ORIGINAL RESEARCH
LESSONS FROM AN ACUPUNCTURE TEACHING CLINIC: PATIENT
CHARACTERISTICS AND PAIN MANAGEMENT EFFECTIVENESS
Kai-Yin Hsu, MAOM, MS-PREP, Lic Ac, Dip. OM, BSN,1# Julie E. Dunn, PhD,2 Ylisabyth S. Bradshaw, DO, MS,3
and Lisa Conboy, MA, MS, ScD1,4
Objectives: To understand the following about patients
using an acupuncture teaching clinic: (1) sociodemographic
characteristics and main complaints and (2) self-reported level
of patient-centered outcomes regarding pain management.
Methods/Design: Retrospective chart review.
Subjects: A total of 458 new patients at NESA clinic during
October 1, 2009 to July 31, 2010 were enrolled in the study.
Interventions: A variety of styles of Oriental medicine, primar-
ily Chinese and Japanese style acupuncture and also heat
treatments (MOXA or heat lamps) and Tui Na (Asia bodywork).
Results: Results from Objective 1 were descriptive (n ¼ 421).
Objective 2 focused on the 59 patients from the larger sample
who completed both an initial and a follow-up Measure
Your Medical Outcome Profile (MYMOP) form and who
used acupuncture for pain management of (1) diseases of the
musculoskeletal system and/or connective tissue or (2) migraine/
headache. Both the symptom severity and activity of daily living/
well-being scales of the MYMOP showed over 15.8% improve-
ment from baseline to at least six weeks of treatment: 28.6% for
Symptom 1, 19.4% for Symptom 2, 35.7% for activities of daily
living, and 25.0% for well-being. The relative majority for each
sociodemographic trait investigated were as follows: female,
INTRODUCTION
Acupuncture is one of the most commonly used comple-
mentary and alternative medicine (CAM) interventions/
1 Department of Research, New England School of Acupuncture,
150 California Street, Newton, MA 02458
2 Friedman School of Nutrition Science and Policy, Tufts University,
Boston, MA
3 Pain Research, Education and Policy Program (PREP), Department
of Public Health and Community Medicine, Tufts University,
School of Medicine Boston, MA
4 Beth Israel Deaconess Medical Center, Harvard Medical School,
Boston, MA
Funding: This project was partially funded by the grant “New
England School of Acupuncture–Harvard Acupuncture Research
Collaborative” (NCCAM U19 AT002022-01)
# Corresponding author:
e-mail: khsu@nesa.edu
284 & 2014 Elsevier Inc. All rights reserved.
ISSN 1550-8307/$36.00
about 40 years old, white, not Hispanic or Latino, married,
highly educated, and employed. Most patients were confident in
acupuncture treatment. Out of the 421 acupuncture patients,
68.2% wanted acupuncture in order to manage pain. Overall,
20.6% of the patients (59, N ¼ 287) who used acupuncture for
pain management for diseases of the musculoskeletal system and/
or connective tissue or migraine or headache completed the
sixth-week follow-up MYMOP form. Of these patients, 57.6%
(34, N ¼ 59) returned during week 6 of the semester for
acupuncture treatment and reported clinical improvement in at
least one MYMOP severity score, and no score got worse.
Conclusions: The information about sociodemographic char-
acteristic and patient-centered outcomes of pain management
can be used for service provision, future study planning, and
marketing. Future studies should address the low follow-up rate,
the quality of self-reported clinic data, and the reasons that
patients chose acupuncture treatments and teaching clinics.
Key words: Acupuncture, complementary medicine, pain
management, medical education and training, pain research,
patient-centered outcomes
(Explore 2014; 10:284-293 & 2014 Elsevier Inc. All rights
reserved.)
therapies according to the 2002 National Health Interview
Survey (NHIS)1 and is becoming more widely accepted by the
American public as well as federal policy makers.2–4 In the
United States, acupuncture is used in many medical special-
ties, for example, complementary and integrative care in
oncology,5 pain management,6,7 public health drug treatment
and community-based HIV care,8,9 and primary care.10,11
Patients who use acupuncture in hospitals or CAM clinics in
the United States are likely to be females who are between the
ages of 18 and 59 years, suffer from pain/musculoskeletal
complaints, and report reduction in pain and improvement in
quality of life.12–14
Many medical research and treatment leaders advocate
patient-centered care and patient-centered outcomes res-
earch.15,16 Since traditional Chinese medicine (TCM) is natu-
rally practiced in an individualized, patient-centered manner, it
fits well into this focus. However, relatively little infor-
mative data has been obtained to date on the characteristics
and patient-centered outcomes of users of TCM teaching clinics
EXPLORE September/October 2014, Vol. 10, No. 5
http://dx.doi.org/10.1016/j.explore.2014.06.004
in the United States. This retrospective chart review, which was
completed at a TCM teaching clinic, the New England School
of Acupuncture (NESA), reports patients' sociodemographic
characteristics and self-reported level of pain and functioning
across a variety of domains. This article also reports symptom
changes using a patient-centered measure used in a TCM
teaching clinic, offering evidence of treatment effectiveness that
can be used for health systems integration and support of
TCM use in the community. As acupuncture continues to
become a more common treatment option, information on use
patterns and patient-centered outcomes in teaching clinics by
observational study can inform the decisions of acupuncture
consumers, healthcare providers, and investigators.
OBJECTIVES
We chose two aspects to study. Objective 1 was to describe
the sociodemographic characteristics and self-reported main
complaints of all new patients (n ¼ 421) at the NESA main
clinic who received acupuncture treatments from a sample
time period: October 1, 2009–July 31, 2010. Objective 2 was
to describe the self-reported level of symptom severity and
functioning across a variety of domains by examining “Meas-
ure Your Medical Outcome Profile” (MYMOP) forms13 for a
subgroup of patients (n ¼ 59) from Objective 1 who used
acupuncture for pain management of diseases of the
musculoskeletal system and/or connective tissue or migraine
or headache. We chose to focus on pain management because
the 2002 National Health Interview Survey12 data indicates
that the most common reasons for seeking acupuncture
treatment in the United States are to relieve low back pain
(34%), joint pain (16%), neck pain (14%), and headache/
migraine (10%). In addition, the 2007 National Health
Interview Survey17 has shown that the number of patients
who have ever used acupuncture increased from 4.2% to
6.3% of the population.
METHODS/DESIGN
We extracted study data from extant patient charts. Self-
reported health history forms (HHFs) and MYMOP forms
were (and still are) regularly completed as part of standard care
at NESA clinics. We abstracted patient data for this project
from the health history and MYMOP forms (see Data
Collection). To analyze patient demographics for Objective
1, we used HHFs from a complete survey of all subjects
meeting eligibility requirements (see Subjects). We described
patients by the following sociodemographic variables: gender,
age, race, ethnicity, marital status, education, and employment
status. We also recorded the following variables of acupuncture
treatment patterns: main complaint, prior exposure to
acupuncture and Chinese herbal medicine, level of con-
fidence in acupuncture treatment, total number of visits in
the study period, treatment duration (days between first and
last treatment), acupuncture style received in the teaching clinic
(Chinese or Japanese), whether Chinese herbal medicine
treatment was also received for the main complaint, and self-
reported main complaints.
Lessons from an Acupuncture Teaching Clinic
To investigate self-reported clinical outcomes for Objective
2, we compared initial and follow-up MYMOP forms for the
subset of patients seeking pain management who had com-
pleted an initial and at least one follow-up MYMOP form
over the study observation period. We examined changes in
symptom severity, physical activity, and well-being scores by
comparing initial and follow-up MYMOP scores.
SETTING/LOCATION
NESA is the oldest acupuncture school in the United States;
this project was completed at NESA's main teaching clinic in
Newton, MA. Annually, about 400 NESA interns give over
10,600 treatments for a variety of conditions at NESA's main
teaching clinic. NESA's teaching clinics are learning environ-
ments where student interns treat patients under the direct
supervision of experienced and licensed acupuncturists.
NESA is committed to making acupuncture and Oriental
medicine affordable and accessible to the greater public,
which results in a heterogeneous patient population in its
clinics that is more representative of the surrounding area
than if treatments were full-cost. Patients receive Oriental
medical care at affordable prices while also participating in
the education and training of future acupuncturists. Patients
choose from Chinese or Japanese acupuncture styles, Chinese
herbal medicine, or a combination approach.
DATA COLLECTION
Patients completed the health history and initial MYMOP
forms before their first treatment at the NESA clinic. Interns
(students in their final year of a 3.5-year training program)
provided their subsequent acupuncture care under close
supervision of a senior practitioner. Following an extensive
intake interview, patients could receive treatment using a
variety of techniques of Oriental medicine, including acu-
puncture, Chinese herbs, heat treatments (MOXA or heat
lamps), and massage. The interns discussed each of these
techniques with subjects prior to any treatment. The clinic
offered patients private treatment rooms. The treatment plans
are individualized, patient-centered, and holistic inter-
ventions.
Generally, subjects saw the same intern or intern pair
(interns sometimes work in pairs if a time slot is popular)
each time. Within the first two appointments, interns recom-
mended a treatment plan of specific dose and duration. To
facilitate scheduling, the front desk staff tried to schedule
appointments for the same day of the week, at the same time,
until the intern-treater determined that it was time to dismiss
subjects from treatment. Ideally, subjects would come in once
a week, but sometimes they would skip some treatments or
stop treatment altogether. Subjects completed follow-up
MYMOP forms when their subsequent visits fell on the sixth
and the 12th week of every semester.
Measure Your Medical Outcome Profile
The MYMOP14 is a patient-centered outcome questionnaire,
first published in the British Medical Journal in 1996.
Originally designed to measure the effects of CAM, use of
EXPLORE September/October 2014, Vol. 10, No. 5 285
this evaluation tool in primary care has become popular in
both biomedicine and CAM literature.13,18–20 It is problem-/
symptom-specific, asking patients to report two symptoms
that are the most bothersome and to rate the severity of those
two symptoms. The MYMOP form requires the respondent,
with some guidance, to specify one or two symptoms that
concern them the most and for which they are seeking
treatment and one activity of daily living that is restricted
or prevented by these symptoms. The respondent then scores
these items on a seven-point scale, 0–6, with 6 being “as bad
as it could be” and 0 being “as good as it could be,” according
to perceived severity in the last week. Subjects also score their
general feeling of well-being over the last week on a similar
scale of 0–6.
The MYMOP form is applicable to all patients who present
with physical, emotional, or social symptoms. It is brief and
simple to administer. Follow-up forms are completed without
help, and the original health concerns are asked about again.13
NESA began implementing the MYMOP form in teaching
clinics in 2008. At follow-ups, the patients are given a new
MYMOP form that already contains the two symptoms given
previously and are asked to rate the severity of those two
symptoms, allowing for the evaluation of changes in the
patients' top two symptoms over time. NESA chose to collect
follow-up MYMOP forms for subsequent (not initial) visits
falling on the sixth and the 12th week of every semester.
Trained work-study students regularly entered the data
from all the HHFs and MYMOP forms into an analyzable
database. A research assistant with an RN license mapped the
HHF complaints and MYMOP symptoms to ICD-9-CM
codes.21 The database was then queried to select relevant data
for Objectives 1 and 2.
Important Note About HHF and MYMOP
In addition to a requested MYMOP form, patients are also
required to fill out a HHF, which collects a more complete
medical history; the interns use this form to determine diagnosis
and treatment. A part of this form requests a main and a
secondary complaint for seeking treatment. For any particular
patient, the main and secondary complaints gathered by the
HHF did not always overlap with the MYMOP forms' first and
second symptoms, henceforth referred to as the HHF Main and
Secondary Complaints and MYMOP Symptoms 1 and 2,
respectively. For example, a patient might say his or her HHF
Main Complaint is lower back pain, whereas their MYMOP
Symptom 1 is weakness in the right leg. This distinction between
Complaints and Symptoms is explained further in Discussion.
We chose to use HHF Main and Secondary Complaints to
describe and categorize our sample for Objective 1 because
these complaints were what motivated the patients to seek
treatment. We also used this HHF data to select the subsample
for Objective 2. At that point, however, we used MYMOP
Symptoms 1 and 2 to consider changes due to treatment
because these MYMOP items offer a quantitative severity rating.
SUBJECTS
The population for Objective 1 was all patients meeting our
four eligibility requirements: (i) who were new patients at the
286 EXPLORE September/October 2014, Vol. 10, No. 5
NESA clinic, (ii) who received acupuncture treatments, (iii)
who were in the observation period: October 1, 2009–July 31,
2010, and (iv) who completed HHFs (N ¼ 421) (Figure 1).
The study population for Objective 2 included all patients
from Objective 1 whose (i) records included an initial and at
least one follow-up MYMOP form and (ii) HHF Complaints
indicated they wanted pain management for (1) diseases of
the musculoskeletal system and/or connective tissue or (2)
migraine or headache (Figure 1).
Confidentiality was carefully preserved. The study was
approved and oversight was provided by the New England
Institutional Review Board (NEIRB). Earlier stages of the
study were also approved by the Tufts University Health
Science Campus' IRB. Subjects could choose to skip parts or
all of the HHF and MYMOP forms if they wished not to
provide the information.
STATISTICAL ANALYSIS
Descriptive statistics were generated for both objectives using
EXCEL 2003 (Microsoft Corp.) and SPSS 22 (IBM Corp.).
Initial (baseline), week six, and week 12 MYMOP scores
(Symptoms 1 and 2, activity of daily living, and well-being)
were summarized by their means and standard deviations. We
calculated changes of each MYMOP score by taking the sixth-
and the 12th-week scores and subtracting the corresponding
initial (baseline) scores, so that a negative change denotes
improvement. Each MYMOP score's percentage change is
calculated by taking mean changes divided by corresponding
initial mean scores. Paired t-tests were used and assessed by
SPSS 22 for comparing the sixth- and the 12th-week scores to
initial MYMOP scores. The change in medication use was
calculated for patients who had recorded this data. Mapping
to ICD-9-CM codes was completed prior to any statistical
analysis, to avoid bias in the mapping.
RESULTS
Figure 1 is a flowchart showing attrition.
Characteristics of Patients at Objective 1 (baseline)
Payment records identified 458 acupuncture patients as poten-
tially new to the NESA main clinic during the study period. In
this survey, 14 patients turned out not to be seen in the study
period, 18 patients were missing entire medical records, and five
patients were missing entire HHFs. Therefore, 421 patients were
included in the final Objective 1 sample (Table 1).
Subjects were mostly female (72.0%). Their mean age was
42.5 years and median age was 39.0 years. Subjects were
predominantly white (82.4%), not Hispanic or Latino
(49.4%), and married (41.6%). Almost half (47.5%) of the
respondents had received at least college education and most
were employed (78.9%). A sizeable minority (46.1%) had
prior exposure to acupuncture or Oriental medicine. Almost
four-fifths (79.1%) had moderate or greater confidence in
acupuncture treatment. Regarding treatment styles, 57
patients (13.5%) received Japanese acupuncture styles (JAS),
296 patients (70.3%) received Chinese acupuncture styles
(CAS), and 68 patients (16.2%) received both JAS and CAS
Lessons from an Acupuncture Teaching Clinic
 Figure 1. Chart selection for Objective 1 and Objective 2.
(Table 2). The mean duration of treatment was 52.7 days,
with a range of zero (one treatment only) to 274 days. The
median duration of treatment was 28 days, but there was a
sharp drop off after three weeks (Table 3).
Using the ICD-9-CM coding system, the top six main
complaints for acupuncture visits were the following: Diseases
of the musculoskeletal system and connective tissue; mental
disorders; symptoms, signs, and ill-defined conditions; nerv-
ous system and sense organs; digestive system; and genito-
urinary system (Table 4).
Lessons from an Acupuncture Teaching Clinic
Treatment Outcomes (Objective 2)
Of the 421 patients, 287 indicated in HHF Main and/or
Secondary Complaints that they sought pain management for
diseases of the musculoskeletal system and/or connective tissue or
migraine or headache. Only 59 patients completed both an initial
and at least one follow-up MYMOP form. Records obtained
from these patients were used to compare pre- and post-treatment
MYMOP scores. (However, recall from Important Note about
Outcome Measures: do not conflate HHF Main and Secondary
Complaints with MYMOP Symptoms 1 and 2.)
EXPLORE September/October 2014, Vol. 10, No. 5 287
Table 1. Baselines and Follow-Up Characteristics of Study Patients Table 1 (continued)
Undergoing Acupuncture Treatment
Characteristic N ¼ 421 % N ¼ 59 %
Characteristic N ¼ 421 % N ¼ 59 %
Missing 17 4.0% 3 5.1%
Gender
Female 303 72.0% 46 78% Prior exposure to
Male 118 28.0% 13 22% acupuncture or
Oriental medicine
Age (mean, 42.46 y/o 39 y/o 47.84 y/o 51 y/o Yes 194 46.1% 27 45.8%
median) No 220 52.3% 31 52.5%
0–20 17 4.0% 2 3.4%
21–30 112 26.6% 10 16.9% Patients' confidence
31–40 88 20.9% 11 18.6% level for TCM
41–50 65 15.4% 6 10.2% treatment
51–60 73 17.3% 15 25.4% Not confident 9 2.1% 1 1.7%
61–70 45 10.7% 11 18.6% Slightly 45 10.7% 7 11.9%
71–100 19 4.5% 4 6.8% confident
Moderately 131 31.1% 18 30.5%
Race confident
American Indian 1 0.2% 0 0% Confident 141 33.5% 22 37.3%
or Alaska native Very confident 61 14.5% 10 16.9%
Asian 26 6.2% 2 3.4%
Black or African 11 2.6% 3 5.1% Use herbs for main
American complaint
Native Hawaiian 1 0.2% 1 1.7% Yes 108 25.7% 17 28.8%
or other Pacific No 310 73.6% 42 71.2%
Islander For other 3 0.7% 0 0%
White 347 82.4% 52 88.1% complaint
Note: Totals may not add up to 421 or 59 because of missing data.
Ethnicity
Hispanic or 24 5.7% 1 1.7%
Latino Table 5 shows treatment outcomes; totals are less than 59
Not Hispanic or 208 49.4% 31 52.5% because when we reviewed these 59 MYMOP forms, we
Latino found missing data in some survey questions. In order to gain
better information, Table 5 breaks down MYMOP Symptoms
Marital status 1 and 2 to two ICD-9 categories and a catchall for everything
Married 175 41.6% 26 44.1% else: diseases of the musculoskeletal system and/or connective
Never married 157 37.3% 16 27.1% tissue; migraine or headache; and all other ailments, such as
Widowed 15 3.6% 1 1.7% stress, vertigo, cough, fatigue, chest tight, and constipation.
Divorced or 61 14.5% 13 22% Table 1 includes the characteristics of these 59 patients.
separated There were no statistical differences (p o .05) on any baseline
measure between patients qualifying for Objective 2 and the
Education 228 patients dismissed from Objective 2.
Grammar 10 2.4% 0 0% MYMOP Symptom 1 severity, MYMOP Symptom
school 2 severity, activity of daily living that MYMOP Symptoms
High school 54 12.8% 12 20.3% 1 or 2 prevent or interfere with, and well-being scales all
showed at least a 13.2% improvement from baseline to six
College 200 47.5% 22 37.3%
weeks (Table 5). Four symptoms showed the minimal
Masters 116 27.6% 20 33.9%
important clinical difference13 of 1.4: week six change score
Doctorate 24 5.7% 2 3.4% in Symptom 1—headache or migraine; week 12 change score
in Symptom 1—musculoskeletal pain; and week 12 change
Employment status
scores in Symptom 1—others and Symptom 2—others. The
Employed 332 78.9% 45 76.3% mean score changes in MYMOP Symptoms 1 and 2 from
Unemployed 29 6.9% 5 8.5% baseline to six weeks vary very little across the ICD-9
Retired 41 9.7% 8 13.6% categories. About half of the patients (30, N ¼ 59) were
Disabled 9 2.1% 1 1.7% taking medication for MYMOP Symptom 1 at their initial

288 EXPLORE September/October 2014, Vol. 10, No. 5 Lessons from an Acupuncture Teaching Clinic
Table 2. Style of Treatment by Number of Visit than number of treatments as a control for natural history.
More recently, NESA added a three-week data collection
Number of Visit in 10 Months Patient Numbers (%) (N ¼ 421) point due to the observation that the median treatment
JAS CAS duration is under a month. This new schedule balances
0 296 (70.3) 57 (13.5) improved quality and quantity of data with time spent to
1–6 95 (22.6) 261 (62.0) obtain such data: Collecting a follow-up every third or every
other visit incurs too much bookkeeping overhead and overly
7–12 22 (5.2) 73 (17.3)
frequent follow-ups risk patients remembering or learning the
13–18 6 (1.4) 19 (4.5)
questions.
19–24 2 (0.5) 6 (1.4) The majority of acupuncture patients wanted pain manage-
25–30 0 3 (0.7) ment at this TCM teaching clinic. Baseline MYMOP scores
430 0 2 (0.5) showed no statistical differences between those who com-
JAS: Japanese acupuncture style; CAS: Chinese acupuncture style. pleted a sixth-week follow-up and those who did not (sixth-
week data could be missing because the subject had no
visit. After six weeks, 10 patients reported taking medication treatment that week, stopped coming in, or skipped filling
for MYMOP Symptom 1, but data were missing for seven out the follow-up MYMOP form). However, the Objective
patients. Many symptoms improved with statistical signifi- 2 analysis supports that after six weeks of acupuncture
cance as determined by unpaired Student's t-tests of pre- treatments, most patients reported improvement in MYMOP
versus post-treatment sample means (Table 5). Symptoms 1 and 2 (when each involved the musculoskeletal
system and/or connective tissue), symptom-related activity
limitations, and general well-being. There were too few
patients for us to evaluate changes in MYMOP symptoms
DISCUSSION
about other issues (neither the musculoskeletal system nor
This project offers a clear profile of acupuncture patients at a
connective tissue). However, a low follow-up rate limited
TCM teaching clinic. The relative majority for each socio-
what could be deduced about pain management effectiveness
demographic trait investigated was as follows: female, 40 years
from this study. Of the patients seeking pain management,
old or younger, white, not Hispanic or Latino, married,
their follow-up MYMOP forms show that at least one in 10
highly educated, and employed. Most patients had confi-
(34/287 is 11.8%) patients returned during week 6 of the
dence in acupuncture treatment. Diseases of the musculoske-
semester for acupuncture treatment and reported clinical
letal system and/or connective tissue were the most common
improvement in MYMOP severity scores.
category treated by acupuncture at the TCM teaching clinic.
The patients' sociodemographic characteristics and condi-
Acupuncture therapy had a mean of six visits and a mean
tions seen in our clinic are similar to other descriptive studies
treatment duration of 53 days. The median duration of
of acupuncture use at teaching clinics,22,23 at regional private
treatment was 28 days. Most patients had one acupuncture
practices,24,25 and nationally1,12,17,26: mostly female, mostly
treatment every nine days.
white, mostly with college undergraduate degree or higher,
When NESA introduced the MYMOP forms, we did not
and mostly seeking treatment for diseases of the musculoske-
know the optimal schedule for administering follow-up
letal system and connective tissue. The primary race in our
forms, so we designed collection timing on semesters (which
clinic reflected the race distribution in Massachusetts, which
are used for intern rotations) and collected follow-up
was comprised of 83.7% whites in 2012.27 Our patients' mean
MYMOP forms at the sixth and the 12th week of every
age was 42.5 ! 16.0 years, which is similar to other
semester. We collected MYMOP forms by timeline rather
findings.22,23 Our data shows that 52.3 % of patients had
not previously experienced TCM, which again is similar to
Table 3. Treatment Duration results from other teaching clinics.22,23 Like us, Marx et al.
Duration (days) Patient Numbers (%) measured confidence in TCM treatment on a five-point scale,
(N ¼ 421) but with different categories, complicating comparison.
Almost half (48.0%) of our patients were “confident” or “very
r21 204 (48.5) confident” and 31.1% were “moderately confident” in TCM
22–42 54 (12.8) treatment, whereas Marx et al.23 found that 70.5% of patients
43–63 27 (6.4) were at least “a bit confident” or “confident.”
64–84 42 (10.0) Besides geographic differences in popularity and/or accept-
85–105 18 (4.3) ance of CAM therapies,28 referrals by students are one of the
106–126 12 (2.9) factors promoting teaching clinics.22 Unfortunately, we do
127–147 16 (3.8) not know how patients in our sample found our teaching
148–168 19 (4.5) clinic. The comparatively low cost of our acupuncture
services is probably one factor. Although NESA is not a
169–210 14 (3.3)
community acupuncture clinic, our treatment fee is equal to
211–252 9 (2.1)
or lower than national acupuncture users' average payment
253–294 6 (1.4) each time. The treatment fee is under $40 at NESA teaching

Lessons from an Acupuncture Teaching Clinic EXPLORE September/October 2014, Vol. 10, No. 5 289
Table 4. Top Six Main Complaint for Acupuncture Visits on Health History Form
ICD-9 Major Disease Categories (21)
Diseases of the musculoskeletal system
and connective tissue (710–739)
Mental disorders (290–319)
Symptoms, signs, and ill-defined
conditions (780–799)
Diseases of the nervous system
and sense organs (320–389)
Diseases of the digestive system (520–
579)
Diseases of the genitourinary system
(580–629)
a
Total number of subjects exceeds N ¼ 421 due to multiple self-reported signs and symptoms per person.
Number of
Self-Reported Signs and Symptoms Examples Subjectsa (%)
“Arm tingling,” “sciatic (sic),” “bursitis,” “arthritis,” “fibromyalgia,” and “SLE” 244 (58.0)
(systemic lupus erythematosus)
“Stress,” “anxiety,” “stress-related insomnia [sic],” “reduce craving cocaine 97 (23.0)
(sic),” “hyperactivity,” “attention,” “smoking,” and “metaphysical tension”
“Over eating (sic),” “high cholesterol,” “insomnia,” “weight loses (sic),” 70 (16.6)
“fatigue,” “lack of energy,” “nausea,” “muscular eye coordination (sic),”
“ringing in ears,” and “fecal incontinence”
“Migraines,” “headache,” “Bells palsy (sic),” “bruxism,” “clenching and 57 (13.5)
grinding,” “vertigo,” “Meniere's disease,” “eyes,” “carpal tunnel (sic),” and
“Parkinson's disease”
“Gastro-sour (sic),” “locked jaw,” “TMJ (temporomandibular joint disorder),” 41 (9.7)
“low abdominal function,” “GB (gallbladder),” “wall edema (probably
abdominal wall edema),” “may be passed a stone (sic),” “colitis,”
“constipation,” “IBS ( irritable bowel syndrome),” and “gluten intolerance”
“Fertility (sic),” “menopause,” “prolonged heavy menstrual bleeding,” “PMS 41 (9.7)
(premenstrual syndrome),” “induction of labor,” “UTI ( urinary tract infection),”
“menstrual problem (clotting, cramp, and irregular bleeding) (sic),”
“endometriosis,” “Candida,” “BPH (benign prostatic hyperplasia),” and
“impotence”

clinics, which is low for the Boston area but not uncommon
nationally; from the NHIS 2007 report, 45% of acupuncture
users paid an average of $40 or less for treatments.24 Finding
patients' motivation for choosing this teaching clinic for their
first TCM treatment is an objective in future studies.
Chao et al.24 found that community acupuncture patients
are more likely to receive frequent acupuncture treatments
(defined as 10 or more treatments in the prior 12 months)
compared to national acupuncture users, regardless of age,
sex, income, education, and health status (48% versus 15%).
Frequent use at our clinic appears to be less than for
community clinic clients but more than national users: in
the 10 months of the study period, only 26.8% of patients
had eight or more treatments, but 21.9% patients already had
10 or more.
The MYMOP form was easy to administer, acceptable to
patients and clinic staff, and provided useful and detailed
quantitative and qualitative patient-centered outcome data.
This matches the experience of other studies.29
One discovery we consider very interesting is that the
MYMOP form allowed patients to provide more detailed/
targeted explanations when their HHF Main or Secondary
Complaints showed they sought pain management for diseases
of the musculoskeletal system and/or connective tissue or
migraine or headache. Such patient-centered outcomes allow
patients to describe in their own words the problems they
considered the most bothersome, and patients provided a wide
variety of MYMOP Symptoms 1 and 2 besides pain, such as
poor appetite, anxiety, insomnia, or lower energy. The improve-
ment of scores for the catchall category for MYMOP Symptoms
is intriguing, but hard to interpret without additional data and
further study. Was acupuncture effective against not only pain
but also associated symptoms or were the associated symptoms
relieved without much impact on pain?
Ours is a unique study that not only reports HHF Main
Complaints by category (ICD-9) but also breaks down
MYMOP outcomes by the same categories. (Many studies
use SF-3630 rather than MYMOP forms, which makes
comparisons with their results less than straightforward.)
Some studies used the MYMOP form to measure outcomes,
but neither reported Main Complaints for acupuncture
treatment nor categorized MYMOP Symptoms 1 or 2.31,32
Many studies equated MYMOP Symptoms 1 and/or 2 to Main
Complaints and categorized them (they used their own system,
not ICD-9) but then did not use those categories to break
down scores for MYMOP Symptoms 1 and 2 when reporting
treatment outcomes.23,33,34 Hull et al.35 used insurance billing
codes to categorize patients' Main Complaint for acupuncture
treatment, but they did not break down scores for MYMOP
Symptoms 1 and 2. Since our study breaks down both HHF
Main Complaints and MYMOP Symptoms by ICD-9 codes,
patterns in each category become analyzable or at least
suggestive despite the small sample size.
Another important distinction of our study is our choice for
Objective 1 to categorize not MYMOP Symptoms 1 or 2 but
the HHF Main Complaint, the “main problem you would like
us to help you with.” We also used the HHF Main Complaint
to select our subjects for Objective 2. We believed that the HHF
Main Complaint captures the reasons patients sought acupunc-
ture treatment more accurately than MYMOP Symptoms. So,

290 EXPLORE September/October 2014, Vol. 10, No. 5 Lessons from an Acupuncture Teaching Clinic
Table 5. Clinical Outcomes From 59 Complete MYMOP Forms
MYMOP Scores at Initial and at Follow-Up

Scale: 0 ¼ “As Good As It Could Be”; 6 ¼ “As Bad As It Could Be”

Initial Week Six Week 12 Paired Mean Change Paired Mean Change Scorea
Follow-Up Follow-Up Scorea (Week Six/Initial) (Week 12/Initial)

Score Mean Score Mean Score Mean Meanb Meanb

N (SD) N (SD) N (SD) N (SD) % Change c
N (SD) % Change c

d e
Symptom 1 57 4.2 (1.1) 55 3.0 (1.5) 12 2.2 (1.3) 55 "1.2 28.6 12 "2.0 47.6
(1.6) (1.5)
Musculoskeletal 40 4.3 (1.1) 39 3.1 (1.4)d 6 2.3 (1.5) 39 "1.1 26.2 6 "1.5 39.5
pain (1.6) (1.9)
Headache or 5 4.8 (1.6) 5 3.0 (2.2) 1 3.0 5 "1.8 37.5 1 "3.0 50.0
migraine (1.6)
Others 12 3.9 (0.8) 11 2.7 (1.6)f 5 1.8 (1.1)f 11 "1.3 32.5 5 "2.4 57.1
(1.8) (1.1)

Symptom 2 50 3.7 (1.3) 44 2.9 (1.2)d 10 1.8 (1.3)f 44 "0.7 19.4 10 "1.3 40.6
(1.2) (1.6)
Musculoskeletal 28 3.6 (1.3) 26 2.8 (1.2)f 6 1.8 (1.2) 26 "0.7 20.0 6 "1.0 35.7
pain (1.3) (1.9)
Headache or 1 3.0 1 1.0 0 1 "2.0 66.7 0
migraine
Others 21 3.8 (1.2) 17 3.2 (1.1) 4 1.8 (1.7) 17 "0.6 15.8 4 "1.7 48.6
(1.1) (1.3)

Activity of daily living 48 4.2 (1.2) 43 2.7 (1.5)d 8 2.3 (0.7)f 43 "1.5 35.7 8 "1.5 39.5
(1.6) (1.4)
Well-being 55 3.2 (1.4) 47 2.4 (1.1)d 9 2.7 (1.3)f 47 "0.8 25.0 9 "1.1 28.9
(1.4) (0.9)
Taking medication 30 13 5
Treatment numbers 51 4.6 (2.2) 12 9.2 (3.4)
a
Note: Six- and 12-week paired mean change scores represent change from initial (baseline).
b
Note: Negative change scores indicate improvement.
c
Note: Percentage change calculated from paired mean change score divided by paired mean initial scores.
d
p o .0001.
e
p o .001.
f
p o .05.

despite the smaller sample size, the quality should be better than
expected from just using MYMOP forms.
Although our observational study design does not have the
internal validity of randomized clinical trials, it provides real-
world evidence about the use of acupuncture and preliminary
evidence of its effectiveness for pain management. This study
supplies actionable information to healthcare providers, patients,
and investigators about characteristics of acupuncture patients at
a TCM teaching clinic; further, it illustrates patients' acupunc-
ture treatment experiences for pain management effectiveness.
Limitations
Limitations of this project include the unavoidable aspects of
pilot studies in real-world situations: small sample size,
quality issues for clinic data, and uncontrolled study design.
The small sample size in Objective 2 is especially disappoint-
ing, since we would like to get outcomes broken down into
ICD-9 categories, or even finer distinctions such as major
disease categories in each ICD-9 category, e.g., low back pain
and neck pain. Nevertheless, our results are still interesting
and, we believe, meaningful. Firstly, ours is a singular study
that breaks down both HHF Main Complaints and also
MYMOP outcomes into categories. Secondly, we (sub)cate-
gorize HHF Main Complaints rather than MYMOP Symp-
toms. Thirdly, and finally, our sample is likely representative
of users of teaching acupuncture clinics. The first two points
were already elaborated upon earlier in Discussion. We go
into more detail for the third point below.
Our sample, because we used a census, is likely very
representative of users of acupuncture teaching clinics.

Lessons from an Acupuncture Teaching Clinic EXPLORE September/October 2014, Vol. 10, No. 5 291
Further, our treatment fee schedule is much more reasonable
(about 50.0% lower) than other acupuncture clinics in the
Greater Boston community, which increases potential patient
diversity. Last, while the sample is self-selected, and thus the
percentage of patients who believe in acupuncture may be
higher than the general population, the belief structure of our
sample is probably very similar to populations of users of
other acupuncture teaching clinic services. So again, we
should expect reasonably good results from analyzing our
sample, despite its small size.
Why was our sample so small? A major factor is that we
had a relatively small fraction of patients who completed
follow-up MYMOP forms. Only 59 patients (20.6%) com-
pleted the sixth-week follow-up MYMOP form out of 287
patients who used acupuncture for pain management for
diseases of the musculoskeletal system and/or connective
tissue or migraine or headache. Unfortunately, we do not
have data for those who discontinued treatment prior to six
weeks. Was it because they got better or because they did not?
We need further research to understand this. However, this
low follow-up rate is similar to other observational study
results. For example, in the 2007 National Health Interview
Survey (NHIS), 25.0% of respondents had only one acu-
puncture visit in the previous 12 months and about 43.8%
had two to five visits.17
The apparent richness of the information from MYMOP
Symptoms versus HHF Complaints merits further consid-
eration. On one hand, with a sufficiently large sample size,
perhaps statistical analysis could illuminate what symptoms
patients care about the most, which may not be the obvious
HHF Complaints. On the other hand, with a small sample
size, clinical significance is weak, such as the Objective 2,
because the MYMOP Symptoms were not directly about
pain and span too wide a range. Furthermore, all this
presupposes that the difference between HHF Complaints
and MYMOP Symptoms is usually meaningful, rather
than because many patients misunderstand the MYMOP
instructions.
CONCLUSION
Despite the limitations enumerated in the previous section,
this study demonstrated that systematic data collection of
sociodemographic characteristics and an observational
patient-centered outcomes study based on pain management
could be implemented in a TCM teaching clinic. This type of
information can be used for service provision, future study
planning, and marketing. The following are several of the
important lessons learned:
1. Most patients at NESA are female and/or seek pain
management. So instructors should make sure students
are well prepared to treat issues involving women's health
and/or pain and help those patients manage their
symptoms.
2. Many patients will show up for at most three visits, so
treatment and education should be planned for the biggest
impact for those early visits. We currently collect follow-
up MYMOP forms at the third week within a semester in
292 EXPLORE September/October 2014, Vol. 10, No. 5
addition to the sixth and 12th weeks to better capture
follow-up data.
3. Future research could explore how confidence in CAM
affects outcomes, and whether it means more aggressive
treatment would be welcome and more effective.
4. Future studies should address the low follow-up rate, the
quality of self-reported clinic data, and the reasons that
patients chose acupuncture treatment and teaching clinics.
5. Future work will include a larger, longer-term, and stand-
ardized data collection to look at sociodemographic
characteristics and patient-centered outcomes of acupunc-
ture in TCM teaching clinical settings.
Acknowledgments
We would like to thank Tufts and NESA faculty and
students, NESA patients, Thomas Yan, Lucy Chen, and
Kenneth Kwan Ho Chui.
REFERENCES
1. Barnes PM, Powell-Griner E, McFann K, Nahin RL. Comple-
mentary and alternative medicine use among adults: United
States, 2002. Adv Data 2004(343):1–19.
2. Diehl DL, Kaplan G, Coulter I, Glik D, Hurwitz EL. Use of
acupuncture by American physicians. J Altern Complement Med.
1997;3(2):119–126.
3. Sherman KJ, Cherkin DC, Eisenberg DM, Erro J, Hrbek A, Deyo
RA. The practice of acupuncture: who are the providers and what
do they do? Ann Fam Med. 2005;3(2):151–158.
4. NIH. Acupuncture. NIH Consensus Statement. 1999;15(5):
1-34.
5. Eisenberg DM, Kessler RC, Foster C, Norlock FE, Calkins DR,
Delbanco TL. Unconventional medicine in the United States.
Prevalence, costs, and patterns of use. N Engl J Med. 1993;328
(4):246–252.
6. Deng G, Cassileth BR. Integrative oncology: complementary
therapies for pain, anxiety, and mood disturbance. CA Cancer J
Clin. 2005;55(2):109–116.
7. Ernst EM, Adrian W, Brian B, Kelly F, Edzard. Meta-analysis:
acupuncture for low back pain. Ann Int Med. 2005;142(8):
651–663.
8. Kolenda J. A brief history of acupuncture for detoxification in
the United States. Acupuncture Today. 2000;01(09). Available at:
http://www.acupuncturetoday.com/mpacms/at/article.php?
id=27686.
9. Porter KS, Beth. Acupuncture: Part of the Public Health
Equation. Acupuncture Today. 2004;05(03). Available at: http://
www.acupuncturetoday.com/mpacms/at/article.php?id=28413.
10. Eisenberg MHC, Andrea H, Roger BD, Steven CS, David M.
Emerging credentialing practices, malpractice liability policies,
and guidelines governing complementary and alternative medical
practices and dietary supplement recommendations: a descriptive
study of 19 integrative health care centers in the United States.
Arch Int Med. 2005;165(3):289–295.
11. Highfield ES, Kaptchuk TJ, Ott MJ, Barnes L, Kemper KJ.
Availability of acupuncture in the hospitals of a major academic
medical center: a pilot study. Complement Ther Med. 2003;11
(3):177–183.
12. Burke A, Upchurch DM, Dye C, Chyu L. Acupuncture use in
the United States: findings from the National Health Interview
Survey. J Altern Complement Med. 2006;12(7):639–648.
Lessons from an Acupuncture Teaching Clinic
13. Paterson C. Measuring outcomes in primary care: a patient
generated measure, MYMOP, compared with the SF-36 health
survey. Br Med J. 1996;312(7037):1016–1020.
14. MYMOP. MYMOP—Intro. Available at: http://sites.pcmd.ac.
uk/mymop/index.php?c=intro/; Accessed 31.03.12.
15. Lusk JM. A concept analysis of patient-centered care. Nurs Forum.
2013;48(2):89–98.
16. Washington AE. The patient-centered outcomes research insti-
tute—promoting better information, decisions, and health. N
Engl J Med. 2011;365(15):e31.
17. Zhang Y, Lao L, Chen H, Ceballos R. Acupuncture use among
American adults: what acupuncture practitioners can learn from
National Health Interview Survey 2007? Evid Based Complement
Altern Med. 2012;2012:710–750.
18. EuroQol. EuroQol—a new facility for the measurement of
health-related quality of life. Health Policy. 1990;16(3):199–208.
19. Paterson C. Complementary practitioners as part of the primary
health care team: consulting patterns, patient characteristics and
patient outcomes. Fam Pract. 1998;14(5):347–354.
20. Paterson C, Britten N. Acupuncture for people with chronic
illness: combining qualitative and quantitative outcome assess-
ment. J Altern Complement Med. 2003;9(5):671–681.
21. ICD—ICD-9-CM—Addenda, Conversion Tables, and Guidelines;
2013.
22. Maiers M, McKenzie E, Evans R, McKenzie M. Patient outcomes
at a traditional Chinese medicine teaching clinic: a prospective
data collection project. J Altern Complement Med. 2008;14(9):
1083–1088.
23. Marx BL, Rubin LH, Milley R, Hammerschlag R, Ackerman DL.
A prospective patient-centered data collection program at an
acupuncture and oriental medicine teaching clinic. J Altern
Complement Med. 2013;19(5):410–415.
24. Chao MT, Tippens KM, Connelly E. Utilization of group-based,
community acupuncture clinics: a comparative study with a
Lessons from an Acupuncture Teaching Clinic
nationally representative sample of acupuncture users. J Altern
Complement Med. 2012;18(6):561–566.
25. Loh YL, Reilly A, Chen W, Coeytaux RR. Incorporating
acupuncture in a university-based family medicine center: lessons
learned. 〈http://dx.doi.org.ezp-prod1.hul.harvard.edu/10.1089/
acm.2008.0319/.〉; 2009.
26. Arnes PM, Bloom B, Nahin RL. Complementary and alternative
medicine use among adults and children: United States, 2007.
Natl Health Stat Report. Available at: http://phstwlp1.partners.
org:2111/pubmed/19361005/.
27. Massachusetts QuickFacts from the US Census Bureau; 2013.
28. Barnes PM, Bloom B, Nahin RL. Complementary and alternative
medicine use among adults and children: United States 2007.
Natl Health Stat Report 2008(12):1–23.
29. Paterson C, Jason, Joire D. Outcomes of traditional Chinese
medicine (traditional acupuncture) treatment for people with
long-term conditions. Complement Ther Clin Pract. 2010;16
(1):3–9.
30. Ware J, Kosinski M, Gandek B, Snow K. SF-36 Health Survey:
Manual and Interpretation Guide. Boston, MA: The Health
Institute, New England Medical Centre; 1993.
31. Fixler M, Ogden C, Moir F, Polley MJ. Patient experience of
acupuncture provision in a GP practice. Complement Ther Clin
Pract. 2012;18(3):140–144.
32. Robinson N, Julie D, Hilary W. Auditing outcomes and costs of
integrated complementary medicine provision—the importance
of length of follow up. Complement Ther Clin Pract. 2006;12
(4):249–257.
33. Day A, Rosie K-S. An audit of acupuncture in general practice.
Acupunct Med. 2004;22(2):87–92.
34. Stellon A. An audit of acupuncture in a single-handed general
practice over one year. Acupunct Med. 2001;19(1):36–42.
35. Hull SK, Page CP, Skinner BD, Linville JC, Coeytaux RR.
Exploring outcomes associated with acupuncture. J Altern Com-
plement Med. 2006;12(3):247–254.
EXPLORE September/October 2014, Vol. 10, No. 5 293