Do As I Do: Exercise Habits of Physical Therapists, Physical Therapist Assistants, and Student Physical Therapists

Julia Chevan, Esther M. Haskvitz


Background Physical therapy practitioners are among the many health care professionals who can counsel their patients to address the public health care concern of physical inactivity. Health care providers who are physically active themselves are more likely to counsel patients on the benefits of activity.

Objective The purposes of this study were: (1) to examine the leisure-time physical activity habits of physical therapists, physical therapist assistants, and student physical therapists in the United States using Centers for Disease Control and Prevention and American College of Sports Medicine (CDC-ACSM) recommendations and (2) to compare these habits with those of the general population and other health care professionals.

Design A cross-sectional survey design was used.

Methods There were 2 data sources. A random sample of American Physical Therapy Association members completed an online survey that included questions about physical activity habits worded in same manner as the leisure-time activities section of the 2005 National Health Interview Survey (NHIS). The final study sample comprised 1,238 participants: 923 physical therapists, 210 student physical therapists, and 105 physical therapist assistants. The 2005 NHIS public use data files were the source for the same information about the general US population and for a subset of health care professionals. Rates of participation in vigorous and moderate physical activity were analyzed.

Results Physical therapists, physical therapist assistants, and student physical therapists exercised at higher rates than adults and health-diagnosing professionals in the 2005 NHIS.

Limitations The study may be limited by sampling and response bias.

Conclusions This study identified that physical therapists, physical therapist assistants, and student physical therapists are meeting CDC-ACSM physical activity guidelines at higher rates than the US adult population and health-diagnosing professionals. These rates exceed the physical activity targets set for adults in Healthy People 2010.

Physical activity is an important component of a healthy lifestyle to reduce a person's risk of developing disease and subsequent disability.1,2 Physical inactivity is a growing public health concern. Among adults 18 years of age and older in the United States, the rate of inactivity remained steady at 40% from 1998 to 2005.3 This concern is being addressed by many professional organizations, including the American Medical Association, the American College of Sports Medicine (ACSM),4 and the American Physical Therapy Association (APTA).5,6 The APTA has provided guidance to physical therapists about their role in promoting physical activity through the development of a plan for physical fitness for special populations, as well as statements on the role of physical therapists in promoting health, wellness, physical education, and physical conditioning.57 The Guide to Physical Therapist Practice specifically states that physical therapists are expected to “promote health, wellness, and fitness.”8 Physical therapists are well suited to provide counseling to patients and clients about the need to participate in physical activity.

Current recommendations for physical activity advocated by the ACSM and the American Heart Association are that “all healthy adults aged 18 to 65 need moderate-intensity aerobic physical activity for a minimum of 30 minutes, 5 days each week or vigorous-intensity aerobic activity for a minimum of 20 minutes, 3 days each week.”9(p1425) The majority of US adults do not meet these guidelines, with 31.6% reporting regular leisure-time physical activity and 38% reporting never engaging in physical activity during leisure time.3,10

Improving levels of physical activity participation in the United States will require effective dissemination of guidelines and acceptance of the importance of physical activity for disease prevention and health enhancement.11 Although the evidence of the effects of behavioral counseling on physical activity is mixed,12 physician and other professional organizations,4 including APTA,7 have developed programs urging professionals to counsel patients about regular physical activity. These professionals also need to be physically active themselves, not only for their own benefit, but also because of their position as role models. The endorsement of an active lifestyle is more credible coming from a professional who is physically active.11 Surveys of physicians and medical students have shown that those who are physically active are more likely to counsel their patients on the benefits of activity.1315 Estimates of the proportion of physicians who participate in leisure-time physical activity have been documented in a number of studies, and these estimates fall in a broad range from 27% to 73%.13,1517 Physical activity participation rates among hospital employees, nursing students, and medical students also have been studied.1822 We were unable to find published studies documenting the physical activity habits of physical therapists.

Physical therapists are trained to be authorities on exercise. Williford et al23 found that when physicians recommended an exercise program to a patient, a physical therapist was the most likely health care professional to whom referral was made. Rea et al24 found that the health-promoting behavior that received the greatest attention from physical therapists was the promotion of physical activity to their patients. Because physicians and medical students who were physically active were more likely to counsel their patients on the benefits of activity than their sedentary counterparts,1315 knowledge of the physical activity habits of physical therapists may be an important indicator of their willingness to engage in counseling activities. In addition to physical therapists, physical therapist assistants are engaged in direct patient care with a potential to counsel. Student physical therapists' physical activity behaviors may provide an indicator for the future of physical activity among those in the profession or a gauge of the level of self-selection of active people into the profession. The purposes of this study were: (1) to develop an estimate of the rate of participation in leisure-time physical activities among physical therapists, physical therapist assistants, and student physical therapists and (2) to compare these estimates to rates of physical activity among US adults and health care professionals identified in the 2005 National Health Interview Survey (NHIS).


Data Sources

We compiled data from 2 sources for this study. The 2005 NHIS was used to assess the physical activity habits of all US adults and of people classified as “health professionals.” A self-designed survey of activity with questions derived from the leisure-time activities portion of the NHIS was administered to a randomly selected sample of APTA members who were physical therapists, physical therapist assistants, and student physical therapists to gather data on their personal exercise habits. The 2005 NHIS was selected as the comparator data source to match the year of administration of the self-designed survey.


The NHIS is a population-based, nationally representative survey conducted annually by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC). The NHIS is one of the tools used to study and measure the health status and health behaviors of the civilian, noninstitutionalized population of the United States. The survey consists of a core module, which includes a family core survey, a survey of a sample adult randomly selected from each household, and a survey of a sample child randomly selected from each household. Public use files are released annually and made available via the CDC Web site. Detailed information about the NHIS and the method used in data collection may be found at the NCHS Web site25 or in published literature.26,27

The 2005 NHIS sample adult data file (SADF) was used to extract data on physical activity for adults and for that subpopulation classified as a health care professional. The subpopulation of health care professionals was selected as the group that would most closely match physical therapists in their occupation, educational background, and social status. The SADF included demographic information and contained variables that documented responses to the survey questions asked about leisure-time activities related to exercise (Appendix). The SADF also contained weighting, primary sampling unit, and strata variables to account for the complex design of the NHIS. These variables are needed to develop national estimates and to enable population-based estimates.

The SADF from 2005 contained 31,428 people aged 18 years or older. In the SADF, 735 people were identified as unable to exercise in any fashion; these people were dropped from the base population for our analysis. Health care professionals were identified from a set of variables that described a person's industry and occupation. Because broad categories are used for the occupation variables, we used the category of “health-treating and -diagnosing professionals.” Among those professionals included in that category are physical therapists, physicians, nurses, and physician assistants. The SADF contained 706 people identified as “health-treating and -diagnosing professionals.” Due to privacy restrictions on public data files, further breakdown of the specific professions associated with these people is not available. Among the SADF respondents who were able to exercise, the mean (SD) age was 45.2 (17.4) years and 47.4 (14.7) years for the health-treating and -diagnosing professionals. The percentage of women among the SADF respondents able to exercise was 56%, whereas for the health care professionals it was 78%.

Survey of Leisure-Time Activities for Physical Therapists.

The self-designed Survey of Leisure-Time Activities for Physical Therapists (SLTAPT) was administered using the SurveyMonkey online survey tool. The survey consisted of 5 background demographic questions followed by sections asking about vigorous activities, moderate activities, and strengthening activities done during leisure time. All of the vigorous, moderate, and strength activities questions were asked using definitions and phrasing identical to the 2005 NHIS (Appendix). By using NHIS phrasing for the survey, we could be sure that we were using questions that previously had been field tested, validated, and assessed for their reliability. All estimates of levels of physical activity in this article and associated statistics presented follow the NCHS guidelines for reliability of data and have a relative standard error of 30% or less (personal communication with NCHS).

The SLTAPT was available online from October 2005 until December 2005. Potential participants were contacted by notices that were posted on physical therapy e-mail listservs for APTA sections or through a direct contact e-mail that contained the link to the survey. Our intent was to generate a representative 1% sample of the APTA membership list, which numbered 69,994 at the time of our study. The Figure shows the flow of respondents in the study.


Flow diagram for respondents to the survey of leisure-time activities for physical therapists. APTA=American Physical Therapy Association.

A random selection algorithm using the entire APTA membership list was applied for the direct contact e-mails, resulting in 2,960 e-mail notices sent. In the algorithm, a random number generator was used to generate a 3-digit number that was matched to letters of the alphabet (eg, A=1, B=2, and so on). Then, all members whose last name started with these 3 letters were sent an e-mail invitation to participate, until 2,960 e-mails were sent by the computer program that was monitoring response rate at the survey Web site. From these e-mails, 466 were returned due to having undeliverable e-mail addresses. We were unable to determine how many of the e-mails actually reached the intended recipients. A follow-up reminder e-mail notice was sent 4 to 6 weeks after the initial e-mail or listserv notice was sent.

At the end of the survey period, data were downloaded for 1,352 survey respondents from the SurveyMonkey Web site. Inclusion criteria for participating in the survey were that respondents must have been currently practicing physical therapy in the United States, have direct patient care, and have billable patient hours. We excluded 114 respondents from the database for not meeting inclusion criteria or for not providing complete information. The final study sample consisted of 1,238 respondents: 923 physical therapists, 105 physical therapist assistants, and 210 student physical therapists.

The ages of the physical therapists, physical therapist assistants, and student physical therapists in our sample appear to be representative of the entire membership of APTA. The mean (SD) age of our student sample was 24.9 (10.7) years compared with 25.2 (4.1) years for the APTA membership. The mean (SD) age of the physical therapists was 40.5 (10.3) years compared with 42.4 (12.3) years for the membership, and the mean (SD) age for the physical therapist assistants was 38.6 (11.2) years compared with 38.7 (10.8) years for the membership. The mean (SD) age in our total sample was 37.7 (11.6) years. Women comprised 73% of our sample compared with 65% of the APTA membership (personal communication, APTA).


Variables to analyze exercise frequency were constructed for both the SADF and the SLTAPT data. Duration of vigorous and moderate leisure-time activities was converted to minutes. We identified a respondent as meeting the guidelines using a moderate level of activity if he or she participated in at least 30 minutes of moderate activity, 5 or more days of the week. A respondent met the vigorous exercise guidelines if he or she participated in at least 20 minutes of vigorous activity, 3 or more days of the week. To meet the guidelines through a combination of moderate and vigorous activity, a respondent had to be engaged in physical activity for at least 5 days of the week, with the vigorous activities encompassing 20 minutes and the moderate activities encompassing 30 minutes. A respondent met the current guidelines for strength training if activities were carried out on at least 2 days of the week.

Data Analysis

Weighted frequencies, percentages, confidence intervals, and standard errors were calculated to examine exercise frequency among all the groups of interest. Tests of proportions were used to compare results among the subgroups. We were most interested in our primary comparison, which was between the physical therapy practitioner groups and the US population. After examining that comparison, we compared the physical therapy practitioner groups with the subpopulation of health care professionals identified in the NHIS data. All data were analyzed using Stata/SE 9.2 for Windows.* Stata survey commands were used to account for the complex survey design.

Wave analysis28 was used to estimate nonresponse bias comparing the earlier and later respondents on both the demographic and exercise frequency variables. Accordingly, both t tests and chi-square analysis were used to compare age, sex, and exercise frequency between people who responded to the survey in the first 2 weeks of survey availability (n=457) and those who responded to the survey in the last 2 weeks of availability (n=47).


Physical therapists, physical therapist assistants, and student physical therapists are all exercising at higher rates than the US noninstitutionalized adults in the 2005 NHIS survey, as well as those identifying themselves as health-diagnosing and -treating professionals. The proportions of those exercising at the ACSM-recommended level are shown in the Table. The rates of activity were significantly different for physical therapists than those from the 2 samples of the NHIS when any physical activity and when vigorous activity were considered (P<.05). The rates of activity also differed significantly when physical therapist assistants and student physical therapists were compared with the 2 samples from the NHIS (P<.05). The rate of moderate activity at the ACSM-recommended level was similar across all study samples. Physical therapists also engaged in strength training at rates that exceeded those of adults and health-diagnosing and -treating professionals in the 2005 NHIS (P<.05).


Physical Activity Comparisons Among Physical Therapists, US Adults, and Health-Diagnosing and -Treating Professionalsa


Physical therapists, physical therapist assistants, and student physical therapists all appear to follow the ACSM recommendations for physical activity at rates higher than those of both the general public and other health care professionals. The higher rates emanate from the high percentage of physical therapists, physical therapist assistants, and student physical therapists who maintain a program of vigorous physical activity. Frank et al22 found that medical students also preferred strenuous exercise as modality of choice for physical activity. One benefit of a program of vigorous activity is that a busy professional can meet the recommendations with only 3 days of activity. Among those we surveyed who met the recommendations, the most frequently reported number of days on which activity was performed was 3 (38.1%), 4 (23.6%), or 5 (16.4%) days.

Jones et al29 found that the prevalence of adults meeting the CDC-ACSM guidelines decreased with age. Our results support that finding. We acknowledge that there are age differences between the sample of therapists and the 2 NHIS samples; however, the 5- to 7-year difference, on average, between the 2 groups would not account for the substantial difference in exercise rates. Jones et al29 showed the greatest reduction in meeting guidelines occurred after the age of 65 years. Among our respondents, student physical therapists had the highest percentage of meeting the recommendations at any activity level. The higher level of physical activity among students may be an indicator of a bias among those who exercise and their occupational preference. We were unable to find any studies that examined this theory.

Level of education also appears to be a factor in engagement in leisure-time physical activity. Research has shown that the prevalence of meeting the CDC-ACSM guidelines is greater with increased educational attainment.29,30 Our results are consistent with those findings. In the studies by Jones et al29 and Brownson et al,30 education level was defined as attaining more or less than 12 years of education. Because physical therapists complete a graduate program and physical therapist assistants complete an associate's degree, they are engaged in educational activities beyond the typical 12 years of secondary education. However, that does not explain why the physical activity rates for physical therapists are higher than those of other health care professionals.

Education has been hypothesized to be an important social influence on health31,32 and likely improves the ability to obtain and interpret health-related information. Perhaps the education related to the benefits of exercise that physical therapists, physical therapist assistants, and student physical therapists receive positively influences their exercise behavior. Another explanation may be that those individuals drawn to the physical therapy profession are inclined to be more active. Because movement plays a central role in the profession of physical therapy,8 individuals drawn to the profession may be more active by nature. Also, those who are active at work are more likely to be active in leisure-time activities.33

The exercise recommendations from the ACSM and American Heart Association include aerobic and muscle strengthening activities. Strength training activities should be performed a minimum of 2 nonconsecutive days each week.9 The Healthy People 2010 target is for 30% of adults aged 18 years and older to be engaged in strength training activities.34 The physical therapists in our survey far exceed this target, with 59.1% participating in strength training activities a minimum of 2 days per week. The 2005 NHIS data indicate that adults are short of the target, with only 20.2% engaged in strength training activities, and that the health-diagnosing and -treating professionals in the 2005 NHIS data are on target with 30.5% engaged.

The strongest determinant of participation in strength training is participation in other aerobic activities.35 Our results support that finding. The physical therapists we surveyed had higher rates of participation in both aerobic and strength training exercise compared with adults and health-diagnosing and -treating professionals surveyed in the 2005 NHIS.

A 1% sample of the APTA membership represents a sufficiently large sample on which to base an estimate. However, several limitations exist in this study, including the potential for sampling and response bias. Sampling APTA members may impart bias because these individuals may be people who are not only more active in their profession but also more active in other venues of their lives. Although response and nonresponse are difficult to track with an electronic survey delivered via e-mail, we dealt with the issue of bias by conducting a wave analysis of early and late responders to the survey.28 These 2 waves did not differ significantly on any demographic or activity variables, mitigating concerns about response bias.

Physical activity has been recognized as an important means to help curb chronic disease and provide substantial health benefits. Recently, the US government issued its first Physical Activity Guidelines for Americans.36 These guidelines are intended to promote physical activity as a preventative measure. Physical therapists can play an important role in helping to promote the benefits of physical activity by counseling their patients to increase their participation.


Based on our sample, most physical therapists appear to be good role models. This finding may have clinical implications related to their counseling of patients about the importance of physical activity. Further research is needed to determine the relationship between personal exercise habits and counseling practices of physical therapists.



Questions and Definitions Used in the Surveys


  • Both authors provided concept/idea/research design, writing, data analysis, and project management. Dr Chevan provided data collection, fund procurement, and facilities/equipment. The authors thank Adam Oyola, Katie Wilby, and Amanda Smith for assistance with data collection.

  • A poster presentation of this research was given at the Combined Sections Meeting of the American Physical therapy Association; February 6–9, 2008; Nashville, Tennessee.

  • This study was funded by the Springfield College Research Fund.

  • * Stata Corp LP, 4905 Lakeway Dr, College Station, TX 77845.

  • Received April 3, 2009.
  • Accepted December 21, 2009.


View Abstract