will enhance the marketability
of pain services.
Determine What to Measure and How
Before choosing an outcomes measurement tool, it is essential
to specify exactly what is to be measured. For example, one can
measure subjective symptoms of pain, objective behavioral consequences
or both. Determinations can be made by the clinician, the patient’s
self-report or both. Special attention needs to be given to patient
satisfaction surveys. Patient satisfaction depends a great deal
on the patient’s expectations and mindset upon referral to
the pain clinic. Since a patient’s expectation is generally
not within the control of the pain physician, it can be problematic
to find a consistent measure of satisfaction.
One can measure pain intensity by using visual analog or numerical
rating scales. These rate pain between 0, “no pain,”
and 10, “unbearable or excruciating pain.” The McGill
Pain Questionnaire adds sensory, affective and evaluative dimensions
to the measurement of pain intensity. Scales showing pictures
of faces representing different degrees of distress can be used
to assess pain intensity in children.
Because pain is inherently a complex, subjective experience,
pain intensity may not be adequate as the sole outcomes measure.
Physicians often turn to objective measures of functional impact
such as quality-of-life or economic indicators. Also, it is important
to recognize both the physical and psychological dimensions of
pain. Physical dimensions include participation in such activities
as hobbies or working. Psychological impact includes affective
components of mood such as depression.
Third-party payers often value health care utilization as an
outcome. Good outcomes are associated with decreased utilization
of resources as measured by number of office visits, emergency
room and hospital admissions or prescription drug utilization.
It can be difficult, however, to make accurate cost and expense
estimates in today’s health care environment, especially
in capitated systems.
The ideal outcomes tool should be easy to use and brief enough
to allow for high patient compliance. By minimizing the time needed
to complete the survey, patients are more likely to provide their
full attention to the task. The ideal outcomes tool also should
be reliable and validated for the population to be measured. This
is particularly important to consider when assessing outcomes
in children or patients who do not speak English.
Select the Right Survey Tools
One can collect outcomes data through telephone polling, written
questionnaires or direct computer entry. Telephone follow-up often
provides a higher degree of compliance but requires increased
personnel effort. Telephone calls must be repeated frequently
if the patient is unavailable and take more time to conduct and
score. Written questionnaires can be used with less expense, although
the compliance rate is often less than telephone surveys. Written
surveys may provide more accurate assessments because more questions
can be asked, and there is less likelihood that the patient will
be influenced to answer one way or another to meet perceived expectations
of the interviewer. Direct computer entry is a promising option
for patients who have the requisite skills. Patients may enter
their responses directly on a computer in the office, over the
Internet through e-mail or by using portable, handheld electronic
devices.
A variety of outcomes survey tools are available. The ASA Committee
on Pain Management in 1997 developed the ASA-Nine outcomes measures
questionnaire [Table 1].
These questions assess key dimensions of pain and outcomes. Eight
of the questions are answered by the patient (or in the case of
pediatric patients, their parents), and one is answered by the
treating physician. Two other tools that are well-validated include
the health status surveys SF-12® and SF-36®, which contain
12 and 36 questions, respectively.
Decide When to Measure and Who
When should outcomes be measured? It is necessary to obtain a
baseline assessment prior to the initial patient visit to establish
a point of reference for subsequent measurements. If the survey
is given shortly after the patient’s appointment, it may
be too soon for the treatment to have its full effect. Measurement
at some later time such as three or six months may provide a more
accurate assessment of the durability of treatment effect. However,
assessing outcomes at later times can raise several problems.
Inquiring about patients’ symptoms at later times may inappropriately
focus a patient on past issues (e.g., pending litigation, disability).
It is common for patients to write back and request further assessment
of chronic problems when prompted by an outcomes survey.
Measuring outcomes at later times also increases the chance that
intervening events may confound the outcome. For example, the
patient may be involved in an accident between his or her visit
to a pain clinic and an outcomes survey at six months. It would
be inappropriate to attribute a poor outcome to an ineffective
pain treatment when that outcome may be due to the intervening
accident. One way around this is to ask patients if they would
recommend similar treatment to a friend or relative who found
themselves in similar circumstances. This approach asks the patient
to distinguish between the treatment given during the visit and
any intervening confounding events.
Which patients should be surveyed? Ideally, all patients seen
in a practice should be asked to complete an outcomes survey.
If resources do not allow this, it would be possible to survey
a smaller number of patients randomly selected from the practice.
Alternatively, it might also be reasonable to survey only a selected
portion of the practice that has a given diagnosis or treatment.
In any case, it is important to recognize that the outcomes one
observes cannot necessarily be causally attributed to the care
rendered. That is to say, the outcomes will be associated with
the care provided, but without a control group, it is impossible
to prove that the outcomes are attributable to the treatments
given.
|
“Ideally all patients seen in a practice should
be asked to complete an outcomes survey. If resources
do not allow this, it would be possible to survey
a smaller number of patients randomly selected from
the practice.”
|
|
At the Mayo Pain Clinic, we use a written survey that includes
19 questions. The questionnaire is given at the initial patient
visit and mailed at three and six months following the visit.
It is therefore possible to observe the change of physical and
psychological function over time. The forms are scanned and scored
by machine. This information is then correlated with patient demographic
data, diagnoses and treatment. We follow outcomes trends continuously
so that we can recognize either improvements or problems associated
with the treatments we give.
Bibliography:
Bech P. Health-related quality of life measurements in the assessment
of pain clinic results. Acta Anaesthesiol Scan. 1999; 43(9):893-896.
Chapman CR, Syrjala KL. Measurement of Pain. In: Bonica’s
Management of Pain. 3rd ed. Loeser JD, ed. Philadelphia: Lippincott
Williams & Wilkins. 2001:310-328.
Gilbert HC. Developing outcomes measures for pain management.
ASA Newsl. 1997; 65(8):10.
Joint Commision on Accreditation of Healthcare Organizations.
Pain Assessment and Management: An Organizational Approach. Oakbrook
Terrace, IL: Joint Commision on Accreditation of Healthcare Organizations.
2000.
Kaegi L. Medical Outcomes Trust Conference presents dramatic
advances in patient-based outcomes assessment and potential applications
in accreditation. Jt Comm J Qual Improv. 1999; 25(4):207-218.
| |
|
David P. Martin,
M.D., Ph.D., is Consultant, Mayo Clinic Department of Anesthesiology
and Assistant Professor of Anesthesiology, Mayo Clinic and
Mayo Foundation, Rochester, Minnesota. |
|
return to top
|