SPECIAL COMMUNICATION
Self-Reported Health Behaviors of Osteopathic Physicians
Joseph P. McNerney, DO
Steven Andes, PhD, CPA
Deborah L. Blackwell, DO
The degree to which osteopathic physicians (DOs) take care
of their own health is of interest not only to the osteopathic
medical community, but also to physicians’ patients. The
American Osteopathic Association (AOA) Committee on
Physician Health asked attendees at the July 2001
AOA House of Delegates Annual Business Meeting in
Chicago, Ill, to complete a one-page survey on their per-
sonal health practices. This survey comprised 22 questions
on such items as vacation and personal time, exercise habits,
weight control, tobacco and alcohol use, and regular physical
examinations and medical screening. Two hundred ninety-
nine attendees completed the survey during the 15 minutes
immediately after the report of the Committee on Physician
Health (response rate: ~75%). The results indicate that DOs
are similar to the proverbial patient in terms of personal
health practices. Although DOs follow some physicians’
orders, they do not follow others. Most DOs take regular
vacations and daily personal time, and they get some of
their recommended physical examinations and medical
screenings. The authors suggest that DOs pay greater atten-
tion to their exercise habits, weight control, and signs of
substance abuse.
J Am Osteopath Assoc. 2007;107:537-546
From the Touro University College of Osteopathic Medicine in Vallejo, Calif
(Drs McNerney and Blackwell) and the American Osteopathic Association
(AOA) Division of Applied Research (Dr Andes).
At the time of the survey, Dr McNerney was chairman of the AOA Com-
mittee on Physician Health. Currently, Dr McNerney is a member of the AOA
Bureau of Osteopathic Education, chairman of the AOA Program and Trainee
Review Council, and vice chairman of the AOA Bureau on International
Osteopathic Medical Education and Affairs.
Address correspondence to Steven Andes, PhD, CPA, AOA Division of
Applied Research, 142 E Ontario St, Chicago, IL 60611-2864.
E-mail: sandes@osteopathic.org
Submitted January 8, 2003; revision received December 29, 2006; accepted Jan-
uary 18, 2007.
McNerney et al • Special Communication
P revious research has suggested that physicians are an
important group to study regarding personal health-
related beliefs and behaviors.1 Patients, of course, look to
their physicians for health advice and medical care. Gross and
coinvestigators1 found that physicians’ own health habits
influence the preventive health counseling they provide to
their patients. Physicians are also an excellent group to study
for healthcare habits because they generally have access to
high-quality medical care, as well as higher-than-average
education and economic status—eliminating factors that are
known barriers in access to healthcare.
Indeed, various studies2,3 have shown how physician
lifestyle and physician health are linked. In a survey of
1040 family practice physicians in Sweden, Sundquist and
Johansson2 found that physicians with high job strain (ie, low
control of their work environments and high work demand)
exhibited a more than threefold increase in risk of impaired
general health, compared with physicians with medium job
strain. Among male physicians, low job strain was associated
with low risk of impaired health.2 These findings highlight
the need for vigilance on physicians’ working conditions.
In a survey of 298 primary care physicians in the United
States, Abramson and coauthors3 found that physicians who
regularly exercise are more likely to counsel their patients to
exercise––sharing the knowledge that regular physical activity
can reduce the incidence and prevalence of many chronic dis-
eases. The authors determined that inadequate time with
patients and limited physician knowledge and experience
regarding exercise are the most common barriers to effective
patient counseling.3
Gross and coauthors1 reported that a physician’s spe-
cialty may influence his or her approach to patient care and per-
sonal use of preventive health services. In this analysis, a
cohort of 915 physicians was surveyed to determine if they had
a regular source of care (RSOC).1 The results of the analysis
showed that 312 (34%) of the surveyed physicians had no
RSOC, and 60 physicians (7%) reported treating themselves.1
When compared with pediatricians and psychiatrists, internists
(odds ratio [OR], 3.26; 95% confidence interval [CI], 1.58-6.74),
pathologists (OR, 5.46; 95% CI, 2.09-14.29), and surgeons
(OR, 2.42; 95% CI, 1.17-5.02) were significantly more likely
not to have an RSOC.1 Earlier studies4 also demonstrated that
physicians with poor personal health practices, such as tobacco
or alcohol use and lack of exercise and seat belt use, are less
JAOA • Vol 107 • No 12 • December 2007 • 537
SPECIAL COMMUNICATION
likely to provide counsel to their patients about those health
practices.
The osteopathic medical profession has long recognized
the importance of physician health. The American Osteopathic
Association (AOA) established the Committee on Impaired
Physicians in 1987, primarily to aid osteopathic physicians
(DOs) in dealing with issues related to chemical and alcohol
dependence and mental and personal conflict (Resolution 61
[M/1988]—Statement of Purpose of the AOA Committee on
Impaired Physicians). In 1999, the AOA renamed this com-
mittee the Committee on Physician Health and expanded its
responsibilities to include all aspects of physician health,
including personal health practices and lifestyle (Resolution 18
[A/1998]—Change of Name of Committee on Impaired Physi-
cians to Committee on Physician Health). The expansion of the
committee’s duties was based on the belief that threats to DOs’
health include not only behaviors such as tobacco and alcohol
use, but also lifestyle issues such as job stress, vacations, and
amount of personal time (Resolution 18 [A/1998]—Change of
Name of Committee on Impaired Physicians to Committee
on Physician Health).
In the early 1990s, John C. Licciardone, DO, MBA, and
Robert D. Hagan, DO,5 analyzed the physical fitness levels of
first-year osteopathic medical students, concluding that a
“greater emphasis on health promotion in the medical cur-
riculum may help students to adopt more healthy behaviors
and, in addition, encourage them to provide preventive med-
ical counseling to their patients.” In the same issue of JAOA—
The Journal of the American Osteopathic Association, then–AOA
Editor in Chief Thomas W. Allen, DO,6 made the following
assertion:
We have learned that role modeling has a very powerful
effect on others. We physicians can, and do, play a significant
part in teaching our patients healthy lifestyle behaviors. A
physically fit physician sets expectations for patients. Can
we not expect, then, the role modeling effect to be positive?
The November 1999 issue of The DO included an article
titled “Practice what you preach: DOs need to apply preven-
tive medicine to their own lives,”7 which emphasized the
importance of DOs taking care of their own health. The article
quoted Richard B. Tancer, DO,7 a then-member of the AOA
Committee on Physician Health, who noted the following:
[Osteopathic physicians] need to remember the osteopathic
medical tenet that health requires wellness of the body, the
mind and the spirit. And DOs need to remind themselves that
osteopathic principles apply to them as well as their patients.
That same issue of The DO also noted the importance of osteo-
pathic medical students tending to their own health.8
In 2002, Ronald R. Gaber, EdS, and Daniel M.
Martin, MA,9 reviewed the Still-Well osteopathic medical stu-
538 • JAOA • Vol 107 • No 12 • December 2007
dent wellness program at A.T. Still University-Kirksville (Mo)
College of Osteopathic Medicine, emphasizing that practicing
proper health maintenance is integral to osteopathic medicine
and medical education. The article noted that the Still-Well
program’s theme of “I am my own first patient” emphasizes
healthy behaviors and physical exercise for osteopathic med-
ical students and DOs.9 Gaber and Martin9 pointed out, “Little
is known about students’ lifestyle commitment to healthy
behaviors. Despite this lack of information, physicians will
often be responsible for their patients’ attitudes regarding
lifestyle and health.”
Most recently, 2007-2008 AOA President Peter B.
Ajluni, DO,10,11 announced that his “presidency [would be]
focused...on health and fitness.” The theme for his three-point
initiative is “DOs: Fit for Life”:
1. Bringing the same sense of commitment [DOs] have to
serving patients to serving [themselves].
2. Striving to keep the AOA and our state societies and
specialty colleges fiscally fit organizations.
3. Ensuring our nation’s [healthcare] delivery system is
healthy. We must reform the dangerously flawed system
currently in place if we truly want our nation to be fit.10
Methods
To evaluate how well DOs protect their own health, the
AOA Committee on Physician Health developed a self-admin-
istered questionnaire about physicians’ personal health behav-
iors.12 The committee created this survey instrument with
technical assistance from the University of Illinois at Chicago
Survey Research Laboratory and School of Public Health. The
survey consisted of 22 items, including questions about lifestyle,
health-related behaviors, and health-promoting actions. The
questionnaire also requested basic demographic data, including
information on practice characteristics.
To encourage a high response rate, the questionnaire was
distributed to all attendees at the July 2001 AOA House of
Delegates Annual Business Meeting in Chicago, Ill, during
the report of the Committee on Physician Health. Participants
were given approximately 15 minutes to complete the survey
after the presentation of the committee’s report. The data were
coded and analyzed using SPSS statistical software (ver-
sions 13.0 and 14.0; SPSS Inc, Chicago, Ill) for both univariate
and multivariate analyses.
Results
Two hundred ninety-nine DOs participated in the AOA Com-
mittee on Physician Health Survey—approximately three-
fourths of the DOs present at the July 2001 AOA House of
Delegates Annual Business Meeting. However, because not all
participants responded to every survey question, the sample
size for many items is less than 299. Sample size also varies for
survey items within reported characteristics for the same reason.
McNerney et al • Special Communication
 SPECIAL COMMUNICATION

The present article first outlines general trends that were One hundred four survey participants (35%) practiced
observed. Then, important differences in health behaviors are in cities, 123 (41%) in suburban areas, 33 (11%) in small towns,
noted according to demographic and practice characteristics. and 37 (12%) in rural areas. Two hundred two DOs (68%)
Finally, patterns of health behaviors among survey respondents reported their medical specialty as primary care. Two hundred
are analyzed. twenty-seven participants (76%) worked in patient care,
31 (10%) in teaching or research, 10 (3%) were retired, and
Respondent Characteristics 26 (9%) were engaged in other medical activities. Ninety-five
Participants represented a cross-section of the osteopathic DOs (32%) were in solo practice, 137 (46%) practiced medicine
medical community in terms of age, sex, practice location, in partnerships or groups, and 50 (17%) practiced in other
medical specialty, and practice characteristics (Table 1). Forty- settings.
eight survey participants (16%) were younger than 40 years,
110 (37%) were between 40 and 49 years, 85 (28%) were Lifestyle Choices
between 50 and 59 years, and 54 (18%) were aged 60 years or Although DOs tend to take vacations regularly, they are much
older. Two hundred forty DOs (80%) were men; 57 (19%) less likely to fulfill other dimensions of a healthy lifestyle
were women. (Figure). Two hundred sixty-seven survey participants (90%)
reported taking annual vacations, but only 147 (50%) scheduled
Table 1 daily personal time. Only 152 participants (51%) exercised
American Osteopathic Association Committee on Physician regularly, and 166 DOs (56%) were more than 10% over their
Health Survey: Characteristics of Respondents (N=299) recommended body weight.
Although practice characteristics and age had some effect
Characteristic No. (%)
on these lifestyle items, we cannot report, based on the survey

Age, y results, that any specific category of DO consistently leads a
▫ <40 48 (16.1)
healthier lifestyle than any other category of DO (Table 2).
▫ 40-49 110 (36.8)
Nevertheless, DOs in teaching and research (24 of 31 [77%])
▫ 50-59 85 (28.4)
▫ 60-65 26 (8.7) were significantly less likely than other DOs (239 of 263 [91%])
▫ >65 28 (9.4) to take annual vacations (␹2=5.505, P<.019). Osteopathic physi-
▫ Missing data* 2 (0.7) cians in solo practice (38 of 95 [40%]) were significantly less

Sex likely than other DOs (106 of 189 [56%]) to exercise regularly
▫ Men 240 (80.3) (␹2=6.144, P<.013). Similarly, DOs practicing medicine in sub-
▫ Women 57 (19.1) urban settings (53 of 123 [43%]) were significantly less likely
▫ Missing data* 2 (0.7) than other DOs (99 of 174 [57%]) to exercise regularly (␹2=5.398,

Practice Location P<.019).
▫ City 104 (34.8) Osteopathic physicians younger than 40 years (39 of
▫ Suburb 123 (41.1)
48 [81%]) were significantly less likely than other DOs (227 of
▫ Small town 33 (11.0)
249 [91%]) to take annual vacations (␹2=4.232, P<.040), and
▫ Rural area 37 (12.4)
▫ Missing data* 2 (0.7) DOs younger than 50 years (65 of 157 [41%]) were less likely

Medical Specialty than other DOs (81 of 139 [58%]) to schedule daily personal
▫ Primary care 202 (67.6) time (␹2=8.397, P<.004) (Table 2). Osteopathic physicians
▫ Specialty care 70 (23.4) younger than 50 years (78 of 157 [50%]) were significantly less
▫ Other 8 (2.7) likely than older DOs (88 of 139 [63%]) to be overweight
▫ Missing data* 19 (6.4) (␹2=5.559, P<.018). This weight correlation was especially true

Practice Type for DOs younger than 40 years (25%), compared with older
▫ Patient care 227 (75.9) DOs (63%) (␹2=22.470, P<.001).
▫ Teaching/research 31 (10.4)
▫ Retired 10 (3.3)
Risky Behaviors
▫ Other 26 (8.7)
Only 20 participating DOs (7%) reported using tobacco, and
▫ Missing data* 5 (1.7)

Practice Category only 57 (19%) reported consuming “more than 2 ounces of
▫ Solo practice 95 (31.8) spirits, 8 ounces of wine, or 24 ounces of beer in a single day,
▫ Partnership/group practice 137 (45.8) more than once a week” (Table 2). Allowing for an average
▫ Other 50 (16.7) underreporting rate of about 13% for self-reported tobacco
▫ Missing data* 17 (5.7) use13—and assuming the same rate of underreporting for
alcohol consumption—we estimate that approximately 8% of
* Indicates number of osteopathic physicians who did not respond to DOs use tobacco and 21% consume more than the recom-
survey question about that characteristic. mended amount of alcohol.

McNerney et al • Special Communication JAOA • Vol 107 • No 12 • December 2007 • 539

SPECIAL COMMUNICATION

Take Annual
Vacation

Schedule Daily
Personal Time

Regular Exercise*

Overweight by >10%

Tobacco Use

Regular Alcohol
Consumption†

Check Cholesterol and

Blood Pressure‡

Physical Examination,
Age <50 y‡

Colonoscopy‡
Men, Age >50 y

Annual Prostate
Examination

Colonoscopy‡ and
Prostate Examination

Annual Mammogram
Women, Age >50 y

Annual
Papanicolaou Smear

DXA‡

Mammogram,
Papanicolaou Smear,
and DXA‡
0 10 20 30 40 50 60 70 80 90 100

540 • JAOA • Vol 107 • No 12 • December 2007 McNerney et al • Special Communication

 SPECIAL COMMUNICATION

Table 2
American Osteopathic Association Committee on Physician Health Survey:
Responses by Characteristic (N=299)*

Characteristic, %
Age <40 y Age <50 y
Survey Item All Respondents, % Yes (n=39) No (n=227) Yes (n=39) No (n=227)

Lifestyle
▫ Take annual vacation 89.6 81.3 91.2† 90.5 88.5
▫ Schedule daily personal time 49.5 40.4 51.0† 41.4 58.3§
▫ Regular exercise‡ 51.0 62.5 49.0† 53.2 48.9
▫ Overweight by >10% 55.9 25.0 62.5§ 47.9 63.3§

Risky Behavior
▫ Tobacco use 6.8 6.3 6.9 7.0 6.6
▫ Regular alcohol consumption// 19.3 18.8 19.4 19.7 18.8

Physical Examination and Medical Screening¶
▫ Check cholesterol and blood pressure 81.9 60.4 85.9§ 77.2 87.1†
▫ Physical examination, age <50 y 66.4 57.8 NA 70.2 NA
▫ Men, age >50 y
– Colonoscopy 42.6 ... ... ... ...
– Annual prostate examination 76.7 ... ... ... ...
– Both colonoscopy and prostate examination 37.7 ... ... ... ...
▫ Women, age >50 y
– Annual mammogram 73.3 ... ... ... ...
– Annual Papanicolaou smear 73.3 ... ... ... ...
– DXA 66.7 ... ... ... ...
– Mammogram, Papanicolaou smear, and DXA 40.0 ... ... ... ...
(continued)
* Sample size (n) for each characteristic is based on the number of osteopathic physicians who responded to that survey item. For data on men older than 50 years, 123
participants answered survey questions about colonoscopy at age 50 years and annual prostate examination. For data on women older than 50 years, 21 respondents
answered the survey question about mammogram; 20, Papanicolaou smear; 15, dual-energy x-ray absorptiometry (DXA) in the past 3 years. In many instances, data
available on women older than 50 years were insufficient for analysis.
† P<.05
‡ Regular exercise was defined as 30 minutes of physical activity three or more times per week.
§ P<.01
// Regular alcohol consumption was defined as “more than 2 ounces of spirits, 8 ounces of wine, or 24 ounces of beer in a single day, more than once a week.”
¶ All survey participants were asked if they had their cholesterol level and blood pressure checked in the past year. Participants younger than 50 years were asked if
they had received a physical examination in the past 3 years. Other survey questions regarding physical examinations and basic medical screening were age-dependent.

Abbreviations: ellipses (...), data insufficient for analysis; NA, not applicable.

Figure. Percentages of osteopathic physicians who reported engaging
in various health-related behaviors in a survey of osteopathic physi-
cians developed by the American Osteopathic Association (AOA)
Committee on Physician Health and distributed to attendees at the
July 2001 AOA House of Delegates Annual Business Meeting in
Chicago, Ill (N=299). *Regular exercise was defined as 30 minutes of
physical activity three or more times per week. †Regular alcohol con-
sumption was defined as “more than 2 ounces of spirits, 8 ounces of
wine, or 24 ounces of beer in a single day, more than once a week.”
‡All survey participants were asked if they had their cholesterol level
and blood pressure checked in the past year. Participants younger than
50 years were asked if they had received a physical examination in the
past 3 years. All other survey questions regarding physical examina-
tions and basic medical screening were age-dependent. Men older
than 50 years were asked if they had a colonoscopy at age 50, while
women in this age group were asked if they had received a bone den-
sity test (dual-energy x-ray absorptiometry or DXA) in the past 3 years.
Demographic and practice characteristics, including age,
sex, and practice location and type, showed no significant
relationship with either tobacco use or alcohol consumption.
Physical Examinations and Medical Screenings
Data from AOA Committee on Physician Health Survey indi-
cate that DOs obtain some of the commonly recommended
physical examinations and medical screenings (Table 2). Two
hundred forty-four participants (82%) reported having their
blood cholesterol levels and blood pressure tested during the
previous year. Osteopathic physicians in primary care (172 of
202 [85%]) were significantly more likely than DOs in other spe-
cialties (50 of 70 [71%]) to have taken these tests (␹2=7.549,
P<.006). In addition, DOs older than 40 years (214 of 249 [86%])
were much more likely than DOs younger than 40 years (29 of

McNerney et al • Special Communication JAOA • Vol 107 • No 12 • December 2007 • 541

SPECIAL COMMUNICATION

Table 2 (continued)
American Osteopathic Association Committee on Physician Health Survey:
Responses by Characteristic (N=299)*

Characteristic, %
Sex Practice Location
Male Female City Suburb Small Town Rural Area
Survey Item (n=236) (n=57) (n=104) (n=123) (n=33) (n=37)

Lifestyle
▫ Take annual vacation 89.4 91.2 86.6 91.9 93.9 86.5
▫ Schedule daily personal time 49.2 48.2 51.0 46.7 48.5 54.1
▫ Regular exercise‡ 50.4 54.4 62.5 43.1 51.5 45.9
▫ Overweight by >10% 58.3 45.6 52.9 55.7 57.6 64.9

Risky Behavior
▫ Tobacco use 7.8 3.5 4.9 7.3 12.5 5.6
▫ Regular alcohol consumption// 20.9 14.0 22.3 20.3 9.4 16.2

Physical Examination and Medical Screening¶
▫ Check cholesterol and blood pressure 82.6 77.2 82.7 76.4 93.9 86.5
▫ Physical examination, age <50 y 64.5 71.8 65.7 61.4 72.7 68.0
▫ Men, age >50 y//
– Colonoscopy ... ... 43.5 40.5 35.7 53.3
– Annual prostate examination ... ... 68.9 85.7 78.6 73.3
– Both colonoscopy and prostate examination ... ... 34.6 46.7 55.6 40.0
(continued)

* Sample size (n) for each characteristic is based on the number of osteopathic physicians who responded to that survey item. For data on men older than 50 years, 123
participants answered survey questions about colonoscopy at age 50 years and annual prostate examination. For data on women older than 50 years, 21 respondents
answered the survey question about mammogram; 20, Papanicolaou smear; 15, dual-energy x-ray absorptiometry (DXA) in the past 3 years. In many instances, data
available on women older than 50 years were insufficient for analysis.
† P<.05
‡ Regular exercise was defined as 30 minutes of physical activity three or more times per week.
§ P<.01
// Regular alcohol consumption was defined as “more than 2 ounces of spirits, 8 ounces of wine, or 24 ounces of beer in a single day, more than once a week.”
¶ All survey participants were asked if they had their cholesterol level and blood pressure checked in the past year. Participants younger than 50 years were asked if
they had received a physical examination in the past 3 years. Other survey questions regarding physical examinations and basic medical screening were age-dependent.

Abbreviations: ellipses (...), data insufficient for analysis; NA, not applicable.

48 [60%]) to have their blood cholesterol and blood pressure
checked (␹2=17.628, P<.001).
In addition, survey results revealed that many DOs have
been getting physical examinations regularly. Ninety-nine of
149 survey respondents (66%) younger than 50 years reported
that they had a physical examination within the previous
3 years. Among male DOs older than 50 years, 89 of 116 (77%)
had an annual prostate examination, and 49 of 115 (43%) had
a colonoscopy at age 50. Among the 15 female DOs older than
50 years, 11 had an annual Papanicolaou smear, 11 also had an
annual mammogram and breast examination, and 10 had
dual-energy x-ray absorptiometry within the previous 3 years.
In contrast to these encouraging findings, the survey
revealed that only 43 of 114 (38%) male DOs older than 50 years
obtained a colonoscopy at age 50 and had a prostate exami-
nation every year thereafter. Likewise, only 6 of 15 female
DOs older than 50 years had an annual Papanicolaou smear,
an annual mammogram, and dual-energy x-ray absorptiom-
etry within the previous 3 years.
Patterns of Health Behaviors
Many health behaviors analyzed in the present study, such as
tobacco use and alcohol consumption, have frequently been
correlated.14 Therefore, examining such behaviors together
can provide more useful information than examining them
separately. Exploratory factor analysis is a statistical method
for identifying constellations (ie, sets) of correlated behaviors,
called factors, that occur together.
A single behavior can be its own factor if it correlates
with no other behavior. The data derived from the health
behavior questions in the AOA Committee on Physician
Health Survey were subjected to an exploratory factor analysis
using the varimax rotation option of SPSS version 14.0
(SPSS Inc, Chicago, Ill). This analysis uncovered four con-
stellations of correlated behaviors (ie, factors) among the
participants in the survey (Table 3). This is an important
finding by itself, because, if DOs lived completely healthy
lives, all health behaviors would be correlated, with the
resulting expectation that there would be only one factor. In

542 • JAOA • Vol 107 • No 12 • December 2007 McNerney et al • Special Communication

 SPECIAL COMMUNICATION

Table 2 (continued)
American Osteopathic Association Committee on Physician Health Survey:
Responses by Characteristic (N=299)*

Characteristic, %
Medical Specialty Practice Type
Primary Care Other Patient Care Teaching/Research Retired Other
Survey Item (n=200) (n=69) (n=225) (n=30) (n=9) (n=26)

Lifestyle
▫ Take annual vacation 91.5 87.0 92.4 76.7 100.0 76.9†
▫ Schedule daily personal time 49.0 43.5 48.4 46.7 88.9 40.0§
▫ Regular exercise‡ 48.5 53.6 47.1 60.0 88.9 61.5
▫ Overweight by >10% 57.3 58.0 57.1 60.0 33.3 50.0

Risky Behavior
▫ Tobacco use 8.1 2.9 6.3 3.4 11.1 11.5
▫ Regular alcohol consumption// 18.7 21.7 19.7 16.7 11.1 23.1

Physical Examination and Medical Screening¶
▫ Check cholesterol and blood pressure 85.0 71.0† 80.4 83.3 100.0 88.5
▫ Physical examination, age <50 y 62.4 71.4 64.4 69.2 73.2 73.3
▫ Men, age >50 y//
– Colonoscopy 38.2 48.3 39.8 46.7 55.6 50.0
– Annual prostate examination 78.7 65.5 74.4 73.3 88.9 90.0
– Both colonoscopy and prostate examination 32.4 44.8 34.6 46.7 55.6 40.0
(continued)

* Sample size (n) for each characteristic is based on the number of osteopathic physicians who responded to that survey item. For data on men older than 50 years, 123
participants answered survey questions about colonoscopy at age 50 years and annual prostate examination. For data on women older than 50 years, 21 respondents
answered the survey question about mammogram; 20, Papanicolaou smear; 15, dual-energy x-ray absorptiometry (DXA) in the past 3 years. In many instances, data
available on women older than 50 years were insufficient for analysis.
† P<.05
‡ Regular exercise was defined as 30 minutes of physical activity three or more times per week.
§ P<.01
// Regular alcohol consumption was defined as “more than 2 ounces of spirits, 8 ounces of wine, or 24 ounces of beer in a single day, more than once a week.”
¶ All survey participants were asked if they had their cholesterol level and blood pressure checked in the past year. Participants younger than 50 years were asked if
they had received a physical examination in the past 3 years. Other survey questions regarding physical examinations and basic medical screening were age-dependent.

Abbreviations: ellipses (...), data insufficient for analysis; NA, not applicable.

the present study, all factors had at least two related variables:

Lifestyle—taking an annual vacation, which was analyzed
independently as its own factor, scheduling daily personal
time, and exercising regularly (ie, physical activity for
30 minutes or more at least three times a week)

Risky behavior—no tobacco use and low alcohol con-
sumption

Physical examination and medical screening—physical
examination within the previous 3 years (if younger than
50 years) and tests of blood cholesterol levels and blood
pressure within the previous year
Various personal characteristics of survey participants,
such as age, sex, and practice type, that affected each of these
factors are as follows:

Lifestyle—Considering all the demographic and practice
variables together indicated that DOs were most likely to
take an annual vacation if any of the following characteris-
tics applied to them: they did not work in patient care; they
practiced in a group setting; or they were unmarried. The
importance of practicing in a group for taking annual vaca-
tions applied especially to DOs who worked in patient care.
In addition, four characteristics of participants had sub-
stantial effects on the remaining characteristics in this factor:
practice type, practice location, retired status, and physi-
cian age. A multivariate analysis, which was used to analyze
these four characteristics simultaneously, presented a more
complete picture of which DOs were most likely to achieve
a healthy lifestyle of daily personal time and regular exer-
cise. They were DOs practicing in a suburban area, retired
DOs, and DOs between the ages of 60 and 65 years. Osteo-
pathic physicians who retire before age 65 years are most
likely to exercise regularly and schedule daily personal time.
Unexpectedly, DOs older than 65 years and younger than
40 years scored lower on this lifestyle factor than did DOs
of other age groups. These findings suggest that neither age

McNerney et al • Special Communication JAOA • Vol 107 • No 12 • December 2007 • 543

SPECIAL COMMUNICATION

Table 2 (continued)
American Osteopathic Association Committee on Physician Health Survey:
Responses by Characteristic (N=299)*

Practice Setting, %*
Survey Item Solo (n=95) Group (n=137) Other (n=50)

Lifestyle
▫ Take annual vacation 86.3 92.7 88.0
▫ Schedule daily personal time 51.6 46.3 48.0
▫ Regular exercise‡ 40.0 53.3 62.0†
▫ Overweight by >10% 61.1 56.6 52.0

Risky Behavior
▫ Tobacco use 6.4 6.7 6.0
▫ Regular alcohol consumption// 15.8 20.7 20.0

Physical Examination and Medical Screening¶
▫ Check cholesterol and blood pressure 85.3 78.8 82.0
▫ Physical examination, age <50 y 62.9 67.9 63.0
▫ Men, age >50 y//
– Colonoscopy 34.0 43.9 54.5
– Annual prostate examination 75.0 71.8 90.9
– Both colonoscopy and prostate examination 34.6 46.7 55.6

* Sample size (n) for each characteristic is based on the number of osteopathic physicians who responded to that survey item. For data on men older than 50 years, 123
participants answered survey questions about colonoscopy at age 50 years and annual prostate examination. For data on women older than 50 years, 21 respondents
answered the survey question about mammogram; 20, Papanicolaou smear; 15, dual-energy x-ray absorptiometry (DXA) in the past 3 years. In many instances, data
available on women older than 50 years were insufficient for analysis.
† P<.05
‡ Regular exercise was defined as 30 minutes of physical activity three or more times per week.
§ P<.01
// Regular alcohol consumption was defined as “more than 2 ounces of spirits, 8 ounces of wine, or 24 ounces of beer in a single day, more than once a week.”
¶ All survey participants were asked if they had their cholesterol level and blood pressure checked in the past year. Participants younger than 50 years were asked if
they had received a physical examination in the past 3 years. Other survey questions regarding physical examinations and basic medical screening were age-dependent.

Abbreviations: ellipses (...), data insufficient for analysis; NA, not applicable.

nor retired status alone lead to the healthy lifestyle behav-
iors of personal time and exercise.

Risky behavior—Results of our survey suggested that
women were less inclined to use tobacco and consume
alcohol, though this finding just missed the level of statistical
significance (P=.051). The same patterns hold for the risky
behavior factor as they do for the individual variables within
that factor. A multivariate analysis did not reveal any indi-
rect patterns between these behaviors and any other demo-
graphic or practice variables in the survey. Thus, it appears
that smoking and drinking are very much individual choices
by DOs, rather than behaviors that are strongly influenced
by other characteristics under investigation in this survey.

Physical examination and medical screening—The per-
centage of DOs who had their blood cholesterol and blood
pressure tested in the previous year did not differ according
to any of the analyzed demographic variables. However,
when all the demographic variables were used to predict this
factor, three characteristics of participants were found to
be the most important: physician age, practice type, and
practice location. Further statistical analysis indicated that
DOs were most likely to receive regular physical examina-
tions and medical screenings if they were younger, practiced
in a group setting, and worked in either a suburban or rural
area.
Comment
Although the findings of the present investigation are impor-
tant and useful for determining the overall health of DOs and
improving patient education, it should be kept in mind that
these results are based on the self-reports of a convenience
sample––rather than on verified data from a random sample.
In addition, members of the AOA House of Delegates tend to
be older, are more likely to be male, and more likely to work
in patient care and solo practice than the general DO popula-
tion.15 For these reasons, the results of this survey cannot be
generalized to all DOs. Nevertheless, the surveyed sample is
diverse, and members of the AOA House of Delegates are
among the “elites” of the osteopathic medical profession.
Therefore, these survey participants would probably be more

544 • JAOA • Vol 107 • No 12 • December 2007 McNerney et al • Special Communication

 SPECIAL COMMUNICATION

Table 3
American Osteopathic Association Committee on Physician Health Survey:
Exploratory Factor Analysis of Reported Behaviors (N=299)*

Factor
Physical Examination/
Behavior Annual Vacation Lifestyle Risky Behavior Medical Screening

Take Annual Vacation 0.8712† -0.0776 -0.0316 0.1266

Schedule Daily Personal Time -0.0172 0.6332† -0.1328 0.1841
Regular Exercise‡ -0.0258 0.8487† 0.0361 0.0625
Overweight by >10% -0.4798 -0.4605 -0.0397 0.4822
Tobacco Use -0.2399 0.0260 0.7949† 0.0155
Regular Alcohol Consumption§ 0.2103 -0.1174 0.7810† 0.0111
Check Cholesterol and Blood Pressure// -0.0643 0.1002 -0.0098 0.7958†
Physical Examination, Age <50 y¶ 0.2736 0.1965 0.0556 0.6748†

* Factors are constellations, or sets, of related behaviors. Analysis based on rotated component matrix, converged in six iterations, using SPSS statistical software
(versions 13.0 and 14.0; SPSS Inc, Chicago, Ill). Extraction method: principal component analysis. Rotation method: varimax with Kaiser normalization.
† Behavior strongly correlated with factor.
‡ Regular exercise was defined as 30 minutes of physical activity three or more times per week.
§ Regular alcohol consumption was defined as “more than 2 ounces of spirits, 8 ounces of wine, or 24 ounces of beer in a single day, more than once a week.”
// All survey participants were asked if they had their cholesterol level and blood pressure checked in the past year.
¶ Participants younger than 50 years were asked if they had received a physical examination in the past 3 years.

aware of the importance of concepts involved in “good health”
than most other DOs. The general findings and policy impli-
cations for the results of the AOA Committee on Physician
Health Survey raise serious issues for the entire osteopathic
medical profession.
Conclusions
In terms of their personal health behaviors, DOs resemble the
stereotypical patient. They are healthy overall, and they carry
out some physician instructions—but not others. Losing weight
and getting more exercise, in particular, are two areas in which
DOs need to “walk the talk” and follow the advice they give
to patients regarding making consistently healthy lifestyle
choices. Controlling weight and getting regular exercise are
problems for DOs regardless of age, sex, or practice type.
Some DOs also need to address their regular use of tobacco
and alcohol.
Osteopathic physicians cannot credibly attribute their
shortcomings in personal healthcare to such commonly cited
reasons as practicing in a solo setting, working in a rural area,
or special gender-related pressures. The results of the
AOA Committee on Physician Health Survey indicate that it
is possible for DOs of both sexes and those who are in solo prac-
tices or practices in rural areas to lead healthy lifestyles.
An often overlooked health risk factor for all physicians
is their treatment of themselves. Canadian physician Sir
William Osler16 wrote, “The physician who treats himself has
a fool for a patient.” Previous studies have demonstrated that
between 42% and 82% of physicians administer healthcare to
themselves in some manner.17 We urge the AOA to conduct
additional studies to determine the prevalence of “self-doc-
toring” throughout the osteopathic medical profession and
to examine other aspects of DOs as patients.
Physicians teach patients by example as much as by their
words. Physicians who ignore their own health encourage
their patients to do likewise. Physicians who convince them-
selves that they are “too busy” to be healthy forget that almost
everyone nowadays faces increased job pressures, extended
workdays and workweeks, and greater demands on time.
Data from the AOA Committee on Physician Health Survey
reveal that DOs need to perform careful self-evaluations of
many aspects of their personal health.
References
1. Gross CP, Mead LA, Ford DE, Klag MJ. Physician, heal thyself? Regular
source of care and use of preventive health services among physicians. Arch
Intern Med. 2000;160:3209-3214. Available at: http://archinte.ama-
assn.org/cgi/content/full/160/21/3209. Accessed September 26, 2007.
2. Sundquist J, Johansson SE. High demand, low control, and impaired gen-
eral health: working conditions in a sample of Swedish general practitioners.
Scand J Public Health. 2000;28:123-131.
3. Abramson S, Stein J, Schaufele M, Frates E, Rogan S. Personal exercise
habits and counseling practices of primary care physicians: a national survey.
Clin J Sport Med. 2000;10:40-48.
4. Wells KB, Lewis CE, Leake B, Ware JE Jr. Do physicians preach what they
practice? A study of physicians’ health habits and counseling practices. JAMA.
1984;252:2846-2848.
5. Licciardone JC, Hagan RD. The physical fitness of first-year osteopathic med-
ical students. J Am Osteopath Assoc. 1992;92:327-333.
6. Allen TW. Physician, heal thyself [editorial]. J Am Osteopath Assoc.
1992;92:268.
(continued)

McNerney et al • Special Communication JAOA • Vol 107 • No 12 • December 2007 • 545

SPECIAL COMMUNICATION

7. Berger J. Practice what you preach: DOs need to apply preventive medicine
to their own lives. The DO. November 1999;40:50-56.
8. Hodges L. Student body: medical students must tend to their own health
first. The DO. November 1999;40:58-61.
9. Gaber RR, Martin DM. Still-Well osteopathic medical student wellness
program. J Am Osteopath Assoc. 2002;102:289-292. Available at:
http://www.jaoa.org/cgi/reprint/102/5/289. Accessed September 26, 2007.
10. Ajluni PB. AOA president’s inaugural speech page. American Osteo-
pathic Association Web site. Available at: http://www.osteopathic.org/
index.cfm/pdf/index.cfm?PageID=aoa_yrfitforlife. Accessed December 12,
2007.
11. Ajluni PB. DOs: fit for life page. American Osteopathic Association Web
site. Available at: https://www.do-online.org/index.cfm?PageID=aoa_
yrfitforlife. Accessed December 12, 2007.
12. Sudman S, Bradburn NM. Asking Questions: A Practical Guide to
Questionnaire Design. San Francisco, Calif: Jossey-Bass; 1982.
13. Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The
validity of self-reporting smoking: a review and meta-analysis. Am J Public
Health. 1994;84:1086-1093.
14. Dierker L, Lloyd-Richardson E, Stolar M, Flay B, Tiffany S, Collins L, et al.
The proximal association between smoking and alcohol use among first year
college students. Drug Alcohol Depend. 2006;81:1-9. Epub July 11, 2005.
15. American Osteopathic Association. Master file: Database on osteopathic
physicians [database online]. Chicago, Ill: American Osteopathic Association;
2001.
16. Osler and rural practice page. University of Nebraska Medical Center
Web site. Available at: http://www.unmc.edu/Community/ruralmeded/
osler.htm. Accessed September 26, 2007.
17. Töyry S, Räsänen K, Kujala S, Aärimaa M, Juntunen J, Kalimo R, et al. Self-
reported health, illness, and self-care among Finnish physicians: a national
survey. Arch Fam Med. 2000;9:1079-1085. Available at: http://archfami.ama-
assn.org/cgi/content/full/9/10/1079. Accessed September 26, 2007.

When my soul is sick unlimited spiritual liberty is given me by the State. Now then, it doesn’t seem logical that the state
shall depart from this great policy, the health of the soul, and change about and take the other position in the matter of smaller
consequences—the health of the body...
Whose property is my body? Probably mine. I so regard it. If I experiment with it, who must be answerable? I, not the State.
If I choose injudiciously, does the State die? Oh, no.
Mark Twain, 1902

546 • JAOA • Vol 107 • No 12 • December 2007 McNerney et al • Special Communication