Bringing Clinicians
together to
IMPROVE

Primary Care!

Overview

The Dartmouth Primary Care Cooperative Information Project (“CO-OP Project”) involves a collaborative research network of 233 primary care clinicians in affiliation with the Department of Community and Family Medicine at Dartmouth Medical School. Recognizing the need for practical ways of assessing patients’ functional status, the group set out to design a set of measures that would not only produce reliable, accurate, easily interpretable, and clinically useful data on a core set of functional dimensions, but would prove itself manageable and efficient in the context of a busy office practice as well.  The result, dubbed the CO-OP Chart System (Pronounced “co-op”), is intended to position the screening of patient functional status among the standard repertoire of vital assessments (such as height, weight, and blood pressure) used in routine practice.

At the heart of the CO-OP System are nine scales, each of which is used to measure a different aspect of patient functional status.  Each scale is presented in the form of a chart that is designed to screen patient functional status in much the same way as Snellen eye Charts are used to screen for vision problems.

Four of the CO-OP Charts focus on specific dimensions of function (physical endurance, emotional health, role function, and social function), three relate to overall well-being (overall health, change in health, level of pain), and two are concerned with quality of life (overall quality of life, and social resources/support).

Each Chart consists of a simple title, one question, and five response choices.  Each possible response is described in words and presented graphically, as a caricature, along a five-point ordinal scale.  High-numbered responses represent unfavorable levels of functioning on all charts.

The Charts can be administered by an interviewer (clinician, office staff), or they can be patient self-administered (paper and pencil task).  Completion has proven to be a user-friendly task; the graphics are engaging, and the instructions are easily understood.  The Charts may be used selectively, and each requires about 30-45 seconds to complete.  Patient answers, or “scores,” on each Chart are simply recorded as-is in the patient’s medical record, on a flow-sheet, or in the progress note for that visit; no scoring algorithm or computing is necessary.

Several non-English versions of the Charts are available; among them are Japanese, Hebrew, Dutch, German, Finnish, and Spanish translations (see box).  These non-English Charts are being developed and tested by a group associated with WONCA (World Organization of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians).  Dr. Jack Froom of SUNY at Stony Brook is a U.S. representative on the WONCA classification.  Meetings to evaluate the Chart translations are being held annually.  Proceedings from the most recent (May 1991) meeting in Holland, as well as copies of all existing translations, are forthcoming in a book being prepared by the Dutch government.

Instrument Evaluation
Reliability and Validity

The CO-OP Chart System has undergone a considerable amount of psychometric testing, particularly as applied to older adult patients in ambulatory care clinic and private office practice settings.  Though firmly establishing the validity of a single-item measure is inherently problematic, developers of the CO-OP Chart system claim that it scores as well on reliability and validity tests as some multidimensional measures.  Additional testing is underway to evaluate the Charts in representative samplings of patients and clinical settings, and to determine if the Charts are sensitive to detecting clinically significant changes in individual patients over time.

Strengths

The greatest advantage of the CO-OP Chart System is its brevity.  The Charts are also simple, easily administered, self-scoring, non-disruptive in routine clinical settings and, because of their use of pictures, useful for patients with limited English or education.

The primary purpose of the CO-OP Charts seems to be to initiate and/or enhance patient-clinician communication.  Such communication can reveal functionally important problems that the clinician might have otherwise missed, hopefully triggering beneficial changes in patient management as a result.

Limitations

Though it is efficient and attractive for use in clinical practice settings, the CO-OP Chart system is also limited in important ways.  While the Charts may, indeed, score as well on reliability and validity tests as other, more well-established measures, the CO-OP Charts will most assuredly lack the sensitivity of a longer instrument.

As a result of this lack of sensitivity, it may be difficult to meaningfully pool Chart-generated data for broad-based analyses of such things as disease progression, the effects of different therapeutic approaches, process-outcomes relationships, overall quality of care, and health care policy.

Finally, as with most instruments, especially those using an interviewer, data reliability may be compromised if the Charts are not administered in a standard way.

In sum, what the CO-OP System gains in brevity it loses in specificity.  Other measures may be better suited for determining in the aggregate which functions, and which patients, are affected under any given set of clinical circumstances.

User Experiences

New London Medical Center.  Dr. Jack Kirk is a physician in private practice in New Hampshire’s New London Medical Center.  He is a member of the Dartmouth Primary Care CO-OP Project.  Dr. Kirk has been using the CO-OP Charts in his practice on a regular, though selective, basis.  He has found the Charts to be helpful, particularly with older patients who have complicated histories and /or who are having trouble communicating.  “The Charts can reveal limitations that a clinician might otherwise miss,” he said, “especially in cases when the physician is unsure about all that a patient might be experiencing.”

In Dr. Kirk’s practice, the Charts are clinician-administered, a process that takes 1-2 minutes.  Responses are recorded in patient medical records.  Dr. Kirk guesses that his practice uses the Charts on about 1-2 patients a month.  “We haven’t had a need to use the Charts more often than that,” he said.  “As a closed practice group, we generally see the same patients we’ve been seeing for years and have gotten to know reasonably well.  I guess we would use the Charts more frequently if we were seeing more new patients.”

The Chicago Institute of Neurosurgery.   James Grutsch, Ph.D., has initiated two projects using the CO-OP Charts at the Chicago Institute of Neurosurgery (not affiliated with the Dartmouth CO-OP Project).  One project focuses on patients with chronic back pain, the other on patients with malignant glioma.

Dr. Grutsch is very optimistic that the CO-OP Charts will tell a great deal about how well patients with chronic back pain function physically and emotionally.  He views the charts as “..a robust set of indices that can be used before, during, and after surgery to assess how fast patients snap back.”  He expects that his group will use the Charts to assess the status of as many as 500 patients with chronic back pain by the end of 1991.  Dr. Grutsch plans to use these CO-OP Chart-generated data to compare the effects of various treatments on patient well-being.

The second project involving CO-OP Charts at the Chicago Institute aims to assist physicians in the functional assessment of patients with malignant glioma.  Although this condition is invariably fatal, patients at the Chicago Institute have been exceeding the expected life span, and sometimes functioning reasonably well in between surgeries.  Chart results in this project will be used to help assess functional status on a case-by-case basis.

The plan is that all nine CO-OP Charts will be self-administered in both of these projects, with clinic staff offering assistance on an as-needed basis.  Charts will be re-administered to each patient four times in the first year following treatment as a regular part of patient care, and twice each year thereafter.  Results will be entered into a computer database separate from the computerized medical record.

At this juncture, Dr. Grutsch does not anticipate any problems using the CO-OP charts.  “They cross language barriers, are simple, theoretically sound, and have proven to be reliable and valid,” he said.  Further, he noted that the CO-OP Chart developers had provided “good guidelines” on how to statistically analyze the Chart results.

Selected Case Studies

The Dartmouth Cooperative reports that the clinical utility of the CO-OP Charts can best be documented by case study examples of their use in clinical practice.  The following four case studies, excerpted from Dartmouth Cooperative materials, begin to provide evidence of the CO-OP Charts’ sensitivity to changes in individuals’ health.

Case 1:  An elderly female patient under treatment for congestive heart failure and hypertension and who was known well by her internist was given the Physical Function Chart.  The patient indicated her activity level was 4 or “very light.”  The surprised physician had assumed that the patient was functioning at a substantially higher level.  He questioned the patient and discovered that she had major limitations caused by her arthritis, which heretofore had gone unrecognized.

Case 2:  A young man in his late twenties was seen by a CO-OP clinician for the first time.  The patient had no known chronic problems, but he complained of severe chest pains.  His Chart scores indicated that he was able to do “very heavy” physical activities, though he had “some difficulty” with daily work and experienced “moderate” pain.  His emotional function had been “extremely” bothersome to him, though he had “quite a bit” of social support.  The clinician noted that the Charts possessed “high” utility at this encounter and provided valuable information for both the doctor and the patient.  although the patient had experienced trouble at home, he was unaware that those problems might be the cause of his chest pains.  The Charts, reported the clinician, “quantified the importance of emotional stress to the patient.”

Case 3:  An elderly female patient under treatment for hiatal hernia fractured her ribs in a fall.  The clinician had anticipated functional limitations but was “considerably surprised” by the scores she reported two weeks later.  Her physical, emotional, and role scores were all between 4 and 5 (poor), yet her social function score was 2 (slightly affected).  He questioned the patient and discovered that prior to the fall she “had been putting on a brave face for the doctor” and was, in fact, “markedly depressed.”

Case 4:  A young woman, 27 years of age, was under treatment for diabetes and kidney problems.  The clinician viewed her overall health “favorably,” however, the patient indicated substantial dysfunction in all areas (physical, social function, overall condition, and quality of life).  The Charts provided the clinician with a more accurate profile of the patient’s health.

Pertinent References
Nelson E. Wasson J, Kirk J, Keller A, Clark D, Dietrich A, Stewart A, Zubkoff M. Assessment of function in routine clinical practice: description of the CO-OP Chart method and preliminary findings. J Chron Dis, 1987, 40 (supplement 1):555-635.

Nelson E, Landgraf J, Hays R, Wasson J, Kirk J. The functional status of patients: how can it be measured in physicians' offices? Medical Care 1990, 28(12):1111-1126.

Dartmouth CO-OP Charts

Note: This is a conceptual outline only. It is not a usuable form of the instrument.