Primary Care Practitioners’ Understanding of Physiatric Practice: Effects on Intention to Refer

Article Outline

• Abstract
• Methods
  • Appropriate referral
  • Knowledge of PM&R skills
  • Knowledge about the specialty
  • Demographic data
• Results
  • Correlation analysis
• Discussion
  • Study limitations
• Conclusions
• Acknowledgment
• Appendix 1. Survey instrument
• Appendix 2. Respondents’ comments
• References
• Copyright

Abstract 

McKenna C, Farber NJ, Eschbach KS, Collier VU. Primary care practitioners’ understanding of physiatric practice: effects on intention to refer.

Objectives

To learn what family practice and internal medicine physicians understand about the scope of practice of physical medicine and rehabilitation (PM&R) and to study what effect that understanding and various demographic variables have on their intention to refer to physiatrists.

Design

Survey-based.

Setting

National survey.

Participants

One thousand internal medicine and family practice physicians were contacted, with 460 respondents.

Interventions

Not applicable.

Main outcome measures

Intention to refer patients to physiatrists using 13 case scenarios (10 appropriate referrals, 4 inappropriate referrals) and self-reported number of referrals per year associated with understanding of 7 skills of physiatrists analyzed by multiple logistic regression analyses.

Results

Although most respondents were likely to refer to physiatrists, a wide variation existed in the types of patients referred. Physicians with a greater understanding of the scope of physiatric practice were more likely to refer (P=.003). Female physicians were more likely to refer than male physicians (P=.003).

Conclusions

There appears to be an association between an understanding of physiatric practice and primary care practitioners’ willingness to refer to PM&R. Primary care physicians should be educated about the benefits of referring patients to physiatrists.

Key words:  Physical medicine , Referral and consultation , Rehabilitation , Surveys

THE PRACTICE OF physical medicine and rehabilitation (PM&R), when applied, has the capacity to reduce human suffering and economic burdens imposed by both acute injury and chronic disability. The awareness that medical students¹, ² and patients³ have about the field of PM&R has been examined. Kirshblum et al¹ reported a low level of knowledge about PM&R among fourth-year medical students before a mandatory clerkship, with an increase in awareness of physiatry after the clerkship. Abramson and Stein,² in a survey of senior medical students at several institutions, asked to which specialty the students would refer for several musculoskeletal disorders. They found that, although many students would refer a patient with chronic low back pain (LBP) to a PM&R physician, the vast majority would not refer such a patient.

Residents in primary care specialty training programs also lack an understanding of the scope of physiatric practice. In 1963, Waylonis and Raptou,³ then PM&R residents, surveyed their fellow house officers in various specialties and found that 56% were unaware of the difference between a physiatrist and a physical therapist. In 1981, Brogan⁴ surveyed physicians in fields likely to see patients who required rehabilitation but found that 25% of the respondents reported seeing no patients in need of rehabilitative services in the preceding 6 months. He concluded that perhaps the physicians were not recognizing their patients’ need for rehabilitation services. This conclusion was reinforced by his choice of survey group: physicians, including family practice physicians and internists, whose patients were likely to benefit from rehabilitation. Students and residents who lack an understanding of what physiatrists can offer patients would then translate that into a lack of knowledge as practicing physicians.

Patients, too, are often unclear about what services are offered by physiatrists, which compounds the physicians’ lack of awareness. Attending physiatrists administered a survey to patients already at their offices and found that only 19% were able to identify the type of doctor they were there to see.⁵ Interviews with patients in a rehabilitation setting have underscored the need for health care providers to understand patients’ perceptions of the expected outcome of rehabilitation.⁶ Studies of inpatient referral patterns have found PM&R to be an underutilized⁷ or inappropriately used service.⁸ Confusion also seems to exist within the medical community about the distinction between physiatry and physical therapy (PT).³, ⁹ Nevertheless, we found no recent data examining primary care physicians’ attitudes and knowledge of PM&R or their likelihood of referring patients to physiatrists.

Of the several types of physicians considered to be primary care specialists (family practitioners, general practitioners, internists, obstetricians and gynecologists, pediatricians), we surveyed family practitioners and internists to examine their likelihood of referrals to and awareness of PM&R. Family practitioners and internists see the most patients with common types of problems that would be referred to a physiatrist. The study examined knowledge of the skill set possessed by physiatrists, knowledge about when to consult a physiatrist, and whether these physicians consult physiatrists appropriately. The association between the physicians’ demographic data and their awareness and utilization of physiatry was analyzed. We hypothesized that a greater understanding of the scope of the PM&R specialty would be associated with a greater likelihood of referral to physiatrists.

Back to Article Outline

Methods 

The survey (appendix 1) was pretested among 50 practicing physicians at Christiana Care Health System for face and content validity. We then conducted a cross-sectional mail survey of randomly selected practicing primary care physicians in the United States, identified through the American Medical Association (AMA) Physician Masterfile, a comprehensive list of US physicians not limited to AMA members. Students, residents, and nonpracticing physicians were excluded. The study was approved by the institutional review board of Christiana Care Health System.

Each physician was sent an anonymous questionnaire along with a $5 incentive. A second questionnaire was mailed to all nonrespondents after 4 weeks. Confidentiality was ensured, because the survey contained no identifying information, and coded envelopes (to determine nonrespondents) were discarded before the coding of the data. All responses received by 4 months after the initial mailing date were included in the analysis.

The survey instrument can be segmented into several parts for the purpose of analysis: appropriateness of referral, physiatric skills, knowledge of PM&R, referral patterns, and demographics.

Appropriate referral 

In this, the first part of the questionnaire, respondents were asked to indicate how likely they are to refer a patient to a physiatrist, given several scenarios. Responses were based on a 4-point Likert scale (from very likely to very unlikely). The clinical vignettes, questions 1 through 13 in the survey, included many situations for which referral to PM&R was appropriate. Only vignettes 2, 5, 6, and 8 were considered inappropriate for referrals to PM&R.

Knowledge of PM&R skills 

Question 14 asked the respondents to indicate their knowledge of the physiatric skill set, selecting from a list of the skills physiatrists may be qualified to perform. Responses were indicated by means of a binomial choice system. The list includes many skills that are taught in physiatry residency and 2 that are not. On this list, only items B (surgery) and I (implantation of deep-brain stimulation devices) are outside the qualifications of most physiatrists. We recognize that it is unlikely that any 1 physiatrist performs all of the physiatric skills listed, but all are skills considered within the realm of physiatric training.

The appropriateness of the scenarios for referral and physiatric skills were determined by a panel of 3 PM&R physicians.

Knowledge about the specialty 

This section (questions 15–19) assesses the respondents’ knowledge about the specialty of PM&R and also the respondents’ annual referral rate of actual patients to physiatrists. Knowledge about the field of PM&R in general, as well as knowledge about the differences between PM&R and PT, were assessed by means of a 4-point Likert scale (from strongly agree to strongly disagree).

Demographic data 

The demographic section (questions 20–29) contains questions that explore the respondents’ exposure to PM&R while in medical school and residency. Geographic information (northeast, south, midwest, west) was also obtained for each respondent. The availability of nearby PM&R services was not examined.

Data were manually entered for analysis by an author, with a cross-check of 30% of the sample; no errors were detected. The number of scenarios deemed as appropriate referrals (total of appropriate referrals and avoiding the 4 inappropriate referrals) was calculated as a separate outcome variable (range, 0–13). The number of skills correctly identified (range, 0–9) with 7 appropriate and 2 inappropriate skills (those not performed by physiatrists) was also calculated as a separate outcome variable. The association between the respondents’ understanding of the scope of physiatric practice and their demographic data considered against the number of appropriate scenario referrals was analyzed by analysis of variance (ANOVA) and multiple logistic regression. Univariate analyses were conducted for the data obtained from the survey instrument. The ANOVA was calculated for each item. Continuous variables were collapsed into a bivariate variable based on the mean. All variances significant at P less than .05 were included in the multivariate logistic regression models. The association between demographic variables and the number of correct responses on physiatric skills in relation to the number of patients referred to PM&R physicians annually was similarly analyzed by ANOVA and logistic regression analysis.

Back to Article Outline

Results 

Of the 1000 questionnaires mailed, 25 could not be delivered and 460 (47%) were returned. Respondents’ demographic and professional characteristics are in table 1. The responding physicians had an average age of 48 years, 69% were men, 71% were in private practice, and 76% had not done a PM&R rotation in medical school. The demographic composition of our survey population was comparable to the demographics of the general internist and family practice physician population in the United States in 2003.¹⁰ Twenty-nine percent had trained at a medical school or residency program where there was a PM&R residency program, whereas 53% had trained where there was no PM&R residency program. Of note, 17% were not sure if there was a PM&R residency program where they trained. Family practice physicians and internists were equally represented, as were the geographic regions of the United States. Physicians spent an average of 88% of their time seeing patients, with 87% of their time in primary care.

Table 1. Demographic Data of Respondents

Demographic	Values
Age (y)	48±11.7
Gender (M/F)⁎	318 (69)/133 (29)
Year graduated from medical school	1983±12.7y
Geographic location⁎
Northeast	107(23)
South	122(27)
Midwest	100(22)
West	100(22)
Locale
Urban	154(36)
Suburban	176(38)
Rural	111(24)
Type of practice†
Private practice	325(71)
HMO	37(8)
Veterans Affairs	10(2)
Hospitalist	27(6)
Other	83(18)
Did you ever do a rotation in PM&R in medical school?⁎
Yes	89(19)
No	351(76)
Unsure	14(3)
Did your medical school or residency training program have a PM&R residency?⁎
Yes	133(29)
No	242(53)
Unsure	80(17)
What is your medical specialty/subspecialty?⁎
Internal medicine	218(47)
Family practice	221(48)
Other	9(2)
What percentage of your practice is devoted to primary care?	87±28.4
What percentage of your professional time is spent seeing patients?	88±20.2
Referral patterns
Have you ever referred a patient to a PM&R physician?
Yes	386(84)
No	71(15)
Average referrals (patients/y)	21±60.3

NOTE. Values are mean ± standard deviation (SD) or n (%).

Abbreviations: F, female; HMO, health maintenance organization; M, male.

The responses to all 13 vignettes are shown in figure 1. The appropriate referrals that PM&R physicians were most likely to receive included referrals for spinal cord injury (SCI) (91%), stroke (86%), cerebral palsy (CP) (85%), workers’ compensation evaluations (71%), LBP (70%), osteoarthritis (OA) (69%), and fibromyalgia (62%). The appropriate referrals PM&R physicians were only moderately likely to receive were for traumatic brain injury (TBI) (46%) and electromyography and nerve conduction studies (NCSs) (46%). The 2 inappropriate referrals were correctly identified by most respondents: substance abuse rehabilitation (85%) and learning disability (73%). However, respondents were likely to refer to physiatrists inappropriately for cauda equina syndrome (51%) and for PT and occupational therapy (OT) (49%).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Referral decisions. The percentage of respondents who reported that they were likely to refer a hypothetical patient in a clinical vignette to a PM&R doctor. The vignettes are found in appendix 1 (questions 1–13). Referrals for the first 9 scenarios (from left to right) are considered appropriate referrals whereas the final 4 items shown are considered inappropriate. The inappropriate and appropriate referrals were interspersed in the actual survey; they are separated here for illustrative purposes. Abbreviations: CP, cerebral palsy; CVA, cerebrovascular accident; EMG, electromyography; OA, osteoarthritis; OT, occupational therapy, SCI, spinal cord injury; TBI, traumatic brain injury.

Figure 2 explores the respondents’ knowledge about skills that physiatrists are trained to execute. Most respondents correctly identified surgery (92%) and implantation of deep brain stimulation devices (90%) as skills that physiatrists do not possess. The physiatric skills correctly identified by a majority of respondents were limb prosthesis evaluation (89%), trigger point injection (88%), electromyography and NCSs (82%), and motor point blocks (75%). Skills correctly attributed to PM&R by significantly fewer respondents included epidural injection (49%), urodynamics (27%), and ventilator management (19%).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Skill-set responses. The percentage of respondents who believed that the skills shown are possessed by physiatrists. See skills list, appendix 1 (question 14). The first 7 skills from left to right are well within the field of PM&R; the final 2 skills are not. The inappropriate and appropriate choices were interspersed in the actual survey; they are separated here for illustrative purposes.

Table 2 shows that 79% of respondents agreed physiatry is a synonym for PM&R and that 95% recognized it as a board-certified specialty. Eighty-four percent believed that they understood the difference between a physiatrist and a physical therapist. Fifty-five percent believed that the conditions PM&R addresses are also treated by neurology, orthopedics, internal medicine, PT, OT, speech therapy, and/or anesthesiology. Eighty-four percent of physicians reported that they referred a patient to PM&R with an average annual referral rate of 21±60.3 patients per year.

Table 2. Perceptions and Definitions of PM&R

Correlation analysis 

Table 3 includes all variables significantly associated with the likelihood of referrals based on univariate analysis. Based on multivariate models, only sex and higher scores on appropriate selection of PM&R referral in clinical vignettes were significant. A positive correlation existed between knowledge about the field of PM&R (as assessed in the skill list [question 14]) and appropriate referral to PM&R in the clinical vignettes section. Respondents who correctly identified 6 or more skills out of a perfect score of 9 on the skill set in question 14 referred appropriately an average of 9.1 times out of a maximum score of 13. Respondents who scored 5 skills or less referred appropriately an average of 8.5 times (P=.003). Female physicians appropriately referred an average of 9.3 hypothetical patients whereas male physicians made 8.8 appropriate referrals (P=.034).

Table 3. Likelihood of Referring Appropriately to Physiatrists, Given Specific Clinical Vignettes

NOTE. See vignettes in appendix 1 (questions 1–13). Univariate analysis: only items where P<.05 are shown. Only these items were subjected to multivariate analyses. After multivariate analysis, only sex and correct identification of skills were statistically significant predictors of likelihood to refer these hypothetical patients. R²=.098.

Only sex and self-reported understanding of the difference between PT and PM&R were significantly associated with respondents’ self-reported annual referrals to physiatrists. Female physicians reported an average of 23.4 annual referrals to PM&R whereas male physicians reported 17.3 (P=.037). Respondents who rated themselves as strongly agreeing with the statement “I understand the difference between PT and PM&R” referred more patients than respondents who did not agree strongly with the statement (P=.036).

No significant associations were seen when comparing the demographic data with the number of correct answers on the question regarding skills that PM&R physicians are qualified to perform. Handwritten text comments by the respondents, although not solicited by the survey, were occasionally included by the respondents. These are included in appendix 2.

Back to Article Outline

Discussion 

It is heartening to know that many of the patients whom physiatrists are trained to treat would be appropriately referred to physiatrists by the primary care physician respondents in the clinical vignettes. These include patients with SCI, stroke, CP, workers’ compensation evaluations, LBP, and OA. Less heartening were the response rates for fibromyalgia, TBI, electromyography, and NCSs, which physiatrists can also treat. The high response rate to perform PT and OT in the vignettes likely indicates confusion between physiatry and PT.

Correctly identifying the skills possessed by a subspecialist is integral to making a decision to refer. As a group, physiatrists can offer skills and services of which primary care physicians may not be aware. Skills within the realm of PM&R that were significantly less likely to be attributed to physiatrists by respondents included motor point blocks, epidural injections, urodynamics, and ventilator management. A statistically significant positive association exists between self-reported referrals and hypothetical patients, on the one hand, and higher score on the number of skills the respondent correctly attributed to physiatrists, on the other hand. The positive association between correctly identifying physiatric skills and referrals (both actual and hypothetical) was expected, because the more one knows about a particular field, the more one is able to use its resources.

Our study supported the findings of earlier studies³, ⁹ that confusion exists about referrals to physiatrists as distinguished from referrals to physical therapists. This confusion is evident in 1 of our vignettes, which asked whether the respondents would send a patient to a physiatrist for the purpose of carrying out preexisting PT and OT orders. We consider this to be an inappropriate referral. However, only 49% of respondents were unlikely to refer a hypothetical patient to carry out preexisting PT and OT orders, although the bulk of respondents (84%) rated themselves as understanding the difference between physiatry and PT.

The confusion between PT and PM&R highlighted by our study is consistent with findings by Waylonis and Raptou in 1963.³ Thirty years have elapsed since their study, yet there remains a gap in medical education concerning the distinction between PT and PM&R.

The association between sex and referral is somewhat puzzling. Female physicians may be inherently more likely to refer (no matter what the field), more likely to elicit complaints from their patients that a physiatrist could assist with, or more likely to have a patient base that is in need of a physiatrist’s services. This final reason presupposes that the patient base is different for male and female physicians, because of a selection bias introduced by patients who use sex as a factor in choosing a physician. Only 29% of the respondents were women. We are cautious about extrapolating the choices made by these 133 women to female primary care physicians at large.

There was no correlation between correctly identifying the role of the physiatrist (in either skill list or vignette form) and physicians’ training in PM&R in medical school or residency. This finding suggests that the training they received during their medical education was not effective in teaching them about the field of PM&R. This finding is in accordance with a previous study,² which found no correlation between the presence of a PM&R department and rate of referral by medical students responding to clinical vignettes, although the frequency of referral correlated with students’ self-rated knowledge about PM&R. Both of these findings are similar to ours.

Study limitations 

The data were self-reports by physicians; no attempts were made to assess the actual behaviors of the respondents. Thus, respondents may have reported that they referred patients to physiatrists when they did not. There is a possibility of nonrespondent bias, given the response rate of 47%. However, this rate is similar to the average response rate for physicians in mailed surveys.¹¹ The average number of 21 yearly referrals seems likely enough as a referral basis but there was a large standard deviation (SD) of 60.3. This large SD may be due to a self-reporting error. It could be due to differing numbers of patients seen overall by the respondents. Respondents were not asked how many patients they see annually. We cannot be certain of the significance of the number of annual referrals.

Back to Article Outline

Conclusions 

The challenge of physiatrists obtaining referrals from internal medicine and family practice physicians is compounded by the lack of knowledge about the services and skills that physiatrists can offer as consultants to the referring physician and to the patient. This void is highlighted by 1 respondent who wrote, “I think you should send a bit of information on what a physiatrist does—what they are best at managing. It would be helpful.”

The burden of educating the primary care physician falls on the PM&R specialty. This outreach should be conducted at every level, from medical school through residency and into practice. How best to effect this education is unclear. Perhaps physiatrists should develop and distribute a curriculum for medical students and residents that would be very clear about the physiatric skill set, the differences between physiatrists and PT and OT, and what disorders are appropriately managed by physiatrists.

What is clear is that the benefits to the field of physiatry of better-educated colleagues would be more referrals and the opportunity to improve more patients’ lives. As 1 respondent stated, “If PM&R docs actually does [sic] all the things listed in the survey, I would gladly refer to them. I plead ignorance on the subject.” Decreasing that ignorance falls to the physiatric community.

Back to Article Outline

Acknowledgment 

We thank the Medical Library Staff at Christiana Hospital and Wilmington Hospital for assistance with literature searching and obtaining articles and Cynthia Clendenin, medical editor at Christiana Care Health System, for review of our manuscript.

Back to Article Outline

Appendix 1. Survey instrument 

Awareness of Physical Medicine and Rehabilitation (PM&R)

We are conducting a survey about the awareness among medical doctors about Physical Medicine and Rehabilitation (PM&R). Your participation in this survey is voluntary; however, should you choose to participate, we ask that you complete all of the questions as fully and as completely as possible.

Please be assured that your responses will remain absolutely confidential, and you will not be identified on this survey instrument. Although not sufficient to reimburse you for your time, please accept the $5.00 bill as a token of our appreciation. This is an academic study and no money came from pharmaceutical companies.

Cristin McKenna, MD, PhD Neil J. Farber, MD

We present to you below a series of scenarios of patients presenting to your office with specific problems. For each of the following hypothetical situations, how likely would you be to refer the patient to a Physical Medicine and Rehabilitation (PM&R) doctor (please select only one answer for each hypothetical patient)?

	Very Likely	Somewhat Likely	Somewhat Unlikely	Very Unlikely
1. To evaluate disability in a workman’s compensation claim
2. To treat a patient who presents with cauda equina syndrome (back pain, leg weakness, incontinence)
3. To address pain management in a fibromyalgia patient
4. To manage spasticity in a patient with a stroke
5. To manage psychological and cognitive symptoms in an adult with a learning disability
6. To oversee drug or alcohol rehabilitation
7. To confirm a diagnosis of Guillain-Barré via EMG and nerve conduction studies
8. To carry out PT/OT orders which you have already written
9. To manage a patient with traumatic brain injury for behavioral problems
10. To improve mobility in a patient with knee osteoarthritis
11. To treat low back pain
12. To maximize mobility in a patient with cerebral palsy
13. To maximize function in a patient with spinal cord injury
14. Which of the following is a PM&R physician qualified to do (check all that apply):
A. nerve conduction and electromyography studies (EMG)
B. surgery
C. motor point blocks
D. trigger point injections
E. epidural injections
F. ventilatory management
G. urodynamics
H. limb prosthesis evaluation
I. implantation of deep brain stimulation devices
Please indicate your agreement or disagreement with each of the following (check one answer for each question):
	Strongly Agree	Somewhat Agree	Somewhat Disagree	Strongly Disagree
15. Physiatry is a synonym for PM&R
16. PM&R is a board-certified specialty
17. I understand the difference between a physiatrist and a physical therapist
18. Everything PM&R addresses is also treated by neurology, orthopedics, internal medicine, physical/occupational/speech therapists, and/or anesthesiology
19. Have you ever referred a patient to a PM&R physician?
YesNo
If yes, how many patients per year on average do you refer to a PM&R physician?________patients
We also need some information about you:
20. Age _________
21. Sex (check one): Male Female
22. Which state do you live in _________?
23. Did you ever do a rotation in PM&R in medical school?
YesNoDon’t remember
24. Did your medical school or residency training program have a PM&R residency?
YesNoDon’t remember
25. What percent of your practice is devoted to primary care? _________ %
26. What percent of your professional time is spent seeing patients? ________%
27. Year graduated from medical school ________
28. How would you characterize the locale in which you practice (check one answer)?
Urban Suburban Rural
29. How would you characterize your type of practice (check all that apply)?
Private practice
HMO practice
VA
Hospitalist
Other _________
30. What is your medical specialty/subspecialty?___________________________
31. Would you have answered this survey without the $5?
YesNoUnsure
Thank you for your assistance.
Abbreviations: EMG, electromyography; HMO, health maintenance organization; OT, occupational therapy; VA, Veterans Administration.

Back to Article Outline

Appendix 2. Respondents’ comments 

Comments were written onto our survey. We excluded only those that were illegible, pertained to question 31, contained material that could identify the author, or were unrelated to the survey. The comments are transcribed exactly as they were written by the respondents on the surveys. No change of spelling, abbreviation, or alteration of vernacular was attempted by the authors.

Back to Article Outline

References 

1. Kirshblum SC , DeLisa JA , Campagnolo DI . Mandatory clerkship in physical medicine and rehabilitation (effect on medical students’ knowledge of physiatry) . Arch Phys Med Rehabil . 1998;79:10–13
   • View In Article
   • Abstract
   • Full-Text PDF (429 KB)
   • CrossRef
2. Abramson S , Stein J . Medical student referral patterns for musculoskeletal disorders . Am J Phys Med Rehabil . 1998;77:235–242
   • View In Article
   • MEDLINE
   • CrossRef
3. Waylonis GW , Raptou AD . A survey of house officer attitudes toward physical medicine and rehabilitation . Arch Phys Med Rehabil . 1963;44:152–156
   • View In Article
   • MEDLINE
4. Brogan DR . Rehabilitation service needs (physicians’ perceptions and referrals) . Arch Phys Med Rehabil . 1981;62:215–219
   • View In Article
   • MEDLINE
5. Finestone HM . Patient awareness of physical medicine and rehabilitation . Am J Phys Med Rehabil . 1992;71:288–290
   • View In Article
   • MEDLINE
   • CrossRef
6. Krevers B , Narvanen AL , Oberg B . Patient evaluation of the care and rehabilitation process in geriatric hospital care . Disabil Rehabil . 2002;24:482–491
   • View In Article
   • MEDLINE
   • CrossRef
7. Musick DW , Nickerson RB , McDowell SM , Gater DR . An exploratory examination of an academic PM&R inpatient consultation service . Disabil Rehabil . 2003;25:354–359
   • View In Article
   • MEDLINE
   • CrossRef
8. Davidoff G , Stolp-Smith KA , Waring WP , Cole TM . Patterns of referral to a university hospital consultation service (failure to accurately predict need for physiatric services) . Arch Phys Med Rehabil . 1988;69:449–450
   • View In Article
   • MEDLINE
9. Rice R . Clinical indicators for a rehabilitation referral . Home Healthcare Nurse . 1994;12:64
   • View In Article
   • MEDLINE
   • CrossRef
10. Pasko T , Smart DR . In: Physician characteristics and distribution in the United States . 2004 ed. Chicago: AMA; 2004;p. 7–8
    • View In Article
11. Asch DA , Jedrziewski MK , Christakis NA . Response rates to mail surveys published in medical journals . J Clin Epidemiol . 1997;50:1129–1136
    • View In Article
    • Abstract
    • Full-Text PDF (1240 KB)
    • CrossRef